From 38f25d85f654073e50d6594d709c254309af66f9 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Sun, 10 Aug 2025 00:03:30 +0300 Subject: [PATCH 01/32] Refines Doxygen style guide for stricter consistency Updates the Doxygen style guide to enforce stricter documentation rules, including mandatory public API documentation, specific tag order, and file headers. Introduces guidelines for templates, concepts, and internal code. Enhances clarity with allowed module groups and formatting rules. Establishes a comprehensive pre-merge checklist to ensure compliance. --- DOXYGENSTYLEGUIDE.md | 189 ++++++++++++++++++++++--------------------- 1 file changed, 96 insertions(+), 93 deletions(-) diff --git a/DOXYGENSTYLEGUIDE.md b/DOXYGENSTYLEGUIDE.md index 6509119f..9101f363 100644 --- a/DOXYGENSTYLEGUIDE.md +++ b/DOXYGENSTYLEGUIDE.md @@ -1,84 +1,67 @@ # Doxygen Style Guide for Modern C++ Projects (Webserv) -This guide defines how documentation must be written across all files in the Webserv project. -It aligns with the Doxyfile configuration, CI integration, and public documentation tooling. +This guide defines **mandatory documentation rules** for the Webserv project. It is strict by design to ensure consistent, maintainable, and navigable documentation across the entire codebase. --- -## 0. Documentation Structure +## 0. Scope & Enforcement -Documentation is automatically generated from: +* Applies to **all** `.hpp`, `.cpp`, `.tpp`, `.ipp` files in `include/`, `src/`, `tests/`. +* Public API **must** be fully documented. +* Pull requests will be **rejected** if: -* `include/`: public headers -* `src/`: implementation files -* `tests/`: documented test helpers -* `README.md`: used as the main page via `USE_MDFILE_AS_MAINPAGE` - -All files are parsed recursively, and `.md` files are included with full Markdown rendering. + * Any public class/function is undocumented. + * Tag order or format does not follow this guide. + * File headers are missing. --- -## 1. General Doxygen Practices +## 1. General Practices -* All `.hpp`, `.cpp`, and `.tpp` files must start with a `@file` header -* Use `/** */` for documentation blocks (not `///` or `//`) -* Use English and present tense ("Initializes", "Returns") -* Leave a blank line between `@brief` and `@details` -* Use structured tags (`@param`, `@return`, etc.) consistently -* Prefer complete documentation on all public headers, types, and methods +* Always use `/** */` for documentation blocks (never `///` or `//`). +* Write in **English** and **present tense** (e.g., "Initializes", not "Initialized"). +* Leave **one blank line** between `@brief` and `@details`. +* Keep **line length < 100 columns**. +* Use **noun phrases** for `@brief`. +* Document **private/internal methods** when behavior is non-obvious (mark with `@internal`). --- -## 2. Mandatory Tags for Public API - -Use the following tags in this order: +## 2. Tag Order (Mandatory) -1. `@brief`: short summary -2. `@details`: long explanation (optional) -3. `@tparam`: template parameters (if any) -4. `@param`: function arguments -5. `@return`: function return value -6. `@retval`: distinct return codes -7. `@throws`: thrown exceptions -8. `@note`: side information or caveats -9. `@warning`: must-know risk -10. `@todo`: pending feature or task +1. `@brief` – short summary (≤ 1 sentence) +2. `@details` – extended description (optional) +3. `@tparam` – template parameters +4. `@param` – parameters (aligned vertically) +5. `@return` – return value +6. `@retval` – distinct return codes (optional) +7. `@throws` – exceptions thrown +8. `@note` – additional info +9. `@warning` – important risk +10. `@todo` – pending task/feature --- -## 3. Example Template Documentation - -```cpp -/** - * @brief Allocates memory from an arena. - * - * @tparam T Type to allocate. - * @param size Number of elements. - * @return Pointer to the allocated memory. - */ -template -T* arenaAlloc(std::size_t size); -``` - ---- - -## 4. File Header Format +## 3. File Header Format ```cpp /** * @file Server.cpp * @brief Implements the Server class. * - * @details Provides TCP server functionality. + * @details Provides TCP server functionality for Webserv. + * @ingroup core */ ``` +* **Must** be the first comment in the file. +* Include `@ingroup` for module classification. + --- -## 5. Modules and Grouping +## 4. Modules & Grouping -Use `@defgroup` and `@ingroup` for logical documentation structure. -These are aliased in the Doxyfile to ensure visibility. +Use `@defgroup` and `@ingroup` to organize documentation. ```cpp /** @@ -95,17 +78,17 @@ class HttpRequest {}; /** @} */ ``` -Suggested groups: +**Allowed groups:** -* `http`: HTTP parsing and routing -* `socket`: socket abstraction -* `config`: configuration and parsing -* `core`: event loop and dispatcher -* `utils`: shared helpers +* `http` – HTTP parsing and routing +* `socket` – socket abstraction +* `config` – configuration and parsing +* `core` – event loop and dispatcher +* `utils` – shared helpers --- -## 6. Class and Member Documentation +## 5. Class & Member Documentation ```cpp /** @@ -116,17 +99,56 @@ Suggested groups: */ class Server { public: - /// @brief Starts the server loop. + /** @brief Starts the server loop. */ void start(); private: - int port_; ///< Port number the server uses. + int port_; ///< Port number used by the server. }; ``` +* Use `///<` for **inline member comments**. +* Use `/** */` for **methods**. + +--- + +## 6. Function Documentation + +```cpp +/** + * @brief Binds a socket to a port. + * + * @param port Port to bind. + * @return `true` if successful. + * @throws std::runtime_error If binding fails. + */ +bool bindSocket(int port); +``` + +* All params **must** be documented. +* If exceptions are thrown, **must** include `@throws`. + +--- + +## 7. Template & Concepts Documentation + +For C++20 concepts or constrained templates: + +```cpp +/** + * @brief Allocates memory from an arena. + * + * @tparam T Type to allocate. + * @param size Number of elements. + * @return Pointer to allocated memory. + */ +template requires std::is_default_constructible_v +T* arenaAlloc(std::size_t size); +``` + --- -## 7. Enum and Struct Documentation +## 8. Enum & Struct Documentation ```cpp /** @@ -142,9 +164,7 @@ enum class HttpMethod { --- -## 8. Internal or Private Code - -Use `@internal` (aliased in Doxyfile) to exclude internal code from public output. +## 9. Internal/Private Code ```cpp /** @@ -155,53 +175,36 @@ static std::map parseQueryInternal(const std::string& query); ``` ---- +* Mark with `@internal` to exclude from public output. -## 9. Formatting Rules +--- -* Use `/** */` style for all documentation blocks -* Align `@param`, `@return`, and `@throws` blocks -* Leave exactly one space after each `*` -* Keep line length under 100 columns -* Use present tense and noun phrases for `@brief` +## 10. Cross-Referencing -Example: +Use `@ref` to link related docs: ```cpp -/** - * @brief Binds a socket to a port. - * - * @param port Port to bind. - * @return `true` if successful. - */ -bool bindSocket(int port); +/// See also: @ref bindSocket ``` --- -## 10. Graphviz and Source Display - -The following are enabled by the Doxyfile: - -* `HAVE_DOT = YES`: enables all graph generation -* `CALL_GRAPH`, `CALLER_GRAPH`: call/caller relationships -* `SOURCE_BROWSER = YES`: annotated source files in HTML -* `VERBATIM_HEADERS = YES`: headers shown with structure - -Install Graphviz to ensure graph rendering. - ---- - -## 11. Final Checklist +## 11. Final Pre-Merge Checklist Before merging: * [ ] Each file has a `@file` header + * [ ] All public classes and methods are documented + * [ ] All parameters and return values are explained + * [ ] Internal helpers use `@internal` + * [ ] All tags are ordered and formatted + * [ ] Logical modules are defined and grouped + * [ ] No undocumented public entities remain --- From dcf5d4bd8d6c2700693f18c730f117181c20371f Mon Sep 17 00:00:00 2001 From: To0nsa Date: Sun, 10 Aug 2025 00:21:48 +0300 Subject: [PATCH 02/32] Adds documentation and refines configuration handling Removes commented-out code for cleanup. --- src/core/main.cpp | 37 +++++++++++++++---------- src/core/webserv.cpp | 66 ++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 86 insertions(+), 17 deletions(-) diff --git a/src/core/main.cpp b/src/core/main.cpp index 8e58a8c1..3e0b06b3 100644 --- a/src/core/main.cpp +++ b/src/core/main.cpp @@ -6,16 +6,37 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:11:30 by irychkov #+# #+# */ -/* Updated: 2025/06/06 10:15:30 by nlouis ### ########.fr */ +/* Updated: 2025/08/10 00:07:50 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file main.cpp + * @brief Entry point for the Webserv application. + * + * @details Initializes and runs the Webserv HTTP server. + * Delegates execution to `runWebserv`, handling all uncaught exceptions. + * @ingroup core + */ + #include "core/webserv.hpp" #include #include #include +/** + * @brief Program entry point. + * + * @details Starts the Webserv server using the provided configuration file path + * or a default configuration if none is given. + * + * @param argc Argument count. + * @param argv Argument vector. The first optional argument is the path to the configuration file. + * @return `EXIT_SUCCESS` on successful shutdown, `EXIT_FAILURE` on error. + * + * @throws std::runtime_error If an unrecoverable error occurs during startup. + */ int main(int argc, char** argv) try { return runWebserv(argc, argv); } catch (const std::exception& e) { @@ -25,17 +46,3 @@ int main(int argc, char** argv) try { std::cerr << "webserv encountered an unexpected error" << std::endl; return EXIT_FAILURE; } - -/* #include "core/webserv.hpp" -#include "config/parser/ConfigParseError.hpp" - -#include -#include -#include - -int main(int argc, char** argv) try { - return runWebserv(argc, argv); -} catch (const ConfigParseError& e) { - std::cerr << "Configuration parse error: " << e.what() << std::endl; - return EXIT_FAILURE; -} */ \ No newline at end of file diff --git a/src/core/webserv.cpp b/src/core/webserv.cpp index 8b7544f1..9795f3da 100644 --- a/src/core/webserv.cpp +++ b/src/core/webserv.cpp @@ -6,10 +6,21 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 21:57:56 by nlouis #+# #+# */ -/* Updated: 2025/05/24 23:25:23 by nlouis ### ########.fr */ +/* Updated: 2025/08/10 00:11:02 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file webserv.cpp + * @brief Webserv bootstrap and runtime orchestration. + * + * @details Resolves the configuration path, loads and parses the configuration file, + * applies normalization and validation, prints the effective config, and + * starts the socket event loop. Helper functions for argument handling and + * file I/O are kept internal to this translation unit. + * @ingroup core + */ + #include "config/Config.hpp" #include "config/normalizeConfig.hpp" #include "config/parser/ConfigParser.hpp" @@ -20,14 +31,34 @@ #include #include #include +#include +#include namespace { + +/** + * @brief Default configuration file path used when no CLI argument is provided. + * @internal + */ inline constexpr std::string_view DEFAULT_CONFIG_PATH{"./configs/default.conf"}; +/** + * @brief Resolves the configuration file path from CLI arguments. + * + * @details Accepts either zero or one user-supplied path. When no path is provided, + * the function returns the default path specified by `DEFAULT_CONFIG_PATH`. + * Passing more than one non-program argument is considered an error. + * + * @param argc Argument count. + * @param argv Argument vector. + * @return The resolved configuration file path. + * @throws std::runtime_error If more than one configuration path is supplied. + * @internal + */ std::string resolveConfigPath(int argc, char** argv) { std::string config_path; if (argc == 1) { - config_path = DEFAULT_CONFIG_PATH; + config_path = std::string(DEFAULT_CONFIG_PATH); } else if (argc == 2) { config_path = argv[1]; } else if (argc > 2) { @@ -36,6 +67,14 @@ std::string resolveConfigPath(int argc, char** argv) { return config_path; } +/** + * @brief Reads the entire configuration file into memory. + * + * @param config_path Path to the configuration file. + * @return The file contents as a single string. + * @throws std::runtime_error If the file cannot be opened. + * @internal + */ std::string extractFileContent(const std::string& config_path) { std::ifstream file(config_path); if (!file) { @@ -47,12 +86,35 @@ std::string extractFileContent(const std::string& config_path) { return buffer.str(); } +/** + * @brief Parses a configuration string into a Config object. + * + * @param fileContent Raw configuration text. + * @return Parsed configuration. + * @throws ConfigParseError On lexical/syntax errors in the configuration text. + * @internal + */ Config loadConfig(const std::string& fileContent) { ConfigParser parser(fileContent); return parser.parseConfig(); } + } // namespace +/** + * @brief Runs the Webserv server. + * + * @details Bootstraps the application by resolving the config path, loading and parsing + * the configuration, normalizing defaults, validating constraints, and launching + * the non-blocking socket manager event loop. + * + * @param argc Argument count. + * @param argv Argument vector. Optional: path to the configuration file. + * @return `EXIT_SUCCESS` on clean shutdown; `EXIT_FAILURE` is returned by `main()` on errors. + * @throws std::runtime_error If CLI arguments are invalid or the config file cannot be opened. + * @throws ConfigParseError If the configuration cannot be tokenized/parsed. + * @throws ValidationError If the resulting configuration fails validation rules. + */ int runWebserv(int argc, char** argv) { std::string configPath = resolveConfigPath(argc, argv); std::string fileContent = extractFileContent(configPath); From a4702d06be567eeebaae04afb6fb9ba4a5c3df14 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 11 Aug 2025 14:43:26 +0300 Subject: [PATCH 03/32] Fix Doxyfile issues and enhance documentation configuration Corrects typos and formatting inconsistencies in Doxyfile, including input encoding and inline member settings. Updates path patterns for exclusions and shortens displayed paths in HTML output. Adds a warning log file and adjusts markdown and style options for improved documentation clarity. Updates `.gitignore` to exclude generated Doxygen HTML files. --- .gitignore | 4 ++++ Doxyfile | 26 ++++++++++++++++---------- 2 files changed, 20 insertions(+), 10 deletions(-) diff --git a/.gitignore b/.gitignore index 8347c4c8..b8689ca1 100644 --- a/.gitignore +++ b/.gitignore @@ -51,3 +51,7 @@ Thumbs.db test_webserv/**/__pycache__/ test_webserv/**/*.pyc test_webserv/**/*.pyo + +# --- Doxygen docs --- + +/docs/html/ diff --git a/Doxyfile b/Doxyfile index 2edd36ec..2ff11cff 100644 --- a/Doxyfile +++ b/Doxyfile @@ -1,11 +1,11 @@ # --- Project Information --- PROJECT_NAME = Webserv -PROJECT_BRIEF = Lightweight HTTP/1.1 server written in modern C++ +PROJECT_BRIEF = "Lightweight HTTP/1.1 server written in modern C++" OUTPUT_DIRECTORY = docs CREATE_SUBDIRS = NO OUTPUT_LANGUAGE = English -_inputENCODING = UTF-8 +INPUT_ENCODING = UTF-8 # --- Input Files --- @@ -13,7 +13,7 @@ INPUT = include/ src/ tests/ README.md RECURSIVE = YES FILE_PATTERNS = *.cpp *.hpp *.tpp *.md USE_MDFILE_AS_MAINPAGE = README.md -EXCLUDE_PATTERNS = /.git/ /build/ +EXCLUDE_PATTERNS = **/.git/** **/build/** # --- Source Code Parsing --- @@ -30,7 +30,7 @@ EXTRACT_PRIVATE = YES EXTRACT_STATIC = YES EXTRACT_LOCAL_METHODS = YES EXTRACT_ANON_NSPACES = YES -IN_lineINHERITED_MEMB = YES +INLINE_INHERITED_MEMB = YES # --- Output Format --- @@ -38,20 +38,26 @@ GENERATE_HTML = YES HTML_OUTPUT = html HTML_FILE_EXTENSION = .html GENERATE_TREEVIEW = YES -HTML_DYNAMIC_SECTIONS = YES +HTML_DYNAMIC_SECTIONS = YES DISABLE_INDEX = NO FULL_PATH_NAMES = YES +# Shorten displayed paths in HTML and include graphs +STRIP_FROM_PATH = . +STRIP_FROM_INC_PATH = include + # --- Markdown and Style --- MARKDOWN_SUPPORT = YES AUTOLINK_SUPPORT = YES +JAVADOC_AUTOBRIEF = NO +MULTILINE_CPP_IS_BRIEF = YES # --- Documentation Layout --- SHOW_INCLUDE_FILES = YES SOURCE_BROWSER = YES -IN_lineSOURCES = NO +INLINE_SOURCES = NO VERBATIM_HEADERS = YES STRIP_CODE_COMMENTS = NO REFERENCED_BY_RELATION = YES @@ -84,15 +90,16 @@ SORT_BY_SCOPE_NAME = YES # --- Aliases --- -ALIASES += "ingroup=\ingroup " -ALIASES += "defgroup=\defgroup " -ALIASES += "internal=\internal " +ALIASES += "ingroup=\ingroup " +ALIASES += "defgroup=\defgroup " +ALIASES += "internal=\internal " # --- Warnings --- WARN_IF_UNDOCUMENTED = YES WARN_IF_DOC_ERROR = YES WARN_NO_PARAMDOC = YES +WARN_LOGFILE = docs/doxygen.warnings # --- Misc --- @@ -101,4 +108,3 @@ GENERATE_LATEX = NO GENERATE_MAN = NO GENERATE_RTF = NO GENERATE_XML = NO -MULTI_lineCPP_IS_BRIEF = YES \ No newline at end of file From 3b9372a3edde4b1b30e814d4c552745a849b5867 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 09:38:42 +0300 Subject: [PATCH 04/32] Add a header containing doxygen groups declaration. --- include/doxygen_groups.hpp | 49 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) create mode 100644 include/doxygen_groups.hpp diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp new file mode 100644 index 00000000..2c802bb8 --- /dev/null +++ b/include/doxygen_groups.hpp @@ -0,0 +1,49 @@ +/** + * @file doxygen_groups.hpp + * @brief Declares main Doxygen documentation groups for Webserv modules. + * + * @details This header contains only `@defgroup` declarations. + * It is not included in the build for logic, only for documentation purposes. + * All source files should use `@ingroup ` to link to these groups. + */ + +/** + * @defgroup config Configuration Parsing + * @brief Configuration structures, parsing, normalization, and validation. + * + * @details Handles reading, tokenizing, and parsing the Webserv configuration file(s), + * validating directives, and preparing normalized configuration objects + * for use by the core server. + */ + +/** + * @defgroup core Core Server Logic + * @brief Main server control flow and application entry point. + * + * @details Contains the event loop, server initialization, and runtime orchestration. + * Ties together configuration, networking, and HTTP processing. + */ + +/** + * @defgroup http HTTP Protocol + * @brief HTTP request/response parsing and handling. + * + * @details Includes classes and functions for parsing HTTP requests, + * building HTTP responses, routing, and applying HTTP rules. + */ + +/** + * @defgroup network Network & Socket Abstraction + * @brief Non-blocking I/O, socket management, and connection handling. + * + * @details Provides abstractions over system calls for listening sockets, + * client connections, and readiness-based multiplexing using poll/kqueue/epoll. + */ + +/** + * @defgroup utils Utilities & Helpers + * @brief Shared helper functions and utility classes. + * + * @details Common helpers for string manipulation, filesystem handling, + * logging, and error reporting used across all modules. + */ From a501acaf0781918b7eda76f89f1f2176b7b20e10 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 09:40:19 +0300 Subject: [PATCH 05/32] Change doxygen group from config to core for Server class. Refine getPort() comment, and add comment for getDefaultServerName(). --- include/core/Server.hpp | 88 ++++++++++++++++++++++++++--------------- src/core/Server.cpp | 10 ++--- 2 files changed, 61 insertions(+), 37 deletions(-) diff --git a/include/core/Server.hpp b/include/core/Server.hpp index 2e5d2db0..ae168a30 100644 --- a/include/core/Server.hpp +++ b/include/core/Server.hpp @@ -3,10 +3,10 @@ /* ::: :::::::: */ /* Server.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:37:06 by irychkov #+# #+# */ -/* Updated: 2025/05/16 08:03:09 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 09:35:27 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,7 +19,7 @@ * error pages, configure body size limits, and contain multiple location * blocks with their own routing rules and behavior. * - * @ingroup config + * @ingroup core */ #pragma once @@ -28,6 +28,8 @@ #include #include #include +#include +#include /** * @brief Represents a virtual server block. @@ -38,7 +40,7 @@ * `server` block in the configuration file and participates in request * routing based on host and port matching. * - * @ingroup config + * @ingroup core */ class Server { private: @@ -46,8 +48,8 @@ class Server { std::string _host; ///< IP address to bind (e.g., "0.0.0.0"). std::vector _server_names; ///< List of server name aliases (Host-based routing). std::map _error_pages; ///< Maps HTTP error codes to custom error page paths. - size_t _client_max_body_size; ///< Maximum allowed body size per request (bytes). - std::vector _locations; ///< Set of location blocks defined for this server. + std::size_t _client_max_body_size; ///< Maximum allowed body size per request (bytes). + std::vector _locations; ///< Set of location blocks defined for this server. public: /////////////////// @@ -59,7 +61,7 @@ class Server { * client max body size (`1 MiB`), and empty location/error blocks. * Intended to be populated via configuration parsing. * - * @ingroup config + * @ingroup core */ Server(); ~Server() = default; @@ -76,9 +78,10 @@ class Server { * the `listen` directive in the configuration file. Validation is done before calling. * * @param port The TCP port number to bind to. - * @ingroup config + * @ingroup core */ void setPort(int port) noexcept; + /** * @brief Sets the IP address to bind this server to. * @@ -87,9 +90,10 @@ class Server { * via the `host` directive in the configuration file. No validation is done here. * * @param host The IP address to bind (e.g., "127.0.0.1" or "0.0.0.0"). - * @ingroup config + * @ingroup core */ void setHost(std::string_view host) noexcept; + /** * @brief Adds a server name alias for this virtual host. * @@ -98,9 +102,10 @@ class Server { * This method appends without deduplication. * * @param name The server name to add (e.g., "example.com"). - * @ingroup config + * @ingroup core */ void addServerName(std::string_view name); + /** * @brief Sets a custom error page for a specific HTTP status code. * @@ -110,9 +115,10 @@ class Server { * * @param code The HTTP error status code to override. * @param path The file path to serve as the custom error page. - * @ingroup config + * @ingroup core */ void setErrorPage(int code, const std::string& path); + /** * @brief Sets the maximum allowed size for the HTTP request body. * @@ -121,9 +127,10 @@ class Server { * should reject it with a 413 Payload Too Large response. * * @param size Maximum body size in bytes. - * @ingroup config + * @ingroup core */ - void setClientMaxBodySize(size_t size) noexcept; + void setClientMaxBodySize(std::size_t size) noexcept; + /** * @brief Adds a location block to this server. * @@ -132,24 +139,23 @@ class Server { * for specific URI prefixes under this server. * * @param location The `Location` instance to add. - * @ingroup config + * @ingroup core */ void addLocation(const Location& location); /////////////// // --- Getters /** - * @brief Sets the port number for this server. - * - * @details This defines which TCP port the server should bind to for incoming - * connections. The value must be in the valid range [0, 65535], and - * is typically configured via the `listen` directive in the config file. - * Validation is performed before this function is called. - * - * @param port The TCP port to bind to. - * @ingroup config - */ + * @brief Returns the port number this server listens on. + * + * @details Reflects the `listen` directive from the configuration. Value is in + * the valid TCP range [0, 65535]. + * + * @return The configured TCP port. + * @ingroup core + */ int getPort() const noexcept; + /** * @brief Returns the configured host IP address for this server. * @@ -158,9 +164,10 @@ class Server { * Typically set via the `host` directive in the configuration file. * * @return Reference to the host IP address string. - * @ingroup config + * @ingroup core */ const std::string& getHost() const noexcept; + /** * @brief Returns the list of server name aliases for this virtual host. * @@ -169,10 +176,24 @@ class Server { * Configured via the `server_name` directive. * * @return Reference to the list of server names. - * @ingroup config + * @ingroup core */ const std::vector& getServerNames() const noexcept; + + /** + * @brief Returns the default server name for this virtual host. + * + * @details If no server names have been configured for this instance, the method + * returns `"localhost"`. Otherwise, it returns the first declared name + * from the configured list of server names. This value is used as the + * fallback when no explicit name match is found during host-based + * request routing. + * + * @return The default server name string. + * @ingroup core + */ const std::string getDefaultServerName() const; + /** * @brief Returns the mapping of HTTP error codes to custom error pages. * @@ -181,7 +202,7 @@ class Server { * Configured via the `error_page` directive. * * @return Reference to the map of error codes to file paths. - * @ingroup config + * @ingroup core */ const std::map& getErrorPages() const noexcept; /** @@ -192,9 +213,10 @@ class Server { * a 413 Payload Too Large error. Configured via the `client_max_body_size` directive. * * @return The maximum request body size in bytes. - * @ingroup config + * @ingroup core */ - size_t getClientMaxBodySize() const noexcept; + std::size_t getClientMaxBodySize() const noexcept; + /** * @brief Returns the list of location blocks defined for this server. * @@ -203,9 +225,10 @@ class Server { * server selects the best-matching location based on the URI. * * @return Reference to the list of `Location` objects. - * @ingroup config + * @ingroup core */ const std::vector& getLocations() const noexcept; + /** * @brief Returns a mutable reference to the server's location blocks. * @@ -214,9 +237,10 @@ class Server { * Use with care to avoid breaking routing logic. * * @return Reference to the list of `Location` objects. - * @ingroup config + * @ingroup core */ std::vector& getLocations() noexcept; + /** * @brief Checks whether the server matches the given server name. * @@ -226,7 +250,7 @@ class Server { * * @param name The server name to check (case-sensitive). * @return `true` if the name matches one of the configured server names. - * @ingroup config + * @ingroup core */ bool hasServerName(std::string_view name) const noexcept; }; diff --git a/src/core/Server.cpp b/src/core/Server.cpp index 8811ac67..45cda85c 100644 --- a/src/core/Server.cpp +++ b/src/core/Server.cpp @@ -3,10 +3,10 @@ /* ::: :::::::: */ /* Server.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:51:19 by irychkov #+# #+# */ -/* Updated: 2025/05/16 08:03:18 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 09:31:08 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,7 +19,7 @@ * error pages, body size limits, and associated location blocks. * It is part of the configuration system and supports parsing and runtime use. * - * @ingroup config + * @ingroup core */ #include "core/Server.hpp" @@ -31,9 +31,9 @@ // --- Constructor Server::Server() - : _port(80) // Default HTTP port + : _port(80) // Default HTTP port , - _host("0.0.0.0") // Default bind address + _host("0.0.0.0") // Default bind address , _client_max_body_size(1 * 1024 * 1024) // 1 MiB { From 837bee1cf8e8ad353d090f267fe2317ef351e076 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 09:42:07 +0300 Subject: [PATCH 06/32] add @copydoc to retrieve document from webserv.cpp --- include/core/webserv.hpp | 7 +++++-- src/core/webserv.cpp | 4 ++-- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/include/core/webserv.hpp b/include/core/webserv.hpp index 2f6e6b12..d1c31dab 100644 --- a/include/core/webserv.hpp +++ b/include/core/webserv.hpp @@ -3,13 +3,16 @@ /* ::: :::::::: */ /* webserv.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 22:47:01 by nlouis #+# #+# */ -/* Updated: 2025/05/20 22:47:24 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 09:23:40 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ #pragma once +/** + * @copydoc runWebserv(int,char**) + */ int runWebserv(int argc, char** argv); \ No newline at end of file diff --git a/src/core/webserv.cpp b/src/core/webserv.cpp index 9795f3da..6a571904 100644 --- a/src/core/webserv.cpp +++ b/src/core/webserv.cpp @@ -3,10 +3,10 @@ /* ::: :::::::: */ /* webserv.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 21:57:56 by nlouis #+# #+# */ -/* Updated: 2025/08/10 00:11:02 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 09:41:11 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ From 9b5d514f759a45ffe4704c881649571bda41e748 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 22:41:59 +0300 Subject: [PATCH 07/32] remove unused doxygen group for utilities and helpers --- include/doxygen_groups.hpp | 8 -------- 1 file changed, 8 deletions(-) diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index 2c802bb8..57338b69 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -39,11 +39,3 @@ * @details Provides abstractions over system calls for listening sockets, * client connections, and readiness-based multiplexing using poll/kqueue/epoll. */ - -/** - * @defgroup utils Utilities & Helpers - * @brief Shared helper functions and utility classes. - * - * @details Common helpers for string manipulation, filesystem handling, - * logging, and error reporting used across all modules. - */ From ccb067749bb38a127030efbacf6435c90495190a Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 22:42:17 +0300 Subject: [PATCH 08/32] Refactor documentation for server selection utilities in core module --- include/core/server_utils.hpp | 26 ++++++++------------------ src/core/server_utils.cpp | 15 +++++++++++++-- 2 files changed, 21 insertions(+), 20 deletions(-) diff --git a/include/core/server_utils.hpp b/include/core/server_utils.hpp index 9add464b..cf89ac7d 100644 --- a/include/core/server_utils.hpp +++ b/include/core/server_utils.hpp @@ -6,20 +6,17 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/02 20:29:58 by nlouis #+# #+# */ -/* Updated: 2025/05/02 22:03:05 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:40:47 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ /** * @file server_utils.hpp - * @brief Provides utility functions for selecting the correct Server instance. + * @brief Declares server selection util for virtual host resolution. * - * @details Contains logic for virtual host resolution based on the request's - * Host header and destination port. These functions are used during HTTP - * request routing to determine which configured server block should handle - * the incoming connection. - * - * @ingroup config + * @details Contains function declaration for selecting the appropriate + * `Server` instance based on listening port and HTTP `Host` header. + * @ingroup core */ #pragma once @@ -28,13 +25,6 @@ #include #include -/** - * @brief Selects the best matching server for a given port and Host header. - * - * @param servers List of all servers loaded from config. - * @param port The port on which the connection was accepted. - * @param host_name The Host header value from the request (e.g. "localhost"). - * @return Reference to the selected Server. - */ -const Server& findMatchingServer(const std::vector& servers, int port, - const std::string& host_name); +const Server& +findMatchingServer(const std::vector& servers, int port, + const std::string& host_name); // documented at src/core/server_utils.cpp diff --git a/src/core/server_utils.cpp b/src/core/server_utils.cpp index bebaa8ce..f5285646 100644 --- a/src/core/server_utils.cpp +++ b/src/core/server_utils.cpp @@ -6,10 +6,21 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/02 20:30:41 by nlouis #+# #+# */ -/* Updated: 2025/05/02 22:10:11 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:39:40 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file server_utils.cpp + * @brief Implements server selection utilities for virtual host resolution. + * + * @details Provides helper function for matching an incoming connection + * to the appropriate `Server` instance based on the listening port + * and the HTTP `Host` header. This util is part of the + * core request-routing mechanism in Webserv. + * @ingroup core + */ + #include "core/server_utils.hpp" #include @@ -43,7 +54,7 @@ const Server& findMatchingServer(const std::vector& servers, int port, } } - // Pass 2: No name match — fall back to the first server that listens on the same port + // Pass 2: No name match, fall back to the first server that listens on the same port for (const Server& server : servers) { if (server.getPort() == port) { return server; // Fallback default server From cecf836fbf1a334007c39cf59b8d725b154a7750 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 22:42:46 +0300 Subject: [PATCH 09/32] Update author information and improve documentation in webserv.hpp and Server.cpp --- include/core/webserv.hpp | 18 ++++++++++++------ src/core/Server.cpp | 10 ++++------ 2 files changed, 16 insertions(+), 12 deletions(-) diff --git a/include/core/webserv.hpp b/include/core/webserv.hpp index d1c31dab..96d0a80c 100644 --- a/include/core/webserv.hpp +++ b/include/core/webserv.hpp @@ -3,16 +3,22 @@ /* ::: :::::::: */ /* webserv.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 22:47:01 by nlouis #+# #+# */ -/* Updated: 2025/08/13 09:23:40 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:36:29 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ -#pragma once - /** - * @copydoc runWebserv(int,char**) + * @file webserv.hpp + * @brief Public entry point for starting the Webserv server. + * + * @details Declares the `runWebserv()` function, which bootstraps the + * configuration loading, validation, and event loop execution. + * @ingroup core */ -int runWebserv(int argc, char** argv); \ No newline at end of file + +#pragma once + +int runWebserv(int argc, char** argv); // documented at ./src/core/webserv.cpp \ No newline at end of file diff --git a/src/core/Server.cpp b/src/core/Server.cpp index 45cda85c..7bc26b36 100644 --- a/src/core/Server.cpp +++ b/src/core/Server.cpp @@ -3,10 +3,10 @@ /* ::: :::::::: */ /* Server.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:51:19 by irychkov #+# #+# */ -/* Updated: 2025/08/13 09:31:08 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:18:12 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -31,10 +31,8 @@ // --- Constructor Server::Server() - : _port(80) // Default HTTP port - , - _host("0.0.0.0") // Default bind address - , + : _port(80), // Default HTTP port + _host("0.0.0.0"), // Default bind address _client_max_body_size(1 * 1024 * 1024) // 1 MiB { } From 0b5a3f03892589c5b5ba2a33de9f3d3cc44aba16 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Wed, 13 Aug 2025 22:43:08 +0300 Subject: [PATCH 10/32] Update documentation in errorUtils and filesystemUtils for clarity and accuracy --- include/utils/errorUtils.hpp | 11 ++++++----- src/utils/errorUtils.cpp | 7 ++++--- src/utils/filesystemUtils.cpp | 4 ++-- 3 files changed, 12 insertions(+), 10 deletions(-) diff --git a/include/utils/errorUtils.hpp b/include/utils/errorUtils.hpp index 2d36a313..c063dd2e 100644 --- a/include/utils/errorUtils.hpp +++ b/include/utils/errorUtils.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/06 19:59:07 by nlouis #+# #+# */ -/* Updated: 2025/05/06 19:59:14 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:16:13 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -15,13 +15,14 @@ * @brief Error formatting utilities. * * @details - * Provides helper functions to standardize how error messages are formatted, - * especially with line and column information, to improve diagnostics. + * Provides helper functions to standardize how error messages are formatted in + * during the parsing of the configuration file especially with line and column + * information, to improve diagnostics. */ /** - * @defgroup ErrorUtils Error message formatting - * @brief Helpers for formatting human-readable error messages. + * @defgroup ParsingErrorUtils Error message formatting + * @brief Helpers for formatting human-readable error messages during parsing. * * @details * These utilities generate consistent and contextual error messages for use diff --git a/src/utils/errorUtils.cpp b/src/utils/errorUtils.cpp index 5797db2f..0702bd68 100644 --- a/src/utils/errorUtils.cpp +++ b/src/utils/errorUtils.cpp @@ -6,17 +6,18 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/05 21:08:50 by nlouis #+# #+# */ -/* Updated: 2025/05/06 20:14:01 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:16:59 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ /** * @file errorUtils.cpp - * @brief Implements error formatting utilities. + * @brief Implements error formatting utilities for parsing. * * @details * Provides the implementation of `formatError`, a helper function that - * generates consistent diagnostic messages with line and column info. + * generates consistent diagnostic messages with line and column info during + * parsing. * @ingroup ErrorUtils */ diff --git a/src/utils/filesystemUtils.cpp b/src/utils/filesystemUtils.cpp index 8b8daef1..52de3ec8 100644 --- a/src/utils/filesystemUtils.cpp +++ b/src/utils/filesystemUtils.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/13 09:39:07 by nlouis #+# #+# */ -/* Updated: 2025/06/10 22:52:30 by nlouis ### ########.fr */ +/* Updated: 2025/08/13 22:21:38 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -36,7 +36,7 @@ std::string resolvePhysicalPath(const HttpRequest& req, const Location& loc) { // 1) Try resolving against the real root std::string rootPath = buildFilePath(req, loc); - // If this isn't a POST *and* the file actually exists under root, use it + // If this isn't a POST and the file actually exists under root, use it if (req.getMethod() != "POST" && !rootPath.empty() && fs::exists(rootPath)) { return rootPath; } From fa7b1ed39fe601c3c12b18c21675a0657b66c47f Mon Sep 17 00:00:00 2001 From: To0nsa Date: Thu, 14 Aug 2025 15:28:43 +0300 Subject: [PATCH 11/32] Refines Doxygen configuration and server code structure Improves Doxygen settings by adjusting options for inline members, documentation visibility, and sorting behavior. Introduces a custom layout file for better navigation in generated documentation. Refactors server-related files for clarity and modularity: renames files for consistency, updates include guards, and enhances documentation grouping. Aligns namespace and grouping annotations with updated Doxygen structure. Fixes minor issues in the build configuration and adds a warnings log to track Doxygen output. Adjusts function signatures and comments to improve readability and maintainability. --- Doxyfile | 13 +- DoxygenLayout.xml | 265 +++++++++++++++++++ Makefile | 6 +- docs/doxygen.warnings | 3 + include/core/Server.hpp | 246 ++++------------- include/core/{webserv.hpp => runWebserv.hpp} | 8 +- include/core/server_utils.hpp | 4 +- include/doxygen_groups.hpp | 31 ++- src/core/Server.cpp | 200 ++++++++++++-- src/core/main.cpp | 8 +- src/core/{webserv.cpp => runWebserv.cpp} | 41 ++- src/core/server_utils.cpp | 4 +- 12 files changed, 574 insertions(+), 255 deletions(-) create mode 100644 DoxygenLayout.xml create mode 100644 docs/doxygen.warnings rename include/core/{webserv.hpp => runWebserv.hpp} (85%) rename src/core/{webserv.cpp => runWebserv.cpp} (76%) diff --git a/Doxyfile b/Doxyfile index 2ff11cff..d01fb474 100644 --- a/Doxyfile +++ b/Doxyfile @@ -30,7 +30,7 @@ EXTRACT_PRIVATE = YES EXTRACT_STATIC = YES EXTRACT_LOCAL_METHODS = YES EXTRACT_ANON_NSPACES = YES -INLINE_INHERITED_MEMB = YES +INLINE_INHERITED_MEMB = NO # --- Output Format --- @@ -41,6 +41,8 @@ GENERATE_TREEVIEW = YES HTML_DYNAMIC_SECTIONS = YES DISABLE_INDEX = NO FULL_PATH_NAMES = YES +TOC_INCLUDE_HEADINGS = 0 +SEPARATE_MEMBER_PAGES = YES # Shorten displayed paths in HTML and include graphs STRIP_FROM_PATH = . @@ -57,12 +59,14 @@ MULTILINE_CPP_IS_BRIEF = YES SHOW_INCLUDE_FILES = YES SOURCE_BROWSER = YES -INLINE_SOURCES = NO +INLINE_SOURCES = YES VERBATIM_HEADERS = YES STRIP_CODE_COMMENTS = NO REFERENCED_BY_RELATION = YES REFERENCES_RELATION = YES REFERENCES_LINK_SOURCE = YES +HIDE_UNDOC_MEMBERS = YES +HIDE_UNDOC_CLASSES = YES # --- Graphs and Diagrams --- @@ -83,8 +87,9 @@ DIRECTORY_GRAPH = YES # --- Sorting and Indexing --- -SORT_MEMBER_DOCS = YES -SORT_BRIEF_DOCS = YES +MEMBER_ORDER = GROUPED +SORT_MEMBER_DOCS = NO +SORT_BRIEF_DOCS = NO SORT_GROUP_NAMES = YES SORT_BY_SCOPE_NAME = YES diff --git a/DoxygenLayout.xml b/DoxygenLayout.xml new file mode 100644 index 00000000..26a114bd --- /dev/null +++ b/DoxygenLayout.xml @@ -0,0 +1,265 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/Makefile b/Makefile index 790a25b2..1a91d49e 100644 --- a/Makefile +++ b/Makefile @@ -6,7 +6,7 @@ # By: nlouis +#+ +:+ +#+ # # +#+#+#+#+#+ +#+ # # Created: 2025/08/09 20:53:27 by nlouis #+# #+# # -# Updated: 2025/08/09 20:59:47 by nlouis ### ########.fr # +# Updated: 2025/08/14 14:43:24 by nlouis ### ########.fr # # # # **************************************************************************** # @@ -50,7 +50,7 @@ SRCS_CORE := \ src/core/Server.cpp \ src/core/main.cpp \ src/core/server_utils.cpp \ - src/core/webserv.cpp + src/core/runWebserv.cpp SRCS_HTTP := \ src/http/HttpRequest.cpp \ @@ -102,7 +102,7 @@ HEADERS := \ include/core/Location.hpp \ include/core/Server.hpp \ include/core/server_utils.hpp \ - include/core/webserv.hpp \ + include/core/runWebserv.hpp \ include/http/HttpRequest.hpp \ include/http/HttpRequestParser.hpp \ include/http/HttpResponse.hpp \ diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings new file mode 100644 index 00000000..9ebb0cb0 --- /dev/null +++ b/docs/doxygen.warnings @@ -0,0 +1,3 @@ +warning: source 'tests' is not a readable file or directory... skipping. +/home/toonsa/myProjects/webserv/include/core/Location.hpp:82: warning: unbalanced grouping commands +/home/toonsa/myProjects/webserv/README.md:103: warning: unable to resolve reference to '/home/toonsa/myProjects/webserv/CONTRIBUTING.md' for \ref command diff --git a/include/core/Server.hpp b/include/core/Server.hpp index ae168a30..4d7f5529 100644 --- a/include/core/Server.hpp +++ b/include/core/Server.hpp @@ -3,10 +3,10 @@ /* ::: :::::::: */ /* Server.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:37:06 by irychkov #+# #+# */ -/* Updated: 2025/08/13 09:35:27 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 14:31:55 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,17 +19,18 @@ * error pages, configure body size limits, and contain multiple location * blocks with their own routing rules and behavior. * - * @ingroup core + * @ingroup server */ #pragma once #include "core/Location.hpp" -#include -#include -#include -#include -#include + +#include // std::size_t for body size type +#include // std::map for _error_pages (HTTP code → error page path) +#include // std::string for host, server names, error page paths +#include // std::string_view for lightweight parameter passing +#include // std::vector for _server_names and _locations /** * @brief Represents a virtual server block. @@ -40,217 +41,60 @@ * `server` block in the configuration file and participates in request * routing based on host and port matching. * - * @ingroup core + * @ingroup server */ class Server { private: - int _port; ///< Port to listen on (0–65535). - std::string _host; ///< IP address to bind (e.g., "0.0.0.0"). - std::vector _server_names; ///< List of server name aliases (Host-based routing). - std::map _error_pages; ///< Maps HTTP error codes to custom error page paths. - std::size_t _client_max_body_size; ///< Maximum allowed body size per request (bytes). - std::vector _locations; ///< Set of location blocks defined for this server. + //=== Data ================================================================ + + int _port; ///< Port to listen on (0–65535). + std::string _host; ///< Bind address (e.g., "0.0.0.0"). + std::vector _server_names; ///< Host-based routing aliases. + std::map _error_pages; ///< HTTP code → error page path. + std::size_t _client_max_body_size; ///< Max request body size (bytes). + std::vector _locations; ///< Location blocks for this server. public: - /////////////////// - // --- Constructor - /** - * @brief Constructs a Server instance with default settings. - * - * @details Initializes default values for host (`"0.0.0.0"`), port (`80`), - * client max body size (`1 MiB`), and empty location/error blocks. - * Intended to be populated via configuration parsing. - * - * @ingroup core - */ + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ Server(); ~Server() = default; Server(const Server& other) = default; Server& operator=(const Server& other) = default; + ///@} - /////////////// - // --- Setters - /** - * @brief Sets the port number this server will listen on. - * - * @details This port must be in the range [0, 65535]. It determines which TCP port - * the server binds to for accepting incoming connections. Typically set via - * the `listen` directive in the configuration file. Validation is done before calling. - * - * @param port The TCP port number to bind to. - * @ingroup core - */ + //=== Configuration Setters ============================================== + + /** @name Configuration setters */ + ///@{ void setPort(int port) noexcept; - - /** - * @brief Sets the IP address to bind this server to. - * - * @details The host determines which network interface(s) the server will listen on. - * A value of `"0.0.0.0"` binds to all interfaces. This is typically configured - * via the `host` directive in the configuration file. No validation is done here. - * - * @param host The IP address to bind (e.g., "127.0.0.1" or "0.0.0.0"). - * @ingroup core - */ void setHost(std::string_view host) noexcept; - - /** - * @brief Adds a server name alias for this virtual host. - * - * @details Server names are used to match the `Host` header of incoming HTTP requests. - * Multiple names can be added to support name-based virtual hosting. - * This method appends without deduplication. - * - * @param name The server name to add (e.g., "example.com"). - * @ingroup core - */ void addServerName(std::string_view name); - - /** - * @brief Sets a custom error page for a specific HTTP status code. - * - * @details Associates an HTTP error code (e.g., 404, 500) with a file path - * that will be served when that error occurs. Overrides the default - * error response. Multiple codes can share the same file path. - * - * @param code The HTTP error status code to override. - * @param path The file path to serve as the custom error page. - * @ingroup core - */ void setErrorPage(int code, const std::string& path); - - /** - * @brief Sets the maximum allowed size for the HTTP request body. - * - * @details Used to limit the size of incoming requests, particularly for - * POST and PUT methods. If a request exceeds this size, the server - * should reject it with a 413 Payload Too Large response. - * - * @param size Maximum body size in bytes. - * @ingroup core - */ void setClientMaxBodySize(std::size_t size) noexcept; - - /** - * @brief Adds a location block to this server. - * - * @details Appends a new `Location` object representing a URI-matching block - * with its own configuration. Locations define routing rules and behavior - * for specific URI prefixes under this server. - * - * @param location The `Location` instance to add. - * @ingroup core - */ void addLocation(const Location& location); + ///@} - /////////////// - // --- Getters - /** - * @brief Returns the port number this server listens on. - * - * @details Reflects the `listen` directive from the configuration. Value is in - * the valid TCP range [0, 65535]. - * - * @return The configured TCP port. - * @ingroup core - */ - int getPort() const noexcept; - - /** - * @brief Returns the configured host IP address for this server. - * - * @details This address determines which local interface(s) the server binds to. - * A value of `"0.0.0.0"` means it will accept connections on all interfaces. - * Typically set via the `host` directive in the configuration file. - * - * @return Reference to the host IP address string. - * @ingroup core - */ - const std::string& getHost() const noexcept; - - /** - * @brief Returns the list of server name aliases for this virtual host. - * - * @details These names are used to match the `Host` header in incoming HTTP requests. - * If none match, the first declared server for the host:port is used as default. - * Configured via the `server_name` directive. - * - * @return Reference to the list of server names. - * @ingroup core - */ - const std::vector& getServerNames() const noexcept; + //=== Queries (Getters) =================================================== - /** - * @brief Returns the default server name for this virtual host. - * - * @details If no server names have been configured for this instance, the method - * returns `"localhost"`. Otherwise, it returns the first declared name - * from the configured list of server names. This value is used as the - * fallback when no explicit name match is found during host-based - * request routing. - * - * @return The default server name string. - * @ingroup core - */ - const std::string getDefaultServerName() const; - - /** - * @brief Returns the mapping of HTTP error codes to custom error pages. - * - * @details This map associates specific HTTP status codes (e.g., 404, 500) - * with file paths to serve instead of default error messages. - * Configured via the `error_page` directive. - * - * @return Reference to the map of error codes to file paths. - * @ingroup core - */ + /** @name Queries (getters) */ + ///@{ + int getPort() const noexcept; + const std::string& getHost() const noexcept; + const std::vector& getServerNames() const noexcept; + const std::string getDefaultServerName() const; const std::map& getErrorPages() const noexcept; - /** - * @brief Returns the maximum allowed size for the request body. - * - * @details This limit applies to the content length of incoming HTTP requests, - * including POST uploads. If exceeded, the server should return - * a 413 Payload Too Large error. Configured via the `client_max_body_size` directive. - * - * @return The maximum request body size in bytes. - * @ingroup core - */ - std::size_t getClientMaxBodySize() const noexcept; - - /** - * @brief Returns the list of location blocks defined for this server. - * - * @details Each location block defines a URI prefix and associated behavior - * (e.g., root, methods, CGI, redirects). During request handling, the - * server selects the best-matching location based on the URI. - * - * @return Reference to the list of `Location` objects. - * @ingroup core - */ - const std::vector& getLocations() const noexcept; - - /** - * @brief Returns a mutable reference to the server's location blocks. - * - * @details Allows in-place modification of the list of `Location` objects, - * typically used during configuration parsing to populate new routes. - * Use with care to avoid breaking routing logic. - * - * @return Reference to the list of `Location` objects. - * @ingroup core - */ - std::vector& getLocations() noexcept; - - /** - * @brief Checks whether the server matches the given server name. - * - * @details Compares the provided name against the configured server names - * for this virtual host. Used during request routing based on the - * `Host` header in the HTTP request. - * - * @param name The server name to check (case-sensitive). - * @return `true` if the name matches one of the configured server names. - * @ingroup core - */ + std::size_t getClientMaxBodySize() const noexcept; + const std::vector& getLocations() const noexcept; + std::vector& getLocations() noexcept; + ///@} + + //=== Matching & Predicates ============================================== + + /** @name Matching & predicates */ + ///@{ bool hasServerName(std::string_view name) const noexcept; + ///@} }; diff --git a/include/core/webserv.hpp b/include/core/runWebserv.hpp similarity index 85% rename from include/core/webserv.hpp rename to include/core/runWebserv.hpp index 96d0a80c..218e32e0 100644 --- a/include/core/webserv.hpp +++ b/include/core/runWebserv.hpp @@ -1,22 +1,22 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* webserv.hpp :+: :+: :+: */ +/* runWebserv.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 22:47:01 by nlouis #+# #+# */ -/* Updated: 2025/08/13 22:36:29 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 15:21:52 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ /** - * @file webserv.hpp + * @file runWebserv.hpp * @brief Public entry point for starting the Webserv server. * * @details Declares the `runWebserv()` function, which bootstraps the * configuration loading, validation, and event loop execution. - * @ingroup core + * @ingroup entrypoint */ #pragma once diff --git a/include/core/server_utils.hpp b/include/core/server_utils.hpp index cf89ac7d..d5b2c8c5 100644 --- a/include/core/server_utils.hpp +++ b/include/core/server_utils.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/02 20:29:58 by nlouis #+# #+# */ -/* Updated: 2025/08/13 22:40:47 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 15:14:24 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -16,7 +16,7 @@ * * @details Contains function declaration for selecting the appropriate * `Server` instance based on listening port and HTTP `Host` header. - * @ingroup core + * @ingroup server */ #pragma once diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index 57338b69..cc0aa5ce 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -17,13 +17,40 @@ */ /** - * @defgroup core Core Server Logic - * @brief Main server control flow and application entry point. + * @defgroup core Core Components + * @brief Main server components, control flow, and application entry point. * * @details Contains the event loop, server initialization, and runtime orchestration. * Ties together configuration, networking, and HTTP processing. */ +/** + * @defgroup server Server Component + * @ingroup core + * @brief Main Server class and its implementation. + * + * @details Contains the `Server` class, which represents a single + * virtual server block in Webserv. + * + * Files: + * - `core/Server.hpp` — class declaration. + * - `core/Server.cpp` — method implementations. + * - `core/server_utils.hpp` — findMatchingServer() + * - `core/server_utils.cpp` — findMatchingServer() + */ + +/** + * @defgroup entrypoint Application Entrypoint + * @ingroup core + * @brief Program entry and top-level orchestration. + * + * @details Contains the `main()` function and the high-level startup/shutdown + * logic for Webserv, including: + * - `runWebserv.hpp` — high-level declarations. + * - `runWebserv.cpp` — main orchestration functions. + * - `main.cpp` — program entry point. + */ + /** * @defgroup http HTTP Protocol * @brief HTTP request/response parsing and handling. diff --git a/src/core/Server.cpp b/src/core/Server.cpp index 7bc26b36..66bad794 100644 --- a/src/core/Server.cpp +++ b/src/core/Server.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:51:19 by irychkov #+# #+# */ -/* Updated: 2025/08/13 22:18:12 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 14:34:04 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,17 +19,29 @@ * error pages, body size limits, and associated location blocks. * It is part of the configuration system and supports parsing and runtime use. * - * @ingroup core + * @ingroup server */ -#include "core/Server.hpp" -#include "utils/stringUtils.hpp" -#include -#include +#include "core/Server.hpp" // Server class declaration +#include "utils/stringUtils.hpp" // toLower() for case‐insensitive server names -/////////////////// -// --- Constructor +#include // std::any_of +#include // std::size_t (body size) +#include // std::map (used in setErrorPage, getErrorPages) +#include // std::string (host, names, paths) +#include // std::string_view parameters +#include // std::vector (server names, locations) +//=== Construction & Special Members ===================================== +/** + * @brief Constructs a Server instance with default settings. + * + * @details Initializes default values for host (`"0.0.0.0"`), port (`80`), + * client max body size (`1 MiB`), and empty location/error blocks. + * Intended to be populated via configuration parsing. + * + * @ingroup server + */ Server::Server() : _port(80), // Default HTTP port _host("0.0.0.0"), // Default bind address @@ -37,74 +49,222 @@ Server::Server() { } -/////////////// -// --- Setters - +//=== Configuration Setters ============================================== +/** + * @brief Sets the port number this server will listen on. + * + * @details This port must be in the range [0, 65535]. It determines which TCP port + * the server binds to for accepting incoming connections. Typically set via + * the `listen` directive in the configuration file. Validation is done before calling. + * + * @param port The TCP port number to bind to. + * @ingroup server + */ void Server::setPort(int port) noexcept { _port = port; } +/** + * @brief Sets the IP address to bind this server to. + * + * @details The host determines which network interface(s) the server will listen on. + * A value of `"0.0.0.0"` binds to all interfaces. This is typically configured + * via the `host` directive in the configuration file. No validation is done here. + * + * @param host The IP address to bind (e.g., "127.0.0.1" or "0.0.0.0"). + * @ingroup server + */ void Server::setHost(std::string_view host) noexcept { _host = host; } +/** + * @brief Adds a server name alias for this virtual host. + * + * @details Server names are used to match the `Host` header of incoming HTTP requests. + * Multiple names can be added to support name-based virtual hosting. + * This method appends without deduplication. + * + * @param name The server name to add (e.g., "example.com"). + * @ingroup server + */ void Server::addServerName(std::string_view name) { // Append the given server name to the list of aliases. _server_names.emplace_back(toLower(std::string(name))); } +/** + * @brief Sets a custom error page for a specific HTTP status code. + * + * @details Associates an HTTP error code (e.g., 404, 500) with a file path + * that will be served when that error occurs. Overrides the default + * error response. Multiple codes can share the same file path. + * + * @param code The HTTP error status code to override. + * @param path The file path to serve as the custom error page. + * @ingroup server + */ void Server::setErrorPage(int code, const std::string& path) { // Map the given HTTP status code to a custom error page path. _error_pages[code] = path; } +/** + * @brief Sets the maximum allowed size for the HTTP request body. + * + * @details Used to limit the size of incoming requests, particularly for + * POST and PUT methods. If a request exceeds this size, the server + * should reject it with a 413 Payload Too Large response. + * + * @param size Maximum body size in bytes. + * @ingroup server + */ void Server::setClientMaxBodySize(std::size_t size) noexcept { _client_max_body_size = size; } +/** + * @brief Adds a location block to this server. + * + * @details Appends a new `Location` object representing a URI-matching block + * with its own configuration. Locations define routing rules and behavior + * for specific URI prefixes under this server. + * + * @param location The `Location` instance to add. + * @ingroup server + */ void Server::addLocation(const Location& location) { // Add a new location block to the server's routing table. _locations.push_back(location); } -/////////////// -// --- Getters +//=== Queries (Getters) =================================================== +/** + * @brief Returns the port number this server listens on. + * + * @details Reflects the `listen` directive from the configuration. Value is in + * the valid TCP range [0, 65535]. + * + * @return The configured TCP port. + * @ingroup server + */ int Server::getPort() const noexcept { return _port; } +/** + * @brief Returns the configured host IP address for this server. + * + * @details This address determines which local interface(s) the server binds to. + * A value of `"0.0.0.0"` means it will accept connections on all interfaces. + * Typically set via the `host` directive in the configuration file. + * + * @return Reference to the host IP address string. + * @ingroup server + */ const std::string& Server::getHost() const noexcept { return _host; } -const std::string Server::getDefaultServerName() const { - return _server_names.empty() ? "localhost" : _server_names.front(); -} - +/** + * @brief Returns the list of server name aliases for this virtual host. + * + * @details These names are used to match the `Host` header in incoming HTTP requests. + * If none match, the first declared server for the host:port is used as default. + * Configured via the `server_name` directive. + * + * @return Reference to the list of server names. + * @ingroup server + */ const std::vector& Server::getServerNames() const noexcept { return _server_names; } +/** + * @brief Returns the default server name for this virtual host. + * + * @details If no server names have been configured for this instance, the method + * returns `"localhost"`. Otherwise, it returns the first declared name + * from the configured list of server names. This value is used as the + * fallback when no explicit name match is found during host-based + * request routing. + * + * @return The default server name string. + * @ingroup server + */ +const std::string Server::getDefaultServerName() const { + return _server_names.empty() ? "localhost" : _server_names.front(); +} + +/** + * @brief Returns the mapping of HTTP error codes to custom error pages. + * + * @details This map associates specific HTTP status codes (e.g., 404, 500) + * with file paths to serve instead of default error messages. + * Configured via the `error_page` directive. + * + * @return Reference to the map of error codes to file paths. + * @ingroup server + */ const std::map& Server::getErrorPages() const noexcept { return _error_pages; } +/** + * @brief Returns the maximum allowed size for the request body. + * + * @details This limit applies to the content length of incoming HTTP requests, + * including POST uploads. If exceeded, the server should return + * a 413 Payload Too Large error. Configured via the `client_max_body_size` directive. + * + * @return The maximum request body size in bytes. + * @ingroup server + */ std::size_t Server::getClientMaxBodySize() const noexcept { return _client_max_body_size; } +/** + * @brief Returns the list of location blocks defined for this server. + * + * @details Each location block defines a URI prefix and associated behavior + * (e.g., root, methods, CGI, redirects). During request handling, the + * server selects the best-matching location based on the URI. + * + * @return Reference to the list of `Location` objects. + * @ingroup server + */ const std::vector& Server::getLocations() const noexcept { return _locations; } +/** + * @brief Returns a mutable reference to the server's location blocks. + * + * @details Allows in-place modification of the list of `Location` objects, + * typically used during configuration parsing to populate new routes. + * Use with care to avoid breaking routing logic. + * + * @return Reference to the list of `Location` objects. + * @ingroup server + */ std::vector& Server::getLocations() noexcept { return _locations; } -///////////////// -// --- Utilities - +//=== Matching & Predicates ============================================== +/** + * @brief Checks whether the server matches the given server name. + * + * @details Compares the provided name against the configured server names + * for this virtual host. Used during request routing based on the + * `Host` header in the HTTP request. + * + * @param name The server name to check (case-sensitive). + * @return `true` if the name matches one of the configured server names. + * @ingroup server + */ bool Server::hasServerName(std::string_view name) const noexcept { // Check if the given name matches any of the configured server names. return std::any_of(_server_names.begin(), _server_names.end(), diff --git a/src/core/main.cpp b/src/core/main.cpp index 3e0b06b3..3fe4ff0e 100644 --- a/src/core/main.cpp +++ b/src/core/main.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:11:30 by irychkov #+# #+# */ -/* Updated: 2025/08/10 00:07:50 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 15:25:54 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -16,10 +16,10 @@ * * @details Initializes and runs the Webserv HTTP server. * Delegates execution to `runWebserv`, handling all uncaught exceptions. - * @ingroup core + * @ingroup entrypoint */ -#include "core/webserv.hpp" +#include "core/runWebserv.hpp" #include #include @@ -36,6 +36,8 @@ * @return `EXIT_SUCCESS` on successful shutdown, `EXIT_FAILURE` on error. * * @throws std::runtime_error If an unrecoverable error occurs during startup. + * + * @ingroup entrypoint */ int main(int argc, char** argv) try { return runWebserv(argc, argv); diff --git a/src/core/webserv.cpp b/src/core/runWebserv.cpp similarity index 76% rename from src/core/webserv.cpp rename to src/core/runWebserv.cpp index 6a571904..3d759f2b 100644 --- a/src/core/webserv.cpp +++ b/src/core/runWebserv.cpp @@ -1,24 +1,24 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* webserv.cpp :+: :+: :+: */ +/* runWebserv.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: nlouis +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 21:57:56 by nlouis #+# #+# */ -/* Updated: 2025/08/13 09:41:11 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 15:22:35 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ /** - * @file webserv.cpp + * @file runWebserv.cpp * @brief Webserv bootstrap and runtime orchestration. * * @details Resolves the configuration path, loads and parses the configuration file, * applies normalization and validation, prints the effective config, and * starts the socket event loop. Helper functions for argument handling and * file I/O are kept internal to this translation unit. - * @ingroup core + * @ingroup entrypoint */ #include "config/Config.hpp" @@ -28,13 +28,24 @@ #include "network/SocketManager.hpp" #include "utils/printInfo.hpp" -#include -#include -#include -#include -#include +#include // std::ifstream for reading the configuration file +#include // std::ostringstream for buffering file content +#include // std::runtime_error for error reporting +#include // std::string for configuration paths and file content +#include // std::string_view for DEFAULT_CONFIG_PATH constant -namespace { +/** + * @namespace bootstrap + * @brief Internal startup helpers for Webserv. + * + * @details Provides functions for resolving the configuration file path, + * reading configuration content, and parsing it into a `Config` + * object. These helpers are used exclusively during application + * initialization by `runWebserv()`. + * @ingroup entrypoint + * @internal + */ +namespace bootstrap { /** * @brief Default configuration file path used when no CLI argument is provided. @@ -114,11 +125,13 @@ Config loadConfig(const std::string& fileContent) { * @throws std::runtime_error If CLI arguments are invalid or the config file cannot be opened. * @throws ConfigParseError If the configuration cannot be tokenized/parsed. * @throws ValidationError If the resulting configuration fails validation rules. + * + * @ingroup entrypoint */ int runWebserv(int argc, char** argv) { - std::string configPath = resolveConfigPath(argc, argv); - std::string fileContent = extractFileContent(configPath); - Config config = loadConfig(fileContent); + std::string configPath = bootstrap::resolveConfigPath(argc, argv); + std::string fileContent = bootstrap::extractFileContent(configPath); + Config config = bootstrap::loadConfig(fileContent); normalizeConfig(config); validateConfig(config); printConfig(config); diff --git a/src/core/server_utils.cpp b/src/core/server_utils.cpp index f5285646..18551596 100644 --- a/src/core/server_utils.cpp +++ b/src/core/server_utils.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/02 20:30:41 by nlouis #+# #+# */ -/* Updated: 2025/08/13 22:39:40 by nlouis ### ########.fr */ +/* Updated: 2025/08/14 15:09:44 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -18,7 +18,7 @@ * to the appropriate `Server` instance based on the listening port * and the HTTP `Host` header. This util is part of the * core request-routing mechanism in Webserv. - * @ingroup core + * @ingroup server */ #include "core/server_utils.hpp" From 65b6e1eeb116c127d5a7fc9a8e3d156381e4df89 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Fri, 15 Aug 2025 23:02:26 +0300 Subject: [PATCH 12/32] Improves documentation and removes legacy ASAN ignore file Enhances documentation across various modules with detailed groupings, descriptions, and usage notes for utility functions. Adds missing file-level summaries and clarifies behavior, error handling, and intended usage for key methods. Fixes minor inconsistencies in formatting and eliminates redundant or outdated comments across utility implementations. --- .asanignore | 21 --- docs/doxygen.warnings | 23 +++ include/doxygen_groups.hpp | 68 ++++++++ include/utils/Logger.hpp | 10 +- include/utils/errorUtils.hpp | 39 +---- include/utils/stringUtils.hpp | 70 +------- src/utils/Logger.cpp | 37 +++++ src/utils/filesystemUtils.cpp | 298 ++++++++++++++++++++++++++++++++-- src/utils/htmlUtils.cpp | 33 +++- src/utils/printInfo.cpp | 19 ++- src/utils/stringUtils.cpp | 79 ++++++++- src/utils/urlUtils.cpp | 82 +++++++++- 12 files changed, 624 insertions(+), 155 deletions(-) delete mode 100644 .asanignore diff --git a/.asanignore b/.asanignore deleted file mode 100644 index d04dc703..00000000 --- a/.asanignore +++ /dev/null @@ -1,21 +0,0 @@ -# Ignore known harmless leaks (exit, TLS keys, system libs) -leak:libc -leak:libstdc++ -leak:libgcc -leak:__cxa -leak:__tls_get_addr -leak:malloc -leak:epoll -leak:poll -leak:kqueue -leak:fork -leak:exec - -# Ignore sanitizer internals -race:__sanitizer - -# Ignore undefined behavior triggered inside system libraries -# Useful if UBSAN reports issues inside glibc or third-party libraries -undefined:libc -undefined:libstdc++ -undefined:malloc diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index 9ebb0cb0..e10b6431 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -1,3 +1,26 @@ warning: source 'tests' is not a readable file or directory... skipping. /home/toonsa/myProjects/webserv/include/core/Location.hpp:82: warning: unbalanced grouping commands +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'value' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'field' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'line' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'column' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'context_provider' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'value' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'field' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'line' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'column' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'context_provider' from the argument list of parseByteSize has multiple @param documentation sections +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'value' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'field' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'line' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'column' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'context_provider' from the argument list of parseInt has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'value' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'field' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'line' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'column' from the argument list of parseByteSize has multiple @param documentation sections +/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'context_provider' from the argument list of parseByteSize has multiple @param documentation sections +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! /home/toonsa/myProjects/webserv/README.md:103: warning: unable to resolve reference to '/home/toonsa/myProjects/webserv/CONTRIBUTING.md' for \ref command diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index cc0aa5ce..45caffa2 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -66,3 +66,71 @@ * @details Provides abstractions over system calls for listening sockets, * client connections, and readiness-based multiplexing using poll/kqueue/epoll. */ + +/** + * @defgroup utils Utility Functions + * @brief Shared helper routines for common operations. + * + * @details Provides generic helper functions reused across modules, including + * string manipulation, date/time handling, and filesystem operations. + */ + +/** + * @defgroup filesystem_utils Filesystem Utilities + * @ingroup utils + * @brief Path manipulation, safe file handling, and directory management. + * + * @details Contains helpers for: + * - Normalizing and joining paths. + * - Mapping URIs to filesystem paths. + * - Sanitizing filenames for uploads. + * - Creating directories recursively. + * - Enforcing upload root boundaries. + * These utilities are designed to prevent directory traversal, + * enforce security constraints, and support Webserv’s upload and + * static file-serving features. + */ + +/** + * @defgroup html_utils HTML Utilities + * @ingroup utils + * @brief Safe HTML encoding and related helpers. + * + * @details Functions for escaping or manipulating HTML content so that + * untrusted input can be embedded safely in a page without being + * interpreted as markup. Prevents common injection vulnerabilities + * like cross-site scripting (XSS) by replacing reserved characters + * with their corresponding HTML entities. + */ + +/** + * @defgroup string_utils String Utilities + * @ingroup utils + * @brief String manipulation, parsing, and formatting helpers. + * + * @details Provides reusable string-related routines including: + * - Case conversion (uppercase/lowercase). + * - Whitespace trimming. + * - Delimited join operations. + * - Size parsing with suffix multipliers (e.g., KiB, MiB). + * - Human-readable byte formatting. + * - Integer parsing with detailed error reporting. + * These utilities are used throughout Webserv for configuration + * parsing, logging, and data presentation. + */ + +/** + * @defgroup url_utils URL Utilities + * @ingroup utils + * @brief Helpers for percent-decoding, form decoding, and safe filename extraction. + * + * @details Provides functions for: + * - Decoding percent-encoded sequences in URIs. + * - Handling `application/x-www-form-urlencoded` form data. + * - Parsing key/value pairs from form bodies. + * - Extracting and validating safe filenames from URI segments. + * + * These functions are typically used during HTTP request parsing, + * particularly for processing query strings, form submissions, + * and safe handling of uploaded filenames. + */ diff --git a/include/utils/Logger.hpp b/include/utils/Logger.hpp index 5d74465f..75fa065e 100644 --- a/include/utils/Logger.hpp +++ b/include/utils/Logger.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/25 12:09:58 by ktieu #+# #+# */ -/* Updated: 2025/06/04 09:48:22 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:00:53 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -22,8 +22,8 @@ class Logger { static void logFrom(LogLevel level, const std::string& from, const std::string& message); private: - Logger() = delete; // Prevent instantiation - ~Logger() = delete; // Prevent instantiation - Logger(const Logger& org) = delete; // Prevent copy - Logger& operator=(const Logger& other) = delete; // Prevent assignment + Logger() = delete; + ~Logger() = delete; + Logger(const Logger& org) = delete; + Logger& operator=(const Logger& other) = delete; }; diff --git a/include/utils/errorUtils.hpp b/include/utils/errorUtils.hpp index c063dd2e..9a5fb0a8 100644 --- a/include/utils/errorUtils.hpp +++ b/include/utils/errorUtils.hpp @@ -6,49 +6,12 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/06 19:59:07 by nlouis #+# #+# */ -/* Updated: 2025/08/13 22:16:13 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 22:59:56 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ -/** - * @file errorUtils.hpp - * @brief Error formatting utilities. - * - * @details - * Provides helper functions to standardize how error messages are formatted in - * during the parsing of the configuration file especially with line and column - * information, to improve diagnostics. - */ - -/** - * @defgroup ParsingErrorUtils Error message formatting - * @brief Helpers for formatting human-readable error messages during parsing. - * - * @details - * These utilities generate consistent and contextual error messages for use - * in exceptions and log output. - * @{ - */ - #pragma once #include -/** - * @brief Formats an error message with line and column information. - * @ingroup ErrorUtils - * - * @details - * Constructs a human-readable error message of the form: - * `"Line X, column Y: "`. Intended for consistent formatting across - * parser or tokenizer errors. - * - * @param msg The error message. - * @param line The line number where the error occurred. - * @param column The column number of the problematic element. - * - * @return A formatted string with line/column context. - */ std::string formatError(const std::string& msg, int line, int column); - -/** @} */ // end of ErrorUtils diff --git a/include/utils/stringUtils.hpp b/include/utils/stringUtils.hpp index 885a4d98..b0d346ad 100644 --- a/include/utils/stringUtils.hpp +++ b/include/utils/stringUtils.hpp @@ -6,30 +6,10 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/05 20:09:44 by nlouis #+# #+# */ -/* Updated: 2025/05/22 22:32:12 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:00:37 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ -/** - * @file stringUtils.hpp - * @brief Utility functions for parsing numeric string values. - * - * @details - * Provides functions to parse integers and human-readable byte sizes from strings. - * Supports contextual error reporting for use in configuration parsing or diagnostics. - */ - -/** - * @defgroup StringUtils String parsing utilities - * @brief Helpers to parse integers and memory sizes from strings. - * - * @details - * This group provides low-level utilities used in configuration parsing for converting - * string representations of numbers and byte sizes into typed values with context-aware - * error handling. - * @{ - */ - #pragma once #include @@ -37,56 +17,12 @@ #include #include -/** - * @brief Parses an integer with detailed error context. - * @ingroup StringUtils - * - * @details - * Parses a non-negative integer and throws a detailed `std::invalid_argument` error - * on failure. Includes the field name, line/column, and optional context string. - * - * @param value The string to parse. - * @param field The name of the field being parsed. - * @param line Line number where the value is found. - * @param column Column number where the value starts. - * @param context_provider A lambda that returns extra context (e.g., the source line). - * - * @return The parsed integer. - */ -int parseInt(const std::string& value, const std::string& field, int line, int column, - const std::function& context_provider); - -/** - * @brief Parses a byte size with detailed error context. - * @ingroup StringUtils - * - * @details - * Accepts size strings with suffixes (K/M/G). On failure, throws `std::invalid_argument` - * with details including field name, line/column, and optional context string. - * - * @param value The string to parse. - * @param field The name of the field being parsed. - * @param line Line number where the value is found. - * @param column Column number where the value starts. - * @param context_provider A lambda that returns extra context (e.g., the source line). - * - * @return The parsed size in bytes. - */ +int parseInt(const std::string& value, const std::string& field, int line, int column, + const std::function& context_provider); std::size_t parseByteSize(const std::string& value, const std::string& field, int line, int column, const std::function& context_provider); -/** - * @brief Converts a string to lowercase. - * - * @details Returns a copy of the input string with all ASCII alphabetic characters - * converted to lowercase using the current locale rules. - * - * @param str The input string to convert. - * @return A lowercase copy of the input string. - */ std::string toLower(const std::string&); std::string toUpper(const std::string& s); std::string formatBytes(std::size_t bytes); std::string joinStrings(const std::vector& list, const std::string& delim = ", "); std::string trim(const std::string& str); - -/** @} */ // end of StringUtils diff --git a/src/utils/Logger.cpp b/src/utils/Logger.cpp index 39c8d4cf..4a82eee2 100644 --- a/src/utils/Logger.cpp +++ b/src/utils/Logger.cpp @@ -1,4 +1,41 @@ +/* ************************************************************************** */ +/* */ +/* ::: :::::::: */ +/* Logger.cpp :+: :+: :+: */ +/* +:+ +:+ +:+ */ +/* By: nlouis +#+ +:+ +#+ */ +/* +#+#+#+#+#+ +#+ */ +/* Created: 2025/08/15 22:54:33 by nlouis #+# #+# */ +/* Updated: 2025/08/15 22:55:35 by nlouis ### ########.fr */ +/* */ +/* ************************************************************************** */ + +/** + * @file Logger.cpp + * @brief Simple color-coded logging utility. + * + * @details Implements the @ref Logger class methods and related helper + * functions for mapping log levels to human-readable strings + * and ANSI terminal colors. Supports standard log levels + * (`DEBUG`, `INFO`, `WARN`, `ERROR`) and can optionally + * prefix messages with a source identifier. + * + * Output is written to `stdout` for non-error messages and + * `stderr` for `ERROR` level messages, ensuring separation + * of normal and error output streams. ANSI escape codes are + * used for color, which may not render correctly in all + * terminals. + * + * @ingroup utils + * + * @note This logger is intended for human-readable output during + * development and runtime diagnostics, not structured logging. + * For production environments or log parsing, consider extending + * it to support formats like JSON or syslog. + */ + #include "utils/Logger.hpp" + #include #include #include diff --git a/src/utils/filesystemUtils.cpp b/src/utils/filesystemUtils.cpp index 52de3ec8..862b6fef 100644 --- a/src/utils/filesystemUtils.cpp +++ b/src/utils/filesystemUtils.cpp @@ -6,10 +6,28 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/13 09:39:07 by nlouis #+# #+# */ -/* Updated: 2025/08/13 22:21:38 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 22:40:36 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file filesystemUtils.cpp + * @brief Implements filesystem utility functions for Webserv. + * + * @details Provides helper routines for safely manipulating paths and + * filesystem resources in the context of an HTTP server. This includes: + * - Canonicalizing and joining URI and filesystem paths. + * - Mapping request URIs to location roots and upload stores. + * - Generating safe filenames for uploads. + * - Creating directories recursively. + * - Verifying file and symlink existence. + * The utilities here are designed to prevent directory traversal, + * enforce upload root boundaries, and integrate cleanly with + * Webserv's request–response pipeline. + * + * @ingroup filesystem_utils + */ + #include "utils/filesystemUtils.hpp" #include "http/responseBuilder.hpp" @@ -28,6 +46,34 @@ namespace fs = std::filesystem; +/** + * @brief Resolves a request URI to a concrete filesystem path. + * + * @details Computes the physical path for a request by: + * 1) Normalizing the request path and attempting resolution under the location's root. + * 2) If the resource is not a POST target and exists, returns that path. + * 3) Otherwise, when uploads are enabled and the URI is within the location prefix, + * rewrites the path into the configured upload store (joining with the remaining + * relative suffix). + * 4) Falls back to the root-based path even if it does not exist, allowing later + * handlers to emit the appropriate HTTP status (e.g., 404). + * + * @ingroup filesystem_utils + * + * @param req Incoming HTTP request (method and normalized URI are consulted). + * @param loc Matched location block providing root, path prefix, and upload settings. + * @return Fully-resolved absolute filesystem path. Returns an empty string when the + * normalized request path is empty (invalid or unsupported path). + * + * @throws std::runtime_error Only if helper utilities (e.g., normalization/join) are + * documented to throw on invalid inputs (none are thrown in the current logic). + * + * @note For POST requests, existing files under the root are intentionally ignored to + * avoid accidental overwrite semantics; uploads are routed to the upload store + * when configured. + * @warning Callers should not assume the returned path exists unless they checked it. + * Existence checking is deliberately limited to the non‑POST, root‑based branch. + */ std::string resolvePhysicalPath(const HttpRequest& req, const Location& loc) { // Normalize the request path std::string requestPath = normalizePath(req.getPath()); @@ -66,25 +112,70 @@ bool isFile(const std::string& path) { return fs::exists(path) && fs::is_regular_file(path); } +/** + * @brief Builds a unique temporary filename in the system temp directory. + * + * @details Generates a candidate path under `std::filesystem::temp_directory_path()` + * using a high‑resolution timestamp and a monotonically increasing counter. + * Intended for write‑to‑temp then atomic rename flows (e.g., buffering + * uploaded request bodies or CGI output before publishing). This function + * returns a name only; callers must create/open the file atomically. + * + * @ingroup filesystem_utils + * + * @param prefix Logical prefix embedded in the filename (e.g., "upload", "cgi"). + * @param counter Monotonic per‑process counter; incremented on each call. + * @return Absolute path string to a temporary file candidate (not created). + * + * @throws std::filesystem::filesystem_error If temp directory discovery fails. + * + * @note Callers should open with exclusive creation flags (e.g., O_CREAT|O_EXCL) + * and retry on collisions to guarantee uniqueness under high concurrency. + * @warning Not thread‑safe w.r.t. the shared @p counter. Use an atomic or pass a + * thread‑local counter if invoked from multiple threads. + * @todo Consider adding PID and a random component (or switch to unique_path) + * to further reduce collision probability. + */ std::string make_temp_name(const std::string& prefix, unsigned& counter) { - // 1) Where to put it (e.g. "/tmp" on Linux, or $TMPDIR) + // 1) Locate the system temp directory (respects platform defaults / env). fs::path tmpdir = fs::temp_directory_path(); - // 2) High-precision timestamp (nanoseconds since epoch) + // 2) High-precision timestamp (nanoseconds since epoch). auto now = std::chrono::high_resolution_clock::now(); auto ns = std::chrono::duration_cast(now.time_since_epoch()).count(); - // 3) Build "___.tmp" + // 3) Build "__.tmp" std::ostringstream ss; ss << prefix << "_" << ns << "_" << counter++ << ".tmp"; return (tmpdir / ss.str()).string(); } +/** + * @brief Canonicalizes a URI-style path and rejects traversal. + * + * @details Collapses repeated slashes, removes "." segments, processes ".." + * by popping the previous segment, and preserves a trailing slash + * (except for root). Always returns an absolute form starting with "/". + * If a leading ".." would escape the root, returns an empty string to + * signal an invalid/unsafe path (caller should treat as 400/403). + * + * @ingroup filesystem_utils + * + * @param path Raw request path or filesystem-like path (may be relative or absolute). + * @return Canonical absolute path (e.g., "/a/b" or "/a/b/"). Returns empty string + * if normalization would traverse above root. + * + * @note Trailing slash is preserved to help callers distinguish "directory intent" + * (e.g., for index resolution or directory listing). + * @warning This function does not perform percent-decoding. Callers should decode + * the URI before normalization and validate disallowed bytes. + */ std::string normalizePath(const std::string& path) { if (path.empty()) return "/"; + // Collapse repeated slashes: "///a//b" -> "/a/b" std::string collapsed; bool prevWasSlash = false; for (char c : path) { @@ -99,10 +190,11 @@ std::string normalizePath(const std::string& path) { } } + // Split and resolve "." and ".." while tracking if caller intended a trailing '/' std::vector segments; std::string segment; std::istringstream stream(collapsed); - bool hadTrailingSlash = collapsed.back() == '/'; + const bool hadTrailingSlash = collapsed.back() == '/'; while (std::getline(stream, segment, '/')) { if (segment.empty() || segment == ".") @@ -111,6 +203,7 @@ std::string normalizePath(const std::string& path) { if (!segments.empty()) { segments.pop_back(); } else { + // Would escape root: signal invalid path to caller return ""; } } else { @@ -118,19 +211,40 @@ std::string normalizePath(const std::string& path) { } } + // Reassemble, preserving trailing slash intent (except for "/") std::string result = "/"; for (std::size_t i = 0; i < segments.size(); ++i) { result += segments[i]; if (i + 1 < segments.size()) result += "/"; } - if (hadTrailingSlash && result != "/") result += "/"; return result; } +/** + * @brief Concatenates a base path and a suffix with a single '/' separator. + * + * @details Ensures that there is exactly one directory separator between the + * base and suffix, regardless of whether the base already ends with + * a slash. Does not normalize or validate the components; callers + * should sanitize inputs beforehand (e.g., via @ref normalizePath). + * Intended for filesystem path assembly when resolving HTTP request + * targets to physical files or upload locations. + * + * @ingroup filesystem_utils + * + * @param base Base path (absolute or relative). May end with '/'. + * @param suffix Path fragment to append (should not begin with '/' + * unless intentional). + * @return Combined path string. + * + * @note If @p base is empty, the function returns @p suffix unchanged. + * @warning This function does not handle '.' or '..' resolution; use only + * with trusted, normalized inputs to avoid directory traversal. + */ std::string joinPath(const std::string& base, const std::string& suffix) { if (base.empty()) return suffix; @@ -139,26 +253,74 @@ std::string joinPath(const std::string& base, const std::string& suffix) { return base + '/' + suffix; } +/** + * @brief Maps a request path to a filesystem path under a location root. + * + * @details Normalizes the incoming request path and the location's configured + * path prefix, verifies that the request path begins with the location + * prefix, and returns the location's root joined with the remaining + * suffix. If the request path does not match the location path, returns + * an empty string to signal no match. + * + * @ingroup filesystem_utils + * + * @param request Incoming HTTP request (URI is read from @ref HttpRequest::getPath()). + * @param loc Matched location block containing `path` (URI prefix) and `root`. + * @return Absolute or relative filesystem path under `loc.root`. Returns an empty + * string if the request path does not begin with the location's path. + * + * @note This function does not check whether the resulting file exists; callers + * may perform existence or type checks before use. + * @warning Callers must ensure that `request.getPath()` and `loc.getPath()` are + * decoded and safe. Normalization via @ref normalizePath helps prevent + * traversal, but upstream input validation is still required. + * @see normalizePath, joinPath + */ std::string buildFilePath(const HttpRequest& request, const Location& loc) { + // Normalize both the request path and the location path prefix. std::string req_path = normalizePath(request.getPath()); std::string loc_path = normalizePath(loc.getPath()); std::string loc_root = loc.getRoot(); + // Ensure the request path starts with the location prefix. if (req_path.rfind(loc_path, 0) != 0) { return ""; } + // Extract the suffix (portion after the location prefix). std::string suffix = req_path.substr(loc_path.size()); if (!suffix.empty() && suffix[0] == '/') suffix.erase(0, 1); + // Join the location's filesystem root with the relative suffix. return joinPath(loc_root, suffix); } +/** + * @brief Splits a path string into non-empty components. + * + * @details Parses the given path by splitting on '/' characters and collects + * all non-empty segments into a vector. Consecutive slashes and empty + * segments are skipped. The returned parts do not include leading or + * trailing separators and preserve the original segment order. + * + * @ingroup filesystem_utils + * + * @param path Path string to split (can be a URI or filesystem path). + * @return Vector of non-empty path segments in order of appearance. + * + * @note This function does not perform normalization; callers should sanitize + * and normalize the path (e.g., via @ref normalizePath) before splitting + * to remove `.` or `..` and collapse repeated slashes. + * @warning Leading or trailing slashes are not included in the result; the + * caller must track whether the path was absolute if needed. + */ static std::vector splitPath(const std::string& path) { std::vector parts; std::stringstream ss(path); std::string part; + + // Extract substrings between '/' and skip empty ones. while (std::getline(ss, part, '/')) { if (!part.empty()) parts.push_back(part); @@ -166,7 +328,7 @@ static std::vector splitPath(const std::string& path) { return parts; } -bool mkdirRecursive(const std::string& path) { +/* bool mkdirRecursive(const std::string& path) { std::vector parts = splitPath(path); std::string current = path[0] == '/' ? "/" : ""; @@ -181,6 +343,43 @@ bool mkdirRecursive(const std::string& path) { } } } + return true; +} */ + +/** + * @brief Ensures that a directory exists, creating missing parents as needed. + * + * @details Checks if the given path already exists. If it exists as a regular + * file, the function fails. If it exists as a directory (or symlink + * to a directory), the function succeeds. Otherwise, it attempts to + * create the directory and all missing parent directories using + * `std::filesystem::create_directories`. + * + * @ingroup filesystem_utils + * + * @param path Filesystem path to ensure as a directory (absolute or relative). + * @return `true` if the directory exists or was created successfully, + * `false` if the path is an existing regular file or creation failed. + * + * @note Uses `std::filesystem` and its error_code overloads to avoid exceptions. + * @warning Does not check for permissions beyond creation attempts; callers + * should handle permission errors via the return value. + */ +bool mkdirRecursive(const std::string& path) { + std::error_code ec; + + // If something exists at path and it's a *file*, fail. + if (fs::exists(path, ec)) { + if (fs::is_regular_file(path)) + return false; + return true; // already a directory (or symlink-to-dir) + } + + // Recursively create any missing directories. + if (!fs::create_directories(path, ec) && ec) { + return false; // creation failed for some reason + } + return true; } @@ -194,6 +393,25 @@ time_t getCurrentTime() { .count(); } +/** + * @brief Generates a timestamp-based fallback filename for uploads. + * + * @details Produces a string of the form `"upload_YYYYMMDDHHMMSS"` based on + * the current UTC time. Intended for use when the client request + * does not provide a valid or safe filename in an upload operation. + * This ensures that the uploaded file can still be stored with a + * unique, deterministic name that reflects its creation time. + * + * @ingroup filesystem_utils + * + * @return Fallback filename string in UTC timestamp format. + * + * @note Uses `std::gmtime`, which is not thread-safe on some platforms. + * If concurrent calls are expected, protect with a mutex or switch + * to `std::gmtime_r` (POSIX) or `std::gmtime_s` (Windows). + * @warning The generated name does not include a file extension; callers + * should append one if required. + */ static std::string makeFallbackName() { auto now = std::chrono::system_clock::now(); auto t = std::chrono::system_clock::to_time_t(now); @@ -202,19 +420,41 @@ static std::string makeFallbackName() { return oss.str(); } +/** + * @brief Produces a safe filename from untrusted input. + * + * @details Strips leading path components, removes any remaining path separators + * (`'/'` or `'\\'`) and control characters, while preserving all other + * bytes (including valid UTF-8 sequences). If the resulting name is + * empty or resolves to `"."` or `".."`, a fallback name from + * @ref makeFallbackName is used instead. Intended for securing uploaded + * filenames against directory traversal and control character injection. + * + * @ingroup filesystem_utils + * + * @param raw Untrusted filename (may include paths, separators, or control chars). + * @return Sanitized filename safe for joining with an upload directory. + * + * @note This function does not alter Unicode codepoints except to remove control + * characters; callers may wish to normalize Unicode for cross-platform + * consistency. + * @warning The sanitized name is safe for use as a single path component, but + * not necessarily unique. Collisions should be handled by the caller. + */ std::string sanitizeFilename(const std::string& raw) { namespace fs = std::filesystem; - // 1) Drop any leading path components + + // 1) Drop any leading path components (e.g., "dir/file.txt" -> "file.txt"). fs::path p(raw); std::string name = p.filename().string(); - // 2) Remove path separators and control chars, keep everything else (including Unicode bytes) + // 2) Remove path separators and control characters; keep other bytes (including Unicode). name.erase( std::remove_if(name.begin(), name.end(), [](unsigned char c) { return c == '/' || c == '\\' || std::iscntrl(c); }), name.end()); - // 3) If that produced empty or “.”/“..”, fallback + // 3) If empty or reserved ".", "..", generate a timestamp-based fallback name. if (name.empty() || name == "." || name == "..") { name = makeFallbackName(); } @@ -222,9 +462,36 @@ std::string sanitizeFilename(const std::string& raw) { return name; } -// Returns empty on any error or if the resolved path would leave uploadRoot. +/** + * @brief Resolves a safe absolute path for an uploaded file. + * + * @details Given a trusted upload root and an untrusted relative path from a + * client request, this function: + * 1. Canonicalizes the upload root path. + * 2. Splits the raw relative path on `'/'` and sanitizes each segment + * via @ref sanitizeFilename to remove dangerous characters. + * 3. Canonicalizes the parent directory (so symlinks and `..` are + * resolved) while preserving the leaf name even if it does not yet exist. + * 4. Ensures that the final candidate path is still within the canonical + * upload root, rejecting any path that would escape it. + * + * @ingroup filesystem_utils + * + * @param uploadRoot Trusted base directory for file uploads. + * @param rawRelativePath Untrusted path from client input (relative to root). + * @return Absolute canonical path string under @p uploadRoot suitable for safe writing. + * Returns an empty string on error, invalid input, or if the resolved + * path would leave @p uploadRoot. + * + * @note Uses `std::filesystem::weakly_canonical` to avoid exceptions and to + * handle partially existing paths. If the leaf file does not exist, + * only the parent path is canonicalized. + * @warning This function only ensures the path is inside @p uploadRoot; callers + * must still create missing directories and open files securely + * (e.g., with `O_CREAT | O_EXCL` to avoid races). + * @see sanitizeFilename + */ std::string makeSafeUploadPath(const std::string& uploadRoot, const std::string& rawRelativePath) { - namespace fs = std::filesystem; std::error_code ec; // 1) Canonicalize uploadRoot @@ -235,7 +502,7 @@ std::string makeSafeUploadPath(const std::string& uploadRoot, const std::string& return {}; } - // 2) Split the rawRelativePath on “/” and sanitize each segment + // 2) Split the rawRelativePath on '/' and sanitize each segment fs::path candidate = root; std::stringstream ss(rawRelativePath); std::string segment; @@ -246,8 +513,7 @@ std::string makeSafeUploadPath(const std::string& uploadRoot, const std::string& candidate /= safeSeg; } - // 3) For a non‐existent leaf, weakly_canonical will strip it off, - // so canonicalize the parent then re‐append the leaf. + // 3) Canonicalize parent while preserving leaf fs::path leaf = candidate.filename(); fs::path canonParent = fs::weakly_canonical(candidate.parent_path(), ec); if (ec) { @@ -256,7 +522,7 @@ std::string makeSafeUploadPath(const std::string& uploadRoot, const std::string& } fs::path canonCandidate = canonParent / leaf; - // 4) Bound‐check: ensure canonCandidate is inside canonRoot + // 4) Bound-check: ensure result is inside canonRoot auto rootStr = canonRoot.generic_string(); auto candStr = canonCandidate.generic_string(); if (candStr.size() < rootStr.size() || candStr.compare(0, rootStr.size(), rootStr) != 0 || diff --git a/src/utils/htmlUtils.cpp b/src/utils/htmlUtils.cpp index 9f4f8d8c..1ebef179 100644 --- a/src/utils/htmlUtils.cpp +++ b/src/utils/htmlUtils.cpp @@ -6,14 +6,41 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/09 22:25:49 by nlouis #+# #+# */ -/* Updated: 2025/06/09 22:39:21 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 22:43:23 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file htmlUtils.cpp + * @brief Implements HTML escaping utilities for Webserv. + * + * @details Provides helper functions for safely embedding arbitrary strings + * into HTML content. These utilities perform entity escaping of + * special characters that would otherwise be interpreted as HTML + * markup, preventing injection vulnerabilities (e.g., XSS). + * + * @ingroup html_utils + */ + #include -// HTML-escape &, <, >, ", ' -// so it's safe to embed in an HTML context. +/** + * @brief Escapes special HTML characters in a string. + * + * @details Replaces the characters `&`, `<`, `>`, `"`, and `'` with their + * corresponding HTML entity representations so that the output + * can be safely embedded in an HTML context without being parsed + * as markup. This is intended to prevent HTML injection and + * cross-site scripting (XSS) when displaying untrusted data. + * + * @ingroup html_utils + * + * @param in Input string that may contain unsafe HTML characters. + * @return Escaped string safe for insertion into HTML content. + * + * @note Only escapes the five most common HTML special characters. If you + * need to handle other entities, extend this mapping accordingly. + */ std::string htmlEscape(const std::string& in) { std::string out; out.reserve(in.size()); diff --git a/src/utils/printInfo.cpp b/src/utils/printInfo.cpp index e2f3dd78..20cfe50c 100644 --- a/src/utils/printInfo.cpp +++ b/src/utils/printInfo.cpp @@ -6,10 +6,27 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 14:01:22 by irychkov #+# #+# */ -/* Updated: 2025/05/24 15:07:38 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 22:58:13 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file printInfo.cpp + * @brief Human-readable printing helpers for Webserv configuration and servers. + * + * @details Implements utilities to format and print usage instructions and a + * structured, readable summary of the loaded configuration: + * - `printUsage()` prints CLI usage hints. + * - `printServerHeader()` prints a banner line for each server block. + * - `printConfig()` renders effective server/location settings + * (limits, methods, error pages, CGI, upload store, etc.). + * + * These helpers are intended for diagnostics during startup and for + * developer/operator visibility of the parsed configuration. + * + * @ingroup utils + */ + #include "utils/printInfo.hpp" #include "utils/stringUtils.hpp" diff --git a/src/utils/stringUtils.cpp b/src/utils/stringUtils.cpp index 6014bc66..5b7bf513 100644 --- a/src/utils/stringUtils.cpp +++ b/src/utils/stringUtils.cpp @@ -6,10 +6,33 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/05 20:09:37 by nlouis #+# #+# */ -/* Updated: 2025/08/09 12:13:42 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 22:57:01 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file stringUtils.cpp + * @brief Utility functions for string parsing, formatting, and case conversion. + * + * @details Implements the @ref string_utils helpers used across Webserv for: + * - Parsing integers and byte sizes from configuration strings with rich error reporting. + * - Converting strings to lowercase or uppercase. + * - Formatting byte counts into human-readable units. + * - Joining lists of strings with a delimiter. + * - Trimming whitespace from strings. + * + * These functions are used in configuration parsing, logging, formatting + * HTTP responses, and general-purpose string handling. + * + * @ingroup string_utils + * + * @note Parsing helpers (`parseInt`, `parseByteSize`) throw @ref ConfigParseError + * with contextual information for configuration errors. + * @warning None of these functions are locale-sensitive beyond standard + * `std::tolower` / `std::toupper` behavior, and they do not perform + * Unicode normalization. + */ + #include "utils/stringUtils.hpp" #include "config/parser/ConfigParseError.hpp" #include "utils/errorUtils.hpp" @@ -20,6 +43,33 @@ #include #include +/** + * @ingroup string_utils + * @brief Parses a non-negative integer from a string with context-aware error reporting. + * + * @details Attempts to parse the entire @p value string into a non-negative integer + * using `std::from_chars` for efficient, allocation-free conversion. + * If parsing fails, the string contains extra characters, or the result + * is negative, a @ref ConfigParseError is thrown. The error includes: + * - The field name for context. + * - The offending value. + * - The line and column numbers. + * - Additional context from @p context_provider, such as a snippet of the + * configuration file being parsed. + * + * @param value String representation of the integer to parse. + * @param field Name of the configuration field being parsed. + * @param line Line number in the configuration source. + * @param column Column number in the configuration source. + * @param context_provider Callable returning a string snippet or context for diagnostics. + * @return Parsed integer value (guaranteed to be non-negative). + * + * @throws ConfigParseError If parsing fails, if extra characters remain after parsing, + * or if the parsed value is negative. + * + * @note This function is typically used when reading numeric configuration values + * (e.g., port numbers, limits) that must be whole, non-negative integers. + */ int parseInt(const std::string& value, const std::string& field, int line, int column, const std::function& context_provider) { int result = 0; @@ -38,6 +88,33 @@ int parseInt(const std::string& value, const std::string& field, int line, int c return result; } +/** + * @ingroup string_utils + * @brief Parses a human-readable byte size string into a byte count. + * + * @details Converts a size string (e.g., `"10K"`, `"5M"`, `"2G"`, or raw bytes `"512"`) + * into a `std::size_t` representing the number of bytes. + * The suffix, if present, is case-insensitive and supports: + * - `K` = kibibytes (× 1024) + * - `M` = mebibytes (× 1024²) + * - `G` = gibibytes (× 1024³) + * Throws a @ref ConfigParseError if: + * - The value is empty. + * - Parsing fails or extra non-numeric characters remain. + * - The computed size exceeds the 4 GiB hard limit from the Webserv spec. + * + * @param value String containing the size to parse (may include suffix). + * @param field Name of the configuration field being parsed. + * @param line Line number in the configuration source. + * @param column Column number in the configuration source. + * @param context_provider Callable returning a string snippet or context for diagnostics. + * @return Parsed size in bytes. + * + * @throws ConfigParseError If the string is empty, cannot be parsed as a size, contains + * leftover characters, or exceeds the maximum allowed size. + * + * @note Used primarily to parse configuration directives like `client_max_body_size`. + */ std::size_t parseByteSize(const std::string& value, const std::string& field, int line, int column, const std::function& context_provider) { if (value.empty()) { diff --git a/src/utils/urlUtils.cpp b/src/utils/urlUtils.cpp index 403c2dc5..e1f3a57d 100644 --- a/src/utils/urlUtils.cpp +++ b/src/utils/urlUtils.cpp @@ -5,11 +5,24 @@ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ -/* Created: 2025/06/06 13:22:39 by nlouis #+# #+# */ -/* Updated: 2025/06/06 21:53:06 by nlouis ### ########.fr */ +/* Created: 2025/08/15 22:56:08 by nlouis #+# #+# */ +/* Updated: 2025/08/15 22:56:09 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file urlUtils.cpp + * @brief URL and form-encoding helpers. + * + * @details Implements percent-decoding (RFC 3986), `application/x-www-form-urlencoded` + * decoding (treating '+' as space), simple key/value parsing for form bodies, + * and a safe filename extractor from a URI segment. These utilities are used + * during request parsing and upload handling to turn encoded inputs into + * validated, safe strings. + * + * @ingroup url_utils + */ + #include #include #include @@ -17,6 +30,22 @@ #include #include +/** + * @brief Decodes percent-encoded octets in a string. + * + * @details Scans the input for sequences of the form `%HH` where `H` is a hex digit, + * converts each pair to a single byte, and returns the decoded result. + * Leaves all non-encoded characters unchanged. + * + * @ingroup url_utils + * + * @param encoded Input possibly containing percent-encoded bytes. + * @return Decoded string with `%HH` sequences replaced by their byte values. + * + * @throws std::invalid_argument If a `%` is incomplete at the end of the string or + * if the two following characters are not hex digits. + * @note This function does not perform UTF‑8 validation; it operates on bytes. + */ std::string decodePercentEncoding(const std::string& encoded) { std::ostringstream result; for (size_t i = 0; i < encoded.length(); ++i) { @@ -42,6 +71,20 @@ std::string decodePercentEncoding(const std::string& encoded) { return result.str(); } +/** + * @brief Decodes `application/x-www-form-urlencoded` field content. + * + * @details First replaces `+` with a space (per form-url-encoded rules), then applies + * percent-decoding to `%HH` sequences. This is suitable for decoding both + * keys and values extracted from a form body. + * + * @ingroup url_utils + * + * @param input Raw form field string (may include `+` and `%HH`). + * @return Decoded string. + * + * @throws std::invalid_argument Propagated from @ref decodePercentEncoding on invalid encodings. + */ std::string percentDecodeForm(const std::string& input) { std::string temp; temp.reserve(input.size()); @@ -52,6 +95,22 @@ std::string percentDecodeForm(const std::string& input) { return decodePercentEncoding(temp); } +/** + * @brief Parses an `application/x-www-form-urlencoded` body into key/value pairs. + * + * @details Splits the body on `&`, then splits each pair on the first `=`. + * Both key and value are decoded using @ref percentDecodeForm. Empty pairs + * are ignored; missing `=` results in the pair being skipped. + * + * @ingroup url_utils + * + * @param body Full form body string (e.g., `"a=1&b=two+words"`). + * @return Map of decoded keys to decoded values. Later duplicates will not overwrite + * earlier ones due to `emplace`; adjust if you want overwrite semantics. + * + * @throws std::invalid_argument Propagated from @ref percentDecodeForm (invalid `%HH`). + * @note If you expect repeated keys, consider using `std::unordered_multimap` instead. + */ std::unordered_map parseFormUrlEncoded(const std::string& body) { std::unordered_map form; size_t start = 0; @@ -79,6 +138,23 @@ std::unordered_map parseFormUrlEncoded(const std::stri return form; } +/** + * @brief Extracts a safe filename from the last URI segment. + * + * @details Takes the substring after the last `/` in @p uri, attempts percent-decoding, + * and validates the result against a conservative allowlist. Rejects empty, + * too-long (>256), or suspicious names (contains `/`, equals `"."` or `".."`, + * starts with `.` or `-`, or fails the regex `^[a-zA-Z0-9._-]+$`). + * + * @ingroup url_utils + * + * @param uri Source URI or path-like string. + * @return A validated filename; returns an empty string if decoding fails or the + * candidate does not pass validation. + * + * @note Designed for deriving a download/upload filename from a URI segment without + * risking directory traversal or confusing special names. + */ std::string extractFilenameFromUri(const std::string& uri) { std::string filename; @@ -103,7 +179,7 @@ std::string extractFilenameFromUri(const std::string& uri) { if (filename == "." || filename == ".." || filename[0] == '.' || filename[0] == '-') return ""; - // Optional: enforce strict pattern + // Enforce strict pattern static const std::regex safePattern(R"(^[a-zA-Z0-9._-]+$)"); if (!std::regex_match(filename, safePattern)) return ""; From 0a66803f690f5699e11ddb4acbcf2eedd8acf8a9 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Fri, 15 Aug 2025 23:16:30 +0300 Subject: [PATCH 13/32] Enhance documentation for Location and Server components, update Doxygen groupings, and refine method descriptions --- docs/doxygen.warnings | 22 --- include/core/Location.hpp | 88 ++++++++---- include/core/Server.hpp | 6 +- include/doxygen_groups.hpp | 22 ++- src/core/Location.cpp | 276 ++++++++++++++++++++++++++++++++++--- src/core/Server.cpp | 36 ++--- src/core/runWebserv.cpp | 14 +- 7 files changed, 367 insertions(+), 97 deletions(-) diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index e10b6431..4fa893b2 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -1,26 +1,4 @@ warning: source 'tests' is not a readable file or directory... skipping. -/home/toonsa/myProjects/webserv/include/core/Location.hpp:82: warning: unbalanced grouping commands -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'value' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'field' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'line' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'column' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'context_provider' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'value' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'field' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'line' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'column' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'context_provider' from the argument list of parseByteSize has multiple @param documentation sections error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'value' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'field' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'line' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'column' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:41: warning: argument 'context_provider' from the argument list of parseInt has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'value' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'field' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'line' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'column' from the argument list of parseByteSize has multiple @param documentation sections -/home/toonsa/myProjects/webserv/include/utils/stringUtils.hpp:60: warning: argument 'context_provider' from the argument list of parseByteSize has multiple @param documentation sections -error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! /home/toonsa/myProjects/webserv/README.md:103: warning: unable to resolve reference to '/home/toonsa/myProjects/webserv/CONTRIBUTING.md' for \ref command diff --git a/include/core/Location.hpp b/include/core/Location.hpp index cbefa78c..0ab417c4 100644 --- a/include/core/Location.hpp +++ b/include/core/Location.hpp @@ -6,28 +6,59 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/02 13:45:05 by nlouis #+# #+# */ -/* Updated: 2025/05/24 15:05:43 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:15:01 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file Location.hpp + * @brief Declares the Location class for route-specific configuration. + * + * @details Represents a configuration block tied to a specific URL path within + * a virtual server. Each Location defines its own root, allowed methods, + * redirection rules, upload storage, index files, and CGI settings. + * This granularity allows per-path customization of behavior and routing. + * + * @ingroup location_component + */ + #pragma once -#include -#include -#include -#include +#include // std::map for CGI interpreter mapping +#include // std::set for allowed HTTP methods +#include // std::string for paths and configuration values +#include // std::vector for index and CGI extension lists +/** + * @brief Encapsulates configuration for a single URL path. + * + * @details Models a location block inside a server configuration. This includes: + * - Path matching rules + * - File system root for serving content + * - Directory listing behavior + * - HTTP method restrictions + * - Optional redirection + * - CGI execution parameters + * - Upload handling + * + * @ingroup location_component + */ class Location { public: - //////////////////////////////// - // --- Constructor + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ Location(); ~Location() = default; Location(const Location& other) = default; Location& operator=(const Location& other) = default; + ///@} - /////////////////// - // --- Setters --- + //=== Configuration Setters ============================================== + + /** @name Configuration setters */ + ///@{ void setPath(const std::string& path); void setRoot(const std::string& root); void setAutoindex(bool enabled); @@ -38,9 +69,12 @@ class Location { void setAllowedMethods(const std::vector& methods); void addIndexFile(const std::string& idx); void addCgiInterpreter(const std::string& ext, const std::string& path); + ///@} + + //=== Queries (Getters) =================================================== - /////////////// - // --- Getters + /** @name Queries (getters) */ + ///@{ const std::string& getPath() const; const std::set& getMethods() const; const std::string& getRoot() const; @@ -55,9 +89,12 @@ class Location { const std::vector& getCgiExtensions() const; std::string getCgiInterpreter(const std::string& ext) const; const std::map& getCgiInterpreterMap() const; + ///@} - ///////////////////// - // --- Logic helpers + //=== Logic & Matching Helpers ============================================ + + /** @name Logic helpers */ + ///@{ bool hasAllowedMethods() const; bool isMethodAllowed(const std::string& method) const; bool matchesPath(const std::string& uri) const; @@ -65,18 +102,19 @@ class Location { bool isUploadEnabled() const; bool isCgiRequest(const std::string& uri) const; std::string getEffectiveIndexPath() const; + ///@} private: - std::string _path; ///< URL path this location matches. - std::set _methods; ///< Set of allowed HTTP methods. - std::string _root; ///< Root directory for file serving. - bool _autoindex; ///< Whether to enable directory listing. - std::string _redirect; ///< Redirection target URL. - int _return_code; ///< HTTP status code for redirection. - std::string _upload_store; ///< Directory for uploaded files. - std::vector _index_files; ///< Ordered list of index files. - std::vector _cgi_extensions; ///< Ordered list of CGI extensions. - std::map _cgi_interpreters; -}; + //=== Data Members ======================================================== -/** @} */ + std::string _path; ///< URL path this location matches. + std::set _methods; ///< Allowed HTTP methods. + std::string _root; ///< Root directory for file serving. + bool _autoindex; ///< Enable/disable directory listing. + std::string _redirect; ///< Target URL for redirection. + int _return_code; ///< HTTP status code for redirection. + std::string _upload_store; ///< Directory for file uploads. + std::vector _index_files; ///< List of default index files. + std::vector _cgi_extensions; ///< File extensions for CGI execution. + std::map _cgi_interpreters; ///< CGI extension → interpreter mapping. +}; diff --git a/include/core/Server.hpp b/include/core/Server.hpp index 4d7f5529..bff52fca 100644 --- a/include/core/Server.hpp +++ b/include/core/Server.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:37:06 by irychkov #+# #+# */ -/* Updated: 2025/08/14 14:31:55 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:07:10 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,7 +19,7 @@ * error pages, configure body size limits, and contain multiple location * blocks with their own routing rules and behavior. * - * @ingroup server + * @ingroup server_component */ #pragma once @@ -41,7 +41,7 @@ * `server` block in the configuration file and participates in request * routing based on host and port matching. * - * @ingroup server + * @ingroup server_component */ class Server { private: diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index 45caffa2..b9131df0 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -25,7 +25,7 @@ */ /** - * @defgroup server Server Component + * @defgroup server_component Server Component * @ingroup core * @brief Main Server class and its implementation. * @@ -39,6 +39,26 @@ * - `core/server_utils.cpp` — findMatchingServer() */ +/** + * @defgroup location_component Location Component + * @ingroup core + * @brief Path-specific routing and configuration within a server. + * + * @details Contains the `Location` class, representing a configuration + * block bound to a specific URI path inside a server. + * Each location can define: + * - A document root. + * - Allowed HTTP methods. + * - Index files and autoindexing. + * - Redirection rules. + * - CGI execution settings. + * - Upload directory configuration. + * + * Files: + * - `core/Location.hpp` — class declaration. + * - `core/Location.cpp` — method implementations. + */ + /** * @defgroup entrypoint Application Entrypoint * @ingroup core diff --git a/src/core/Location.cpp b/src/core/Location.cpp index 4b947f91..b0536163 100644 --- a/src/core/Location.cpp +++ b/src/core/Location.cpp @@ -6,124 +6,304 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:45:32 by irychkov #+# #+# */ -/* Updated: 2025/05/31 14:49:34 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:14:13 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ -#include "core/Location.hpp" -#include "utils/filesystemUtils.hpp" -#include "utils/stringUtils.hpp" -#include -#include +/** + * @file Location.cpp + * @brief Implements the Location class for route-specific configuration. + * + * @details Defines all methods for @ref Location, including setters, getters, + * and logic helpers for request routing and filesystem resolution. + * A Location manages path matching, allowed methods, redirection, + * CGI parameters, index files, autoindexing, and upload storage. + * + * @ingroup location_component + */ + +#include "core/Location.hpp" // Class declaration +#include "utils/filesystemUtils.hpp" // normalizePath(), joinPath() +#include "utils/stringUtils.hpp" // toLower() +#include // std::find +#include // std::filesystem::path #include #include -/////////////////////// -// --- Constructor --- +//=== Construction & Special Members ===================================== +/** + * @brief Constructs a Location with safe defaults. + * + * @details Initializes autoindex to `false` and return code to `0`. + * Other members are left empty until populated by configuration. + * + * @ingroup location_component + */ Location::Location() : _autoindex(false), _return_code(0) { } -/////////////// -// --- Setters +//=== Configuration Setters ============================================== +/** + * @brief Sets the URI path this location matches. + * + * @param path Canonical route prefix (e.g., "/images"). + * + * @ingroup location_component + */ void Location::setPath(const std::string& path) { _path = path; } +/** + * @brief Adds a single allowed HTTP method. + * + * @param method Method name (e.g., "GET", "POST"). + * + * @ingroup location_component + */ void Location::addMethod(const std::string& method) { _methods.insert(method); } +/** + * @brief Replaces the set of allowed HTTP methods. + * + * @param methods List of allowed methods. + * + * @ingroup location_component + */ void Location::setAllowedMethods(const std::vector& methods) { _methods.clear(); - for (std::vector::const_iterator it = methods.begin(); it != methods.end(); ++it) { + for (std::vector::const_iterator it = methods.begin(); it != methods.end(); ++it) addMethod(*it); - } } +/** + * @brief Sets the filesystem root for this location. + * + * @param root Directory path. + * + * @ingroup location_component + */ void Location::setRoot(const std::string& root) { _root = root; } +/** + * @brief Appends an index file candidate in preference order. + * + * @param idx Filename (e.g., "index.html"). + * + * @ingroup location_component + */ void Location::addIndexFile(const std::string& idx) { _index_files.push_back(idx); } +/** + * @brief Enables or disables directory listing. + * + * @param enabled `true` to enable, `false` to disable. + * + * @ingroup location_component + */ void Location::setAutoindex(bool enabled) { _autoindex = enabled; } +/** + * @brief Configures an HTTP redirect. + * + * @param target Target URL. + * @param code HTTP status code (default 301). + * + * @ingroup location_component + */ void Location::setRedirect(const std::string& target, int code) { _redirect = target; _return_code = code; } +/** + * @brief Sets the upload store directory. + * + * @param path Filesystem path for uploaded files. + * + * @ingroup location_component + */ void Location::setUploadStore(const std::string& path) { _upload_store = path; } +/** + * @brief Adds a CGI extension. + * + * @param ext Extension (e.g., ".php"). + * + * @ingroup location_component + */ void Location::addCgiExtension(const std::string& ext) { _cgi_extensions.push_back(ext); } +/** + * @brief Associates a CGI extension with its interpreter. + * + * @param ext Extension (e.g., ".py"). + * @param path Interpreter path. + * + * @ingroup location_component + */ void Location::addCgiInterpreter(const std::string& ext, const std::string& path) { _cgi_interpreters[ext] = path; } -/////////////// -// --- Getters +//=== Queries (Getters) =================================================== +/** + * @brief Returns the path this location matches. + * + * @return Reference to the path string. + * + * @ingroup location_component + */ const std::string& Location::getPath() const { return _path; } +/** + * @brief Returns the allowed HTTP methods. + * + * @return Set of method strings. + * + * @ingroup location_component + */ const std::set& Location::getMethods() const { return _methods; } +/** + * @brief Returns the root directory path. + * + * @return Reference to root string. + * + * @ingroup location_component + */ const std::string& Location::getRoot() const { return _root; } +/** + * @brief Returns the first index filename, or empty string. + * + * @return Reference to filename or empty string. + * + * @ingroup location_component + */ const std::string& Location::getIndex() const { static const std::string empty; return _index_files.empty() ? empty : _index_files.front(); } +/** + * @brief Returns all index filenames. + * + * @return Vector of filenames. + * + * @ingroup location_component + */ const std::vector& Location::getIndexFiles() const { return _index_files; } +/** + * @brief Checks if autoindexing is enabled. + * + * @return `true` if enabled. + * + * @ingroup location_component + */ bool Location::isAutoindexEnabled() const { return _autoindex; } +/** + * @brief Checks if a redirect is configured. + * + * @return `true` if redirect is set. + * + * @ingroup location_component + */ bool Location::hasRedirect() const { return !_redirect.empty(); } +/** + * @brief Returns the redirect target. + * + * @return Reference to redirect URL. + * + * @ingroup location_component + */ const std::string& Location::getRedirect() const { return _redirect; } +/** + * @brief Returns the redirect HTTP status code. + * + * @return HTTP code. + * + * @ingroup location_component + */ int Location::getReturnCode() const { return _return_code; } +/** + * @brief Returns the upload store directory. + * + * @return Reference to upload path. + * + * @ingroup location_component + */ const std::string& Location::getUploadStore() const { return _upload_store; } +/** + * @brief Returns the first CGI extension, or empty string. + * + * @return Reference to extension string. + * + * @ingroup location_component + */ const std::string& Location::getCgiExtension() const { static const std::string empty; return _cgi_extensions.empty() ? empty : _cgi_extensions.front(); } +/** + * @brief Returns all CGI extensions. + * + * @return Vector of extensions. + * + * @ingroup location_component + */ const std::vector& Location::getCgiExtensions() const { return _cgi_extensions; } +/** + * @brief Returns the interpreter for a CGI extension. + * + * @param ext Extension (case-insensitive, optional dot). + * @return Interpreter path or empty string. + * + * @ingroup location_component + */ std::string Location::getCgiInterpreter(const std::string& ext) const { std::string key = toLower(ext); if (!key.empty() && key[0] != '.') @@ -133,32 +313,64 @@ std::string Location::getCgiInterpreter(const std::string& ext) const { return (it != _cgi_interpreters.end()) ? it->second : ""; } +/** + * @brief Returns the CGI extension → interpreter map. + * + * @return Map of extension to interpreter path. + * + * @ingroup location_component + */ const std::map& Location::getCgiInterpreterMap() const { return _cgi_interpreters; } -///////////////////// -// --- Logic Helpers +//=== Logic Helpers ======================================================= +/** + * @brief Checks if allowed methods list is non-empty. + * + * @return `true` if at least one method is allowed. + * + * @ingroup location_component + */ bool Location::hasAllowedMethods() const { return !_methods.empty(); } +/** + * @brief Checks if a method is allowed. + * + * @param method Method name. + * @return `true` if allowed. + * + * @ingroup location_component + */ bool Location::isMethodAllowed(const std::string& method) const { return _methods.count(method) > 0; } -/* bool Location::matchesPath(const std::string& uri) const { - return normalizePath(uri).rfind(normalizePath(_path), 0) == 0; -} */ - +/** + * @brief Checks if a URI matches this location's path. + * + * @param uri Request URI. + * @return `true` if URI starts with location path. + * + * @ingroup location_component + */ bool Location::matchesPath(const std::string& uri) const { std::string cleanUri = normalizePath(uri); std::string locPath = normalizePath(_path); - return cleanUri.rfind(locPath, 0) == 0; } +/** + * @brief Resolves an absolute filesystem path from a URI. + * + * @param uri Request URI. + * @return Joined filesystem path, or empty on mismatch. + * + * @ingroup location_component + */ std::string Location::resolveAbsolutePath(const std::string& uri) const { std::string cleanUri = normalizePath(uri); if (!matchesPath(cleanUri) || _path.length() > cleanUri.length()) @@ -166,15 +378,37 @@ std::string Location::resolveAbsolutePath(const std::string& uri) const { return joinPath(_root, cleanUri.substr(_path.length())); } +/** + * @brief Checks if uploads are enabled. + * + * @return `true` if upload store is set. + * + * @ingroup location_component + */ bool Location::isUploadEnabled() const { return !_upload_store.empty(); } +/** + * @brief Checks if a path refers to a CGI request. + * + * @param path Filesystem or URI path. + * @return `true` if extension is in CGI list. + * + * @ingroup location_component + */ bool Location::isCgiRequest(const std::string& path) const { std::string ext = std::filesystem::path(path).extension().string(); return std::find(_cgi_extensions.begin(), _cgi_extensions.end(), ext) != _cgi_extensions.end(); } +/** + * @brief Returns absolute path to first index file. + * + * @return `joinPath(root, index)` or empty if none. + * + * @ingroup location_component + */ std::string Location::getEffectiveIndexPath() const { if (_index_files.empty()) return ""; diff --git a/src/core/Server.cpp b/src/core/Server.cpp index 66bad794..54464af0 100644 --- a/src/core/Server.cpp +++ b/src/core/Server.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 09:51:19 by irychkov #+# #+# */ -/* Updated: 2025/08/14 14:34:04 by nlouis ### ########.fr */ +/* Updated: 2025/08/15 23:08:18 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -19,7 +19,7 @@ * error pages, body size limits, and associated location blocks. * It is part of the configuration system and supports parsing and runtime use. * - * @ingroup server + * @ingroup server_component */ #include "core/Server.hpp" // Server class declaration @@ -40,7 +40,7 @@ * client max body size (`1 MiB`), and empty location/error blocks. * Intended to be populated via configuration parsing. * - * @ingroup server + * @ingroup server_component */ Server::Server() : _port(80), // Default HTTP port @@ -58,7 +58,7 @@ Server::Server() * the `listen` directive in the configuration file. Validation is done before calling. * * @param port The TCP port number to bind to. - * @ingroup server + * @ingroup server_component */ void Server::setPort(int port) noexcept { _port = port; @@ -72,7 +72,7 @@ void Server::setPort(int port) noexcept { * via the `host` directive in the configuration file. No validation is done here. * * @param host The IP address to bind (e.g., "127.0.0.1" or "0.0.0.0"). - * @ingroup server + * @ingroup server_component */ void Server::setHost(std::string_view host) noexcept { _host = host; @@ -86,7 +86,7 @@ void Server::setHost(std::string_view host) noexcept { * This method appends without deduplication. * * @param name The server name to add (e.g., "example.com"). - * @ingroup server + * @ingroup server_component */ void Server::addServerName(std::string_view name) { // Append the given server name to the list of aliases. @@ -102,7 +102,7 @@ void Server::addServerName(std::string_view name) { * * @param code The HTTP error status code to override. * @param path The file path to serve as the custom error page. - * @ingroup server + * @ingroup server_component */ void Server::setErrorPage(int code, const std::string& path) { // Map the given HTTP status code to a custom error page path. @@ -117,7 +117,7 @@ void Server::setErrorPage(int code, const std::string& path) { * should reject it with a 413 Payload Too Large response. * * @param size Maximum body size in bytes. - * @ingroup server + * @ingroup server_component */ void Server::setClientMaxBodySize(std::size_t size) noexcept { _client_max_body_size = size; @@ -131,7 +131,7 @@ void Server::setClientMaxBodySize(std::size_t size) noexcept { * for specific URI prefixes under this server. * * @param location The `Location` instance to add. - * @ingroup server + * @ingroup server_component */ void Server::addLocation(const Location& location) { // Add a new location block to the server's routing table. @@ -147,7 +147,7 @@ void Server::addLocation(const Location& location) { * the valid TCP range [0, 65535]. * * @return The configured TCP port. - * @ingroup server + * @ingroup server_component */ int Server::getPort() const noexcept { return _port; @@ -161,7 +161,7 @@ int Server::getPort() const noexcept { * Typically set via the `host` directive in the configuration file. * * @return Reference to the host IP address string. - * @ingroup server + * @ingroup server_component */ const std::string& Server::getHost() const noexcept { return _host; @@ -175,7 +175,7 @@ const std::string& Server::getHost() const noexcept { * Configured via the `server_name` directive. * * @return Reference to the list of server names. - * @ingroup server + * @ingroup server_component */ const std::vector& Server::getServerNames() const noexcept { return _server_names; @@ -191,7 +191,7 @@ const std::vector& Server::getServerNames() const noexcept { * request routing. * * @return The default server name string. - * @ingroup server + * @ingroup server_component */ const std::string Server::getDefaultServerName() const { return _server_names.empty() ? "localhost" : _server_names.front(); @@ -205,7 +205,7 @@ const std::string Server::getDefaultServerName() const { * Configured via the `error_page` directive. * * @return Reference to the map of error codes to file paths. - * @ingroup server + * @ingroup server_component */ const std::map& Server::getErrorPages() const noexcept { return _error_pages; @@ -219,7 +219,7 @@ const std::map& Server::getErrorPages() const noexcept { * a 413 Payload Too Large error. Configured via the `client_max_body_size` directive. * * @return The maximum request body size in bytes. - * @ingroup server + * @ingroup server_component */ std::size_t Server::getClientMaxBodySize() const noexcept { return _client_max_body_size; @@ -233,7 +233,7 @@ std::size_t Server::getClientMaxBodySize() const noexcept { * server selects the best-matching location based on the URI. * * @return Reference to the list of `Location` objects. - * @ingroup server + * @ingroup server_component */ const std::vector& Server::getLocations() const noexcept { return _locations; @@ -247,7 +247,7 @@ const std::vector& Server::getLocations() const noexcept { * Use with care to avoid breaking routing logic. * * @return Reference to the list of `Location` objects. - * @ingroup server + * @ingroup server_component */ std::vector& Server::getLocations() noexcept { return _locations; @@ -263,7 +263,7 @@ std::vector& Server::getLocations() noexcept { * * @param name The server name to check (case-sensitive). * @return `true` if the name matches one of the configured server names. - * @ingroup server + * @ingroup server_component */ bool Server::hasServerName(std::string_view name) const noexcept { // Check if the given name matches any of the configured server names. diff --git a/src/core/runWebserv.cpp b/src/core/runWebserv.cpp index 3d759f2b..22be713c 100644 --- a/src/core/runWebserv.cpp +++ b/src/core/runWebserv.cpp @@ -28,11 +28,11 @@ #include "network/SocketManager.hpp" #include "utils/printInfo.hpp" -#include // std::ifstream for reading the configuration file -#include // std::ostringstream for buffering file content -#include // std::runtime_error for error reporting -#include // std::string for configuration paths and file content -#include // std::string_view for DEFAULT_CONFIG_PATH constant +#include // std::ifstream for reading the configuration file +#include // std::ostringstream for buffering file content +#include // std::runtime_error for error reporting +#include // std::string for configuration paths and file content +#include // std::string_view for DEFAULT_CONFIG_PATH constant /** * @namespace bootstrap @@ -110,7 +110,7 @@ Config loadConfig(const std::string& fileContent) { return parser.parseConfig(); } -} // namespace +} // namespace bootstrap /** * @brief Runs the Webserv server. @@ -125,7 +125,7 @@ Config loadConfig(const std::string& fileContent) { * @throws std::runtime_error If CLI arguments are invalid or the config file cannot be opened. * @throws ConfigParseError If the configuration cannot be tokenized/parsed. * @throws ValidationError If the resulting configuration fails validation rules. - * + * * @ingroup entrypoint */ int runWebserv(int argc, char** argv) { From 701635341ec78befe7a00c3831ebada99742ab51 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 14:10:33 +0300 Subject: [PATCH 14/32] Enhances configuration parsing and validation with detailed documentation Adds comprehensive docstrings and inline comments across configuration-related components, improving readability and maintainability. --- README.md | 10 - include/config/Config.hpp | 37 +- include/config/normalizeConfig.hpp | 31 +- include/config/parser/ConfigParseError.hpp | 38 +- include/config/parser/ConfigParser.hpp | 55 ++- .../config/parser/directive_handler_table.hpp | 46 +- include/config/tokenizer/Tokenizer.hpp | 109 ++++- include/config/tokenizer/token.hpp | 29 +- include/config/validateConfig.hpp | 15 +- include/doxygen_groups.hpp | 47 +- src/config/Config.cpp | 35 +- src/config/normalizeConfig.cpp | 60 ++- src/config/parser/ConfigParseError.cpp | 30 +- src/config/parser/ConfigParser.cpp | 450 ++++++++++++++---- src/config/parser/directive_handler_table.cpp | 306 +++++++++++- src/config/tokenizer/Tokenizer.cpp | 256 ++++++++-- src/config/tokenizer/token.cpp | 52 +- src/config/validateConfig.cpp | 392 ++++++++++++++- src/utils/filesystemUtils.cpp | 6 +- 19 files changed, 1757 insertions(+), 247 deletions(-) diff --git a/README.md b/README.md index fe59470f..3aa2ecdf 100644 --- a/README.md +++ b/README.md @@ -78,16 +78,6 @@ On each push or pull request to `main` or `dev`, the following jobs are run auto All configurations rely on the project `Makefile` and follow the project's coding style. -### 🧼 Sanitizer Suppressions - -To reduce noise in sanitizer reports, the `.asanignore` file suppresses: - -- Known benign leaks from `libstdc++`, `libc`, and dynamic allocators -- Internal race conditions in `__sanitizer` symbols -- Undefined behavior in standard library internals - -These help CI focus on bugs *in your code*, not external sources. - ### 📚 Documentation - Doxygen generates HTML docs from source code and Markdown (`README.md` is the main page) diff --git a/include/config/Config.hpp b/include/config/Config.hpp index 60657e5a..bebbb000 100644 --- a/include/config/Config.hpp +++ b/include/config/Config.hpp @@ -6,32 +6,57 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 10:34:50 by irychkov #+# #+# */ -/* Updated: 2025/05/21 21:36:40 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 13:20:32 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file Config.hpp + * @brief Declares the Config aggregate for parsed virtual servers. + * + * @details Owns the collection of @ref Server objects produced by the parser + * and subsequently normalized/validated before runtime use. + * + * @ingroup config + */ + #pragma once #include "core/Server.hpp" #include +/** + * @brief Top-level configuration aggregate. + * + * @details Simple container holding all parsed @ref Server instances from the + * configuration input. Acts as the handoff object between parsing and + * later normalization/validation stages. + * + * @ingroup config + */ class Config { public: - //////////////////////////////// - // --- Constructor + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ Config() = default; ~Config() = default; Config(const Config&) = default; Config& operator=(const Config&) = default; Config(Config&&) noexcept = default; Config& operator=(Config&&) noexcept = default; + ///@} + + //=== Public API ========================================================== - ////////////////// - // --- Public API + /** @name Public API */ + ///@{ void addServer(const Server& server); const std::vector& getServers() const; std::vector& getServers(); + ///@} private: - std::vector _servers; ///< List of all parsed servers + std::vector _servers; ///< List of all parsed servers. }; diff --git a/include/config/normalizeConfig.hpp b/include/config/normalizeConfig.hpp index fd5afb16..2b3a0ce4 100644 --- a/include/config/normalizeConfig.hpp +++ b/include/config/normalizeConfig.hpp @@ -1,17 +1,42 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* normalizeConfig.hpp :+: :+: :+: */ +/* normalizeConfig.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 21:02:21 by nlouis #+# #+# */ -/* Updated: 2025/05/20 22:37:22 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 13:26:36 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file normalizeConfig.hpp + * @brief Declares normalization utilities for parsed configuration. + * + * @details Applies canonical defaults and fills in missing values after parsing, + * so downstream components can rely on predictable, complete settings. + * Typical defaults include body size limits, error pages, roots, indices, + * and allowed methods. + * + * @ingroup config_normalizing + */ + #pragma once #include "config/Config.hpp" -void normalizeConfig(Config& config); \ No newline at end of file +/** + * @brief Applies in-place normalization to a parsed configuration. + * + * @details Iterates over all servers/locations and fills defaults for: + * - client_max_body_size + * - error_page mappings + * - location root and index files + * - allowed HTTP methods + * + * @param config Parsed configuration to normalize (modified in place). + * + * @ingroup config_normalizing + */ +void normalizeConfig(Config& config); diff --git a/include/config/parser/ConfigParseError.hpp b/include/config/parser/ConfigParseError.hpp index 30f49cb5..28d2322a 100644 --- a/include/config/parser/ConfigParseError.hpp +++ b/include/config/parser/ConfigParseError.hpp @@ -6,27 +6,53 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/06 13:13:44 by nlouis #+# #+# */ -/* Updated: 2025/05/21 21:37:09 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 12:01:14 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file ConfigParseError.hpp + * @brief Declares parse/validation error types for the config subsystem. + * + * @details Provides a single exception class used for tokenizer, parser, and + * validator errors. The class formats a human-friendly message and + * optionally appends a contextual source line for diagnostics. + * + * @ingroup config_parse_error + */ + #pragma once #include #include +/** + * @brief Base exception for configuration processing errors. + * + * @details Carries a formatted message and an optional context line (e.g., + * the offending source line). All config-related errors (tokenizer, + * parser, unexpected token, validation) alias this class. + * + * @ingroup config_parse_error + */ class ConfigParseError : public std::exception { public: + /** @name Construction & interface */ + ///@{ ConfigParseError(const std::string& message, const std::string& context = ""); - const char* what() const noexcept override; + ///@} protected: - std::string _context; ///< Optional contextual line - std::string _fullMessage; ///< Formatted error with context + std::string _context; ///< Optional contextual source line shown after the message. + std::string _fullMessage; ///< Final formatted message (message + context arrow). }; -using SyntaxError = ConfigParseError; +/** @brief Alias for syntax errors (tokenizer/parser). @ingroup config_parse_error */ +using SyntaxError = ConfigParseError; +/** @brief Alias for unexpected token errors. @ingroup config_parse_error */ using UnexpectedToken = ConfigParseError; -using TokenizerError = ConfigParseError; +/** @brief Alias for tokenizer-specific errors. @ingroup config_parse_error */ +using TokenizerError = ConfigParseError; +/** @brief Alias for configuration validation errors.@ingroup config_parse_error */ using ValidationError = ConfigParseError; diff --git a/include/config/parser/ConfigParser.hpp b/include/config/parser/ConfigParser.hpp index c40f139b..3e495933 100644 --- a/include/config/parser/ConfigParser.hpp +++ b/include/config/parser/ConfigParser.hpp @@ -6,10 +6,21 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/05 15:06:31 by nlouis #+# #+# */ -/* Updated: 2025/05/22 21:03:44 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 12:10:47 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file ConfigParser.hpp + * @brief Declares the ConfigParser for syntactic/semantic directive parsing. + * + * @details Consumes a token stream from @ref Tokenizer and builds a concrete + * configuration model (@ref Config, @ref Server, @ref Location) by + * dispatching recognized directives to handler tables. + * + * @ingroup config_parsing + */ + #pragma once #include "config/Config.hpp" @@ -22,25 +33,38 @@ #include #include +/** + * @brief Syntactic parser for the webserv configuration language. + * + * @details Validates block structure (`server` → `location`) and delegates + * directive semantics to the handler tables. Produces a fully + * constructed @ref Config tree or throws on errors. + * + * @ingroup config_parsing + */ class ConfigParser { public: - ////////////////// - // --- Public API + //=== Public API ========================================================== + /** @name Public API */ + ///@{ ConfigParser(std::string source); Config parseConfig(); + ///@} private: - //////////////////////////// - // --- Server Block Parsing + //=== Server Block Parsing =============================================== + /** @name Server block parsing */ + ///@{ Server parseServer(); void parseServerDirective(Server& server); Location parseLocation(); void parseLocationDirective(Location& location); + ///@} - //////////////////////// - // --- Token Navigation - + //=== Token Navigation ==================================================== + /** @name Token navigation */ + ///@{ const Token& current() const; const Token& previous() const; const Token& peek(std::size_t offset = 1) const; @@ -51,13 +75,20 @@ class ConfigParser { void expect(TokenType expected, const std::string& context); Token expectOneOf(std::initializer_list types, const std::string& context); std::vector collectArgs(std::span validTypes); + ///@} - ///////////////////// - // --- Error Context + //=== Error Context ======================================================= + /** @name Error context */ + ///@{ std::string getLineSnippet() const; + ///@} private: - Tokenizer _tokenizer; ///< Tokenizer instance used to produce the token stream. - std::vector _tokens; ///< Flattened list of tokens extracted from the source input. + //=== Internal State ====================================================== + /** @name Internal state */ + ///@{ + Tokenizer _tokenizer; ///< Produces the token stream from the raw source. + std::vector _tokens; ///< Flattened token list consumed by the parser. std::size_t _pos = 0; ///< Current index in the token stream. + ///@} }; diff --git a/include/config/parser/directive_handler_table.hpp b/include/config/parser/directive_handler_table.hpp index 778c0da6..8a8505a5 100644 --- a/include/config/parser/directive_handler_table.hpp +++ b/include/config/parser/directive_handler_table.hpp @@ -6,10 +6,23 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/08 17:10:31 by nlouis #+# #+# */ -/* Updated: 2025/05/21 10:45:39 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 12:36:11 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file directive_handler_table.hpp + * @brief Declares handler maps for server/location directives. + * + * @details Provides the callable signatures and lookup tables used by + * @ref ConfigParser to apply parsed directives to @ref Server + * and @ref Location instances. Each entry maps a directive + * keyword (e.g., "listen", "root") to a function that mutates + * the target object or throws on invalid arguments. + * + * @ingroup config_parsing + */ + #pragma once #include @@ -22,14 +35,35 @@ namespace directive { +/** + * @brief Handler signature for server-level directives. + * + * @param s Target server to mutate. + * @param args Directive arguments as strings (already token-collected). + * @param line Source line for diagnostics. + * @param column Source column for diagnostics. + * @param ctx Contextual source snippet to append in error messages. + * + * @ingroup config_parsing + */ using ServerHandler = std::function& args, int line, - int column, const std::string& getLineSnippet)>; + int column, const std::string& ctx)>; -using LocationHandler = - std::function& args, int line, int column, - const std::string& getLineSnippet)>; +/** + * @brief Handler signature for location-level directives. + * + * @param loc Target location to mutate. + * @param args Directive arguments as strings (already token-collected). + * @param line Source line for diagnostics. + * @param column Source column for diagnostics. + * @param ctx Contextual source snippet to append in error messages. + * + * @ingroup config_parsing + */ +using LocationHandler = std::function& args, + int line, int column, const std::string& ctx)>; const std::unordered_map& serverHandlers(); const std::unordered_map& locationHandlers(); -} // namespace directive \ No newline at end of file +} // namespace directive diff --git a/include/config/tokenizer/Tokenizer.hpp b/include/config/tokenizer/Tokenizer.hpp index 5c3d8ff1..1716fec5 100644 --- a/include/config/tokenizer/Tokenizer.hpp +++ b/include/config/tokenizer/Tokenizer.hpp @@ -6,89 +6,146 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/04 13:17:47 by nlouis #+# #+# */ -/* Updated: 2025/05/21 21:36:52 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 11:45:36 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file Tokenizer.hpp + * @brief Declares the Tokenizer for config lexical analysis. + * + * @details Converts raw configuration text into a typed token stream with + * accurate line/column offsets for diagnostics. Recognizes keywords, + * identifiers, numbers with single-letter size suffixes, strings, + * punctuation, and skips comments/whitespace/BOM. + * + * @ingroup config_tokenizing + */ + #pragma once -#include "token.hpp" +#include "token.hpp" // Token, TokenType #include #include +/** + * @brief Stateful lexer for the configuration language. + * + * @details Produces a vector of @ref Token from an input string. The lexer is + * single-pass and records source positions for error reporting. It is + * intentionally strict (e.g., rejects multi-letter numeric suffixes). + * + * @ingroup config_tokenizing + */ class Tokenizer { public: - //////////////////// - // --- Constructors + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ explicit Tokenizer(std::string input); Tokenizer(const Tokenizer&) = delete; Tokenizer& operator=(const Tokenizer&) = delete; Tokenizer(Tokenizer&&) noexcept = default; Tokenizer& operator=(Tokenizer&&) noexcept = default; + ///@} - //////////////// - // --- Main API + //=== Main API ============================================================ + + /** @name Main API */ + ///@{ [[nodiscard]] std::vector tokenize(); + ///@} + + //=== Token Access Helpers =============================================== - //////////////////// - // --- Token Access + /** @name Token access helpers */ + ///@{ std::string extractLine(std::size_t offset) const; + ///@} private: - ///////////////////////// - // --- Core Cursor Logic + //=== Core Cursor Logic =================================================== + + /** @name Core cursor logic */ + ///@{ bool isAtEnd() const noexcept; bool match(char expected) noexcept; unsigned char peek() const noexcept; unsigned char peekNext() const noexcept; unsigned char advance() noexcept; + ///@} + + //=== Classification Logic =============================================== - //////////////////////////// - // --- Classification Logic + /** @name Classification logic */ + ///@{ inline bool isIdentifierStart(unsigned char c) const; inline bool isIdentifierChar(unsigned char c) const; + ///@} + + //=== High-Level Parsers ================================================== - ////////////////////////// - // --- High-Level Parsers + /** @name High-level parsers */ + ///@{ void skipUtf8BOM(); void skipWhitespaceAndComments(); TokenType resolveKeywordType(const std::string& word); void scanIdentifier(); void validateIdentifier(std::size_t start); Token parseIdentifierOrKeyword(); + ///@} - ///////////////////////////// - // --- Number & Unit Parsing + //=== Number & Unit Parsing ============================================== + + /** @name Number & unit parsing */ + ///@{ void scanDigits(); void scanOptionalUnitSuffix(); Token parseNumberOrUnit(); + ///@} + + //=== String Parsing ====================================================== - ////////////////////// - // --- String Parsing + /** @name String parsing */ + ///@{ void throwUnterminatedString(const std::string& reason); Token parseStringLiteral(); + ///@} + + //=== Whitespace & Comment Helpers ======================================= - //////////////////////////////////// - // --- Whitespace & Comment Helpers + /** @name Whitespace & comment helpers */ + ///@{ void skipCR(); void skipNewline(); void skipOtherWhitespace(); void skipHashComment(); + ///@} - ////////////////////// - // --- Token Dispatch + //=== Token Dispatch ====================================================== + + /** @name Token dispatch */ + ///@{ bool looksLikeIpAddress() const; void dispatchToken(); + ///@} + + //=== Token Creation ====================================================== - //////////////////////////////////// - // --- Token Creation + /** @name Token creation */ + ///@{ Token makeToken(TokenType type, const std::string& value) const; + ///@} + + //=== Internal State ====================================================== - ////////////////////// - // --- Internal State + /** @name Internal state */ + ///@{ std::string _input; ///< Raw input string to tokenize. std::vector _tokens; ///< Accumulated list of emitted tokens. std::size_t _pos = 0; ///< Current byte offset in the input. int _line = 1; ///< Current line number (1-based). int _column = 1; ///< Current column number (1-based). + ///@} }; diff --git a/include/config/tokenizer/token.hpp b/include/config/tokenizer/token.hpp index 45315e71..033acf84 100644 --- a/include/config/tokenizer/token.hpp +++ b/include/config/tokenizer/token.hpp @@ -6,14 +6,29 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 00:55:06 by nlouis #+# #+# */ -/* Updated: 2025/05/21 21:36:18 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 11:58:30 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file token.hpp + * @brief Token and TokenType declarations for the config lexer. + * + * @details Defines the token enum and POD struct used by the Tokenizer to + * represent lexemes with their source location (line/column/offset). + * + * @ingroup config_tokenizing + */ + #pragma once #include +/** + * @brief Token categories recognized by the configuration lexer. + * + * @ingroup config_tokenizing + */ enum class TokenType { IDENTIFIER, ///< A generic identifier (directive name or argument) NUMBER, ///< A numeric literal (may include optional size suffix) @@ -39,11 +54,19 @@ enum class TokenType { KEYWORD_CGI_EXTENSION ///< `cgi_extension` directive }; +/** + * @brief Plain-old-data token with source coordinates. + * + * @details Carries the token type, lexical value, and its position within the + * original input (line, column, byte offset) for diagnostics. + * + * @ingroup config_tokenizing + */ struct Token { TokenType type; ///< Type of the token (identifier, keyword, etc.) std::string value; ///< Lexical string value of the token - int line; ///< Line number where the token begins - int column; ///< Column offset (zero-based) + int line; ///< Line number where the token begins (1-based) + int column; ///< Column number where the token begins (1-based) std::size_t offset; ///< Byte offset in the original input string Token(TokenType t, const std::string& v, int l, int c, std::size_t o); diff --git a/include/config/validateConfig.hpp b/include/config/validateConfig.hpp index 4a2bddcb..5f50be8b 100644 --- a/include/config/validateConfig.hpp +++ b/include/config/validateConfig.hpp @@ -1,15 +1,26 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* validateConfig.hpp :+: :+: :+: */ +/* validateConfig.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/05 00:11:16 by nlouis #+# #+# */ -/* Updated: 2025/05/20 23:11:03 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 14:05:46 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file validateConfig.hpp + * @brief Declares validation for normalized configuration. + * + * @details Performs semantic checks after parsing/normalization to guarantee + * a runnable configuration (e.g., port/host collisions, invalid paths, + * duplicate server names on same listen, illegal method sets, etc.). + * + * @ingroup config_validating + */ + #pragma once #include "config/Config.hpp" diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index b9131df0..a2657dd8 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -9,13 +9,58 @@ /** * @defgroup config Configuration Parsing - * @brief Configuration structures, parsing, normalization, and validation. + * @brief Configuration structures, parsing, normalization, and validation. * * @details Handles reading, tokenizing, and parsing the Webserv configuration file(s), * validating directives, and preparing normalized configuration objects * for use by the core server. */ +/** + * @defgroup config_tokenizing Tokenizing + * @ingroup config + * @brief Lexical analysis of configuration text into tokens. + * + * @details Converts raw configuration input into a token stream (types, values, positions) + * consumed by the parser. + */ + +/** + * @defgroup config_parsing Parsing + * @ingroup config + * @brief Syntactic parsing and directive dispatch. + * + * @details Builds in-memory config objects from tokens and applies directive handlers + * to populate `Server`/`Location` structures. + */ + +/** + * @defgroup config_parse_error Parsing Errors + * @ingroup config + * @brief Error types and helpers for tokenizer/parser diagnostics. + * + * @details Structured exceptions carrying human-friendly messages and context snippets + * for syntax/lexing errors. + */ + +/** + * @defgroup config_normalizing Normalizing + * @ingroup config + * @brief Post-parse defaulting and canonicalization. + * + * @details Fills in default values (methods, error pages, indices, sizes) and + * normalizes paths/settings for predictable downstream behavior. + */ + +/** + * @defgroup config_validation Validating + * @ingroup config + * @brief Configuration validation passes. + * + * @details Static checks for structure, duplicates, domain/port uniqueness, allowed methods, + * CGI mappings, and filesystem preconditions (roots, upload stores). + */ + /** * @defgroup core Core Components * @brief Main server components, control flow, and application entry point. diff --git a/src/config/Config.cpp b/src/config/Config.cpp index 18e629dc..74d440ea 100644 --- a/src/config/Config.cpp +++ b/src/config/Config.cpp @@ -6,24 +6,51 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/04/30 10:36:29 by irychkov #+# #+# */ -/* Updated: 2025/05/21 14:59:06 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 13:20:53 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file Config.cpp + * @brief Implements the Config aggregate for parsed servers. + * + * @details Minimal container methods for appending and accessing the list of + * @ref Server instances created during parsing. + * + * @ingroup config + */ + #include "config/Config.hpp" #include "config/parser/ConfigParseError.hpp" -////////////////// -// --- Public API +//=== Public API ============================================================ +/** + * @brief Appends a parsed server to the configuration. + * + * @param server The server instance to add. + * @ingroup config + */ void Config::addServer(const Server& server) { - _servers.push_back(server); + _servers.push_back(server); // Store by value (copy/move elision applies) } +/** + * @brief Returns a mutable reference to the list of servers. + * + * @return Vector reference for in-place modifications. + * @ingroup config + */ std::vector& Config::getServers() { return _servers; } +/** + * @brief Returns a const reference to the list of servers. + * + * @return Read-only vector reference. + * @ingroup config + */ const std::vector& Config::getServers() const { return _servers; } diff --git a/src/config/normalizeConfig.cpp b/src/config/normalizeConfig.cpp index d40bf5a3..a8865fbd 100644 --- a/src/config/normalizeConfig.cpp +++ b/src/config/normalizeConfig.cpp @@ -1,28 +1,61 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* normalizeConfig.cpp :+: :+: :+: */ +/* normalizeConfig.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ -/* Created: 2025/05/11 21:02:41 by nlouis #+# #+# */ -/* Updated: 2025/05/20 22:37:46 by nlouis ### ########.fr */ +/* Created: 2025/08/18 13:12:00 by nlouis #+# #+# */ +/* Updated: 2025/08/18 13:26:53 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file normalizeConfig.cpp + * @brief Implements post-parse normalization of configuration objects. + * + * @details Establishes default values and canonical forms for servers and + * locations so that runtime logic can assume non-empty, consistent + * configuration fields. + * + * @ingroup config_normalizing + */ + #include "config/normalizeConfig.hpp" -constexpr size_t DEFAULT_CLIENT_MAX_BODY_SIZE = 1 * 1024 * 1024; -const std::string DEFAULT_ERROR_PAGE_PATH = "/error.html"; -const std::string DEFAULT_ROOT = "/var/www"; -const std::string DEFAULT_INDEX = "index.html"; -const std::vector DEFAULT_METHODS = {"GET", "POST", "DELETE"}; +/// 1 MiB default for request body limit. +constexpr std::size_t DEFAULT_CLIENT_MAX_BODY_SIZE = 1 * 1024 * 1024; +/// Default error page path used for common error codes. +const std::string DEFAULT_ERROR_PAGE_PATH = "/error.html"; +/// Default filesystem root for locations missing an explicit root. +const std::string DEFAULT_ROOT = "/var/www"; +/// Default index file for the root location if none provided. +const std::string DEFAULT_INDEX = "index.html"; +/// Default allowed HTTP methods when none are specified. +const std::vector DEFAULT_METHODS = {"GET", "POST", "DELETE"}; +/** + * @brief Normalizes a single server and its locations. + * + * @details + * - Ensures a default `client_max_body_size` if unset (1 MiB). + * - Injects a default set of `error_page` mappings if none were provided. + * - For each location: + * - Sets a default `root` when missing. + * - Adds a default `index` when the location is `/` and index list is empty. + * - Ensures a default set of allowed HTTP methods when unspecified. + * + * @param server Server to normalize (modified in place). + * + * @ingroup config_normalizing + */ static void normalizeServer(Server& server) { + // Fill default body size if not specified if (server.getClientMaxBodySize() == 0) { server.setClientMaxBodySize(DEFAULT_CLIENT_MAX_BODY_SIZE); } + // Install a basic set of error pages if none provided if (server.getErrorPages().empty()) { server.setErrorPage(500, DEFAULT_ERROR_PAGE_PATH); server.setErrorPage(404, DEFAULT_ERROR_PAGE_PATH); @@ -30,21 +63,32 @@ static void normalizeServer(Server& server) { server.setErrorPage(502, DEFAULT_ERROR_PAGE_PATH); } + // Normalize each location for (Location& loc : server.getLocations()) { + // Root path default if (loc.getRoot().empty()) { loc.setRoot(DEFAULT_ROOT); } + // Provide an index for "/" if none given if (loc.getIndexFiles().empty() && loc.getPath() == "/") { loc.addIndexFile(DEFAULT_INDEX); } + // Ensure at least a default method set if (!loc.hasAllowedMethods()) { loc.setAllowedMethods(DEFAULT_METHODS); } } } +/** + * @brief Applies normalization to all servers/locations in the config. + * + * @param config Parsed configuration to normalize (modified in place). + * + * @ingroup config_normalizing + */ void normalizeConfig(Config& config) { for (Server& s : config.getServers()) { normalizeServer(s); diff --git a/src/config/parser/ConfigParseError.cpp b/src/config/parser/ConfigParseError.cpp index 2946fbc4..245407b9 100644 --- a/src/config/parser/ConfigParseError.cpp +++ b/src/config/parser/ConfigParseError.cpp @@ -5,15 +5,35 @@ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ -/* Created: 2025/05/05 20:44:40 by nlouis #+# #+# */ -/* Updated: 2025/06/09 10:28:12 by nlouis ### ########.fr */ +/* Created: 2025/08/18 12:05:00 by nlouis #+# #+# */ +/* Updated: 2025/08/18 12:01:48 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file ConfigParseError.cpp + * @brief Implements configuration error construction and reporting. + * + * @details Formats error messages (optionally including a contextual source + * snippet) and exposes them through `what()`. + * + * @ingroup config_parse_error + */ + #include "config/parser/ConfigParseError.hpp" #include #include +/** + * @brief Builds a formatted error message with optional context. + * + * @details If `context` is provided, appends it on a new line prefixed with + * an arrow (`"→ "`), producing a compact, readable diagnostic. + * + * @param message Primary error message. + * @param context Optional offending line or snippet (may be empty). + * @ingroup config_parse_error + */ ConfigParseError::ConfigParseError(const std::string& message, const std::string& context) : _context(context) { if (_context.empty()) @@ -22,6 +42,12 @@ ConfigParseError::ConfigParseError(const std::string& message, const std::string _fullMessage = message + "\n→ " + _context; } +/** + * @brief Returns the formatted error string. + * + * @return Null-terminated C string owned by the exception. + * @ingroup config_parse_error + */ const char* ConfigParseError::what() const noexcept { return _fullMessage.c_str(); } diff --git a/src/config/parser/ConfigParser.cpp b/src/config/parser/ConfigParser.cpp index c970aaf9..c4d87ce6 100644 --- a/src/config/parser/ConfigParser.cpp +++ b/src/config/parser/ConfigParser.cpp @@ -5,11 +5,22 @@ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ -/* Created: 2025/05/09 08:46:22 by nlouis #+# #+# */ -/* Updated: 2025/08/09 12:11:52 by nlouis ### ########.fr */ +/* Created: 2025/08/18 12:18:00 by nlouis #+# #+# */ +/* Updated: 2025/08/18 12:29:57 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file ConfigParser.cpp + * @brief Implements parsing of servers, locations, and directives. + * + * @details Walks the token stream, enforces grammar, and dispatches each + * directive to handler maps to populate @ref Server and @ref Location. + * Provides precise diagnostics with contextual line snippets. + * + * @ingroup config_parsing + */ + #include "config/parser/ConfigParser.hpp" #include "config/Config.hpp" #include "config/parser/ConfigParseError.hpp" @@ -33,65 +44,154 @@ namespace { -static const std::unordered_set kRepeatableServerDirectives = {"error_page"}; -static const std::unordered_set kRepeatableLocationDirectives = {"methods", - "cgi_interpreter"}; -static constexpr std::array kArgTokenTypes = {TokenType::STRING, TokenType::NUMBER, - TokenType::IDENTIFIER}; - +/** + * @brief Directives allowed multiple times inside a server block. + * @ingroup config_parsing + */ +static const std::unordered_set kRepeatableServerDirectives = {"error_page"}; + +/** + * @brief Directives allowed multiple times inside a location block. + * @ingroup config_parsing + */ +static const std::unordered_set kRepeatableLocationDirectives = { + "methods", + "cgi_interpreter", +}; + +/** + * @brief Token types accepted as directive arguments (ordered by frequency). + * @ingroup config_parsing + */ +static constexpr std::array kArgTokenTypes = { + TokenType::STRING, + TokenType::NUMBER, + TokenType::IDENTIFIER, +}; + +/** + * @brief Checks whether a directive name is allowed (not duplicated). + * + * @details Tracks encountered directive names within a block (`server` or `location`). + * If the directive is marked as repeatable, it is always accepted. + * Otherwise, the name is inserted into the `seen` set: + * - If insertion succeeds, the directive is new and valid. + * - If insertion fails, the directive is a duplicate and rejected. + * + * @param name The directive keyword being checked. + * @param seen Set of already encountered directive names in the current block. + * @param repeatable Set of directive names allowed to appear multiple times. + * + * @return `true` if the directive is valid (new or repeatable), + * `false` if it is a duplicate and not allowed. + * + * @ingroup config_parser + */ bool checkDuplicateDirective(const std::string& name, std::unordered_set& seen, const std::unordered_set& repeatable) { return repeatable.contains(name) || seen.insert(name).second; } +/** + * @brief Dispatches a configuration directive to its registered handler. + * + * @details Looks up the directive keyword (from the current token) in the + * provided handler table. If a matching handler is found, it is + * invoked with the target object (Server or Location), parsed values, + * source position (line/column), and contextual snippet. If no handler + * exists for the directive name, a @ref SyntaxError is thrown. + * + * @tparam T The type of the configuration target (e.g., `Server` or `Location`). + * @tparam HandlerMap A mapping from directive name (`std::string`) to a handler + * function/lambda capable of applying the directive. + * + * @param target The configuration object being populated (Server/Location). + * @param token The token representing the directive keyword. + * @param values List of string arguments parsed for the directive. + * @param handlers Map of directive names to handler functions. + * @param line Line number in the source config where the directive occurs. + * @param column Column number in the source config where the directive occurs. + * @param ctx A snippet of the source line for contextual error messages. + * + * @throws SyntaxError If the directive name is unknown or unhandled. + * + * @ingroup config_parser + */ template void parseDirective(T& target, const Token& token, std::vector& values, const HandlerMap& handlers, int line, int column, const std::string& ctx) { const std::string& directiveName = token.value; + // Look up the directive in the handler table auto handlerIt = handlers.find(directiveName); if (handlerIt == handlers.end()) { throw SyntaxError(formatError("Unknown directive: '" + token.value + "'", line, column), ctx); } + // Dispatch to the registered handler function auto& handler = handlerIt->second; handler(target, values, line, column, ctx); } } // namespace -////////////////// -// --- Public API - -ConfigParser::ConfigParser(std::string source) - : // Initialize tokenizer with the raw config string - _tokenizer(std::move(source)) { - // Immediately tokenize the input and store the token list +//=== Public API ========================================================== + +/** + * @brief Constructs a ConfigParser instance from raw configuration text. + * + * @details Initializes the internal tokenizer with the provided configuration + * source string. Immediately tokenizes the input and stores the resulting + * token sequence for parsing. This makes subsequent parsing methods + * (`parseConfig`, `parseServer`, etc.) operate directly on a prebuilt + * token stream rather than raw text. + * + * @param source Raw configuration string to be parsed. + * + * @ingroup config_parser + */ +ConfigParser::ConfigParser(std::string source) : _tokenizer(std::move(source)) { _tokens = _tokenizer.tokenize(); } +/** + * @brief Parses the full configuration input into a Config object. + * + * @details Expects the top-level configuration to consist of one or more + * `server` blocks. Ensures that the input is non-empty, validates + * the presence of `server` keywords, and parses each block in turn. + * If invalid structure or unexpected tokens are encountered, a + * `SyntaxError` is thrown with contextual information. + * + * @return A fully constructed Config object containing all parsed servers. + * + * @throws SyntaxError If the configuration is empty, missing `server` blocks, + * or contains unexpected tokens after a block. + * + * @ingroup config_parser + */ Config ConfigParser::parseConfig() { Config config; - // Reject completely empty configuration input + // Reject completely empty configuration if (isAtEnd()) { throw SyntaxError(formatError("Empty configuration", 1, 1), getLineSnippet()); } - // Top-level config must consist of one or more `server` blocks + // Parse one or more server blocks while (!isAtEnd()) { - // If the current token is not `server`, fail early + // Each top-level block must start with the 'server' keyword if (current().type != TokenType::KEYWORD_SERVER) { throw SyntaxError( formatError("Expected 'server' block", current().line, current().column), getLineSnippet()); } - // Parse a full server block and append it to the config + // Parse the full server block and add it to the Config config.addServer(parseServer()); - // After a server block, only EOF or another server block is allowed + // After a server block, only EOF or another 'server' is allowed if (!isAtEnd() && current().type != TokenType::KEYWORD_SERVER && current().type != TokenType::END_OF_FILE) { throw SyntaxError(formatError("Unexpected token after server block", current().line, @@ -103,57 +203,109 @@ Config ConfigParser::parseConfig() { return config; } -//////////////////////////// -// --- Server Block Parsing - +//=== Server Block Parsing =============================================== + +/** + * @brief Parses a single `server` block into a Server object. + * + * @details Ensures the block starts with the `server` keyword and opening brace, + * then iterates through its directives and nested `location` blocks. + * Each directive is validated against duplicates (unless repeatable) + * and dispatched to the appropriate handler. + * + * @return A fully populated Server object representing the parsed block. + * + * @throws SyntaxError If the block is malformed, contains duplicate directives, + * or is missing the expected closing brace. + * + * @ingroup config_parser + */ Server ConfigParser::parseServer() { - expect(TokenType::KEYWORD_SERVER, "server block"); // Ensure block starts with 'server' - expect(TokenType::LBRACE, "start of server block"); // Expect opening brace '{' + // Validate required "server {" start sequence + expect(TokenType::KEYWORD_SERVER, "server block"); + expect(TokenType::LBRACE, "start of server block"); - Server server; - std::unordered_set seen; // Track directives to detect duplicates + Server server; // New Server instance + std::unordered_set seen; // Track seen directives (for duplicates) - // Loop until closing '}' or end of file + // Loop until closing brace or EOF while (!isAtEnd() && current().type != TokenType::RBRACE) { if (current().type == TokenType::KEYWORD_LOCATION) { - // Parse and attach a location block to the server + // Nested location block -> parse recursively server.addLocation(parseLocation()); } else { - // Normalize directive name to lowercase and check for duplicates + // Check for duplicate directives unless repeatable const std::string name = current().value; if (!checkDuplicateDirective(name, seen, kRepeatableServerDirectives)) { throw SyntaxError(formatError("Duplicate directive: '" + name + "'", current().line, current().column), getLineSnippet()); } - // Parse the directive and apply it to the server + // Parse directive and apply it to the server parseServerDirective(server); } } - expect(TokenType::RBRACE, "end of server block"); // Validate closing '}' + // Ensure proper block termination + expect(TokenType::RBRACE, "end of server block"); return server; } +/** + * @brief Parses and applies a single directive inside a `server` block. + * + * @details Reads the current directive token (e.g., `listen`, `host`, etc.), + * collects its arguments, enforces a terminating semicolon, and then + * dispatches the directive to the appropriate handler function from + * the server handler table. + * + * @param server Reference to the Server object being populated. + * + * @throws SyntaxError If the directive is malformed, missing arguments, or + * missing its terminating semicolon. + * + * @ingroup config_parser + */ void ConfigParser::parseServerDirective(Server& server) { - Token key = current(); // Capture the directive token (e.g., "listen", "host", etc.) - advance(); // Consume the directive name - std::vector values = collectArgs(kArgTokenTypes); // Parse directive arguments - expect(TokenType::SEMICOLON, "semicolon after server directive"); // Enforce `;` terminator - parseDirective(server, key, values, - directive::serverHandlers(), // Dispatch to the correct handler - key.line, key.column, getLineSnippet()); -} + Token key = current(); // Capture the directive keyword token + advance(); // Consume the keyword (move to its arguments) + + // Collect directive arguments (strings, numbers, or identifiers) + std::vector values = collectArgs(kArgTokenTypes); -////////////////////////////// -// --- Location Block Parsing + // Ensure directive ends with a semicolon + expect(TokenType::SEMICOLON, "semicolon after server directive"); + // Dispatch directive to its handler, applying it to the server + parseDirective(server, key, values, directive::serverHandlers(), key.line, key.column, + getLineSnippet()); +} + +//=== Location Block Parsing ============================================= + +/** + * @brief Parses a single `location` block inside a `server` block. + * + * @details Ensures the block starts with the `location` keyword, validates and + * sets the location path, then iterates over its contained directives. + * Each directive is validated for duplicates (unless repeatable) and + * dispatched to the appropriate handler. The block must end with a + * closing brace `}`. + * + * @return A fully constructed `Location` object populated with directives. + * + * @throws SyntaxError If the block is malformed, the path is missing or invalid, + * contains duplicate directives, or is missing required + * braces. + * + * @ingroup config_parser + */ Location ConfigParser::parseLocation() { - expect(TokenType::KEYWORD_LOCATION, "location block"); // Ensure block starts with 'location' + expect(TokenType::KEYWORD_LOCATION, "location block"); // Must start with `location` Location location; - // Validate path token exists and is of correct type + // Validate that the next token is a valid path (string or identifier) TokenType type = current().type; if (type != TokenType::STRING && type != TokenType::IDENTIFIER) { throw SyntaxError(formatError("Expected location path after 'location', but got '" + @@ -162,6 +314,7 @@ Location ConfigParser::parseLocation() { getLineSnippet()); } + // Extract and validate the path token Token pathTok = current(); if (pathTok.value.empty() || pathTok.value[0] != '/') { throw SyntaxError( @@ -170,13 +323,13 @@ Location ConfigParser::parseLocation() { getLineSnippet()); } location.setPath(pathTok.value); - advance(); // consume the path token + advance(); // Consume the path token - expect(TokenType::LBRACE, "start of location block"); // Expect opening brace '{' + expect(TokenType::LBRACE, "start of location block"); // Require opening `{` std::unordered_set seen; // Track encountered directives to catch duplicates - // Parse directives until closing brace + // Parse directives until closing brace `}` while (!isAtEnd() && current().type != TokenType::RBRACE) { const std::string name = current().value; if (!checkDuplicateDirective(name, seen, kRepeatableLocationDirectives)) { @@ -184,46 +337,85 @@ Location ConfigParser::parseLocation() { current().column), getLineSnippet()); } - // Parse and apply a directive to the Location object - parseLocationDirective(location); + parseLocationDirective(location); // Apply directive to the location } - expect(TokenType::RBRACE, "end of location block"); // Expect closing brace '}' + expect(TokenType::RBRACE, "end of location block"); // Ensure closing `}` return location; } +/** + * @brief Parses a single directive inside a `location` block. + * + * @details Reads the directive keyword, collects its arguments, enforces that + * it is terminated by a semicolon `;`, and dispatches it to the + * appropriate handler function via the directive table. + * + * @param location Reference to the `Location` object where parsed directive + * data will be applied. + * + * @throws SyntaxError If arguments are invalid, missing, or the directive + * is not followed by a semicolon. + * + * @ingroup config_parser + */ void ConfigParser::parseLocationDirective(Location& location) { - Token key = current(); // Capture the directive token (e.g., "root", "index", etc.) + Token key = current(); // Capture directive keyword token (e.g., "root", "index", etc.) advance(); // Consume the directive keyword - std::vector values = collectArgs(kArgTokenTypes); // Parse directive arguments - expect(TokenType::SEMICOLON, "semicolon after location directive"); // Ensure `;` terminator - parseDirective(location, key, values, // Dispatch to handler - directive::locationHandlers(), key.line, key.column, getLineSnippet()); + + // Collect arguments that follow the directive + std::vector values = collectArgs(kArgTokenTypes); + + // Ensure the directive is properly terminated by a semicolon + expect(TokenType::SEMICOLON, "semicolon after location directive"); + + // Dispatch the directive to the correct handler function from the table + parseDirective(location, key, values, directive::locationHandlers(), key.line, key.column, + getLineSnippet()); } -//////////////////////// -// --- Token Navigation +//=== Token Navigation ==================================================== +/** + * @brief Returns the current token (bounds-checked). + * @ingroup config_parsing + */ const Token& ConfigParser::current() const { - // Return the token at the current parsing position return _tokens.at(_pos); } +/** + * @brief Returns the previous token or the first token if at start. + * @ingroup config_parsing + */ const Token& ConfigParser::previous() const { if (_pos == 0) { - return _tokens.at(0); // fallback to first token + return _tokens.at(0); } return _tokens.at(_pos - 1); } +/** + * @brief Peeks a token at an offset ≥ 1 from current. + * + * @param offset Relative index (defaults to 1). + * @return Token at index or EOF token if out of range. + * @ingroup config_parsing + */ const Token& ConfigParser::peek(std::size_t offset) const { std::size_t index = _pos + offset; if (index < _tokens.size()) { return _tokens.at(index); } - return _tokens.back(); // Fallback to EOF token + return _tokens.back(); } +/** + * @brief Looks behind by a given offset, or returns an EOF dummy. + * + * @param offset Distance to look back (defaults to 1). + * @ingroup config_parsing + */ const Token& ConfigParser::lookBehind(std::size_t offset) const { static Token dummy(TokenType::END_OF_FILE, "", 0, 0, 0); @@ -233,104 +425,194 @@ const Token& ConfigParser::lookBehind(std::size_t offset) const { return dummy; } +/** + * @brief Advances to the next token and returns it. + * + * @return New current token after increment (or EOF if at end). + * @ingroup config_parsing + */ const Token& ConfigParser::advance() { if (!isAtEnd()) - ++_pos; // Move to the next token if not already at the end - return current(); // Return the new current token + ++_pos; + return current(); } +/** + * @brief Checks if the parser has consumed all tokens or reached EOF. + * + * @return `true` if no more tokens are available. + * @ingroup config_parsing + */ bool ConfigParser::isAtEnd() const { - // True if we've reached the end of the token stream or explicitly hit the EOF token return _pos >= _tokens.size() || _tokens[_pos].type == TokenType::END_OF_FILE; } +/** + * @brief Consumes the current token if its type matches. + * + * @param type Expected token type. + * @return `true` if matched/consumed. + * @ingroup config_parsing + */ bool ConfigParser::match(TokenType type) { - // If at end or current token doesn't match the expected type, do nothing if (isAtEnd() || _tokens[_pos].type != type) return false; - ++_pos; // Consume the token if it matches - return true; // Indicate successful match and consumption + ++_pos; + return true; } +/** + * @brief Ensures that the current token matches a specific expected type. + * + * @details This method validates that the parser is currently positioned + * at a token of type @p expected. If the token matches, it is consumed + * (the parser advances by one). If not, an `UnexpectedToken` error is thrown + * with details about what was expected versus what was found. + * + * @param expected The exact token type required at this position. + * @param context Human-readable context for error reporting + * (unused here but provided for consistency). + * + * @throws UnexpectedToken If the current token does not match @p expected. + * + * @ingroup config_parser + */ void ConfigParser::expect(TokenType expected, const std::string& context) { - (void) context; + (void) context; // Currently unused, but kept for interface consistency + // If at end or wrong type → prepare error if (isAtEnd() || _tokens[_pos].type != expected) { const Token* actual; if (isAtEnd()) { - actual = &_tokens.back(); // fallback to last token (should be EOF) + actual = &_tokens.back(); // fallback to EOF token } else { - actual = &_tokens[_pos]; + actual = &_tokens[_pos]; // offending token } + // Build and throw detailed error throw UnexpectedToken(formatError("Expected " + debugTokenType(expected) + ", but got " + debugTokenType(actual->type), actual->line, actual->column), getLineSnippet()); } - ++_pos; // Consume the token if it matched + ++_pos; // Consume token if matched } +/** + * @brief Ensures the current token matches one of several expected types. + * + * @details This method is used when a directive or grammar rule allows + * multiple token types in the same position (e.g., `STRING` or `IDENTIFIER`). + * It checks if the current token matches any type in @p types: + * + * - If a match is found, the token is consumed via `advance()` and returned. + * - If none match, an `UnexpectedToken` exception is thrown with a + * descriptive error message listing all expected types and the actual one. + * + * @param types List of acceptable token types (initializer list). + * @param context Human-readable context for error reporting. + * + * @return The matched token, already consumed from the stream. + * + * @throws UnexpectedToken If the current token does not match any expected type. + * + * @ingroup config_parser + */ Token ConfigParser::expectOneOf(std::initializer_list types, const std::string& context) { - TokenType actual = current().type; + TokenType actual = current().type; // Current token type - // Check if the current token matches any of the expected types + // Try matching against the expected set for (TokenType expected : types) { if (actual == expected) { - return advance(); // If matched, consume and return it + return advance(); // Consume & return if found } } - // If no match, build an error message listing all expected types + // Build detailed error message if no match std::ostringstream msg; msg << "Expected "; for (auto it = types.begin(); it != types.end(); ++it) { if (it != types.begin()) { - msg << " or "; + msg << " or "; // Separate multiple expected types } - msg << debugTokenType(*it); + msg << debugTokenType(*it); // Append readable token type } msg << " for " << context << ", but got " << debugTokenType(actual); - // Throw a detailed syntax error with contextual highlighting + // Throw with contextual line snippet throw UnexpectedToken(formatError(msg.str(), current().line, current().column), getLineSnippet()); } +/** + * @brief Collects a list of argument values following a directive. + * + * @details This function consumes tokens that belong to the directive's argument list. + * It accepts only tokens of specific types (`STRING`, `NUMBER`, `IDENTIFIER`), + * provided in the @p validTypes span. Arguments may be separated by commas. + * Example: + * ``` + * root /var/www; + * methods GET, POST, DELETE; + * ``` + * + * - Stops parsing when a token does not match a valid argument type. + * - Supports comma-separated lists of arguments. + * - Throws if a comma is not followed by a valid argument. + * + * @param validTypes A span of allowed token types for directive arguments. + * + * @return A vector of string values representing the collected arguments. + * + * @throws SyntaxError If a comma is not followed by a valid argument type. + * + * @ingroup config_parser + */ std::vector ConfigParser::collectArgs(std::span validTypes) { - std::vector values; + std::vector values; // Accumulate argument values here + // Continue as long as we are not at end-of-file while (!isAtEnd()) { TokenType t = current().type; + // If current token type is not among the allowed ones, stop collecting if (std::find(validTypes.begin(), validTypes.end(), t) == validTypes.end()) break; + // Append the argument value (string/number/identifier) values.push_back(current().value); - advance(); + advance(); // Move to the next token + // Handle optional comma-separated arguments while (match(TokenType::COMMA)) { + // A comma must be followed by another valid argument if (isAtEnd() || std::find(validTypes.begin(), validTypes.end(), current().type) == validTypes.end()) { throw SyntaxError( formatError("Expected value after comma", current().line, current().column), getLineSnippet()); } + + // Append the value after the comma values.push_back(current().value); - advance(); + advance(); // Consume argument token } } - return values; + return values; // Return the collected argument list } -///////////////////// -// --- Error Context +//=== Error Context ======================================================= +/** + * @brief Extracts the full source line of the last consumed token. + * + * @return Source line for diagnostic display. + * @ingroup config_parsing + */ std::string ConfigParser::getLineSnippet() const { - // Extract the full line of source text where the previous token is located return _tokenizer.extractLine(previous().offset); } diff --git a/src/config/parser/directive_handler_table.cpp b/src/config/parser/directive_handler_table.cpp index eb8696c7..e44e0403 100644 --- a/src/config/parser/directive_handler_table.cpp +++ b/src/config/parser/directive_handler_table.cpp @@ -3,13 +3,32 @@ /* ::: :::::::: */ /* directive_handler_table.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/08 17:14:27 by nlouis #+# #+# */ -/* Updated: 2025/06/14 10:27:30 by irychkov ### ########.fr */ +/* Updated: 2025/08/18 12:47:33 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file directive_handler_table.cpp + * @brief Implements directive handler maps for `server` and `location` blocks. + * + * @details Defines the concrete handler tables returned by + * `directive::serverHandlers()` and `directive::locationHandlers()`. + * Each handler validates its arguments, reports precise diagnostics + * (line/column + contextual snippet), and mutates the target + * configuration object (@ref Server or @ref Location). + * + * This file also contains small internal helpers used by handlers: + * - `resolveToAbsolute()` — canonicalizes filesystem paths. + * - `requireArgCount()` / `requireMinArgCount()` — arity checks. + * - `validateIPv4Address()` — IPv4 format validation (with `localhost` allowed). + * - `validateCgiExtension()` — enforces dot-prefixed alnum extensions. + * + * @ingroup config_parsing + */ + #include "config/parser/directive_handler_table.hpp" #include "config/parser/ConfigParseError.hpp" #include "core/Location.hpp" @@ -24,15 +43,55 @@ namespace directive { +/** + * @brief Resolves a filesystem path to its absolute, normalized form. + * + * @details Converts the given raw path string into an absolute path + * using `std::filesystem::absolute()`, then applies lexical + * normalization (removing redundant `.` and `..` components + * and duplicate separators). This ensures the result is a + * canonicalized path string safe for use in configuration + * directives (e.g., `root`, `upload_store`). + * + * @param rawPath The raw input path (may be relative or contain `.`/`..`). + * @return A string representing the absolute, lexically normalized path. + * + * @throw std::filesystem::filesystem_error If the path cannot be resolved + * due to invalid characters or inaccessible filesystem. + * + * @ingroup config_parser + */ static std::string resolveToAbsolute(const std::string& rawPath) { std::filesystem::path abs = std::filesystem::absolute(rawPath); return abs.lexically_normal().string(); } +/** + * @brief Ensures that a directive has exactly the expected number of arguments. + * + * @details Used during configuration parsing to validate that a directive + * (e.g., `listen`, `root`, `client_max_body_size`) is provided with + * the correct number of arguments. If the count does not match + * the expectation, a `SyntaxError` is thrown with contextual + * information about the directive, line, column, and source snippet. + * + * @param args The list of argument strings parsed for the directive. + * @param expected The exact number of arguments required. + * @param directive The name of the directive being validated. + * @param line The line number in the config file where the directive appears. + * @param column The column number in the config file where the directive appears. + * @param ctx A snippet of the surrounding source text for error reporting. + * + * @throw SyntaxError If the directive does not have exactly `expected` arguments. + * + * @ingroup config_parser + */ static void requireArgCount(const std::vector& args, std::size_t expected, const std::string& directive, int line, int column, const std::string& ctx) { + // Check if the argument count matches the expected number if (args.size() != expected) { + // If not, throw a SyntaxError with detailed context throw SyntaxError(formatError("Directive '" + directive + "' takes exactly " + std::to_string(expected) + " argument(s), but got " + std::to_string(args.size()), @@ -41,10 +100,32 @@ static void requireArgCount(const std::vector& args, std::size_t ex } } +/** + * @brief Ensures that a directive has at least the required number of arguments. + * + * @details Used during configuration parsing to validate directives that + * accept a variable number of arguments (e.g., `server_name`, + * `error_page`, `index`). If the number of arguments is below + * the minimum threshold, a `SyntaxError` is thrown with detailed + * information about the directive, its location, and the input context. + * + * @param args The list of arguments parsed for the directive. + * @param min The minimum number of arguments required. + * @param directive The name of the directive being validated. + * @param line The line number in the config file where the directive appears. + * @param column The column number in the config file where the directive appears. + * @param ctx A snippet of the surrounding config source text for context. + * + * @throw SyntaxError If the directive has fewer than `min` arguments. + * + * @ingroup config_parser + */ static void requireMinArgCount(const std::vector& args, std::size_t min, const std::string& directive, int line, int column, const std::string& ctx) { + // Check if number of parsed arguments is below the minimum if (args.size() < min) { + // Throw a SyntaxError with detailed error message and context throw SyntaxError(formatError("Directive '" + directive + "' requires at least " + std::to_string(min) + " argument(s), but got " + std::to_string(args.size()), @@ -53,29 +134,58 @@ static void requireMinArgCount(const std::vector& args, std::size_t } } +/** + * @brief Validates that a given string is a valid IPv4 address. + * + * @details This function ensures that the provided `ip` string is either + * `"localhost"` (special case allowed) or a well-formed IPv4 address + * in dotted-decimal notation (e.g., `127.0.0.1`, `192.168.1.42`). + * It splits the input into octets, parses each into an integer, and + * validates that: + * - There are exactly 4 octets. + * - Each octet is between 0 and 255. + * - No extra segments exist. + * + * @param ip The IP address string to validate. + * @param line Line number in the config file where the directive appears. + * @param column Column number in the config file where the directive appears. + * @param context_provider Lazy-evaluated function returning a source context snippet. + * + * @throw SyntaxError If the IP address is malformed (wrong number of octets, + * non-numeric values, or values outside 0–255). + * + * @ingroup config_parser + */ static void validateIPv4Address(const std::string& ip, int line, int column, const std::function& context_provider) { std::istringstream iss(ip); std::string segment; int count = 0; + // Special case: allow "localhost" without further validation if (ip == "localhost") return; + // Split the string by '.' and validate each segment while (std::getline(iss, segment, '.')) { if (++count > 4) { + // More than 4 parts → invalid IPv4 throw SyntaxError(formatError("Too many octets in IP address: " + ip, line, column), context_provider()); } + // Convert segment to integer, throws if non-numeric int octet = parseInt(segment, "host", line, column, context_provider); + // Check range validity (0–255) if (octet < 0 || octet > 255) { throw SyntaxError( formatError("Invalid IP octet '" + segment + "' in host: " + ip, line, column), context_provider()); } } + + // Must have exactly 4 octets if (count != 4) { throw SyntaxError( formatError("Invalid IP address format (expected 4 octets): " + ip, line, column), @@ -83,13 +193,45 @@ static void validateIPv4Address(const std::string& ip, int line, int column, } } +/** + * @brief Validates the syntax of a CGI file extension. + * + * @details This function ensures that a given CGI extension string is valid. + * A valid extension must: + * - Be non-empty. + * - Begin with a dot (`.`). + * - Contain only alphanumeric characters after the dot. + * - Not be just `"."` with no following characters. + * + * Examples of valid extensions: + * - `.php` + * - `.py` + * - `.cgi` + * + * Examples of invalid extensions: + * - `""` (empty string) + * - `"."` (dot only, no extension name) + * - `"php"` (missing leading dot) + * - `.sh!` (contains invalid character `!`) + * + * @param ext The extension string to validate (e.g., ".php"). + * @param line Line number in the config file for error reporting. + * @param column Column number in the config file for error reporting. + * @param context_provider Lazy-evaluated function returning a source context snippet. + * + * @throw SyntaxError If the extension is malformed or contains invalid characters. + * + * @ingroup config_parser + */ static void validateCgiExtension(const std::string& ext, int line, int column, const std::function& context_provider) { + // Must not be empty, a lone '.', and must start with a dot if (ext.empty() || ext == "." || ext[0] != '.') { throw SyntaxError(formatError("Invalid CGI extension: '" + ext + "'", line, column), context_provider()); } + // Check that all characters after '.' are alphanumeric for (std::size_t i = 1; i < ext.size(); ++i) { char c = ext[i]; if (!std::isalnum(c)) { @@ -100,70 +242,157 @@ static void validateCgiExtension(const std::string& ext, int line, int column, } } +/** + * @brief Returns the mapping of server-level directives to their handler functions. + * + * @details + * This function builds (once, statically) an `unordered_map` that associates each + * directive keyword valid inside a `server { ... }` block with a corresponding + * validation and handler lambda. Each handler is responsible for: + * - Checking argument count and validity. + * - Validating the argument format (IP, port, byte sizes, etc.). + * - Updating the `Server` object with the parsed value. + * + * Supported directives: + * - **listen ;** + * Validates port number (0–65535) and stores it in the server object. + * - **host ;** + * Validates IPv4 address (or "localhost") and assigns it to the server. + * - **server_name ;** + * Accepts one or more hostnames; each is registered as a valid server name. + * - **client_max_body_size ;** + * Validates a byte-size expression (e.g., "10M") and sets the upload limit. + * - **error_page ;** + * Accepts one or more HTTP error codes followed by a URI. + * Maps each error code to the provided URI. + * + * @return A reference to the immutable directive-to-handler map. + * + * @ingroup config_parser + */ const std::unordered_map& serverHandlers() { + // Static so the map is created once and reused on subsequent calls. static const std::unordered_map map = { + // ─────────────────────────────── + // `listen ;` + // Validates that the port number is within range [0, 65535]. {"listen", [](Server& s, const auto& v, int line, int column, const std::string& ctx) { - requireArgCount(v, 1, "listen", line, column, ctx); + requireArgCount(v, 1, "listen", line, column, ctx); // must have exactly one argument int port = parseInt(v[0], "listen", line, column, [&]() { return ctx; }); if (port < 0 || port > 65535) { throw SyntaxError( formatError("Port number out of valid range (0-65535): " + v[0], line, column), ctx); } - s.setPort(port); + s.setPort(port); // assign validated port to server }}, + + // ─────────────────────────────── + // `host ;` + // Validates IPv4 address (or "localhost"). {"host", [](Server& s, const auto& v, int line, int column, const std::string& ctx) { - requireArgCount(v, 1, "host", line, column, ctx); + requireArgCount(v, 1, "host", line, column, ctx); // must have exactly one argument const std::string& ip = v[0]; - validateIPv4Address(ip, line, column, [&]() { return ctx; }); - s.setHost(ip); + validateIPv4Address(ip, line, column, [&]() { return ctx; }); // ensure valid format + s.setHost(ip); // store host in server }}, + + // ─────────────────────────────── + // `server_name ;` + // Allows one or more server names (aliases). {"server_name", [](Server& s, const auto& v, int line, int column, const std::string& ctx) { - requireMinArgCount(v, 1, "server_name", line, column, ctx); + requireMinArgCount(v, 1, "server_name", line, column, ctx); // at least one name for (const auto& name : v) { - s.addServerName(name); + s.addServerName(name); // register each alias } }}, + // ─────────────────────────────── + // `client_max_body_size ;` + // Validates a byte-size string (e.g., "10M") and sets upload limit. {"client_max_body_size", [](Server& s, const auto& v, int line, int column, const std::string& ctx) { - requireArgCount(v, 1, "client_max_body_size", line, column, ctx); - s.setClientMaxBodySize( - parseByteSize(v[0], "client_max_body_size", line, column, [&]() { return ctx; })); + requireArgCount(v, 1, "client_max_body_size", line, column, ctx); // one argument + s.setClientMaxBodySize(parseByteSize(v[0], "client_max_body_size", line, column, + [&]() { return ctx; })); // convert and set }}, + + // ─────────────────────────────── + // `error_page ;` + // Maps one or more error codes to a redirect URI. {"error_page", [](Server& s, const auto& v, int line, int column, const std::string& ctx) { - requireMinArgCount(v, 2, "error_page", line, column, ctx); - std::string uri = v.back(); + requireMinArgCount(v, 2, "error_page", line, column, ctx); // at least one code + URI + std::string uri = v.back(); // last argument is the URI for (std::size_t i = 0; i + 1 < v.size(); ++i) { int code = parseInt(v[i], "error_page", line, column, [&]() { return ctx; }); - s.setErrorPage(code, uri); + s.setErrorPage(code, uri); // map each code to the URI } }}, }; + return map; } +/** + * @brief Returns the mapping of location-level directives to their handler functions. + * + * @details + * This function builds (once, statically) an `unordered_map` that associates each + * directive keyword valid inside a `location { ... }` block with its validation + * and handler lambda. Each handler checks argument counts, validates syntax, + * and updates the `Location` object accordingly. + * + * Supported directives: + * - **root ;** + * Sets the filesystem root for this location (absolute, normalized path). + * - **index ;** + * Specifies one or more default index files (supports comma-separated values). + * - **autoindex on|off;** + * Enables or disables directory listing. + * - **methods ;** + * Restricts allowed HTTP methods (GET, POST, DELETE). + * - **upload_store ;** + * Defines the directory for file uploads (absolute, normalized path). + * - **cgi_extension <.ext...>;** + * Lists extensions handled by CGI (e.g., `.php`, `.py`). + * - **cgi_interpreter <.ext> ;** + * Associates a CGI extension with an executable interpreter. + * - **return ;** + * Configures a redirect with HTTP status code. + * + * @return A reference to the immutable directive-to-handler map. + * + * @ingroup config_parser + */ const std::unordered_map& locationHandlers() { static const std::unordered_map map = { + // ─────────────────────────────── + // `root ;` + // Sets the root directory for this location. {"root", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { - requireArgCount(v, 1, "root", line, column, ctx); - loc.setRoot(resolveToAbsolute(v[0])); + requireArgCount(v, 1, "root", line, column, ctx); // must have exactly one path + loc.setRoot(resolveToAbsolute(v[0])); // normalize to absolute path }}, + + // ─────────────────────────────── + // `index ;` + // Defines one or more index files (comma-separated allowed). {"index", [](Location& loc, const auto& args, int line, int column, const std::string& ctx) { - requireMinArgCount(args, 1, "index", line, column, ctx); + requireMinArgCount(args, 1, "index", line, column, ctx); // at least one file for (const std::string& raw : args) { size_t start = 0, end; + // Split on commas while ((end = raw.find(',', start)) != std::string::npos) { std::string idx = raw.substr(start, end - start); if (!idx.empty()) - loc.addIndexFile(idx); + loc.addIndexFile(idx); // add each file start = end + 1; } std::string idx = raw.substr(start); @@ -171,6 +400,10 @@ const std::unordered_map& locationHandlers() { loc.addIndexFile(idx); } }}, + + // ─────────────────────────────── + // `autoindex on|off;` + // Enables or disables directory listing. {"autoindex", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { requireArgCount(v, 1, "autoindex", line, column, ctx); @@ -183,6 +416,10 @@ const std::unordered_map& locationHandlers() { formatError("Invalid value for 'autoindex': " + v[0], line, column), ctx); } }}, + + // ─────────────────────────────── + // `methods ;` + // Restricts allowed HTTP methods (must be GET, POST, or DELETE). {"methods", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { if (v.empty()) { @@ -196,19 +433,28 @@ const std::unordered_map& locationHandlers() { if (!valid_methods.count(m)) { throw SyntaxError(formatError("Invalid HTTP method: " + m, line, column), ctx); } - loc.addMethod(m); + loc.addMethod(m); // store validated method } }}, + + // ─────────────────────────────── + // `upload_store ;` + // Sets the directory where file uploads will be saved. {"upload_store", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { requireArgCount(v, 1, "upload_store", line, column, ctx); - loc.setUploadStore(resolveToAbsolute(v[0])); + loc.setUploadStore(resolveToAbsolute(v[0])); // normalize path }}, + + // ─────────────────────────────── + // `cgi_extension <.ext...>;` + // Declares CGI file extensions (e.g., `.php`). {"cgi_extension", [](Location& loc, const auto& args, int line, int column, const std::string& ctx) { requireMinArgCount(args, 1, "cgi_extension", line, column, ctx); for (const std::string& raw : args) { size_t start = 0, end; + // Split multiple extensions separated by commas while ((end = raw.find(',', start)) != std::string::npos) { std::string ext = raw.substr(start, end - start); if (!ext.empty()) { @@ -224,12 +470,17 @@ const std::unordered_map& locationHandlers() { } } }}, + + // ─────────────────────────────── + // `cgi_interpreter <.ext> ;` + // Maps a CGI extension to its interpreter binary. {"cgi_interpreter", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { - requireArgCount(v, 2, "cgi_interpreter", line, column, ctx); + requireArgCount(v, 2, "cgi_interpreter", line, column, ctx); // ext + path required const std::string& ext = v[0]; const std::string& path = v[1]; - validateCgiExtension(ext, line, column, [&]() { return ctx; }); + + validateCgiExtension(ext, line, column, [&]() { return ctx; }); // must be valid ext if (!std::filesystem::is_regular_file(path) || access(path.c_str(), X_OK) != 0) { throw SyntaxError( formatError("Interpreter not executable or not found: " + path, line, column), @@ -241,15 +492,20 @@ const std::unordered_map& locationHandlers() { column), ctx); } - loc.addCgiInterpreter(ext, path); + loc.addCgiInterpreter(ext, path); // store extension → interpreter mapping }}, + + // ─────────────────────────────── + // `return ;` + // Defines a redirect (e.g., `return 301 /newpath;`). {"return", [](Location& loc, const auto& v, int line, int column, const std::string& ctx) { requireArgCount(v, 2, "return", line, column, ctx); int code = parseInt(v[0], "return", line, column, [&]() { return ctx; }); - loc.setRedirect(v[1], code); + loc.setRedirect(v[1], code); // assign redirect URI + status code }}, }; + return map; } diff --git a/src/config/tokenizer/Tokenizer.cpp b/src/config/tokenizer/Tokenizer.cpp index 79e7e131..c6c72194 100644 --- a/src/config/tokenizer/Tokenizer.cpp +++ b/src/config/tokenizer/Tokenizer.cpp @@ -6,10 +6,22 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 01:06:09 by nlouis #+# #+# */ -/* Updated: 2025/05/23 10:14:39 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 12:32:37 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file Tokenizer.cpp + * @brief Implements the Tokenizer used for config lexical analysis. + * + * @details Defines all methods of the Tokenizer, including cursor management, + * classification helpers, high-level parsers (identifiers, numbers, + * strings), comment/whitespace skipping, token dispatch, and token + * creation with accurate source locations. + * + * @ingroup config_tokenizing + */ + #include "config/tokenizer/Tokenizer.hpp" #include "config/parser/ConfigParseError.hpp" #include "utils/errorUtils.hpp" @@ -21,15 +33,32 @@ #include #include -//////////////////////////////// -// --- Constructors - +//=== Construction & Special Members ===================================== + +/** + * @brief Constructs a Tokenizer over an input string. + * + * @details Initializes cursor position and source coordinates to the start of + * the input buffer. + * + * @param input Raw configuration text to tokenize. Ownership is moved in. + * @ingroup config_tokenizing + */ Tokenizer::Tokenizer(std::string input) : _input(std::move(input)), _pos(0), _line(1), _column(1) { } -//////////////// -// --- Main API - +//=== Main API ============================================================ + +/** + * @brief Converts the input buffer into a sequence of tokens. + * + * @details Skips BOM, whitespace, and comments; emits recognized tokens and a + * final END_OF_FILE token. Pre-reserves capacity for fewer reallocs. + * + * @return Vector of tokens in source order (always ends with END_OF_FILE). + * @throws TokenizerError On malformed constructs (e.g., bad strings/suffixes). + * @ingroup config_tokenizing + */ std::vector Tokenizer::tokenize() { skipUtf8BOM(); // Skip UTF-8 BOM if present at the beginning @@ -47,14 +76,26 @@ std::vector Tokenizer::tokenize() { return _tokens; } -///////////////////////// -// --- Core Cursor Logic +//=== Core Cursor Logic =================================================== +/** + * @brief Tests if the cursor reached the end of the input. + * + * @return `true` when no more characters are available. + * @ingroup config_tokenizing + */ bool Tokenizer::isAtEnd() const noexcept { // True if the cursor has reached or passed end of input return _pos >= _input.size(); } +/** + * @brief Consumes a specific character if present. + * + * @param expected Character to match. + * @return `true` if consumed; `false` otherwise. + * @ingroup config_tokenizing + */ bool Tokenizer::match(char expected) noexcept { // Return false if at end or current char does not match if (isAtEnd() || _input[_pos] != expected) { @@ -66,11 +107,23 @@ bool Tokenizer::match(char expected) noexcept { return true; } +/** + * @brief Peeks the current character without consuming it. + * + * @return Current character (undefined if at end). + * @ingroup config_tokenizing + */ unsigned char Tokenizer::peek() const noexcept { // Return current character without consuming it return static_cast(_input[_pos]); } +/** + * @brief Peeks the next character without consuming it. + * + * @return Next character, or `'\0'` at end. + * @ingroup config_tokenizing + */ unsigned char Tokenizer::peekNext() const noexcept { if (isAtEnd()) { return '\0'; @@ -78,6 +131,15 @@ unsigned char Tokenizer::peekNext() const noexcept { return static_cast(_input[_pos + 1]); } +/** + * @brief Advances by one character, updating line/column. + * + * @details Increments line and resets column on newline; otherwise increments + * column. + * + * @return The consumed character. + * @ingroup config_tokenizing + */ unsigned char Tokenizer::advance() noexcept { char c = _input[_pos++]; // Consume current character and move cursor forward if (c == '\n') { @@ -89,24 +151,41 @@ unsigned char Tokenizer::advance() noexcept { return c; // Return the consumed character } -//////////////////////////// -// --- Classification Logic +//=== Classification Logic =============================================== +/** + * @brief Checks if a byte is a valid identifier start. + * + * @param c Character to test. + * @return `true` if valid start. + * @ingroup config_tokenizing + */ inline bool Tokenizer::isIdentifierStart(unsigned char c) const { // Valid first char for identifiers return std::isalpha(c) || c == '_' || c == '/' || c == '.' || c == '-' || c == ':'; } +/** + * @brief Checks if a byte is a valid identifier continuation. + * + * @param c Character to test. + * @return `true` if valid continuation. + * @ingroup config_tokenizing + */ inline bool Tokenizer::isIdentifierChar(unsigned char c) const { // Valid body char for identifiers return std::isalnum(c) || c == '_' || c == '/' || c == '.' || c == '-' || c == ':'; } -////////////////////////// -// --- High-Level Parsers +//=== High-Level Parsers ================================================== -//////////////////////////////// -// --- Skip BOM +/** + * @brief Skips a UTF-8 Byte Order Mark at input start. + * + * @details If present, advances cursor past `0xEF 0xBB 0xBF`. + * + * @ingroup config_tokenizing + */ void Tokenizer::skipUtf8BOM() { static const std::string BOM = "\xEF\xBB\xBF"; // UTF-8 Byte Order Mark if (_input.compare(0, BOM.size(), BOM) == 0) { @@ -114,8 +193,14 @@ void Tokenizer::skipUtf8BOM() { } } -////////////////////////////////// -// --- Skip Whitespace & Comments +/** + * @brief Skips whitespace and hash (`#`) comments. + * + * @details Treats CR/LF/newlines and isspace() as whitespace. Hash comments run + * until end-of-line. + * + * @ingroup config_tokenizing + */ void Tokenizer::skipWhitespaceAndComments() { while (!isAtEnd()) { unsigned char c = peek(); @@ -133,9 +218,14 @@ void Tokenizer::skipWhitespaceAndComments() { } } } -/////////////////////////////////// -// ---Identifier & Keyword Parsing +/** + * @brief Resolves an identifier to a keyword token type when applicable. + * + * @param word Candidate identifier (case-insensitive). + * @return The keyword's TokenType or IDENTIFIER. + * @ingroup config_tokenizing + */ TokenType Tokenizer::resolveKeywordType(const std::string& word) { // Static map of all recognized configuration keywords (lowercase only) static const std::unordered_map keywords = { @@ -170,6 +260,13 @@ TokenType Tokenizer::resolveKeywordType(const std::string& word) { return TokenType::IDENTIFIER; } +/** + * @brief Scans an identifier body after a valid start. + * + * @details Consumes identifier characters until a non-identifier byte. + * + * @ingroup config_tokenizing + */ void Tokenizer::scanIdentifier() { // Assumes the current character is a valid identifier start (checked beforehand) while (!isAtEnd() && isIdentifierChar(peek())) { @@ -177,6 +274,13 @@ void Tokenizer::scanIdentifier() { } } +/** + * @brief Validates the last scanned identifier. + * + * @param start Byte offset where the identifier started. + * @throws TokenizerError If empty, contains '$', or non-printable chars. + * @ingroup config_tokenizing + */ void Tokenizer::validateIdentifier(std::size_t start) { // Reject empty identifiers if (_pos == start) { @@ -212,6 +316,12 @@ void Tokenizer::validateIdentifier(std::size_t start) { } } +/** + * @brief Parses an identifier and resolves to keyword when applicable. + * + * @return Token of type IDENTIFIER or specific KEYWORD_*. + * @ingroup config_tokenizing + */ Token Tokenizer::parseIdentifierOrKeyword() { std::size_t start = _pos; // Record start position of the identifier scanIdentifier(); // Consume all valid identifier characters @@ -221,8 +331,13 @@ Token Tokenizer::parseIdentifierOrKeyword() { return makeToken(type, word); } -///////////////////////////// -// --- Number & Unit Parsing +//=== Number & Unit Parsing ============================================== + +/** + * @brief Consumes consecutive decimal digits. + * + * @ingroup config_tokenizing + */ void Tokenizer::scanDigits() { // Consume characters as long as they are digits (0–9) while (!isAtEnd() && std::isdigit(peek())) { @@ -230,6 +345,15 @@ void Tokenizer::scanDigits() { } } +/** + * @brief Optionally consumes a single-letter size suffix. + * + * @details Accepts exactly one alphabetic suffix (e.g., k/m/g). Rejects + * multi-letter forms (e.g., "mb", "MiB"). + * + * @throws TokenizerError On multi-letter suffix usage. + * @ingroup config_tokenizing + */ void Tokenizer::scanOptionalUnitSuffix() { if (isAtEnd()) return; // Nothing to scan @@ -248,6 +372,12 @@ void Tokenizer::scanOptionalUnitSuffix() { } } +/** + * @brief Parses a NUMBER token with optional single-letter suffix. + * + * @return NUMBER token constructed from the scanned span. + * @ingroup config_tokenizing + */ Token Tokenizer::parseNumberOrUnit() { std::size_t start = _pos; // Record start of numeric token @@ -258,14 +388,31 @@ Token Tokenizer::parseNumberOrUnit() { return makeToken(TokenType::NUMBER, _input.substr(start, _pos - start)); } -////////////////////// -// --- String Parsing +//=== String Parsing ====================================================== + +/** + * @brief Throws a formatted unterminated string error. + * + * @param reason Short explanation (e.g., "unexpected newline"). + * @throws TokenizerError Always; provides formatted context line. + * @ingroup config_tokenizing + */ void Tokenizer::throwUnterminatedString(const std::string& reason) { // Throw a TokenizerError with a contextual reason and source line throw TokenizerError(formatError("Invalid string literal (" + reason + ")", _line, _column), extractLine(_pos)); } +/** + * @brief Parses a quoted string literal. + * + * @details Supports both single and double quotes. Enforces a 64 KiB maximum + * payload and forbids embedded newlines. + * + * @return STRING token containing the unescaped contents. + * @throws TokenizerError On newline, overflow, or EOF before closing quote. + * @ingroup config_tokenizing + */ Token Tokenizer::parseStringLiteral() { static const std::size_t MAX_STRING_LITERAL_LENGTH = 64 * 1024; // 64 KiB @@ -298,24 +445,39 @@ Token Tokenizer::parseStringLiteral() { return Token(TokenType::STRING, "", _line, _column, 0); // Unreachable } -//////////////////////////////////// -// --- Whitespace & Comment Helpers +//=== Whitespace & Comment Helpers ======================================= +/** + * @brief Skips a carriage return (`'\r'`). + * @ingroup config_tokenizing + */ void Tokenizer::skipCR() { ++_pos; } +/** + * @brief Skips a newline and advances line counter. + * @ingroup config_tokenizing + */ void Tokenizer::skipNewline() { ++_pos; ++_line; _column = 1; } +/** + * @brief Skips a single whitespace character (except newline). + * @ingroup config_tokenizing + */ void Tokenizer::skipOtherWhitespace() { ++_pos; ++_column; } +/** + * @brief Skips a `#`-style comment to the end of the line. + * @ingroup config_tokenizing + */ void Tokenizer::skipHashComment() { ++_pos; ++_column; @@ -325,8 +487,16 @@ void Tokenizer::skipHashComment() { } } -////////////////////// -// --- Token Dispatch +//=== Token Dispatch ====================================================== + +/** + * @brief Heuristically checks if the remaining slice looks like an IPv4 address. + * + * @details Counts dots while digits continue; returns true for exactly 3 dots. + * + * @return `true` if the upcoming token resembles `a.b.c.d`. + * @ingroup config_tokenizing + */ bool Tokenizer::looksLikeIpAddress() const { std::size_t len = _input.size(); std::size_t i = _pos, dots = 0; @@ -344,10 +514,19 @@ bool Tokenizer::looksLikeIpAddress() const { } } - // A valid IPv4 address has at least two dots (e.g., 127.0.0.1) + // A valid IPv4 address has at least two dots (e.g. 127.0.0.1) return dots == 3; } +/** + * @brief Reads and emits the next token from the input. + * + * @details Dispatches to number/identifier/string parsers or emits punctuation + * tokens. Raises on unknown characters. + * + * @throws TokenizerError On illegal bytes or malformed lexemes. + * @ingroup config_tokenizing + */ void Tokenizer::dispatchToken() { unsigned char c = peek(); @@ -356,7 +535,7 @@ void Tokenizer::dispatchToken() { unsigned char next = peekNext(); // If it looks like an IPv4 address (e.g. 127.0.0.1), or - // if the next character is *not* a digit but *is* a valid identifier char + // if the next character is not a digit but is a valid identifier char // ("1index.html" or "/api/v2a3"), then parse as an identifier/keyword if (looksLikeIpAddress() || (!std::isdigit(next) && isIdentifierChar(next))) { _tokens.push_back(parseIdentifierOrKeyword()); @@ -402,8 +581,16 @@ void Tokenizer::dispatchToken() { } } -//////////////////////////////////// -// --- Token Creation & Source Info +//=== Token Creation & Source Info ======================================= + +/** + * @brief Creates a token with current source coordinates. + * + * @param type Token type to emit. + * @param value Token payload (unescaped for strings). + * @return Token with line/column and byte offset. + * @ingroup config_tokenizing + */ Token Tokenizer::makeToken(TokenType type, const std::string& value) const { // Estimate starting column based on current column and token length int col = _column - static_cast(value.length()); @@ -415,6 +602,13 @@ Token Tokenizer::makeToken(TokenType type, const std::string& value) const { return Token(type, value, _line, col, off); } +/** + * @brief Extracts the full source line containing a given byte offset. + * + * @param offset Byte offset into the input buffer. + * @return The line text (without trailing newline). + * @ingroup config_tokenizing + */ std::string Tokenizer::extractLine(std::size_t offset) const { // Find the last newline before (or at) the offset to locate the start of the line std::size_t start = _input.rfind('\n', offset); diff --git a/src/config/tokenizer/token.cpp b/src/config/tokenizer/token.cpp index 00338087..a8bf5e76 100644 --- a/src/config/tokenizer/token.cpp +++ b/src/config/tokenizer/token.cpp @@ -5,19 +5,50 @@ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ -/* Created: 2025/05/04 00:19:57 by nlouis #+# #+# */ -/* Updated: 2025/05/20 23:40:05 by nlouis ### ########.fr */ +/* Created: 2025/08/18 11:58:00 by nlouis #+# #+# */ +/* Updated: 2025/08/18 11:58:59 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file token.cpp + * @brief Implements Token helpers for the config lexer. + * + * @details Provides the Token constructor and lightweight debug helpers used to + * print token kinds and annotated token strings for diagnostics. + * + * @ingroup config_tokenizing + */ + #include "config/tokenizer/token.hpp" #include +//=== Construction ======================================================== + +/** + * @brief Constructs a Token with type, value and source coordinates. + * + * @param t Token type. + * @param v Lexeme string (unescaped for STRING). + * @param l 1-based line number where the token begins. + * @param c 1-based column number where the token begins. + * @param off Byte offset into the original input. + * @ingroup config_tokenizing + */ Token::Token(TokenType t, const std::string& v, int l, int c, std::size_t off) : type(t), value(v), line(l), column(c), offset(off) { } +//=== Debug Helpers ======================================================= + +/** + * @brief Returns a compact string name for a TokenType. + * + * @param type Token category. + * @return Short constant name (e.g., `"IDENTIFIER"`, `"LBRACE"`). + * @ingroup config_tokenizing + */ std::string debugTokenType(TokenType type) { switch (type) { case TokenType::IDENTIFIER: @@ -32,6 +63,8 @@ std::string debugTokenType(TokenType type) { return "RBRACE"; case TokenType::SEMICOLON: return "SEMICOLON"; + case TokenType::COMMA: + return "COMMA"; case TokenType::END_OF_FILE: return "END_OF_FILE"; @@ -61,21 +94,28 @@ std::string debugTokenType(TokenType type) { return "KEYWORD_CLIENT_MAX_BODY_SIZE"; case TokenType::KEYWORD_CGI_EXTENSION: return "KEYWORD_CGI_EXTENSION"; - default: return "UNKNOWN"; } } +/** + * @brief Builds a human-readable token string (for logs and errors). + * + * @details Omits overly large payloads to avoid noisy logs. + * + * @param token Token to stringify. + * @return A string like: `[Token type=IDENTIFIER value="root" line=3 column=5]`. + * @ingroup config_tokenizing + */ std::string debugToken(const Token& token) { // Avoid logging large token payloads to prevent excessive output if (token.value.size() > 1024 * 1024) { - return "[Token ]"; // Return a placeholder for large tokens + return "[Token ]"; } std::ostringstream oss; - // Build the string representation of the token with its type, value, line, and column oss << "[Token type=" << debugTokenType(token.type) << " value=\"" << token.value << "\" line=" << token.line << " column=" << token.column << "]"; - return oss.str(); // Return the formatted token string + return oss.str(); } diff --git a/src/config/validateConfig.cpp b/src/config/validateConfig.cpp index c7fe5fde..9f5ce9f4 100644 --- a/src/config/validateConfig.cpp +++ b/src/config/validateConfig.cpp @@ -6,10 +6,40 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 23:23:50 by nlouis #+# #+# */ -/* Updated: 2025/08/09 22:17:01 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 14:03:41 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file validateConfig.cpp + * @brief Implements validation logic for parsed configuration objects. + * + * @details This file defines the validation pipeline for the Webserv configuration + * after parsing and normalization. It enforces semantic correctness of + * `Server` and `Location` blocks, ensuring that directives are consistent, + * non-conflicting, and point to valid filesystem resources. + * + * The validation routines cover: + * - Presence of at least one location per server + * - Correct syntax and uniqueness of location paths + * - Required defaults (root or return directive) + * - Validity of `server_name` (format, duplicates, uniqueness per host:port) + * - Port and host binding conflicts + * - Error page code ranges + * - Redirect codes restricted to standard 3xx values + * - Allowed HTTP methods (`GET`, `POST`, `DELETE`) + * - Non-zero client body size limits + * - CGI extensions and interpreter mappings + * - Existence and type of root and upload directories + * - Index file usage with valid roots + * + * Together, these checks guarantee that the configuration is safe to + * use at runtime, preventing invalid states and improving user feedback + * with actionable error messages. + * + * @ingroup config_validation + */ + #include "config/validateConfig.hpp" #include "config/parser/ConfigParseError.hpp" #include "utils/errorUtils.hpp" @@ -27,9 +57,25 @@ namespace { namespace fs = std::filesystem; +/** + * @brief Ensures that each server has at least one location block. + * + * @details Iterates through all parsed servers and checks whether they + * contain at least one `Location` directive. + * A server without locations cannot properly handle requests, + * so this function throws a `ValidationError` if any server is + * missing location blocks. + * + * @param servers List of parsed server configurations to validate. + * + * @throws ValidationError If a server has no associated location blocks. + * + * @ingroup config_validation + */ void validateHasLocation(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { if (servers[serverIndex].getLocations().empty()) { + // Invalid: No location blocks defined for this server throw ValidationError("Missing location blocks in server #" + std::to_string(serverIndex + 1) + "\n→ Add at least one 'location' block to handle requests"); @@ -37,6 +83,25 @@ void validateHasLocation(const std::vector& servers) { } } +/** + * @brief Validates location paths inside each server block. + * + * @details This function ensures that: + * - Every location path is non-empty and begins with a '/'. + * - No path segment starts with '.' (disallows '.', '..', hidden or malformed segments). + * - Each location path within a server is unique (no duplicates allowed). + * + * Violations result in a `ValidationError` with a detailed message. + * + * @param servers List of parsed server configurations to validate. + * + * @throws ValidationError If: + * - A location path is empty or does not start with '/', + * - A path segment begins with '.', + * - A duplicate location path is found within the same server. + * + * @ingroup config_validation + */ void validateLocationPaths(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -45,11 +110,13 @@ void validateLocationPaths(const std::vector& servers) { for (const Location& loc : server.getLocations()) { const std::string& path = loc.getPath(); + // Must be non-empty and start with '/' if (path.empty() || path[0] != '/') { throw ValidationError("Invalid location path '" + path + "' in server #" + std::to_string(serverIndex + 1) + "\n→ Must start with '/'"); } + // Validate each segment of the path (no '.'-prefixed parts) fs::path locPath(path); for (const auto& part : locPath) { const std::string seg = part.string(); @@ -61,6 +128,7 @@ void validateLocationPaths(const std::vector& servers) { } } + // Ensure path uniqueness within the same server if (!seenPaths.insert(path).second) { throw ValidationError("Duplicate location path '" + path + "' in server #" + std::to_string(serverIndex + 1)); @@ -69,6 +137,26 @@ void validateLocationPaths(const std::vector& servers) { } } +/** + * @brief Validates default requirements for each location block. + * + * @details This function enforces two critical invariants for location definitions: + * - Each location must define at least one of the following: + * - A `root` directive (filesystem path for content), + * - OR a `return` directive (HTTP redirection). + * - A location cannot define both CGI behavior (`cgi_extension`) and a redirection + * (`return`) at the same time, since these are mutually exclusive. + * + * Violations result in a `ValidationError` with descriptive guidance. + * + * @param servers List of parsed server configurations to validate. + * + * @throws ValidationError If: + * - A location block has neither `root` nor `return`, + * - A location block has both `return` and `cgi_extension` defined. + * + * @ingroup config_validation + */ void validateLocationDefaults(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -86,7 +174,7 @@ void validateLocationDefaults(const std::vector& servers) { "\n→ Each location must have at least a 'root' or a 'return'"); } - // Cannot combine CGI behavior and redirect + // Cannot combine CGI behavior with redirect if (location.hasRedirect() && !location.getCgiExtensions().empty()) { throw ValidationError( "Location '" + path + "' in server #" + std::to_string(serverIndex + 1) + @@ -97,7 +185,18 @@ void validateLocationDefaults(const std::vector& servers) { } } -// Utility: Split string by '.' +/** + * @brief Splits a domain name into its individual labels. + * + * @details A domain name like `"example.com"` is split into + * `{"example", "com"}`. The splitting uses `.` as the separator, + * and empty labels are preserved if consecutive dots appear. + * + * @param name Domain name string to split. + * @return A vector of labels (substrings between dots). + * + * @ingroup config_validation + */ static std::vector splitLabels(const std::string& name) { std::vector labels; std::size_t start = 0; @@ -111,7 +210,19 @@ static std::vector splitLabels(const std::string& name) { return labels; } -// Validate a single label (RFC 1035 rules) +/** + * @brief Validates a single domain label according to RFC 1035. + * + * @details Checks that: + * - Label is non-empty and no longer than 63 characters. + * - Does not begin or end with a hyphen (`-`). + * - Contains only alphanumeric characters (`a-z`, `A-Z`, `0-9`) or hyphens. + * + * @param label The label string to validate. + * @return `true` if the label is valid, `false` otherwise. + * + * @ingroup config_validation + */ bool isValidLabel(const std::string& label) { if (label.empty() || label.size() > 63) return false; @@ -124,7 +235,19 @@ bool isValidLabel(const std::string& label) { return true; } -// Full domain validation +/** + * @brief Validates a full server name against domain format rules. + * + * @details Enforces RFC 1035-style constraints: + * - Name must be non-empty and no longer than 253 characters. + * - Cannot contain consecutive dots (`..`), which would create empty labels. + * - Each label (between dots) must satisfy @ref isValidLabel. + * + * @param name Full domain/server name to validate. + * @return `true` if the name follows domain rules, `false` otherwise. + * + * @ingroup config_validation + */ bool isServerNameValid(const std::string& name) { if (name.empty() || name.size() > 253) return false; @@ -141,18 +264,50 @@ bool isServerNameValid(const std::string& name) { return true; } +/** + * @brief Validates the format of all `server_name` directives across servers. + * + * @details Ensures that every declared `server_name` in the configuration + * adheres to domain naming conventions and does not contain invalid + * characters. Specifically, it checks: + * + * - Non-empty names: Empty strings are not allowed. + * - Printable characters only: No control characters (e.g., newlines, tabs). + * - No whitespace: Spaces are not valid inside domain names. + * - RFC 1035 compliance via @ref isServerNameValid: + * - Maximum length 253 characters. + * - No empty labels (e.g. `"example..com"`). + * - Labels up to 63 characters, only alphanumeric and `-`. + * - No leading or trailing `-` in a label. + * + * If any of these conditions are violated, a @ref ValidationError is thrown + * with a descriptive message indicating the invalid server name and the server + * block index where it was found. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if an invalid `server_name` is detected. + * + * @ingroup config_validation + * + * @see isServerNameValid + * @see isValidLabel + * @see splitLabels + */ void validateServerNameFormat(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const std::vector& serverNames = servers[serverIndex].getServerNames(); for (std::size_t nameIndex = 0; nameIndex < serverNames.size(); ++nameIndex) { const std::string& name = serverNames[nameIndex]; + // must not be empty if (name.empty()) { throw ValidationError("Empty server_name is not allowed in server #" + std::to_string(serverIndex + 1) + "\n→ Specify a non-empty string for server_name"); } + // must contain only printable characters for (std::size_t charIndex = 0; charIndex < name.size(); ++charIndex) { unsigned char c = static_cast(name[charIndex]); if (!std::isprint(c)) { @@ -164,6 +319,7 @@ void validateServerNameFormat(const std::vector& servers) { } } + // no whitespace allowed if (name.find(' ') != std::string::npos) { throw ValidationError( "Whitespace not allowed in server_name in server #" + @@ -171,18 +327,44 @@ void validateServerNameFormat(const std::vector& servers) { "'\n→ Use valid domain-like names without spaces or line breaks"); } + // must follow domain name rules (RFC 1035) if (!isServerNameValid(name)) { throw ValidationError( "Invalid domain format in server_name in server #" + std::to_string(serverIndex + 1) + ": '" + name + "'\n→ Must follow domain format (RFC 1035): labels may only contain a-z, 0-9, " - "dashes; " - "no empty labels, no leading/trailing dashes, and max 253 characters total"); + "dashes; no empty labels, no leading/trailing dashes, and max 253 characters " + "total"); } } } } +/** + * @brief Ensures uniqueness of `server_name` directives within each server block. + * + * @details This function validates that no single `server` block declares the same + * `server_name` more than once. While multiple distinct names are allowed + * (for virtual hosting and aliases), each must be unique within the same + * server definition. + * + * Validation rules: + * - A `server_name` must not appear more than once in the same server block. + * - Duplicate names within the same block trigger a @ref ValidationError. + * - Cross-server duplicates are allowed if they bind to different host:port + * combinations (checked separately in @ref validateUniquePorts). + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if a duplicate `server_name` is found inside the + * same server block. + * + * @ingroup config_validation + * + * @see validateServerNameFormat Validates syntax/format of each server_name. + * @see validateUniquePorts Ensures uniqueness of (host, port, server_name) + * tuples across servers. + */ void validateUniqueServerNames(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const std::vector& names = servers[serverIndex].getServerNames(); @@ -198,6 +380,31 @@ void validateUniqueServerNames(const std::vector& servers) { } } +/** + * @brief Ensures uniqueness of (host, port, server_name) bindings across servers. + * + * @details This function validates that no two servers conflict on the same + * `(host, port, server_name)` tuple, which would otherwise cause + * ambiguous routing at runtime. + * + * Validation rules: + * - Each `(host, port, server_name)` combination must be unique globally. + * - If a server has no `server_name`, it is considered the default for that + * `(host, port)` pair. Only one default server is allowed per host:port. + * - Duplicate `server_name` values across different servers are allowed only + * if they bind to different `(host, port)` pairs. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - Two or more servers define the same `(host, port, server_name)` combination. + * - More than one default server (no `server_name`) exists on the same host:port. + * + * @ingroup config_validation + * + * @see validateUniqueServerNames Ensures names are unique inside one server block. + * @see validateServerNameFormat Checks domain-like syntax of each server_name. + */ void validateUniquePorts(const std::vector& servers) { using HostPort = std::pair; using NameSet = std::unordered_set; @@ -231,6 +438,21 @@ void validateUniquePorts(const std::vector& servers) { } } +/** + * @brief Validates that all configured error_page directives use valid HTTP status codes. + * + * @details Iterates over each server’s `error_page` mappings and ensures that every + * status code falls within the HTTP error range **400–599**. + * These codes represent client (4xx) and server (5xx) errors only. + * Other codes (e.g., 200, 301) are not valid for error pages. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - Any error_page status code is outside the 400–599 range. + * + * @ingroup config_validation + */ void validateErrorPageCodes(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -250,6 +472,25 @@ void validateErrorPageCodes(const std::vector& servers) { } } +/** + * @brief Validates that all `return` directives use valid HTTP redirection codes. + * + * @details This function iterates through all locations of each server + * and checks the status codes defined in the `return` directive. + * Only the following redirection status codes are allowed: + * 301, 302, 303, 307, 308. + * + * These represent permanent or temporary redirects and are standard + * in HTTP/1.1. Any other code (e.g., 200, 404, 500) is invalid + * when used with `return`. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - A `return` directive uses a status code not in {301, 302, 303, 307, 308}. + * + * @ingroup config_validation + */ void validateRedirectCodes(const std::vector& servers) { static const std::array ALLOWED = {{301, 302, 303, 307, 308}}; for (std::size_t i = 0; i < servers.size(); ++i) { @@ -268,6 +509,24 @@ void validateRedirectCodes(const std::vector& servers) { } } +/** + * @brief Validates that all HTTP methods declared in `methods` directives are allowed. + * + * @details This function iterates through every location of each server and + * ensures that the configured HTTP methods are valid. + * Only the following methods are supported by this server: + * **GET, POST, DELETE**. + * + * If any other method (e.g., PUT, PATCH, OPTIONS, HEAD) is specified, + * the configuration is considered invalid and a @ref ValidationError is thrown. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - A location declares an unsupported HTTP method. + * + * @ingroup config_validation + */ void validateAllowedMethods(const std::vector& servers) { static const std::set validMethods = {"GET", "POST", "DELETE"}; @@ -288,6 +547,24 @@ void validateAllowedMethods(const std::vector& servers) { } } +/** + * @brief Validates that `client_max_body_size` is set to a positive value. + * + * @details This function checks each server’s configured maximum request body size + * (`client_max_body_size`) to ensure it is not set to **zero**. + * A value of `0` would effectively disable all request bodies, + * which is considered an invalid configuration. + * + * The size is typically specified in bytes or with human-readable units + * like `k`, `m`, or `g` (e.g., `1m` = 1 megabyte). + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - `client_max_body_size` is set to `0`. + * + * @ingroup config_validation + */ void validateClientMaxBodySize(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -302,6 +579,28 @@ void validateClientMaxBodySize(const std::vector& servers) { } } +/** + * @brief Validates the format of CGI extensions in all server locations. + * + * @details Iterates over every server and its `location` blocks to ensure + * that configured CGI extensions follow a valid format: + * - Must **not** be empty. + * - Must begin with a dot (`.`). + * - Must contain at least one alphanumeric character after the dot + * (e.g., `.php`, `.py`, `.pl`). + * + * This ensures that CGI extensions are properly declared before they + * are mapped to interpreters using the `cgi_interpreter` directive. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - The extension is empty (`""`). + * - The extension is only `"."`. + * - The extension does not start with a dot (`.`). + * + * @ingroup config_validation + */ void validateCgiExtensions(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -323,6 +622,22 @@ void validateCgiExtensions(const std::vector& servers) { } } +/** + * @brief Validates `index` directives in all server locations. + * + * @details Iterates over every server and its `location` blocks to ensure + * that any declared `index` directive is meaningful. An index file + * (e.g., `index.html`) is only valid if the location also specifies + * a `root` directory, since the server must resolve the index file + * path relative to that root. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - A `location` declares an `index` directive but has no `root`. + * + * @ingroup config_validation + */ void validateIndexFiles(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -394,6 +709,23 @@ void validateIndexFiles(const std::vector& servers) { } } */ +/** + * @brief Validates consistency between `cgi_extension` and `cgi_interpreter` directives. + * + * @details Ensures that every declared CGI extension has a corresponding + * interpreter and that no extra interpreters exist for undeclared + * extensions. This prevents misconfigured mappings where requests + * could not be executed correctly or could point to unused binaries. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - A `cgi_extension` exists without a matching `cgi_interpreter`. + * - A `cgi_interpreter` is defined for an extension that is not listed in + * `cgi_extension` (orphaned mapping). + * + * @ingroup config_validation + */ void validateCgiInterpreters(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; @@ -430,31 +762,58 @@ void validateCgiInterpreters(const std::vector& servers) { } } +/** + * @brief Validates that all `root` directives point to existing directories. + * + * @details Iterates over all server locations and verifies that: + * - Locations without a `return` directive must have a `root`. + * - The specified root path must exist on the filesystem. + * - The path must be a directory (not a file or invalid path). + * + * This ensures that each location serving files has a valid, accessible root + * directory, preventing runtime errors due to missing or invalid paths. + * + * @param servers Vector of configured Server objects to validate. + * + * @throws ValidationError if: + * - A location missing a `return` directive has no `root`. + * - The specified root path does not exist. + * - The root path exists but is not a directory. + * + * @ingroup config_validation + */ void validateRootsExist(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { const Server& server = servers[serverIndex]; for (const Location& loc : server.getLocations()) { - // If location is a redirect, root is not required for serving + // Redirect-only locations do not require a filesystem root if (loc.hasRedirect()) continue; const std::string& root = loc.getRoot(); + + // Defensive check: non-redirecting locations must define a root + // (Should already be enforced by validateLocationDefaults) if (root.empty()) { - // Should already be prevented by validateLocationDefaults, but double-guard throw ValidationError("Location '" + loc.getPath() + "' in server #" + std::to_string(serverIndex + 1) + " is missing a 'root' (required when no 'return' is set)"); } + // Probe filesystem status; capture errors via std::error_code std::error_code ec; const fs::file_status st = fs::status(root, ec); + + // Path must exist if (ec || !fs::exists(st)) { throw ValidationError("Root path '" + root + "' does not exist for location '" + loc.getPath() + "' in server #" + std::to_string(serverIndex + 1) + "\n→ Create the directory or update the 'root' path."); } + + // Path must be a directory (not a regular file/symlink/etc.) if (!fs::is_directory(st)) { throw ValidationError("Root path '" + root + "' is not a directory for location '" + loc.getPath() + "' in server #" + @@ -465,6 +824,21 @@ void validateRootsExist(const std::vector& servers) { } } +/** + * @brief Validates upload store directories for all servers and locations. + * + * @details Ensures that whenever `upload_store` is enabled in a location block: + * - The configured path exists in the filesystem. + * - The path points to a directory (not a file or special node). + * + * @param servers Collection of server configuration objects to validate. + * + * @throws ValidationError If: + * - The configured upload store path does not exist. + * - The path exists but is not a directory. + * + * @ingroup config_validation + */ void validateUploadStores(const std::vector& servers) { namespace fs = std::filesystem; diff --git a/src/utils/filesystemUtils.cpp b/src/utils/filesystemUtils.cpp index 862b6fef..abbd635d 100644 --- a/src/utils/filesystemUtils.cpp +++ b/src/utils/filesystemUtils.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/13 09:39:07 by nlouis #+# #+# */ -/* Updated: 2025/08/15 22:40:36 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 12:05:15 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -315,7 +315,7 @@ std::string buildFilePath(const HttpRequest& request, const Location& loc) { * @warning Leading or trailing slashes are not included in the result; the * caller must track whether the path was absolute if needed. */ -static std::vector splitPath(const std::string& path) { +/* static std::vector splitPath(const std::string& path) { std::vector parts; std::stringstream ss(path); std::string part; @@ -326,7 +326,7 @@ static std::vector splitPath(const std::string& path) { parts.push_back(part); } return parts; -} +} */ /* bool mkdirRecursive(const std::string& path) { std::vector parts = splitPath(path); From 8e4a7bfea89e0da3564e0794b08f468a4e5e193f Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 14:56:03 +0300 Subject: [PATCH 15/32] Remove outdated CONTRIBUTING guidelines and enhance README with configuration parsing details The outdated CONTRIBUTING.md file has been removed, as its content is no longer relevant or applicable. Key sections were either redundant or replaced by newer practices. README.md has been updated to include detailed documentation on the configuration parsing flow, covering tokenization, parsing, normalization, validation, and error handling. These additions provide a comprehensive overview of the configuration pipeline and its rules. This update improves clarity for contributors by consolidating essential information into the README and reducing redundancy across the repository. --- CONTRIBUTING.md | 669 ------------------------------------------------ README.md | 172 +++++++++++-- 2 files changed, 148 insertions(+), 693 deletions(-) delete mode 100644 CONTRIBUTING.md diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md deleted file mode 100644 index d03599d1..00000000 --- a/CONTRIBUTING.md +++ /dev/null @@ -1,669 +0,0 @@ -# Contributing to Webserv - -Thanks for being part of the team! This document outlines our workflow, standards, and best practices to ensure smooth collaboration. - ---- - -## Branching Strategy - -Our Git branching model keeps the codebase **stable**, **collaborative**, and **easy to maintain**. - -- **`main`**: - Stable, production-ready code only. - No direct pushes. Only PRs from `dev` are merged after review and validation. - -- **`dev`**: - Latest working version. - All new features, fixes, and experiments integrate here first. - -- **Feature branches**: - Created from `dev`. - Naming: `feature/` or `bugfix/`. - PRs must target `dev`, not `main`. - -> **Note:** TODO CI would automatically open a PR from `dev` to `main` when `dev` is green. - -### Typical Workflow - -```bash -git checkout dev -git pull origin dev -git checkout -b feature/ -# Work on your feature... -git add . -git commit -m "feature: implement basic GET request handling" -git push origin feature/ -# Open a PR from feature/ → dev -``` - ---- - -## 📄 Issues & Pull Requests - -We use GitHub templates: - -- **[Bug / Feature Requests](.github/ISSUE_TEMPLATE/issue_template)** -- **[Pull Requests](.github/PULL_REQUEST_TEMPLATE/pull_request_template.md)** - - -### PR Title Convention - -```text -: short description -``` - -Example: `http: support chunked transfer decoding`. - ---- - -## Pre-Push Checklist - -- `make` must succeed without warnings. -- `./webserv` must start without errors. -- Run : - - `make asan` - - `make tsan` - - `make ubsan` - -Open a PR only when all local tests pass. - ---- - -## Commit Guidelines - -Good commits = good collaboration. - -### Commit Structure - -```text -: short summary (max 50 chars) - -Optional longer explanation (wrap at 72 chars). -``` - -- **Module**: `http`, `server`, `config`, `utils`, etc. -- **Short Summary**: Clear and specific. -- **Longer Explanation** (optional): Extra context, tricky parts, fixes. - -### Rules - -- Fix commits must mention the bug (`cgi: fix PATH_INFO handling`). -- One commit = one purpose. -- Imperative mode: “implement”, not “implemented”. -- No WIP commits. Before opening a PR, clean up your commit history using: (`git rebase -i origin/dev`). - ---- - -## Pull Requests - -### How to do a Pull Request - -**1.** Start from dev - -```bash -git checkout dev -git pull origin dev -``` - -**2.** Create your feature branch from `dev` - -```bash -git checkout -b feature/my-feature -``` - -**3.** Work and commit -Make your changes, then commit them following the commit guidelines: - -```bash -git add . -git commit -m "module: short, meaningful description" -``` -Repeat as needed. - -**4.** Push your branch - -```bash -git push origin feature/my-feature -``` - -**5.** Open the Pull Request on GitHub - -- Go to your repo on GitHub. - - You’ll see a prompt: “Compare & pull request” — click it -Or go to Pull Requests → New Pull Request - -- Set the base to dev -- Set the compare to feature/my-feature - -- Fill in: - - Title: follow the format: module: short description - - Description: - - What was changed - - Why it was needed - - Anything specific to review/test - -- Submit as a Draft if it's not finished yet, or as a regular PR if it's ready - -**6.** After Review - -- Make changes if requested. -- Push new commits: they are automatically added to the same PR -- Once approved, you or the reviewer can Squash & Merge the PR into dev - -### Opening a PR - -- Always target `dev`, **never** `main`. -- Open a **Draft PR** for early feedback. -- Mark **Ready for Review** once finalized. -- Use the PR template. -- Follow commit/PR title conventions. - -### A PR must - -- Contain one logical change. -- Include a clear title and detailed description: - - What changed? - - Why was it needed? - - Anything special to watch during review. -- Pass local builds/tests. -- Be cleaned up (no debug prints, no commented-out code). -- Be rebased on the latest `dev`. - ---- - -### Code Review - -**1.** You open a Pull Request (PR) -- Once you’ve pushed your feature branch and opened a PR targeting dev, it enters review phase. - -**2.** Teammates are notified -- Anyone watching the repo or specifically requested as a reviewer gets notified. -- You can explicitly assign reviewers via the PR sidebar. - -**3.** Reviewers leave comments -- Reviewers can: - - Approve the PR - - Request changes - - Comment (neutral feedback) - -- They can: - - Leave inline comments on specific lines - - Leave general comments at the top or bottom - - Suggest changes with GitHub's suggestion UI - - Discuss architecture, bugs, naming, clarity, etc. - -**4.** You respond to feedback -- You can: - - Reply to comments - - Commit fixes and push to the same branch — the PR auto-updates - - Mark comments as resolved when done - - GitHub tracks which comments are resolved vs. still active. - -**5.** Once approved, you merge -- After: - - All reviewers approve - - CI passes - - You’ve rebased & cleaned history - -- Then: - - Use Squash & Merge or Rebase & Merge - - Delete the feature branch if no longer needed - -#### Tips for Efficient Review -- Be responsive to feedback -- Keep changes small and scoped -- Avoid "monster PRs" -- Use Draft PRs early to gather feedback -- Address every comment, even if just to explain why no change is needed -- At least **one teammate** must review and approve. -- Clarify anything that’s ambiguous. - ---- - -## Merge Strategies: Rebase & Merge vs Squash & Merge - -When collaborating on GitHub, choosing the right merge strategy affects your repository's history clarity, bisectability, and maintainability. Here's a clear comparison between **Rebase & Merge** and **Squash & Merge**, with guidance for when to use each. - ---- - -### Rebase & Merge - -#### What it does - -* Replays each commit from the PR branch **on top of** the target branch (e.g., `dev`). -* **Preserves** all individual commits. -* Avoids merge commits. - -#### Resulting history - -``` -Before: -main --- A --- B - \ - C --- D (feature) - -After: -main --- A --- B --- C' --- D' (C and D rebased on B) -``` - -#### Pros and Cons - -| Pros | Cons | -| -------------------------------------------- | ------------------------------------------------------- | -| Clean, linear history | Rewrites commit SHAs (not suitable for shared branches) | -| Preserves logical commit structure | Requires clean, well-structured commits | -| Great for detailed `git blame` and bisecting | | - -#### Use when - -* The PR has clean, meaningful commits -* You want to preserve granular history -* Your team follows good commit hygiene - -#### Example Workflow - -```bash -git checkout dev -git pull origin dev -git checkout -b feature/add-auth -echo "// implement auth" > auth.cpp -git add auth.cpp -git commit -m "auth: implement token-based authentication" -git push origin feature/add-auth -# Open a PR to dev -# After review, rebase interactively to clean history: -git rebase -i dev -git push --force-with-lease -# On GitHub: Use "Rebase & Merge" to merge the PR -``` - ---- - -### Squash & Merge - -#### What it does - -* Squashes **all PR commits into one**. -* Commits that were WIP, fixups, etc., are flattened. - -#### Resulting history - -``` -Before: -main --- A --- B - \ - C --- D (feature) - -After: -main --- A --- B --- E (one squashed commit) -``` - -#### Pros and Cons - -| Pros | Cons | -| ------------------------------------ | ------------------------------------------- | -| Extremely clean and concise history | Loses individual commit information | -| Easy to revert a single change | Harder to trace how a feature was developed | -| Great when PR contains noisy commits | | - -#### Use when - -* PR has WIP/fix commits -* You care more about trunk clarity than commit history -* You want one commit per feature/fix - -#### Example Workflow - -```bash -git checkout dev -git pull origin dev -git checkout -b bugfix/fix-upload-path -echo "// fix path bug" > upload.cpp -git add upload.cpp -git commit -m "fix: wrong upload directory" -echo "// debug" >> upload.cpp -git commit -am "debug: temporary logging" -echo "// cleanup" >> upload.cpp -git commit -am "cleanup: remove debug" -git push origin bugfix/fix-upload-path -# Open a PR to dev -# On GitHub: Use "Squash & Merge" and write a clean commit message -``` - ---- - -### Summary Table - -| Feature | Rebase & Merge | Squash & Merge | -| ---------------------------- | -------------------- | -------------- | -| Preserves individual commits | Yes | No | -| Cleans messy commit history | No | Yes | -| Easy to revert feature | Maybe (many commits) | Yes (1 commit) | -| Great for `git blame` | Yes | Depends | -| Suitable for WIP commits | No | Yes | -| Maintains linear history | Yes | Yes | - ---- - -### Best Practices - -* Rebase locally to keep up-to-date: `git pull --rebase` -* Clean up commits before PR: `git rebase -i` -* Use squash for small fixes, doc updates, or messy PRs -* Use rebase when you want full commit fidelity - ---- - -### Suggested Strategy for `webserv` - -| PR Type | Suggested Merge | -| ----------------- | --------------- | -| Clean feature PR | Rebase & Merge | -| Bugfix with noise | Squash & Merge | -| Docs/Chores | Squash & Merge | - -Maintain a balance: clear main branch history with squash, but keep fidelity where it helps future devs understand changes. - - ---- - -### Cleanup Rules - -- No commented-out code. -- Always run `make clean && make fclean` before pushing. -- Make sure `.gitignore` correctly excludes build artifacts and editor files. - ---- - -## clang-tidy Usage and Best Practices - -`clang-tidy` is a powerful static analysis tool for C++ that helps catch bugs, enforce coding standards, and suggest modern best practices. It integrates cleanly into CI/CD pipelines and local development workflows. - ---- - -### What is clang-tidy? - -`clang-tidy` is part of the LLVM/Clang toolchain. It performs static analysis on C++ code using a set of configurable checks. It supports: - -* Bug detection (memory issues, undefined behavior, etc.) -* Style enforcement (naming conventions, brace usage, etc.) -* Modernization (C++11–C++20 idioms) -* Performance improvements (loop transformations, unneeded copies) - ---- - -### Project Integration - -In the Webserv project: - -* We enforce best practices via `clang-tidy` -* Configuration is located in the root of the repository: [`.clang-tidy`](.clang-tidy) - ---- - -### Configuration Overview (`.clang-tidy`) - -```yaml -Checks: > - -*, - bugprone-*, - modernize-*, - performance-*, - readability-*, - clang-analyzer-*, - cppcoreguidelines-*, - misc-* - -WarningsAsErrors: '*' - -HeaderFilterRegex: '.*' - -FormatStyle: file - -CheckOptions: - - key: modernize-use-nullptr.NullMacros - value: 'NULL' - - key: modernize-use-override.CheckSpelling - value: true - - key: readability-identifier-naming.VariableCase - value: lower_case - - key: readability-identifier-naming.ClassCase - value: PascalCase - - key: readability-identifier-naming.FunctionCase - value: camelCase - - key: readability-identifier-naming.PrivateMemberPrefix - value: _ -``` - ---- - -### Enforced Rule Categories - -The configuration above enables the following major rule sets: - -* **`bugprone-*`**: Detects code that is likely to be incorrect or prone to bugs (e.g., uninitialized values, copy-paste errors). -* **`modernize-*`**: Promotes modern C++ usage (e.g., `nullptr`, `override`, range-based loops, smart pointers). -* **`performance-*`**: Highlights inefficient code patterns and suggests optimized alternatives (e.g., unnecessary copies). -* **`readability-*`**: Improves clarity and style consistency (e.g., naming, magic numbers, redundant expressions). -* **`clang-analyzer-*`**: Static path-sensitive analysis to detect dead code, memory leaks, and undefined behavior. -* **`cppcoreguidelines-*`**: Enforces the C++ Core Guidelines to encourage safe, maintainable, and modern C++. -* **`misc-*`**: A miscellaneous set of useful checks not covered by other categories. - -In addition: - -* **All warnings are treated as errors** -* **Naming conventions** are strictly enforced - ---- - -### Naming Conventions Enforced - -`clang-tidy` enforces identifier naming across your codebase using the `readability-identifier-naming` checks. The enforced rules include: - -| Entity Type | Convention | Example | -| --------------- | ------------ | ---------------------- | -| Variables | `lower_case` | `timeout`, `buffer_id` | -| Functions | `camelCase` | `handleRequest()` | -| Classes/Structs | `PascalCase` | `HttpServer`, `Config` | -| Private Members | `_` suffix | `connection_`, `port_` | - -These conventions help maintain a consistent, readable codebase. Violations will be flagged and must be corrected before submission. - ---- - -### How to Run clang-tidy Locally - -#### Prerequisites - -* `clang-tidy` installed (typically via `clang-tools` or `llvm` package) -* Compilation database (`compile_commands.json`) in your build directory - -Generate the compilation database (with CMake): - -```bash -cmake -DCMAKE_EXPORT_COMPILE_COMMANDS=ON -S . -B build -``` - -Or if using CMake presets: - -```bash -cmake --preset debug -``` - -Then symlink or copy it to project root: - -```bash -ln -s build/compile_commands.json . -``` - -#### Manual Run - -```bash -clang-tidy src/Server.cpp -- -Iinclude -``` - -> Use `-- -Iinclude` to pass include paths to the compiler properly. - -#### Run on all files (parallel) - -```bash -find src/ -name '*.cpp' | xargs -P$(nproc) -n1 clang-tidy -- -Iinclude -``` - ---- - -### Fixing Issues - -Use fix mode to auto-apply some suggestions: - -```bash -clang-tidy src/Server.cpp -fix -- -Iinclude -``` - -Always review changes with `git diff` before committing. - ---- - -### CI Integration - -GitHub Actions automatically runs `clang-tidy` on pushes to `dev` and `main`. If any issue is found: - -* CI will fail -* You must fix all warnings before merging or submitting a PR - -Use `clang-tidy` locally to catch issues early. - ---- - -### IDE Integration - -Using an IDE that supports `clang-tidy` greatly improves your workflow: - -### Visual Studio Code (VSCode) - -* Requires extensions: `clangd` or `LLVM`, and optionally `CodeLLDB` -* Make sure `compile_commands.json` is generated -* Set up `c_cpp_properties.json` or use `clangd` with `--compile-commands-dir` -* Configure diagnostics via `.clang-tidy` - -### Other Editors - -* Most modern C++ IDEs (Eclipse CDT, Qt Creator, etc.) can integrate with `clang-tidy` -* Look for settings to point to `.clang-tidy` and the compilation database - -Benefits of IDE integration: - -* Instant feedback while coding -* Fix suggestions as you type -* Fewer CI surprises -* Encourages early cleanup - ---- - -### Development Tips - -* Use `.clang-tidy` overrides via command line: `-checks=*,-some-check` -* For language version issues: `--extra-arg=-std=c++20` -* Add `-p build` if `compile_commands.json` is in a subdirectory -* Use `clang-tidy -export-fixes fixes.yaml` to export fix suggestions - ---- - -### Summary - -* `clang-tidy` enforces consistency and quality -* All warnings are treated as errors -* Used in CI and should be part of local workflow -* IDE integration makes it faster and easier to use -* Configured to match the Webserv style guide -* Covers bug-prone patterns, modern C++ practices, performance, readability, and naming -* Enforces strict identifier naming conventions to ensure clean and uniform code - ---- - -For full documentation: [https://clang.llvm.org/extra/clang-tidy/](https://clang.llvm.org/extra/clang-tidy/) - - ---- - -## Code Style - -Follow the **[Modern C++ Style Guide](STYLEGUIDE.md)** based on **LLVM style**, adapted for us: - -- 4 spaces indentation, no tabs. -- Max 100 columns. -- K&R braces (`if (...) {`). -- `*` and `&` bind to type (`int* ptr`). -- Includes sorted alphabetically but logical grouping preserved. -- No `using namespace std;` at file scope. -- Use smart pointers (`std::unique_ptr`, `std::shared_ptr`) over raw pointers. -- Follow Rule of 0 / Rule of 5. -- Prefer exceptions and `std::optional` / `std::expected` for errors. -- Doxygen comments for all public entities. - ---- - -### Documentation - -The project documentation is automatically generated using **Doxygen** and deployed to **GitHub Pages**: - -- [📚 View Online Documentation](https://to0nsa.github.io/webserv/) - -Documentation is generated from the `include/`, `src/`, and `tests/` directories, and can be regenerated manually using Doxygen. - -- [📄 Local Documentation Guide](docs/README.md) — Explains the structure, regeneration process, and GitHub Pages deployment. - -> Documentation includes call graphs and caller graphs if [Graphviz](https://graphviz.gitlab.io/) is installed. - -#### Documentation Style - -All public classes, functions, and modules must follow the project's [Doxygen Style Guide](DOXYGENSTYLEGUIDE.md). - -This guide defines: - -- Mandatory sections (`@brief`, `@details`, `@param`, `@return`, etc.) -- File headers for all `.hpp`, `.cpp`, and `.tpp` files -- Logical module grouping (`@defgroup`, `@ingroup`) -- Formatting rules and tag order -- Use of present tense and alignment of tags -- Handling of private/internal documentation (`@internal`) - ---- - -## Editor Configuration - -We ship an [`.editorconfig`](.editorconfig): - -- UTF-8 encoding -- 4-space indentation -- Unix (LF) line endings -- End files with newline -- No trailing whitespace (Markdown exempt) -- 100 characters recommended line limit - -Configure your editor accordingly. - ---- - -## Git Attributes - -Our [`.gitattributes`](.gitattributes) ensures: - -- LF line endings everywhere -- Correct binary file handling (e.g., `.png`, `.jpg`, `.gif`) - ---- - -## Git Ignore - -Our [`.gitignore`](.gitignore) excludes: - -- Build artifacts (`build/`, `objs/`, `bin/`) -- Compiler outputs (`*.o`, `*.out`, `*.exe`, `*.a`) -- IDE configs (`.vscode/`, `.idea/`, `*.swp`) -- OS files (`.DS_Store`, `Thumbs.db`) -- Logs (`*.log`) - -**Never** commit ignored files. - ---- diff --git a/README.md b/README.md index 3aa2ecdf..b8763f3c 100644 --- a/README.md +++ b/README.md @@ -12,6 +12,103 @@ ___ +## Configuration Parsing Flow + +This section describes how the configuration parsing logic of **Webserv** works, including the step‑by‑step pipeline and the rules applied during parsing and validation. + +
+See Details + +### 1. Tokenization + +* **Component:** `Tokenizer` +* **Goal:** Convert raw configuration text into a structured list of tokens. +* **Steps:** + + * Skip UTF‑8 BOM if present. + * Ignore whitespace, line breaks, and comments (`# ...`). + * Classify tokens into categories: + + * **Keywords:** `server`, `location`, `listen`, `host`, `root`, `index`, `autoindex`, `methods`, `upload_store`, `return`, `error_page`, `client_max_body_size`, `cgi_extension`. + * **Identifiers:** Alphanumeric strings with `-`, `.`, `/`, `:` allowed. + * **Numbers & Units:** Digits with optional single‑letter suffix (`k`, `m`, `g`). + * **Strings:** Quoted values (single `'` or double `"`). + * **Symbols:** `{`, `}`, `;`, `,`. + * Detect and reject invalid characters, control characters, or malformed identifiers. + +### 2. Parsing + +* **Component:** `ConfigParser` +* **Goal:** Transform token stream into structured objects (`Config`, `Server`, `Location`). +* **Rules:** + + * **Block structure:** Curly braces `{ ... }` delimit `server` and `location` blocks. + * **Directives:** Each directive must end with `;` unless it opens a block. + * **Directive placement:** Certain directives are only valid at specific levels: + + * Server level: `listen`, `host`, `server_name`, `error_page`, `client_max_body_size`. + * Location level: `root`, `index`, `autoindex`, `methods`, `upload_store`, `return`, `cgi_extension`, `cgi_interpreter`. + * **Nesting:** Locations may not contain other `server` blocks. + +### 3. Configuration Objects + +* **Server:** Represents a virtual host. + + * Holds host, port, server names, error pages, body size limits, and `Location` blocks. +* **Location:** Defines behavior for a URI path prefix. + + * Includes root directory, index file(s), autoindex flag, allowed methods, redirects, CGI settings, and upload store. + +### 4. Normalization + +* After parsing, the configuration is **normalized** to ensure consistency and defaults: + + * Missing `client_max_body_size` → default = **1 MB**. + * Missing `error_page` → add defaults for common errors (403, 404, 500, 502 → `/error.html`). + * Missing `methods` → defaults to **GET, POST, DELETE**. + * Locations without `root` → fallback to `/var/www` (unless redirected). + * Root location (`/`) without `index` → defaults to **index.html**. +* Normalization guarantees that later validation and runtime logic operate on a **complete and uniform** model. + +### 5. Validation + +* **Component:** `validateConfig` +* **Goal:** Enforce semantic correctness beyond syntax. +* **Checks applied:** + + * **Presence checks:** At least one `location` per `server`. + * **Path rules:** Location paths must start with `/` and not contain segments beginning with `.`. + * **Defaults:** Each location must define either a `root` or `return` (but not both with CGI). + * **Server names:** Must be unique per host\:port, valid per RFC 1035 (no spaces, no control chars, no empty labels). + * **Ports:** Only one unnamed default server per host\:port pair. + * **Error pages:** Codes restricted to 400–599. + * **Redirects:** Only 301, 302, 303, 307, 308 allowed. + * **Methods:** Only `GET`, `POST`, `DELETE` permitted. + * **Client body size:** Must be > 0. + * **CGI:** Extensions must start with a dot, interpreters must map 1‑to‑1 with declared extensions. + * **Roots & Upload stores:** Must exist and be directories. + * **Index:** Requires a valid `root`. + +### 6. Error Handling + +* **Tokenizer:** Throws `TokenizerError` with line/column context when encountering invalid tokens. +* **Parser:** Throws `ConfigParseError` on invalid structure or misplaced directives. +* **Validator:** Throws `ValidationError` with descriptive guidance on fixing invalid configurations. + +
+ +### Flow Overview + +1. **Tokenizer** → breaks input into tokens. +2. **ConfigParser** → builds in‑memory `Config` with `Server` & `Location` objects. +3. **normalizeConfig** → fills missing defaults (sizes, error pages, roots, index, methods). +4. **validateConfig** → applies semantic checks. +5. **Runtime** → validated configuration is passed to the server for request routing. + +The configuration pipeline guarantees that only syntactically valid, normalized, and semantically correct configurations are accepted. This ensures the server runs with predictable defaults, strong validation, and developer-friendly diagnostics. + +___ + ## Build & Test Instructions ### Build with Makefile @@ -65,41 +162,70 @@ ___ ## Continuous Integration & Documentation -> This project uses **GitHub Actions** to automate building, testing, and documentation deployment. +This project leverages **GitHub Actions** to ensure code quality, stability, and up-to-date documentation. + +
+See Details -### ✅ CI Pipeline +### CI Pipeline -On each push or pull request to `main` or `dev`, the following jobs are run automatically: +* Runs automatically on pushes and pull requests to `main` and `dev`. +* Includes manual triggers (`workflow_dispatch`) and dependency checks after successful builds. -| Job | Purpose | -|---|---| -| 🧪 Build (Release) | Builds the project using the provided `Makefile`. | -| 📄 Doxygen Docs | Generates and deploys Doxygen documentation to GitHub Pages. | +**Jobs Overview:** -All configurations rely on the project `Makefile` and follow the project's coding style. +| Job | Description | +| ------------ | ------------------------------------------------------------------------------------------------ | +| 🔨 **Build** | Compiles the project using the provided `Makefile` to ensure successful builds. | +| 🧪 **Test** | Builds the server, runs Python test suite against a live instance, and captures logs on failure. | +| 📚 **Docs** | Generates Doxygen documentation (with Graphviz diagrams) and deploys it to **GitHub Pages**. | -### 📚 Documentation +
-- Doxygen generates HTML docs from source code and Markdown (`README.md` is the main page) -- Graphviz is enabled for call graphs, class diagrams, and source browser -- Documentation is deployed automatically via GitHub Pages from the `docs/html` directory +Every code change is built, tested, and documented automatically, ensuring a robust development workflow and always-available reference docs. ___ -## Contributing +## Documentation + +This section describes how project documentation is generated, structured, and published. + +
+See Details + +### 1. Doxygen-Powered + +* Documentation is generated automatically from **source code comments** and **Markdown files**. +* `README.md` serves as the **entry point**, offering an overview and links to modules. + +### 2. Graphical Support + +* **Graphviz** integration produces: + + * **Class diagrams** to illustrate object hierarchies. + * **Call graphs** to visualize execution flow. + * **Dependency graphs** to map relationships between modules. +* These visuals improve comprehension of the server’s architecture. + +### 3. Navigation & Browsing + +* The source browser cross-references **functions, classes, and files**. +* Each documented entity links directly to its definition in the codebase. +* Groups (`@defgroup`, `@ingroup`) provide thematic navigation across modules (e.g., `config`, `core`, `http`). + +### 4. Deployment -Contribution guidelines and workflow standards are detailed in the dedicated document: +* Documentation is built in **CI/CD pipelines**. +* Published automatically via **GitHub Pages** from the `docs/html` directory. +* Ensures the latest version is always available for contributors and maintainers. -- [📚 View Contributing Guide](CONTRIBUTING.md) +### 5. Best Practices -This document explains: +* Consistent **Doxygen-style headers** across `.hpp` and `.cpp` files. +* Markdown files complement code documentation with **high-level design notes** and **workflow explanations**. +* Together, these guarantee both **low-level API reference** and **high-level architectural guidance**. -- The coding style -- The branching strategy (main, dev, feature branches) -- The commit message conventions (module: short description) -- How to structure pull requests properly -- The review and merge process -- Cleanup and quality rules before pushing code +
___ @@ -116,8 +242,6 @@ webserv ├── 📁 tests/ # Unit tests for various modules ├── 📁 configs/ # Test configuration files for parser/tokenizer ├── 📁 docs/ # Markdown documentation (DOCS.md, guides, etc.) -├── 📁 scripts/ # Helper scripts to run tests and sanitizer builds -├── .asanignore # Suppression rules for AddressSanitizer (e.g. libc++ internals) ├── .clang-format # Enforces formatting rules (4-space indent, K&R braces, etc.) ├── .editorconfig # Shared IDE/editor config for consistent style ├── .gitattributes # Defines merge/diff rules for Git (e.g. binary files) From 64ae82a91fd31ab1b3cb8867ab78038e4c4f24ab Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 19:36:25 +0300 Subject: [PATCH 16/32] Refines Doxygen setup and removes obsolete docs Cleans up Doxygen configuration by removing unused directories, files, and outdated documentation references. Updates settings to limit graph node count, reduce threads, and disable multi-targets for consistency. Enhances documentation generation workflow by clearing previous output and adjusts input sources for accuracy. Deletes redundant comments and unused helper functions for streamlined codebase. --- .github/workflows/docs.yml | 3 ++ .gitignore | 1 + Doxyfile | 7 +-- README.md | 1 - docs/DOCS.md | 37 --------------- docs/doxygen.warnings | 45 ++++++++++++++++++- include/config/normalizeConfig.hpp | 15 +------ src/config/parser/directive_handler_table.cpp | 17 ++----- src/config/tokenizer/Tokenizer.cpp | 4 +- src/config/validateConfig.cpp | 14 +++--- src/utils/filesystemUtils.cpp | 21 +-------- 11 files changed, 63 insertions(+), 102 deletions(-) delete mode 100644 docs/DOCS.md diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml index 1e9c7a7d..f84948fd 100644 --- a/.github/workflows/docs.yml +++ b/.github/workflows/docs.yml @@ -16,6 +16,9 @@ jobs: - name: Install Doxygen and Graphviz run: sudo apt-get update && sudo apt-get install -y doxygen graphviz + - name: Clean previous HTML output + run: rm -rf docs/html || true + - name: Generate Doxygen Documentation run: doxygen Doxyfile diff --git a/.gitignore b/.gitignore index b8689ca1..b8edd1c8 100644 --- a/.gitignore +++ b/.gitignore @@ -55,3 +55,4 @@ test_webserv/**/*.pyo # --- Doxygen docs --- /docs/html/ +/docs/doxygen.warnings diff --git a/Doxyfile b/Doxyfile index d01fb474..9dcc44d9 100644 --- a/Doxyfile +++ b/Doxyfile @@ -9,7 +9,7 @@ INPUT_ENCODING = UTF-8 # --- Input Files --- -INPUT = include/ src/ tests/ README.md +INPUT = include/ src/ README.md RECURSIVE = YES FILE_PATTERNS = *.cpp *.hpp *.tpp *.md USE_MDFILE_AS_MAINPAGE = README.md @@ -71,10 +71,11 @@ HIDE_UNDOC_CLASSES = YES # --- Graphs and Diagrams --- HAVE_DOT = YES -DOT_NUM_THREADS = 4 +DOT_NUM_THREADS = 1 CALL_GRAPH = YES CALLER_GRAPH = YES -DOT_MULTI_TARGETS = YES +DOT_MULTI_TARGETS = NO +DOT_GRAPH_MAX_NODES = 100 CLASS_DIAGRAMS = YES DOT_IMAGE_FORMAT = svg INTERACTIVE_SVG = YES diff --git a/README.md b/README.md index b8763f3c..c3703b0a 100644 --- a/README.md +++ b/README.md @@ -247,7 +247,6 @@ webserv ├── .gitattributes # Defines merge/diff rules for Git (e.g. binary files) ├── .gitignore # Files and folders ignored by Git (e.g. build/, *.o) ├── ACTIONPLAN.md # Project-level planning or roadmap -├── CONTRIBUTING.md # Guidelines for contributing to the project ├── DOXYGENSTYLEGUIDE.md # Doxygen conventions for documenting code ├── Doxyfile # Main config for Doxygen documentation generation ├── LICENSE # Project license (e.g. MIT, GPL) diff --git a/docs/DOCS.md b/docs/DOCS.md deleted file mode 100644 index d4656a15..00000000 --- a/docs/DOCS.md +++ /dev/null @@ -1,37 +0,0 @@ -# Documentation - -This directory contains the generated project documentation. - -## Structure - -- `html/` — The HTML output generated by [Doxygen](https://www.doxygen.nl/). - - Main entry point: [`index.html`](html/index.html) - -## How to regenerate documentation - -If you modify or add code comments, you can regenerate the documentation locally: - -```bash -doxygen Doxyfile -``` - -This will update the `html/` folder automatically. - -## GitHub Pages - -The contents of `docs/html/` are published automatically via GitHub Pages. - -The documentation is automatically regenerated and deployed whenever a `git push` is made to the `main` branch. - -You can view the live documentation at: - -**[https://to0nsa.github.io/webserv/](https://to0nsa.github.io/webserv/)** - ---- - -## Notes - -- The documentation is generated from `include/`, `src/`, and `tests/` directories. -- Graphs (call graphs, caller graphs) are included if [Graphviz](https://graphviz.gitlab.io/) is installed. - ---- diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index 4fa893b2..bc2af0fc 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -1,4 +1,45 @@ -warning: source 'tests' is not a readable file or directory... skipping. +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:262: warning: no matching file member found for +const std::unordered_map< std::string, ServerHandler > & directive::serverHandlers +Possible candidates: + 'const std::unordered_map< std::string, ServerHandler > & serverHandlers()' at line 262 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:361: warning: no matching file member found for +const std::unordered_map< std::string, LocationHandler > & directive::locationHandlers +Possible candidates: + 'const std::unordered_map< std::string, LocationHandler > & locationHandlers()' at line 361 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/validateConfig.cpp:351: warning: unable to resolve reference to 'validateUniquePorts' for \ref command error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! -/home/toonsa/myProjects/webserv/README.md:103: warning: unable to resolve reference to '/home/toonsa/myProjects/webserv/CONTRIBUTING.md' for \ref command +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command diff --git a/include/config/normalizeConfig.hpp b/include/config/normalizeConfig.hpp index 2b3a0ce4..010bccd1 100644 --- a/include/config/normalizeConfig.hpp +++ b/include/config/normalizeConfig.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 21:02:21 by nlouis #+# #+# */ -/* Updated: 2025/08/18 13:26:36 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 15:58:50 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -26,17 +26,4 @@ #include "config/Config.hpp" -/** - * @brief Applies in-place normalization to a parsed configuration. - * - * @details Iterates over all servers/locations and fills defaults for: - * - client_max_body_size - * - error_page mappings - * - location root and index files - * - allowed HTTP methods - * - * @param config Parsed configuration to normalize (modified in place). - * - * @ingroup config_normalizing - */ void normalizeConfig(Config& config); diff --git a/src/config/parser/directive_handler_table.cpp b/src/config/parser/directive_handler_table.cpp index e44e0403..c9779aae 100644 --- a/src/config/parser/directive_handler_table.cpp +++ b/src/config/parser/directive_handler_table.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/08 17:14:27 by nlouis #+# #+# */ -/* Updated: 2025/08/18 12:47:33 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 19:35:02 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -15,10 +15,10 @@ * @brief Implements directive handler maps for `server` and `location` blocks. * * @details Defines the concrete handler tables returned by - * `directive::serverHandlers()` and `directive::locationHandlers()`. + * directive::serverHandlers and directive::locationHandlers. * Each handler validates its arguments, reports precise diagnostics * (line/column + contextual snippet), and mutates the target - * configuration object (@ref Server or @ref Location). + * configuration object (Server or Location). * * This file also contains small internal helpers used by handlers: * - `resolveToAbsolute()` — canonicalizes filesystem paths. @@ -203,17 +203,6 @@ static void validateIPv4Address(const std::string& ip, int line, int column, * - Contain only alphanumeric characters after the dot. * - Not be just `"."` with no following characters. * - * Examples of valid extensions: - * - `.php` - * - `.py` - * - `.cgi` - * - * Examples of invalid extensions: - * - `""` (empty string) - * - `"."` (dot only, no extension name) - * - `"php"` (missing leading dot) - * - `.sh!` (contains invalid character `!`) - * * @param ext The extension string to validate (e.g., ".php"). * @param line Line number in the config file for error reporting. * @param column Column number in the config file for error reporting. diff --git a/src/config/tokenizer/Tokenizer.cpp b/src/config/tokenizer/Tokenizer.cpp index c6c72194..53f7357e 100644 --- a/src/config/tokenizer/Tokenizer.cpp +++ b/src/config/tokenizer/Tokenizer.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 01:06:09 by nlouis #+# #+# */ -/* Updated: 2025/08/18 12:32:37 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 16:28:41 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -448,7 +448,7 @@ Token Tokenizer::parseStringLiteral() { //=== Whitespace & Comment Helpers ======================================= /** - * @brief Skips a carriage return (`'\r'`). + * @brief Skips a carriage return (`'\\r'`). * @ingroup config_tokenizing */ void Tokenizer::skipCR() { diff --git a/src/config/validateConfig.cpp b/src/config/validateConfig.cpp index 9f5ce9f4..9d37e338 100644 --- a/src/config/validateConfig.cpp +++ b/src/config/validateConfig.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/20 23:23:50 by nlouis #+# #+# */ -/* Updated: 2025/08/18 14:03:41 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 16:31:16 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -241,7 +241,7 @@ bool isValidLabel(const std::string& label) { * @details Enforces RFC 1035-style constraints: * - Name must be non-empty and no longer than 253 characters. * - Cannot contain consecutive dots (`..`), which would create empty labels. - * - Each label (between dots) must satisfy @ref isValidLabel. + * - Each label (between dots) must satisfy isValidLabel. * * @param name Full domain/server name to validate. * @return `true` if the name follows domain rules, `false` otherwise. @@ -274,13 +274,13 @@ bool isServerNameValid(const std::string& name) { * - Non-empty names: Empty strings are not allowed. * - Printable characters only: No control characters (e.g., newlines, tabs). * - No whitespace: Spaces are not valid inside domain names. - * - RFC 1035 compliance via @ref isServerNameValid: + * - RFC 1035 compliance via isServerNameValid: * - Maximum length 253 characters. * - No empty labels (e.g. `"example..com"`). * - Labels up to 63 characters, only alphanumeric and `-`. * - No leading or trailing `-` in a label. * - * If any of these conditions are violated, a @ref ValidationError is thrown + * If any of these conditions are violated, a ValidationError is thrown * with a descriptive message indicating the invalid server name and the server * block index where it was found. * @@ -289,10 +289,6 @@ bool isServerNameValid(const std::string& name) { * @throws ValidationError if an invalid `server_name` is detected. * * @ingroup config_validation - * - * @see isServerNameValid - * @see isValidLabel - * @see splitLabels */ void validateServerNameFormat(const std::vector& servers) { for (std::size_t serverIndex = 0; serverIndex < servers.size(); ++serverIndex) { @@ -518,7 +514,7 @@ void validateRedirectCodes(const std::vector& servers) { * **GET, POST, DELETE**. * * If any other method (e.g., PUT, PATCH, OPTIONS, HEAD) is specified, - * the configuration is considered invalid and a @ref ValidationError is thrown. + * the configuration is considered invalid and a ValidationError is thrown. * * @param servers Vector of configured Server objects to validate. * diff --git a/src/utils/filesystemUtils.cpp b/src/utils/filesystemUtils.cpp index abbd635d..976a88b7 100644 --- a/src/utils/filesystemUtils.cpp +++ b/src/utils/filesystemUtils.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/13 09:39:07 by nlouis #+# #+# */ -/* Updated: 2025/08/18 12:05:15 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 15:58:05 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -296,25 +296,6 @@ std::string buildFilePath(const HttpRequest& request, const Location& loc) { return joinPath(loc_root, suffix); } -/** - * @brief Splits a path string into non-empty components. - * - * @details Parses the given path by splitting on '/' characters and collects - * all non-empty segments into a vector. Consecutive slashes and empty - * segments are skipped. The returned parts do not include leading or - * trailing separators and preserve the original segment order. - * - * @ingroup filesystem_utils - * - * @param path Path string to split (can be a URI or filesystem path). - * @return Vector of non-empty path segments in order of appearance. - * - * @note This function does not perform normalization; callers should sanitize - * and normalize the path (e.g., via @ref normalizePath) before splitting - * to remove `.` or `..` and collapse repeated slashes. - * @warning Leading or trailing slashes are not included in the result; the - * caller must track whether the path was absolute if needed. - */ /* static std::vector splitPath(const std::string& path) { std::vector parts; std::stringstream ss(path); From 333cbe0214c42c32c9c42e6e6b209ba1a8e236f0 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 20:52:04 +0300 Subject: [PATCH 17/32] Refactors Makefile structure and streamlines build process Reorganizes and updates the Makefile setup, including a new dedicated Makefile for to0nsa. Adds `.pytest_cache/` to `.gitignore` and includes a categorized help section for Makefile targets in the README. Removes redundant test and script files, consolidating functionality into the primary Makefile. Improves clarity, efficiency, and maintainability of the build and test workflows. --- .gitignore | 4 + Makefile => Makefile.hive | 10 +- Makefile.to0nsa | 212 +++++++++++++++++++++++++++++++++++ README.md | 84 ++++++-------- scripts/run_all_tests.sh | 71 ------------ scripts/run_webserv.sh | 44 -------- scripts/run_webserv_asan.sh | 17 --- scripts/run_webserv_tsan.sh | 17 --- scripts/run_webserv_ubsan.sh | 17 --- 9 files changed, 254 insertions(+), 222 deletions(-) rename Makefile => Makefile.hive (97%) create mode 100644 Makefile.to0nsa delete mode 100755 scripts/run_all_tests.sh delete mode 100644 scripts/run_webserv.sh delete mode 100755 scripts/run_webserv_asan.sh delete mode 100755 scripts/run_webserv_tsan.sh delete mode 100644 scripts/run_webserv_ubsan.sh diff --git a/.gitignore b/.gitignore index b8edd1c8..cdc22ccd 100644 --- a/.gitignore +++ b/.gitignore @@ -29,6 +29,7 @@ compile_commands.json __pycache__/ *.pyc *.pyo +.pytest_cache/ # --- Virtual environments --- .venv/ @@ -56,3 +57,6 @@ test_webserv/**/*.pyo /docs/html/ /docs/doxygen.warnings + +# --- Makefile for To0nsa config --- +Makefile diff --git a/Makefile b/Makefile.hive similarity index 97% rename from Makefile rename to Makefile.hive index b67e7612..c86c33e3 100644 --- a/Makefile +++ b/Makefile.hive @@ -1,12 +1,12 @@ # **************************************************************************** # # # # ::: :::::::: # -# Makefile :+: :+: :+: # +# Makefile.hive :+: :+: :+: # # +:+ +:+ +:+ # # By: nlouis +#+ +:+ +#+ # # +#+#+#+#+#+ +#+ # # Created: 2025/08/09 20:53:27 by nlouis #+# #+# # -# Updated: 2025/08/18 19:43:01 by nlouis ### ########.fr # +# Updated: 2025/08/18 20:44:34 by nlouis ### ########.fr # # # # **************************************************************************** # @@ -53,7 +53,7 @@ SRCS_CORE := \ src/core/Server.cpp \ src/core/main.cpp \ src/core/server_utils.cpp \ - src/core/runWebserv.cpp + src/core/webserv.cpp SRCS_HTTP := \ src/http/HttpRequest.cpp \ @@ -105,7 +105,7 @@ HEADERS := \ include/core/Location.hpp \ include/core/Server.hpp \ include/core/server_utils.hpp \ - include/core/runWebserv.hpp \ + include/core/webserv.hpp \ include/http/HttpRequest.hpp \ include/http/HttpRequestParser.hpp \ include/http/HttpResponse.hpp \ @@ -207,4 +207,4 @@ help: @echo " $(GREEN)make help$(RESET) → Show this help message 📚" .PHONY: all clean fclean re test format help install_test_deps --include $(DEPS) +-include $(DEPS) \ No newline at end of file diff --git a/Makefile.to0nsa b/Makefile.to0nsa new file mode 100644 index 00000000..a0110a42 --- /dev/null +++ b/Makefile.to0nsa @@ -0,0 +1,212 @@ +# **************************************************************************** # +# # +# ::: :::::::: # +# Makefile :+: :+: :+: # +# +:+ +:+ +:+ # +# By: nlouis +#+ +:+ +#+ # +# +#+#+#+#+#+ +#+ # +# Created: 2025/08/09 20:53:27 by nlouis #+# #+# # +# Updated: 2025/08/18 20:05:41 by nlouis ### ########.fr # +# # +# **************************************************************************** # + +# Compiler settings +CXX := c++ +CXXFLAGS := -Wall -Wextra -Werror -I include -std=c++20 \ + -O3 -DNDEBUG -flto -march=native +# We use -flto (link time optimization) for better performance. +# -O3 is for optimization, -DNDEBUG disables debug assertions. + +# Executable output +NAME := webserv +BINDIR := bin +TARGET := $(BINDIR)/$(NAME) + +# Object dir (deps live next to objects: .o + .d) +OBJDIR := objs + +# Colors +GREEN := \033[0;32m +CYAN := \033[0;36m +YELLOW := \033[1;33m +RED := \033[0;31m +RESET := \033[0m + +.DEFAULT_GOAL := all + +SRCS_CONFIG := \ + src/config/Config.cpp \ + src/config/normalizeConfig.cpp \ + src/config/validateConfig.cpp + +SRCS_CONFIG_PARSER := \ + src/config/parser/ConfigParseError.cpp \ + src/config/parser/ConfigParser.cpp \ + src/config/parser/directive_handler_table.cpp + +SRCS_CONFIG_TOKENIZER := \ + src/config/tokenizer/Tokenizer.cpp \ + src/config/tokenizer/token.cpp + +SRCS_CORE := \ + src/core/Location.cpp \ + src/core/Server.cpp \ + src/core/main.cpp \ + src/core/server_utils.cpp \ + src/core/runWebserv.cpp + +SRCS_HTTP := \ + src/http/HttpRequest.cpp \ + src/http/HttpRequestParser.cpp \ + src/http/HttpResponse.cpp \ + src/http/handleCgi.cpp \ + src/http/methodsHandler/handleDelete.cpp \ + src/http/methodsHandler/handleGet/generateAutoindex.cpp \ + src/http/methodsHandler/handleGet/handleGet.cpp \ + src/http/methodsHandler/handlePost/handleMultipartForm.cpp \ + src/http/methodsHandler/handlePost/handlePost.cpp \ + src/http/requestRouter.cpp \ + src/http/responseBuilder.cpp + +SRCS_NETWORK := \ + src/network/SocketManager.cpp \ + src/network/SocketManagerRequest.cpp \ + src/network/SocketManagerResponse.cpp \ + src/network/SocketManagerTimeouts.cpp \ + src/network/SocketManagerUtils.cpp + +SRCS_UTILS := \ + src/utils/Logger.cpp \ + src/utils/errorUtils.cpp \ + src/utils/filesystemUtils.cpp \ + src/utils/htmlUtils.cpp \ + src/utils/printInfo.cpp \ + src/utils/stringUtils.cpp \ + src/utils/urlUtils.cpp + +SRCS := \ + $(SRCS_CONFIG) \ + $(SRCS_CONFIG_PARSER) \ + $(SRCS_CONFIG_TOKENIZER) \ + $(SRCS_CORE) \ + $(SRCS_HTTP) \ + $(SRCS_NETWORK) \ + $(SRCS_UTILS) + +HEADERS := \ + include/config/Config.hpp \ + include/config/normalizeConfig.hpp \ + include/config/parser/ConfigParseError.hpp \ + include/config/parser/ConfigParser.hpp \ + include/config/parser/directive_handler_table.hpp \ + include/config/tokenizer/Tokenizer.hpp \ + include/config/tokenizer/token.hpp \ + include/config/validateConfig.hpp \ + include/core/Location.hpp \ + include/core/Server.hpp \ + include/core/server_utils.hpp \ + include/core/runWebserv.hpp \ + include/http/HttpRequest.hpp \ + include/http/HttpRequestParser.hpp \ + include/http/HttpResponse.hpp \ + include/http/Url.hpp \ + include/http/handleCgi.hpp \ + include/http/methodsHandler.hpp \ + include/http/requestRouter.hpp \ + include/http/responseBuilder.hpp \ + include/network/SocketManager.hpp \ + include/utils/Logger.hpp \ + include/utils/errorUtils.hpp \ + include/utils/filesystemUtils.hpp \ + include/utils/htmlUtils.hpp \ + include/utils/printInfo.hpp \ + include/utils/stringUtils.hpp \ + include/utils/urlUtils.hpp + +# ---- Derived files ----------------------------------------------------------- + +# Map each .cpp in $(SRCS) to its corresponding .o in $(OBJDIR). +# Note: patsubst here does NOT perform any filesystem wildcard expansion +# it just does a string substitution on the fixed list in $(SRCS). +# This is safe and deterministic, unlike $(wildcard) or $(shell find ...). +OBJS := $(patsubst src/%.cpp,$(OBJDIR)/%.o,$(SRCS)) +DEPS := $(OBJS:.o=.d) + +# ---- Rules ------------------------------------------------------------------ + +all: $(TARGET) + +$(TARGET): $(OBJS) + @mkdir -p $(BINDIR) + @$(CXX) $(CXXFLAGS) $^ -o $@ + @echo "$(CYAN)🚀 Built executable:$(RESET) $(TARGET)" + +$(OBJDIR)/%.o: src/%.cpp + @mkdir -p $(dir $@) + $(CXX) $(CXXFLAGS) -MMD -MP -c $< -o $@ + @echo "$(GREEN)🛠️ Compiled:$(RESET) $<" + +# Cleaning +clean: + @rm -rf $(OBJDIR) + @echo "$(YELLOW)🧹 Cleaned object and dependency files.$(RESET)" + +fclean: clean + @rm -rf $(TARGET) $(BINDIR) build + @echo "$(YELLOW)🗑️ Completely removed executables, binaries, build/.$(RESET)" + +re: fclean all + +# Python test dependencies +install_test_deps: + @echo "$(CYAN)📦 Creating isolated Python venv for tests...$(RESET)" + @python3 -m venv .venv || { \ + echo "$(RED)python3-venv is missing. Install it: sudo apt-get install -y python3-venv$(RESET)"; \ + exit 1; \ + } + @.venv/bin/python -m pip install --upgrade pip + @.venv/bin/pip install -r requirements-test.txt + +# Tests +test: all install_test_deps + @echo "$(CYAN)🧪 Launching web server in background...$(RESET)" + @./$(TARGET) ./test_webserv/tester/config/tester.conf & echo $$! > .webserv_test.pid + @sleep 1 + @echo "$(CYAN)🧪 Running Python test suite...$(RESET)" + @.venv/bin/python run_test.py || { \ + echo "$(RED)❌ Tests failed.$(RESET)"; \ + kill `cat .webserv_test.pid` >/dev/null 2>&1 || true; \ + rm -f .webserv_test.pid; \ + exit 1; \ + } + @echo "$(CYAN)🧹 Shutting down test server...$(RESET)" + @kill `cat .webserv_test.pid` >/dev/null 2>&1 || true + @rm -f .webserv_test.pid + @echo "$(GREEN)🏆 All tests passed successfully!$(RESET)" + +# Formatting (explicit file lists; no find/globs) +format: + @echo "$(CYAN)🎨 Formatting source files...$(RESET)" + @clang-format -i $(SRCS) $(HEADERS) + +# Help +help: + @echo "$(CYAN)📦 Build Targets:$(RESET)" + @echo " $(GREEN)make$(RESET) → Build the project 🚀" + @echo " $(GREEN)make re$(RESET) → Clean and rebuild everything 🔁" + @echo "" + @echo "$(CYAN)🧪 Test Targets:$(RESET)" + @echo " $(GREEN)make test$(RESET) → Build, install Python deps, run server + Python tests 🧪" + @echo "" + @echo "$(CYAN)🧹 Cleaning Targets:$(RESET)" + @echo " $(GREEN)make clean$(RESET) → Remove object and dependency files 🧹" + @echo " $(GREEN)make fclean$(RESET)→ Remove everything including binaries and build dirs 🗑️" + @echo "" + @echo "$(CYAN)🧹 Code Quality Targets:$(RESET)" + @echo " $(GREEN)make format$(RESET)→ Format all source files 🎨" + @echo "" + @echo "$(CYAN)📚 Other:$(RESET)" + @echo " $(GREEN)make help$(RESET) → Show this help message 📚" + +.PHONY: all clean fclean re test format help install_test_deps +-include $(DEPS) diff --git a/README.md b/README.md index 9a86f377..98b5306a 100644 --- a/README.md +++ b/README.md @@ -109,57 +109,6 @@ The configuration pipeline guarantees that only syntactically valid, normalized, ___ -## Build & Test Instructions - -### Build with Makefile - -```bash -make -./bin/webserv configs/default.conf -``` - -Available Makefile targets: - -### Build Modes - -| Command | Description | -|----|----| -| `make` | Build in release mode (optimized) | -| `make debug` | Build in debug mode (with `-g` and no optimizations) | -| `make debug_asan` | Build in debug mode with AddressSanitizer | -| `make debug_ubsan`| Build in debug mode with UndefinedBehaviorSanitizer | -| `make fast` | Fast build without dependency tracking (development only) | - -### Code quality - -| Command | Description | -|-----|----| -| `make format` | Format all `.cpp` and `.hpp` files using `clang-format` | - -### Run and Test - -| Command | Description | -|-----|----| -| `make run` | Build and run the web server | -| `make test` | Build and run all test binaries from `tests/` folder | -| `make sanitize` | Build and run under all sanitizers (ASAN, TSAN, UBSAN) | - -### Cleaning - -| Command | Description | -|-----|----| -| `make clean` | Remove all object files and dependency files | -| `make fclean`| Remove everything: binaries, builds, tests | -| `make re` | Full clean and rebuild | - -### Help - -| Command | Description | -|-----|---| -| `make help` | Displays a categorized list of all available `Makefile` targets | - -___ - ## Continuous Integration & Documentation This project leverages **GitHub Actions** to ensure code quality, stability, and up-to-date documentation. @@ -258,6 +207,39 @@ webserv ___ +## Build & Test Instructions + +### Build with Makefile + +```bash +make +./bin/webserv +``` + +> The default goal is `all`. The binary is produced at `bin/webserv`. + +### Available Makefile Targets + +| Command | Description | +| ------------------------ | ------------------------------------------------------------------------------------------------------------------------------- | +| `make` | Build the project in release mode (C++20, `-O3 -flto -DNDEBUG -march=native`). | +| `make re` | Clean everything and rebuild from scratch. | +| `make clean` | Remove object files and dependency files in `objs/`. | +| `make fclean` | Remove the executable, `bin/`, and all build artifacts (also runs `clean`). | +| `make install_test_deps` | Create a local Python venv in `.venv/` and install `requirements-test.txt`. | +| `make test` | Build, start the server in background with `./test_webserv/tester/config/tester.conf`, run `run_test.py`, then stop the server. | +| `make format` | Run `clang-format -i` on all listed sources and headers. | +| `make help` | Print a categorized list of available targets. | + +### Notes + +* Objects and auto-generated deps are stored under `objs/` (built via `-MMD -MP`). +* The build uses explicit source lists (no wildcards) for deterministic builds. +* The test rule writes the PID to `.webserv_test.pid` and cleans it up on success/failure. +* Ensure `python3-venv` and `clang-format` are installed on your system. + +___ + ## License This project is licensed under the terms of the [MIT License](LICENSE). diff --git a/scripts/run_all_tests.sh b/scripts/run_all_tests.sh deleted file mode 100755 index 1f443331..00000000 --- a/scripts/run_all_tests.sh +++ /dev/null @@ -1,71 +0,0 @@ -#!/bin/bash -# set -euo pipefail: -# -e : Exit immediately if any command exits with a non-zero status. -# -u : Treat unset variables as an error and exit immediately. -# -o pipefail : Return the exit status of the last command in a pipeline that failed -set -euo pipefail - -# Colors -GREEN="\033[0;32m" -RED="\033[0;31m" -CYAN="\033[0;36m" -YELLOW="\033[1;33m" -RESET="\033[0m" - -# Path to suppression file -SUPPRESSION_FILE="$(dirname "$0")/.asanignore" - -function run_tests() { - local preset="$1" - local build_dir="build/${preset}" - - echo -e "${CYAN}🔧 Preparing build for preset: ${preset}${RESET}" - - # Clean previous build if it exists - if [ -d "$build_dir" ]; then - echo -e "${YELLOW}🧹 Cleaning previous build directory: ${build_dir}${RESET}" - rm -rf "$build_dir" - fi - - echo -e "${CYAN}⚙️ Configuring with CMake preset: ${preset}${RESET}" - cmake --preset "$preset" -DBUILD_TESTING=ON - - echo -e "${CYAN}🏗️ Building project with preset: ${preset}${RESET}" - cmake --build --preset "$preset" - - echo -e "${CYAN}🧪 Running tests for preset: ${preset}${RESET}" - - # Set sanitizer suppression environment variables dynamically - if [[ "$preset" == "asan" ]]; then - export ASAN_OPTIONS="detect_leaks=1:suppressions=${SUPPRESSION_FILE}:exitcode=42:fast_unwind_on_malloc=0" - elif [[ "$preset" == "tsan" ]]; then - export TSAN_OPTIONS="suppressions=${SUPPRESSION_FILE}:exitcode=42" - elif [[ "$preset" == "ubsan" ]]; then - export UBSAN_OPTIONS="suppressions=${SUPPRESSION_FILE}:print_stacktrace=1:exitcode=42" - else - unset ASAN_OPTIONS TSAN_OPTIONS UBSAN_OPTIONS - fi - - if ! ctest --preset "$preset" --output-on-failure; then - echo -e "${RED}❌ Tests failed for preset: ${preset}${RESET}" - exit 1 - fi - - echo -e "${GREEN}✅ Tests passed for preset: ${preset}${RESET}\n" -} - -start_time=$(date +%s) - -echo -e "${CYAN}🚀 Starting full test run for all configurations...${RESET}\n" - -# List of presets to test -presets=("debug" "asan" "tsan" "ubsan") - -for preset in "${presets[@]}"; do - run_tests "$preset" -done - -end_time=$(date +%s) -elapsed=$((end_time - start_time)) - -echo -e "${GREEN}🏆 All builds and tests completed successfully in ${elapsed} seconds!${RESET}" diff --git a/scripts/run_webserv.sh b/scripts/run_webserv.sh deleted file mode 100644 index 2325147a..00000000 --- a/scripts/run_webserv.sh +++ /dev/null @@ -1,44 +0,0 @@ -#!/bin/bash -set -euo pipefail - -# Check if the default server binary exists -if [ ! -x "./bin/webserv" ]; then - echo "❌ Error: ./bin/webserv not found or not executable." - exit 1 -fi - -# If no arguments are provided, run the default server -if [ $# -eq 0 ]; then - exec ./bin/webserv -fi - -# Dispatch based on first argument -case "$1" in - asan) - if [ ! -x "./run_webserv_asan.sh" ]; then - echo "❌ Error: ./run_webserv_asan.sh not found or not executable." - exit 1 - fi - shift - exec ./run_webserv_asan.sh "$@" - ;; - tsan) - if [ ! -x "./run_webserv_tsan.sh" ]; then - echo "❌ Error: ./run_webserv_tsan.sh not found or not executable." - exit 1 - fi - shift - exec ./run_webserv_tsan.sh "$@" - ;; - ubsan) - if [ ! -x "./run_webserv_ubsan.sh" ]; then - echo "❌ Error: ./run_webserv_ubsan.sh not found or not executable." - exit 1 - fi - shift - exec ./run_webserv_ubsan.sh "$@" - ;; - *) - exec ./bin/webserv "$@" - ;; -esac diff --git a/scripts/run_webserv_asan.sh b/scripts/run_webserv_asan.sh deleted file mode 100755 index 3b1cbedd..00000000 --- a/scripts/run_webserv_asan.sh +++ /dev/null @@ -1,17 +0,0 @@ -#!/bin/bash -set -euo pipefail - -# Check if the default server binary exists -if [ ! -x "./bin/webserv" ]; then - echo "❌ Error: ./bin/webserv not found or not executable." - exit 1 -fi - -# Directory of this script (even if called from elsewhere) -SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)" - -# Export ASAN options -export ASAN_OPTIONS="detect_leaks=1:leak_check_at_exit=1:fast_unwind_on_malloc=0:abort_on_error=1:suppressions=${SCRIPT_DIR}/.asanignore" - -# Run webserv -exec "${SCRIPT_DIR}/bin/webserv" "$@" diff --git a/scripts/run_webserv_tsan.sh b/scripts/run_webserv_tsan.sh deleted file mode 100755 index 82f7d32a..00000000 --- a/scripts/run_webserv_tsan.sh +++ /dev/null @@ -1,17 +0,0 @@ -#!/bin/bash -set -euo pipefail - -# Check if the default server binary exists -if [ ! -x "./bin/webserv" ]; then - echo "❌ Error: ./bin/webserv not found or not executable." - exit 1 -fi - -# Directory of this script (even if called from elsewhere) -SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)" - -# Export TSAN options -export TSAN_OPTIONS="suppressions=${SCRIPT_DIR}/.asanignore:halt_on_error=1:exitcode=66" - -# Run webserv -exec "${SCRIPT_DIR}/bin/webserv" "$@" diff --git a/scripts/run_webserv_ubsan.sh b/scripts/run_webserv_ubsan.sh deleted file mode 100644 index 117c5f56..00000000 --- a/scripts/run_webserv_ubsan.sh +++ /dev/null @@ -1,17 +0,0 @@ -#!/bin/bash -set -euo pipefail - -# Check if the default server binary exists -if [ ! -x "./bin/webserv" ]; then - echo "❌ Error: ./bin/webserv not found or not executable." - exit 1 -fi - -# Directory of this script (even if called from elsewhere) -SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)" - -# Export UBSAN options -export UBSAN_OPTIONS="suppressions=${SCRIPT_DIR}/.asanignore:print_stacktrace=1:halt_on_error=1:exitcode=77" - -# Run webserv -exec "${SCRIPT_DIR}/bin/webserv" "$@" From 448d2b29d08e2630779d40cb0b2f176536b45bed Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 21:00:28 +0300 Subject: [PATCH 18/32] Update project structure and remove unused dependencies Refines project directory structure for clarity by renaming the unit test folder, updating documentation descriptions, and adding a Python test entry point. --- README.md | 15 ++++++++------- requirements.txt | 5 ----- 2 files changed, 8 insertions(+), 12 deletions(-) delete mode 100644 requirements.txt diff --git a/README.md b/README.md index 98b5306a..c7bedce0 100644 --- a/README.md +++ b/README.md @@ -188,20 +188,21 @@ webserv │ └── docs.yml # Doxygen documentation generation & GitHub Pages deploy ├── 📁 include/ # All public project headers, grouped by module (config, http, core, etc.) ├── 📁 src/ # Source files, mirrors the include/ structure -├── 📁 tests/ # Unit tests for various modules -├── 📁 configs/ # Test configuration files for parser/tokenizer -├── 📁 docs/ # Markdown documentation (DOCS.md, guides, etc.) +├── 📁 test_webserv/ # Unit tests +├── 📁 configs/ # Default config file +├── 📁 docs/ # Documentation generated by doxygen ├── .clang-format # Enforces formatting rules (4-space indent, K&R braces, etc.) ├── .editorconfig # Shared IDE/editor config for consistent style ├── .gitattributes # Defines merge/diff rules for Git (e.g. binary files) ├── .gitignore # Files and folders ignored by Git (e.g. build/, *.o) -├── ACTIONPLAN.md # Project-level planning or roadmap +├── ACTIONPLAN.md # Project-level planning/roadmap ├── DOXYGENSTYLEGUIDE.md # Doxygen conventions for documenting code ├── Doxyfile # Main config for Doxygen documentation generation -├── LICENSE # Project license (e.g. MIT, GPL) -├── Makefile # Build system entry point (defines targets like all, clean, fclean) -├── README.md # Main README shown on GitHub (overview, build, usage, etc.) +├── LICENSE # Project license +├── Makefile # Build system entry point +├── README.md # Main README ├── STYLEGUIDE.md # Coding conventions for naming, layout, formatting +├── run_test.py # Entrypoint for python tests ├── webserv.subject.pdf # Original subject specification for the project ``` diff --git a/requirements.txt b/requirements.txt deleted file mode 100644 index 7a4d7c6c..00000000 --- a/requirements.txt +++ /dev/null @@ -1,5 +0,0 @@ -requests -pytest -pytest-asyncio -aiohttp -asyncio From 07cb39614354bead85572a9d0acd9addb1981094 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 21:18:30 +0300 Subject: [PATCH 19/32] Improves README with server architecture details Adds a detailed overview of the core components and program entrypoint of Webserv, including responsibilities of `Server`, `Location`, and `runWebserv`. --- README.md | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index c7bedce0..e86bd6ad 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,4 @@ -# webserv +# Webserv [![Build Status](https://github.com/to0nsa/webserv/actions/workflows/build.yml/badge.svg)](https://github.com/to0nsa/webserv/actions/workflows/build.yml) [![Docs Status](https://github.com/to0nsa/webserv/actions/workflows/docs.yml/badge.svg?branch=main)](https://to0nsa.github.io/webserv/) @@ -12,6 +12,27 @@ ___ +## Core of Webserv + +| Component | Responsibility | +| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| **`Server`** | Represents a virtual host configuration. Manages binding (host + port), server names, error pages, body size limits, and a collection of `Location` blocks. | +| **`Location`** | Encapsulates route-specific configuration. Defines path matching, allowed HTTP methods, root directories, index files, redirects, CGI interpreters, and upload stores. | +| **`runWebserv`** | Orchestrates the execution of the web server: initializes `Server` objects from the parsed configuration, launches sockets, and enters the event loop. | + +___ + +### Program Entrypoint + + 1. Parse CLI arguments or fallback to default configuration. + 2. Load and normalize configuration (via config parser). + 3. Validate `Server` and `Location` objects. + 4. Call **`runWebserv()`** to start the server runtime. + +The program is designed so that **configuration and validation are complete before runtime begins**, ensuring that only consistent and safe server objects are passed to the execution loop. + +___ + ## Configuration Parsing Flow This section describes how the configuration parsing logic of **Webserv** works, including the step‑by‑step pipeline and the rules applied during parsing and validation. @@ -97,7 +118,9 @@ This section describes how the configuration parsing logic of **Webserv** works, -### Flow Overview +___ + +## Flow Overview 1. **Tokenizer** → breaks input into tokens. 2. **ConfigParser** → builds in‑memory `Config` with `Server` & `Location` objects. From ccb9d5d652e2247e3c1de6ff9b8d8c217aa74d48 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 22:54:23 +0300 Subject: [PATCH 20/32] Enhances SocketManager with detailed documentation Documents SocketManager's core functionality, including connection management, timeout handling, CGI integration, and response processing. --- include/doxygen_groups.hpp | 2 +- include/network/SocketManager.hpp | 192 ++++++++++++---- src/network/SocketManager.cpp | 272 ++++++++++++++++++++--- src/network/SocketManagerRequest.cpp | 305 ++++++++++++++++++++++++-- src/network/SocketManagerResponse.cpp | 167 +++++++++++++- src/network/SocketManagerTimeouts.cpp | 208 +++++++++++++++++- src/network/SocketManagerUtils.cpp | 151 ++++++++++++- 7 files changed, 1179 insertions(+), 118 deletions(-) diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index a2657dd8..53432c23 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -125,7 +125,7 @@ */ /** - * @defgroup network Network & Socket Abstraction + * @defgroup socker_mananager Network & Socket Abstraction * @brief Non-blocking I/O, socket management, and connection handling. * * @details Provides abstractions over system calls for listening sockets, diff --git a/include/network/SocketManager.hpp b/include/network/SocketManager.hpp index e6548038..cf82552b 100644 --- a/include/network/SocketManager.hpp +++ b/include/network/SocketManager.hpp @@ -3,13 +3,37 @@ /* ::: :::::::: */ /* SocketManager.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: ktieu +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:51:47 by irychkov #+# #+# */ -/* Updated: 2025/06/09 01:29:01 by ktieu ### ########.fr */ +/* Updated: 2025/08/18 22:51:45 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManager.hpp + * @brief Declares the SocketManager and per-client state for non-blocking I/O. + * + * @details The SocketManager owns the event loop (based on `poll(2)`) and + * multiplexes: + * - listening sockets (bind/listen for each configured host:port), + * - accepted client sockets (read → parse → route → respond), + * - CGI subprocess pipes (lifecycle, timeouts, finalization), + * - write-side backpressure and keep-alive. + * + * Each connected client is tracked with a @ref ClientInfo structure + * that aggregates request/response buffers, timers, CGI state, and + * file-stream handles for zero-copy file responses. + * + * The implementation guarantees: + * - single `poll()` over all FDs, read & write monitored together, + * - non-blocking I/O for sockets and CGI pipes, + * - accurate HTTP status codes and default error pages, + * - resilient behavior under stress and strict timeouts. + * + * @ingroup socker_mananager + */ + #pragma once #include "core/Server.hpp" // for Server @@ -26,55 +50,118 @@ #include // for size_t, time_t #include // for vector +//=== Tunables & limits ======================================================= + +/** + * @name Socket and protocol timeouts / limits + * @brief Runtime guards for robustness and resource control. + */ +///@{ + +/** @brief Generic inactivity timeout (seconds) for body recv / idle / send. */ #define TIMEOUT 20 + +/** @brief Max delay to complete request headers once bytes start arriving. */ #define HEADER_TIMEOUT_SECONDS 6 + +/** @brief Minimal sane header size to consider a request "non-empty". */ #define HEADER_MIN_LENGTH 15 + +/** @brief Maximum allowed header bytes before replying 431. */ #define HEADER_MAX_LENGTH 8192 + +/** @brief Receive buffer size for a single `recv()` call. */ #define RECV_BUFFER HEADER_MAX_LENGTH * 2 + +/** @brief Hard cap on the number of concurrent tracked client FDs. */ #define MAX_CLIENTS 1024 + +/** @brief Max time a CGI is allowed to run without finishing (seconds). */ #define CGI_TIMEOUT_SECONDS 45 +///@} + +// Forward declaration to avoid header cycles. + +/** + * @brief Per-client runtime state tracked by the SocketManager. + * + * @details Holds identifiers, rolling counters, parsing flags, request and + * response queues, CGI process handle, and file streaming context. + * All fields are mutated only by the owning SocketManager on the + * poll thread. + * + * @ingroup network + */ +struct ClientInfo { + //=== Core identifiers and timing ======================================== -class Location; + int client_fd; ///< Accepted client socket FD. + time_t lastRequestTime; ///< Last time we received any bytes from client. + time_t connectionStartTime; ///< First byte time for the *current* request header. + time_t lastSendAttemptTime; ///< Last time we attempted to `send()` bytes. -struct ClientInfo { - // Core identifiers and timing - int client_fd; - time_t lastRequestTime; - time_t connectionStartTime; - time_t lastSendAttemptTime; - - // Byte tracking - size_t headerBytesReceived; - size_t bodyBytesReceived; - size_t bytes_sent; - - // Request/response flow - bool headerComplete; - std::string requestBuffer; - std::string current_raw_response; - std::vector serversOnPort; - std::queue pendingRequests; - std::queue responses; - - // CGI - std::optional cgiProcess; - bool isCgiProcessRunning; - HttpRequest currentCgiRequest; - - // File I/O - std::ifstream file_stream; + //=== Byte tracking ======================================================= + + std::size_t headerBytesReceived; ///< Count of header bytes seen so far. + std::size_t bodyBytesReceived; ///< Count of body bytes seen so far. + std::size_t bytes_sent; ///< Bytes already sent from the current raw buffer. + + //=== Request/response flow ============================================== + + bool headerComplete; ///< True once `\r\n\r\n` found for the current request. + std::string requestBuffer; ///< Inbound raw buffer (may hold pipelined requests). + std::string current_raw_response; ///< Outbound raw response header/body (when not file). + + std::vector serversOnPort; ///< Virtual servers sharing the same listen FD. + std::queue pendingRequests; ///< Parsed but not yet processed requests. + std::queue responses; ///< Prepared responses waiting to be sent. + + //=== CGI state =========================================================== + + std::optional cgiProcess; ///< Active CGI process/pipe set (if any). + bool isCgiProcessRunning; ///< Convenience flag while CGI is alive. + HttpRequest currentCgiRequest; ///< Request currently handled by CGI. + + //=== File I/O for sendfile-like streaming =============================== + + std::ifstream file_stream; ///< Opened file for body streaming (CGI temp or static file). }; +/** + * @brief Poll-driven socket and CGI orchestrator. + * + * @details + * Responsibilities: + * - Create and bind all listening sockets for configured servers. + * - Accept clients and maintain a single `poll()` set for all descriptors. + * - Receive, enforce limits, parse (supports pipelining), and route requests. + * - Spawn/manage CGI, enforce CGI timeout, collect output, finalize response. + * - Stream responses (raw / file) with keep-alive and backpressure handling. + * - Apply precise timeout policy (idle/header/body/send). + * + * Error handling strategy: + * - Never block; on I/O errors or protocol violations, push an error response + * and cleanly close when required. + * - Protect the loop with wide catch blocks—FDs are closed on failure paths. + * + * @ingroup network + */ class SocketManager { public: + //=== Ctors / Dtor / Special members ===================================== SocketManager(void) = delete; SocketManager(const std::vector& servers); ~SocketManager(void); SocketManager(const SocketManager& other) = delete; SocketManager& operator=(const SocketManager& other) = delete; + //=== Main loop =========================================================== void run(); + + /** + * @brief Exception type for socket-level setup/runtime failures. + */ class SocketError : public std::exception { private: std::string _msg; @@ -85,28 +172,49 @@ class SocketManager { }; private: + //=== Poll & indices ====================================================== + std::vector _poll_fds; ///< Monitored file descriptors for poll(). - std::map> - _listen_map; ///< Maps listen fds to their corresponding servers - std::map _client_info; /// Stores all information about each client - std::map _fd_to_cgi; ///< Maps CGI stdout fds to client fds - // Setup & connection + /** + * @brief Listen FD → vhost set. + * + * @details For a given listen FD (host:port), all matching `Server` objects + * are stored here to support name-based routing. + */ + std::map> _listen_map; + + /** + * @brief Client FD → per-client state. + * + * @details Contains request/response queues, timers, CGI, and file state. + */ + std::map _client_info; + + /** + * @brief CGI stdout FD → owning client FD. + * + * @details Used to ignore/route events on CGI pipe descriptors inside the + * main poll loop. + */ + std::map _fd_to_cgi; + + //=== Setup & connection ================================================== void setupSockets(const std::vector& servers); void handleNewConnection(int listen_fd); - // Event handling + //=== Event handling ====================================================== bool handleClientData(int client_fd, size_t index); void handleCgiPollEvents(); void sendResponse(int client_fd, size_t index); void handlePollError(int fd, size_t index, short revents); - // Response helpers + //=== Response helpers ==================================================== void logResponseStatus(int status, int fd); bool sendFileResponse(int fd, size_t index, HttpResponse& response); bool sendRawResponse(int fd, size_t index, HttpResponse& response); - // Client lifecycle + //=== Client lifecycle ==================================================== void initializeClientInfo(int client_fd, int listen_fd); void cleanupClientConnectionClose(int client_fd, size_t index); void removePollFd(size_t index); @@ -114,25 +222,25 @@ class SocketManager { void cleanupCgiForClient(int client_fd); void resetRequestState(int client_fd); - // Request processing + //=== Request processing ================================================== bool receiveFromClient(int fd, size_t index); bool checkRequestLimits(int fd); bool parseAndQueueRequests(int client_fd); void processPendingRequests(int client_fd); - // Timeout checks + //=== Timeout checks ====================================================== bool checkClientTimeouts(int client_fd, size_t index); bool isHeaderTimeout(int fd, time_t now); bool isBodyTimeout(int fd, time_t now); bool isSendTimeout(int fd, time_t now); bool isIdleTimeout(int fd, time_t now); - // Request routing and CGI + //=== Request routing and CGI ============================================ bool handleCgiRequest(int client_fd, const HttpRequest& request, const Server& server, const Location& location); bool handleRequestErrorIfAny(int fd, int code, HttpRequest& req, const Server& server); bool shouldSpawnCgi(const HttpRequest& req, const Location& location); - // Error utility + //=== Error utility ======================================================= void respondError(int fd, int status_code); }; diff --git a/src/network/SocketManager.cpp b/src/network/SocketManager.cpp index a411b44e..e37b6eee 100644 --- a/src/network/SocketManager.cpp +++ b/src/network/SocketManager.cpp @@ -3,13 +3,31 @@ /* ::: :::::::: */ /* SocketManager.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: ktieu +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:51:20 by irychkov #+# #+# */ -/* Updated: 2025/06/09 01:29:07 by ktieu ### ########.fr */ +/* Updated: 2025/08/18 22:51:52 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManager.cpp + * @brief Implements the SocketManager core loop and server socket setup. + * + * @details + * Provides the main event-driven networking logic of Webserv. + * - Sets up listening sockets for configured servers. + * - Runs the non-blocking poll() loop to multiplex all I/O. + * - Handles new connections, client requests, and responses. + * - Integrates CGI handling, timeout checks, and error recovery. + * + * This file contains the **control flow backbone** of the web server: + * it ties together sockets, HTTP parsing, response building, CGI execution, + * and cleanup into a single poll-driven loop. + * + * @ingroup socker_mananager + */ + #include "network/SocketManager.hpp" #include "utils/Logger.hpp" // for LogLevel, Logger #include "utils/filesystemUtils.hpp" // for getCurrentTime @@ -25,25 +43,74 @@ #include // for close #include // for pair, make_pair +/// @internal +/// @brief Global flag controlling the server loop. +/// @details Set to `0` when SIGINT is received, causing the main loop to exit. static volatile sig_atomic_t running = 1; +/** + * @internal + * @brief Signal handler for process-level interrupts. + * + * @param signum Signal number received. + * + * @details + * - On `SIGINT` (Ctrl+C), sets the global `running` flag to `0` + * to trigger a graceful shutdown. + * - `SIGPIPE` is ignored in the constructor (see @ref SocketManager), + * to prevent crashes when clients disconnect abruptly. + */ static void signalHandler(int signum) { if (signum == SIGINT) running = 0; } -// Constructor: sets up sockets for each server defined in the config +/** + * @brief Constructs a new SocketManager. + * + * @details + * - Installs signal handlers (`SIGINT` for shutdown, `SIGPIPE` ignored). + * - Initializes listening sockets for all configured servers. + * + * @param servers List of configured @ref Server instances. + * + * @throws SocketError if socket creation, binding, or listening fails. + */ SocketManager::SocketManager(const std::vector& servers) { signal(SIGINT, signalHandler); signal(SIGPIPE, SIG_IGN); setupSockets(servers); } +/** + * @brief Destructor for SocketManager. + * + * @details + * Closes all file descriptors tracked in the poll set to + * ensure a clean shutdown. + */ SocketManager::~SocketManager() { for (const pollfd& pfd : _poll_fds) close(pfd.fd); } +/** + * @brief Handles error or hangup events reported by poll(). + * + * @details + * Interprets the poll revents mask for a given client file descriptor + * and logs the corresponding error type: + * - `POLLNVAL`: Invalid poll event on this fd. + * - `POLLERR`: Generic socket error. + * - `POLLHUP`: Client disconnected. + * + * After logging, the client connection is closed and cleaned up + * via @ref cleanupClientConnectionClose. + * + * @param fd The file descriptor that triggered the error event. + * @param index Index of the fd in the internal poll list. + * @param revents The bitmask of poll events received for the fd. + */ void SocketManager::handlePollError(int fd, size_t index, short revents) { if (revents & POLLNVAL) { Logger::logFrom(LogLevel::ERROR, "SocketManager", @@ -67,112 +134,171 @@ const char* SocketManager::SocketError::what() const throw() { return (_msg.c_str()); } -// Set up sockets for each server (host:port) +/** + * @brief Initializes and binds listening sockets for all configured servers. + * + * @details + * Iterates over the provided @ref Server instances and: + * - Ensures each (host, port) pair is only bound once. + * - Creates a non-blocking TCP socket (`SOCK_STREAM | SOCK_NONBLOCK`). + * - Enables `SO_REUSEADDR` to allow quick restarts on the same port. + * - Binds the socket to the host:port. + * - Puts the socket into listening mode with `listen()`. + * - Registers the socket in the internal poll list for event monitoring. + * - Groups servers sharing the same host:port into a virtual host list. + * + * @param servers List of servers to initialize listening sockets for. + * + * @throws SocketError If socket creation, setsockopt, bind, or listen fails. + * + * @note + * - "localhost" is explicitly mapped to `127.0.0.1`. + * - Duplicate host:port entries are skipped to avoid double-binding. + */ void SocketManager::setupSockets(const std::vector& servers) { - std::set> bound; + std::set> bound; // Track already-bound host:port pairs + for (size_t i = 0; i < servers.size(); ++i) { const std::string& host = servers[i].getHost(); int port = servers[i].getPort(); + // Skip if this host:port has already been bound std::pair key = std::make_pair(host, port); if (bound.count(key)) - continue; // Already bound, skip + continue; bound.insert(key); + // Create a non-blocking TCP socket int fd = socket(AF_INET, SOCK_STREAM | SOCK_NONBLOCK, 0); if (fd < 0) throw SocketError("socket() failed: " + std::string(std::strerror(errno))); - int opt = 1; // Reuse same address/port if the server is restarted quickly + // Allow socket address reuse to restart quickly after shutdown + int opt = 1; if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &opt, sizeof(opt)) < 0) { close(fd); throw SocketError("setsockopt() failed: " + std::string(std::strerror(errno))); } + // Prepare socket address structure sockaddr_in addr; addr.sin_family = AF_INET; addr.sin_port = htons(port); // Convert port to network byte order - // Convert hostname to IP address + + // Map "localhost" to loopback explicitly, otherwise use given host string if (host == "localhost") addr.sin_addr.s_addr = inet_addr("127.0.0.1"); else addr.sin_addr.s_addr = inet_addr(host.c_str()); - if (bind(fd, (sockaddr*) &addr, sizeof(addr)) < 0) { // Bind socket to IP:port + // Bind socket to the host:port + if (bind(fd, (sockaddr*) &addr, sizeof(addr)) < 0) { close(fd); throw SocketError("bind() failed on " + host + ":" + std::to_string(port) + ": " + strerror(errno)); } - if (listen(fd, SOMAXCONN) < 0) { // Start listening for incoming connections + // Put socket into listening mode + if (listen(fd, SOMAXCONN) < 0) { close(fd); throw SocketError("listen() failed: " + std::string(std::strerror(errno))); } - // Register fd in poll list + // Register fd in poll list for event monitoring _poll_fds.push_back((pollfd){fd, POLLIN, 0}); - // Collect all servers for this host:port + + // Collect all servers configured on this host:port (virtual hosts) std::vector vhosts; for (size_t j = 0; j < servers.size(); ++j) { if (servers[j].getHost() == host && servers[j].getPort() == port) vhosts.push_back(servers[j]); } + // Associate this listening fd with its vhost set _listen_map[fd] = vhosts; + Logger::logFrom(LogLevel::INFO, "SocketManager", "Listening on " + host + ":" + std::to_string(port)); } } +/** + * @brief Runs the main event loop of the server. + * + * @details + * This method drives the **non-blocking, poll()-based event loop** of Webserv. + * - Uses a single `poll()` call to multiplex all I/O (listening sockets, client FDs, CGI pipes). + * - Periodically checks for CGI activity and client timeouts. + * - Handles: + * - New connections (on listening sockets), + * - Read events (incoming client data), + * - Write events (pending responses), + * - Error/hangup events (POLLERR, POLLHUP, POLLNVAL). + * - Ensures resilience with try/catch around per-FD handling. + * + * The loop continues until the global `running` flag is cleared (SIGINT). + * + * @throws SocketError If `poll()` fails unexpectedly (other than EINTR). + */ void SocketManager::run() { while (running) { + // Wait up to 1s for activity on any registered FD int n = poll(&_poll_fds[0], _poll_fds.size(), 1000); if (n < 0) { if (errno == EINTR) { + // Interrupted by signal → trigger graceful shutdown running = 0; continue; } throw SocketError("poll() failed: " + std::string(std::strerror(errno))); } + // First, handle any active CGI processes (timeout, completion, cleanup) handleCgiPollEvents(); - // Then handle sockets — but skip *all* CGI FDs before doing error/HUP checks + // Process poll events for each file descriptor in reverse order for (size_t i = _poll_fds.size(); i-- > 0;) { try { - short revents = _poll_fds[i].revents; + short revents = _poll_fds[i].revents; // events that fired int current_fd = _poll_fds[i].fd; + + // Check if this client has hit a timeout (idle/header/body/send) if (checkClientTimeouts(current_fd, i)) { - _poll_fds[i].events |= - POLLOUT; // If connection keep-alive but client idle we close + // Mark for POLLOUT → we will close after sending error/timeout response + _poll_fds[i].events |= POLLOUT; } - // **FIX**: skip CGI pipe FDs entirely + // Skip CGI pipe descriptors entirely — they’re handled in handleCgiPollEvents() if (_fd_to_cgi.contains(current_fd)) { continue; } - // Now error/hangup on *client* sockets + // Handle socket-level errors or hangups if (revents & POLLERR || revents & POLLHUP || revents & POLLNVAL) { handlePollError(current_fd, i, revents); continue; } + // Incoming data available if (revents & POLLIN) { if (_listen_map.count(current_fd)) { + // This is a listening socket → accept new connection handleNewConnection(current_fd); } else { + // Existing client → read request if (!handleClientData(current_fd, i)) - continue; - // we have a response queued, request poll‐out + continue; // client closed or error, skip further handling + // At least one response is now queued → request POLLOUT _poll_fds[i].events |= POLLOUT; } } + // Ready to write response if ((revents & POLLOUT) && !_client_info[current_fd].responses.empty()) { sendResponse(current_fd, i); } } catch (const std::exception& e) { + // Per-FD error resilience: log and cleanup Logger::logFrom(LogLevel::ERROR, "SocketManager", "Exception in poll loop for fd: " + std::to_string(_poll_fds[i].fd) + ": " + e.what()); @@ -189,70 +315,154 @@ void SocketManager::run() { Logger::logFrom(LogLevel::INFO, "SocketManager", "Shutting down server"); } +/** + * @brief Initializes tracking information for a newly accepted client. + * + * @details + * Creates or resets a @ref ClientInfo entry for the given client file descriptor: + * - Stores the client FD for later reference. + * - Initializes timestamps (`lastRequestTime`, `connectionStartTime`). + * - Resets request/response state (header/body counters, completion flags). + * - Clears send progress tracking (`bytes_sent`). + * - Associates the client with all servers (virtual hosts) listening on the + * same port as the accept()'ed socket. + * + * @param client_fd File descriptor of the newly accepted client socket. + * @param listen_fd File descriptor of the listening socket that accepted this client. + */ void SocketManager::initializeClientInfo(int client_fd, int listen_fd) { - ClientInfo& info = _client_info[client_fd]; + ClientInfo& info = _client_info[client_fd]; // Create/lookup client state entry + + // Basic identification + info.client_fd = client_fd; - info.client_fd = client_fd; + // Initialize timers for connection and activity info.lastRequestTime = getCurrentTime(); info.connectionStartTime = getCurrentTime(); + + // Reset counters and parsing state info.headerBytesReceived = 0; info.bodyBytesReceived = 0; info.headerComplete = false; - info.bytes_sent = 0; - info.serversOnPort = _listen_map[listen_fd]; + + // Reset response send tracking + info.bytes_sent = 0; + + // Assign list of servers bound to the same listen_fd (vhost support) + info.serversOnPort = _listen_map[listen_fd]; } -// Accept new client and add to poll list +/** + * @brief Accepts and registers a new client connection. + * + * @details + * Called when a listening socket signals readiness (`POLLIN`): + * - Calls `accept()` to retrieve a new client socket FD. + * - Handles resource exhaustion gracefully: + * - `EMFILE` / `ENFILE`: process/system FD limit reached → log and drop. + * - `MAX_CLIENTS` limit: too many clients → reject and close FD. + * - Initializes the new client state via @ref initializeClientInfo. + * - Adds the client FD to the poll set, monitoring for read events. + * + * @param listen_fd The listening socket FD on which the connection was accepted. + */ void SocketManager::handleNewConnection(int listen_fd) { + // Accept a pending connection on the listening socket int client_fd = accept(listen_fd, NULL, NULL); if (client_fd < 0) { if (errno == EMFILE || errno == ENFILE) { - // We’ve hit the per‐process or system FD limit. + // System or process has run out of file descriptors Logger::logFrom( LogLevel::ERROR, "SocketManager", "Out of file descriptors (accept failed: " + std::string(strerror(errno)) + ")"); return; } + // Generic accept() failure Logger::logFrom(LogLevel::ERROR, "SocketManager", "accept() failed: " + std::string(std::strerror(errno))); return; } + // Enforce application-level connection limit if (_poll_fds.size() >= MAX_CLIENTS) { Logger::logFrom(LogLevel::WARN, "SocketManager", "Maximum client limit reached. Rejecting connection."); - close(client_fd); // Optionally send HTTP error before closing (nice but optional) 503 + close(client_fd); // Could optionally send an HTTP 503 before closing return; } + // Initialize per-client state tracking initializeClientInfo(client_fd, listen_fd); + + // Register client FD into poll list, listening for incoming data _poll_fds.push_back((pollfd){client_fd, POLLIN, 0}); + + // Debug log: report accepted client and total open connections Logger::logFrom(LogLevel::kDEBUG, "SocketManager", "Accept returned fd: " + std::to_string(client_fd) + " | current open clients: " + std::to_string(_client_info.size())); } +/** + * @brief Processes incoming data for a client connection. + * + * @details + * Called when a client socket is readable (`POLLIN`): + * 1. Reads available data from the client using @ref receiveFromClient. + * - If the client disconnects or a read error occurs → return `false` (FD will be closed). + * 2. Attempts to parse one or more complete HTTP requests from the buffer + * via @ref parseAndQueueRequests. + * - If parsing is incomplete, waits for more data. + * - If a parse error occurs, queues an error response. + * 3. Processes all queued requests (except those blocked by CGI execution) + * via @ref processPendingRequests. + * + * Exception safety: + * - Catches `std::bad_alloc` (memory exhaustion), `std::exception`, and unknown errors. + * - On error: disables further POLLIN events for this client and queues a `500 Internal Server + * Error`. + * + * @param client_fd File descriptor of the client socket. + * @param index Index of the client FD in the internal poll list. + * + * @return `true` if processing completed (even with errors queued), + * `false` if the connection should be closed. + */ bool SocketManager::handleClientData(int client_fd, size_t index) { try { + // 1) Receive raw bytes from the client socket if (!receiveFromClient(client_fd, index)) - return false; + return false; // Client disconnected or fatal recv() error + + // 2) Parse requests and queue them for processing if (!parseAndQueueRequests(client_fd)) - return false; + return false; // Incomplete request → wait for more data + + // 3) Execute queued requests (routing, CGI, response building) processPendingRequests(client_fd); - return (true); + + return true; } catch (const std::bad_alloc& e) { + // Out-of-memory while processing this client Logger::logFrom(LogLevel::ERROR, "SocketManager", "Memory allocation failed while handling client " + std::to_string(client_fd) + ": " + e.what()); } catch (const std::exception& e) { + // Generic runtime error Logger::logFrom(LogLevel::ERROR, "SocketManager", "Exception while handling client " + std::to_string(client_fd) + ": " + e.what()); } catch (...) { + // Any other unknown error Logger::logFrom(LogLevel::ERROR, "SocketManager", "Unknown exception while handling client " + std::to_string(client_fd)); } + + // Disable further reads from this client (avoid repeated errors) _poll_fds[index].events &= ~POLLIN; + + // Queue an internal server error response respondError(client_fd, 500); - return (true); + + return true; // Keep connection alive long enough to send error response } diff --git a/src/network/SocketManagerRequest.cpp b/src/network/SocketManagerRequest.cpp index ea63db9d..04c26cad 100644 --- a/src/network/SocketManagerRequest.cpp +++ b/src/network/SocketManagerRequest.cpp @@ -6,10 +6,33 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 14:53:38 by irychkov #+# #+# */ -/* Updated: 2025/06/09 00:47:10 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 22:51:58 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManagerRequest.cpp + * @brief Implements request handling logic for SocketManager. + * + * @details + * This file contains all methods of @ref SocketManager related to + * **client request processing**: + * - Receiving raw data from clients (@ref receiveFromClient). + * - Parsing HTTP requests into structured @ref HttpRequest objects + * (@ref parseAndQueueRequests). + * - Executing and routing pending requests, including CGI support + * (@ref processPendingRequests, @ref handleCgiRequest). + * - Managing CGI process polling, timeouts, and finalization + * (@ref handleCgiPollEvents). + * - Handling request errors and generating error responses + * (@ref handleRequestErrorIfAny). + * + * It forms the **read-path of the event loop**, turning client bytes into + * application-level requests and responses. + * + * @ingroup socker_mananager + */ + #include "core/Location.hpp" // for Location #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest @@ -35,44 +58,81 @@ #include // for pair #include // for vector +/** + * @brief Receives raw data from a client socket. + * + * @details + * - Performs a non-blocking `recv()` on the given client FD. + * - Handles connection termination cases: + * - `bytes == 0`: client closed the connection → cleanup and return `false`. + * - `bytes < 0`: read error → cleanup and return `false`. + * - Appends received data to the client’s `requestBuffer`. + * - Tracks request progress: + * - Updates header vs. body byte counters. + * - Marks `headerComplete` when the `\r\n\r\n` delimiter is first encountered. + * + * @param client_fd The client socket file descriptor. + * @param index Index of the client FD in the poll list (needed for cleanup). + * + * @return `true` if data was successfully received and buffered, + * `false` if the connection should be closed. + * + * @throws None. Errors are logged, and cleanup is performed internally. + */ bool SocketManager::receiveFromClient(int client_fd, size_t index) { char buffer[RECV_BUFFER]; + + // Update last activity timestamp (used for idle timeout detection) _client_info[client_fd].lastRequestTime = getCurrentTime(); + + // Non-blocking read from the client int bytes = recv(client_fd, buffer, sizeof(buffer) - 1, MSG_DONTWAIT); + if (bytes == 0) { + // Client performed an orderly shutdown Logger::logFrom(LogLevel::INFO, "SocketManager", "Client fd " + std::to_string(client_fd) + " disconnected."); cleanupClientConnectionClose(client_fd, index); return false; } if (bytes < 0) { + // Fatal read error Logger::logFrom(LogLevel::ERROR, "SocketManager", std::string("recv() failed: ") + std::strerror(errno)); cleanupClientConnectionClose(client_fd, index); return false; } + + // Null-terminate buffer for safe string construction buffer[bytes] = '\0'; + // Append the new chunk to the client's request buffer std::string single_msg(buffer, bytes); _client_info[client_fd].requestBuffer += single_msg; + + // Detect end of HTTP headers ("\r\n\r\n") size_t headerEndPos = _client_info[client_fd].requestBuffer.find("\r\n\r\n"); if (headerEndPos == std::string::npos) { + // Still receiving headers if (_client_info[client_fd].headerBytesReceived == 0) { + // First time seeing header data → mark connection start _client_info[client_fd].connectionStartTime = getCurrentTime(); } _client_info[client_fd].headerBytesReceived += bytes; } else { if (!_client_info[client_fd].headerComplete) { + // Header just completed in this recv() size_t fullHeaderSize = headerEndPos + 4; size_t oldBufferSize = _client_info[client_fd].requestBuffer.size() - bytes; size_t headerBytesThisTime = std::max((ssize_t) 0, (ssize_t) (fullHeaderSize - oldBufferSize)); + _client_info[client_fd].headerBytesReceived += headerBytesThisTime; _client_info[client_fd].bodyBytesReceived += (bytes - headerBytesThisTime); _client_info[client_fd].headerComplete = true; } else { - // Header already counted, this must be body + // Header already completed → this is pure body data _client_info[client_fd].bodyBytesReceived += bytes; } } @@ -80,36 +140,84 @@ bool SocketManager::receiveFromClient(int client_fd, size_t index) { return true; } +/** + * @brief Finds the best-matching Location for a given request path. + * + * @details + * Iterates through all @ref Location objects configured in the given @ref Server + * and selects the one whose `path` prefix matches the request path with the + * **longest match length** (longest-prefix match). + * + * Matching logic: + * - A location matches if `normalizePath(loc.getPath())` is a prefix of `path`. + * - If multiple locations match, the one with the longest normalized path is chosen. + * - If no location matches, returns `nullptr`. + * + * @param path The normalized request target path (e.g. `/images/foo.jpg`). + * @param server The server whose configured locations should be searched. + * + * @return Pointer to the best matching @ref Location, or `nullptr` if none match. + */ static const Location* findMatchingLocation(const std::string& path, const Server& server) { const Location* best = nullptr; size_t max = 0; + + // Iterate over all configured locations for this server for (const Location& loc : server.getLocations()) { + // Check if location path is a prefix of the request path if (path.rfind(normalizePath(loc.getPath()), 0) == 0 && normalizePath(loc.getPath()).size() > max) { + // Keep track of the longest prefix match so far best = &loc; max = normalizePath(loc.getPath()).size(); } } - return best; + + return best; // nullptr if no location matched } +/** + * @brief Processes all pending HTTP requests for a client. + * + * @details + * Runs through the client's queued @ref HttpRequest objects and: + * 1. Skips processing if a CGI process is currently running + * (only one CGI may run per client at a time). + * 2. Checks for parse errors in the request: + * - If present, generates an error response via @ref handleRequestErrorIfAny. + * 3. Resolves the request target to the best-matching @ref Location + * using @ref findMatchingLocation. + * - If no location matches, responds with 404. + * 4. If the request must be served by a CGI: + * - Marks client as running CGI, initializes the process with + * @ref handleCgiRequest, and defers further processing. + * 5. Otherwise, handles the request normally via @ref handleRequest and + * queues the @ref HttpResponse. + * + * The loop continues until: + * - No more pending requests, OR + * - A CGI request is started, OR + * - A response requires connection close. + * + * @param client_fd File descriptor of the client whose requests are processed. + */ void SocketManager::processPendingRequests(int client_fd) { ClientInfo& client = _client_info[client_fd]; - // As long as there is at least one pending request AND no CGI is currently running: + // Process while there are queued requests and no active CGI process while (!client.pendingRequests.empty() && !client.isCgiProcessRunning) { HttpRequest nextReq = client.pendingRequests.front(); const Server& server = client.serversOnPort[nextReq.getMatchedServerIndex()]; - // nextReq.printRequest(); - + // Handle requests with parsing errors int code = nextReq.getParseErrorCode(); if (code != 0) { if (handleRequestErrorIfAny(client_fd, code, nextReq, server)) - return; - continue; + return; // stop if error requires closing connection + continue; // otherwise, process next pending request } + // Match request path against server locations const Location* location = findMatchingLocation(normalizePath(nextReq.getPath()), server); if (!location) { if (handleRequestErrorIfAny(client_fd, 404, nextReq, server)) @@ -117,28 +225,62 @@ void SocketManager::processPendingRequests(int client_fd) { continue; } + // If request matches a CGI location → launch CGI process if (shouldSpawnCgi(nextReq, *location)) { client.currentCgiRequest = nextReq; client.isCgiProcessRunning = true; bool ok = handleCgiRequest(client_fd, nextReq, server, *location); client.pendingRequests.pop(); if (!ok) - continue; - return; + continue; // CGI init failed → continue with next request + return; // CGI started → stop until it finishes } + // Otherwise, handle request normally and enqueue response HttpResponse resp = handleRequest(nextReq, server); client.responses.push(resp); client.pendingRequests.pop(); + + // If response signals "Connection: close" → stop processing if (resp.isConnectionClose()) return; } } +/** + * @brief Parses buffered client data into HTTP requests and queues them. + * + * @details + * This function repeatedly attempts to parse complete HTTP requests + * from a client’s `requestBuffer`: + * + * 1. **Check request limits:** + * - If headers exceed configured limits, clears buffer and pushes + * a `431 Request Header Fields Too Large` error. + * + * 2. **Parse next request:** + * - Uses @ref HttpRequestParser::parse to decode from the buffer. + * - On success: consumes parsed bytes, resets state, and queues the request. + * - On incomplete parse (`errorCode == 0`): returns `false` to wait for more data. + * - On parse error (`400`, `411`, `413`, `415`, etc.): + * - Marks the request with the error code. + * - Either clears the whole buffer (fatal errors) or erases consumed bytes. + * - Resets state and queues the error request. + * + * 3. **Loop continuation:** + * - If more data remains in `requestBuffer` (and headers are present), + * parsing continues until buffer is exhausted or incomplete. + * + * @param client_fd File descriptor of the client whose buffer is parsed. + * + * @return `true` if parsing succeeded or error requests were queued, + * `false` if parsing is incomplete and more data is required. + */ bool SocketManager::parseAndQueueRequests(int client_fd) { ClientInfo& client = _client_info[client_fd]; while (true) { + // 1) enforce request size/limit rules if (checkRequestLimits(client_fd)) { client.requestBuffer.clear(); resetRequestState(client_fd); @@ -149,34 +291,41 @@ bool SocketManager::parseAndQueueRequests(int client_fd) { int errorCode = 0; std::size_t consumedBytes = 0; + // 2) attempt to parse one HTTP request from the buffer bool ok = HttpRequestParser::parse(request, client.requestBuffer, client.serversOnPort, errorCode, consumedBytes); if (!ok) { if (errorCode == 0) - return false; // incomplete + return false; // Incomplete request → wait for more data + // Fatal parse error (e.g. invalid headers, bad content length) request.setParseErrorCode(errorCode); request.printRequest(); + // Clear buffer for hard errors, else drop only consumed part if (errorCode == 415 || errorCode == 411 || errorCode == 400 || errorCode == 413) client.requestBuffer.clear(); else client.requestBuffer.erase(0, consumedBytes); + // Reset parsing state and queue error request resetRequestState(client_fd); client.pendingRequests.push(request); + // Stop if no headers left in buffer if (client.requestBuffer.find("\r\n\r\n") == std::string::npos) break; continue; } + // Success: remove consumed bytes, reset state, queue request client.requestBuffer.erase(0, consumedBytes); resetRequestState(client_fd); client.pendingRequests.push(request); + // Stop if buffer doesn’t contain another full header if (client.requestBuffer.find("\r\n\r\n") == std::string::npos) break; } @@ -184,25 +333,57 @@ bool SocketManager::parseAndQueueRequests(int client_fd) { return true; } +/** + * @brief Monitors and manages active CGI processes for all clients. + * + * @details + * Iterates over all connected clients and checks if a CGI process + * is associated with them. For each active CGI: + * + * 1. **Timeout check:** + * - If the process has been idle longer than @ref CGI_TIMEOUT_SECONDS, + * it is killed and a `504 Gateway Timeout` response is queued. + * + * 2. **Completion check:** + * - If @ref CGI::tryTerminateCgi indicates the process is done, + * finalize output with @ref CGI::finalizeCgi, push response, + * cleanup process state, and re-enable `POLLOUT` to send data. + * - After cleanup, if more requests remain queued for the client, + * continue processing them with @ref processPendingRequests. + * + * 3. **Exception safety:** + * - If any error occurs during CGI handling, logs the error, + * queues a `500 Internal Server Error`, and cleans up CGI state. + * + * Integration: + * - This method is called once per poll loop iteration before + * handling socket-level events. + * - It ensures CGI subprocesses do not block the server’s main loop. + * + * @see processPendingRequests + */ void SocketManager::handleCgiPollEvents() { for (auto& [client_fd, client] : _client_info) { if (!client.cgiProcess) - continue; + continue; // Skip clients without an active CGI try { CgiProcess& cgi = *client.cgiProcess; const Server& server = client.serversOnPort[client.currentCgiRequest.getMatchedServerIndex()]; + // 1. Timeout check if (getCurrentTime() - cgi.last_activity > CGI_TIMEOUT_SECONDS) { Logger::logFrom(LogLevel::WARN, "CGI", "Timeout. Killing CGI process for fd: " + std::to_string(client_fd)); client.responses.push(ResponseBuilder::generateError(504, server, {})); - CGI::errorOnCgi(cgi); - client.cgiProcess.reset(); + CGI::errorOnCgi(cgi); // Kill process with error + client.cgiProcess.reset(); // Drop process handle client.isCgiProcessRunning = false; client.currentCgiRequest = HttpRequest(); + + // Ensure POLLOUT so timeout response gets sent for (auto& pfd : _poll_fds) { if (pfd.fd == client_fd) { pfd.events |= POLLOUT; @@ -212,14 +393,16 @@ void SocketManager::handleCgiPollEvents() { continue; } + // 2. Completion check if (CGI::tryTerminateCgi(cgi)) { HttpResponse resp = CGI::finalizeCgi(cgi, server, client.currentCgiRequest); client.responses.push(resp); - CGI::cleanupCgi(cgi); + CGI::cleanupCgi(cgi); // Free temp files/resources client.cgiProcess.reset(); client.isCgiProcessRunning = false; client.currentCgiRequest = HttpRequest(); + // Mark socket ready for sending CGI response for (auto& pfd : _poll_fds) { if (pfd.fd == client_fd) { pfd.events |= POLLOUT; @@ -227,6 +410,7 @@ void SocketManager::handleCgiPollEvents() { } } + // Resume processing queued requests (pipelined) size_t idx = 0; for (; idx < _poll_fds.size(); ++idx) { if (_poll_fds[idx].fd == client_fd) @@ -236,11 +420,14 @@ void SocketManager::handleCgiPollEvents() { processPendingRequests(client_fd); } } catch (const std::exception& e) { + // 3. Exception handling: fail safe Logger::logFrom(LogLevel::ERROR, "SocketManager", "Exception during CGI handling for fd " + std::to_string(client_fd) + ": " + e.what()); - respondError(client_fd, 500); - cleanupCgiForClient(client_fd); + respondError(client_fd, 500); // Queue internal error + cleanupCgiForClient(client_fd); // Force cleanup + + // Mark socket for POLLOUT to flush error response for (auto& pfd : _poll_fds) { if (pfd.fd == client_fd) { pfd.events |= POLLOUT; @@ -256,38 +443,112 @@ bool hasFullChunkedBody(const std::string& buffer, size_t bodyStart) { return end != std::string::npos; } +/** + * @brief Initializes and starts a CGI process for a client request. + * + * @details + * - Allocates a new @ref CgiProcess in the client state. + * - Attempts to initialize the CGI environment via @ref CGI::initCgiProcess. + * - On success: leaves the process active and returns `true`. + * - On failure: + * - Logs an error with details. + * - Queues an error @ref HttpResponse (e.g. `500`, or errorCode returned). + * - Cleans up CGI state and clears the current request. + * - Returns `false`. + * + * @param client_fd The client file descriptor. + * @param request The HTTP request to be executed by CGI. + * @param server The server context for this request. + * @param location The location context where CGI execution is configured. + * + * @return `true` if CGI initialization succeeded, + * `false` if an error response was queued instead. + * + * @see handleCgiPollEvents, processPendingRequests + */ bool SocketManager::handleCgiRequest(int client_fd, const HttpRequest& request, const Server& server, const Location& location) { ClientInfo& client = _client_info[client_fd]; - client.cgiProcess.emplace(); + client.cgiProcess.emplace(); // Allocate a new CGI process slot - int errorCode = 500; + int errorCode = 500; // Default fallback error code if (!CGI::initCgiProcess(*client.cgiProcess, request, server, location, _poll_fds, errorCode)) { + // Failed to spawn CGI → log and queue error response Logger::logFrom(LogLevel::ERROR, "SocketManager", "[CGI] Failed to initialize CGI process for client_fd " + std::to_string(client_fd) + " with script: " + location.getPath()); + HttpResponse err = ResponseBuilder::generateError(errorCode, server, request); _client_info[client_fd].responses.push(err); + + // Reset CGI state so client can recover client.cgiProcess.reset(); client.isCgiProcessRunning = false; - client.currentCgiRequest = HttpRequest(); // clears request - return false; // error response queued + client.currentCgiRequest = HttpRequest(); + + return false; // Error response queued instead of running CGI } - return true; // handled as CGI + return true; // CGI successfully started, will be managed by handleCgiPollEvents() } +/** + * @brief Handles an HTTP request error by generating and queuing an error response. + * + * @details + * - Builds an error @ref HttpResponse using @ref ResponseBuilder::generateError. + * - Pushes the response into the client’s response queue. + * - Removes the faulty request from the pending request queue. + * - Returns whether the response requires closing the connection (e.g. HTTP/1.0 without + * keep-alive). + * + * @param fd File descriptor of the client connection. + * @param code HTTP status code to return (e.g. 400, 404, 413). + * @param req The offending HTTP request (may be partially parsed). + * @param server The server context used for error page resolution. + * + * @return `true` if the error response requires closing the connection, + * `false` otherwise (keep-alive). + */ bool SocketManager::handleRequestErrorIfAny(int fd, int code, HttpRequest& req, const Server& server) { + // Generate error response for this request HttpResponse err = ResponseBuilder::generateError(code, server, req); + + // Queue the response to be sent back to the client _client_info[fd].responses.push(err); + + // Drop the faulty request from the pending queue _client_info[fd].pendingRequests.pop(); + // Decide if the connection must be closed after sending this response return err.isConnectionClose(); } +/** + * @brief Determines whether a request should be handled by a CGI process. + * + * @details + * A request qualifies for CGI execution if: + * - The path can be resolved to a valid absolute file path (`resolveAbsolutePath` not empty). + * - The HTTP method is either `GET` or `POST` (other methods are not CGI-eligible). + * - The resolved path matches a CGI-enabled location (`isCgiRequest`). + * + * If all conditions are met, returns `true` so the request is delegated + * to @ref handleCgiRequest. Otherwise, it should be served as a normal static/dynamic response. + * + * @param req The incoming HTTP request to evaluate. + * @param location The matched @ref Location context for the request path. + * + * @return `true` if the request should spawn a CGI process, + * `false` if it should be handled normally. + */ bool SocketManager::shouldSpawnCgi(const HttpRequest& req, const Location& location) { + // Resolve absolute filesystem path of the requested resource std::string resolved = location.resolveAbsolutePath(req.getPath()); + + // Only GET/POST methods are supported for CGI execution, + // and only if the path belongs to a CGI-enabled location. return !resolved.empty() && (req.getMethod() == "GET" || req.getMethod() == "POST") && location.isCgiRequest(normalizePath(req.getPath())); } diff --git a/src/network/SocketManagerResponse.cpp b/src/network/SocketManagerResponse.cpp index 6d410573..39fe5eda 100644 --- a/src/network/SocketManagerResponse.cpp +++ b/src/network/SocketManagerResponse.cpp @@ -3,13 +3,38 @@ /* ::: :::::::: */ /* SocketManagerResponse.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 15:00:54 by irychkov #+# #+# */ -/* Updated: 2025/06/08 18:59:11 by irychkov ### ########.fr */ +/* Updated: 2025/08/18 22:52:05 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManagerResponse.cpp + * @brief Implements response sending logic for SocketManager. + * + * @details + * This file contains all methods of @ref SocketManager related to + * **writing HTTP responses back to clients**: + * - Logging response status codes (@ref logResponseStatus). + * - Sending file-backed responses in chunks (@ref sendFileResponse). + * - Sending in-memory/raw responses (@ref sendRawResponse). + * - Driving the high-level send path for a client (@ref sendResponse). + * + * Responsibilities: + * - Stream headers and body data over non-blocking sockets. + * - Support both static files and CGI-generated temporary files. + * - Handle partial writes (tracking progress via `bytes_sent`). + * - Manage connection lifecycle (`keep-alive` vs `close`). + * - Ensure resource cleanup (closing file streams, unlinking CGI temp files). + * + * This file implements the **write-path** of the event loop, + * complementing @ref SocketManagerRequest.cpp which handles the read-path. + * + * @ingroup socker_mananager + */ + #include "http/HttpResponse.hpp" // for HttpResponse #include "http/handleCgi.hpp" // for unlinkWithErrorLog #include "network/SocketManager.hpp" // for ClientInfo, SocketManager @@ -30,6 +55,22 @@ #include // for pair #include // for vector +/** + * @brief Logs the HTTP response status being sent to a client. + * + * @details + * - Formats a message indicating the status code and target file descriptor. + * - Logs at different severity levels depending on status code range: + * - `< 400`: Informational (success / redirection). + * - `400–499`: Warning (client error). + * - `>= 500`: Error (server error). + * + * This function is called before sending a response to provide + * visibility into the server’s behavior and help with debugging. + * + * @param status The HTTP status code of the response. + * @param fd The client file descriptor receiving the response. + */ void SocketManager::logResponseStatus(int status, int fd) { std::string message = "Sending HTTP " + std::to_string(status) + " → fd " + std::to_string(fd); if (status < 400) @@ -40,9 +81,54 @@ void SocketManager::logResponseStatus(int status, int fd) { Logger::logFrom(LogLevel::ERROR, "SocketManager sendResponse", message); } +/** + * @brief Sends a file-based HTTP response to a client. + * + * @details + * Handles responses that stream a file (static resource or CGI temp file) + * to the client in a **non-blocking, incremental fashion**: + * + * Workflow: + * 1. **First call (headers not sent yet):** + * - Opens the response’s file path in binary mode. + * - If CGI body offset is set, seeks past CGI headers. + * - Builds and stores the HTTP status line and headers in + * `client.current_raw_response`. + * + * 2. **Send headers (if not fully sent yet):** + * - Writes remaining header bytes using `send(MSG_DONTWAIT)`. + * - Updates offset tracking (`bytes_sent`). + * - Returns early if partial write occurs (retry later). + * + * 3. **Send file body (chunked):** + * - Reads next chunk (8 KB) from the file into a buffer. + * - Sends it to the client socket. + * - Returns early if partial write occurs. + * + * 4. **Completion:** + * - On EOF (or 0 bytes read), closes file stream. + * - Clears `current_raw_response` and resets offset. + * - If response came from a CGI temp file, deletes it. + * - Pops the response from the queue. + * - Handles connection policy: + * - If `Connection: close` → closes client FD. + * - If keep-alive and no more responses → disables POLLOUT. + * + * Error handling: + * - On file open failure → queues a `500 Internal Server Error`. + * - On socket send failure → logs error, closes the connection. + * + * @param fd Client socket file descriptor. + * @param index Index of the client FD in the poll list. + * @param response The HTTP response object to send. + * + * @return `true` if more data remains to be sent, + * `false` if connection was closed or error occurred. + */ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& response) { ClientInfo& client = _client_info[fd]; + // Step 1: open the file and build headers (first call) if (!client.file_stream.is_open()) { client.file_stream.open(response.getFilePath(), std::ios::binary); if (!client.file_stream.is_open()) { @@ -52,15 +138,18 @@ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& respons if (response.getCgiBodyOffset() > 0) client.file_stream.seekg(response.getCgiBodyOffset()); + // Build HTTP response headers std::ostringstream head; head << "HTTP/1.1 " << response.getStatusCode() << " " << response.getStatusMessage() << "\r\n"; for (const auto& header : response.getHeaders()) head << header.first << ": " << header.second << "\r\n"; head << "\r\n"; + client.current_raw_response = head.str(); } + // Step 2: send HTTP headers if not finished std::string& raw = client.current_raw_response; size_t& offset = client.bytes_sent; if (offset < raw.size()) { @@ -74,9 +163,10 @@ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& respons offset += sent; client.lastSendAttemptTime = getCurrentTime(); if (offset < raw.size()) - return true; + return true; // partial header write → retry later } + // Step 3: stream file body in chunks char buffer[8192]; client.file_stream.read(buffer, sizeof(buffer)); std::streamsize bytes_read = client.file_stream.gcount(); @@ -89,9 +179,10 @@ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& respons return false; } client.lastSendAttemptTime = getCurrentTime(); - return true; + return true; // still more file left } + // Step 4: completion (EOF reached) if (client.file_stream.eof() || bytes_read == 0) { Logger::logFrom(LogLevel::INFO, "SocketManager", "[DONE] We sent full FILE RESPONSE to fd:" + std::to_string(fd)); @@ -99,10 +190,13 @@ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& respons client.current_raw_response.clear(); offset = 0; + // Cleanup CGI temp file if response came from CGI if (response.isCgiTempFile()) { CGI::unlinkWithErrorLog(response.getCgiTempFile(), "out temp file"); response.setCgiTempFile(""); } + + // Pop response and handle connection policy bool shouldClose = response.isConnectionClose(); client.responses.pop(); if (shouldClose) { @@ -113,22 +207,49 @@ bool SocketManager::sendFileResponse(int fd, size_t index, HttpResponse& respons } else if (client.responses.empty()) { Logger::logFrom(LogLevel::INFO, "SocketManager", "Connection: keep-alive - keeping the connection open"); - _poll_fds[index].events &= ~POLLOUT; // Disable POLLOUT for keep-alive + _poll_fds[index].events &= ~POLLOUT; // disable POLLOUT until more data } } return true; } +/** + * @brief Sends an in-memory (raw) HTTP response to a client. + * + * @details + * Handles responses that are already fully materialized in memory: + * 1) On first call, serialize the @ref HttpResponse to a byte string + * (`toHttpString()`) and reset the per-client send offset. + * 2) Perform non-blocking `send()` of the remaining bytes; update `bytes_sent`. + * 3) When all bytes are sent: + * - Pop the response from the queue, + * - Clear the staging buffer and reset offset, + * - Honor connection policy: + * * `Connection: close` → close FD, + * * keep-alive → if no more responses, drop POLLOUT to avoid busy loops. + * + * Error handling: + * - On `send()` failure, logs the error and closes the connection. + * + * @param fd Client socket file descriptor. + * @param index Index of the client FD in the poll list. + * @param response The response object to serialize and send. + * + * @return `true` if the socket remains usable (may need more sends), + * `false` if the connection was closed or a fatal error occurred. + */ bool SocketManager::sendRawResponse(int fd, size_t index, HttpResponse& response) { ClientInfo& client = _client_info[fd]; size_t& offset = client.bytes_sent; + // First invocation for this response: serialize headers + body into a single string. if (client.current_raw_response.empty()) { client.current_raw_response = response.toHttpString(); offset = 0; } + // Attempt to send the remaining bytes (non-blocking). std::string& raw = client.current_raw_response; if (offset < raw.size()) { ssize_t sent = send(fd, raw.c_str() + offset, raw.size() - offset, MSG_DONTWAIT); @@ -142,6 +263,7 @@ bool SocketManager::sendRawResponse(int fd, size_t index, HttpResponse& response client.lastSendAttemptTime = getCurrentTime(); } + // If everything is sent, finalize and apply connection policy. if (offset >= raw.size()) { bool shouldClose = response.isConnectionClose(); client.responses.pop(); @@ -149,46 +271,73 @@ bool SocketManager::sendRawResponse(int fd, size_t index, HttpResponse& response Logger::logFrom(LogLevel::INFO, "SocketManager", "[DONE] We sent full RESPONSE to fd:" + std::to_string(fd)); client.current_raw_response.clear(); + if (shouldClose) { Logger::logFrom(LogLevel::INFO, "SocketManager", "Connection: close - closing the connection"); cleanupClientConnectionClose(fd, index); return false; } else if (client.responses.empty()) { + // No more data to write → stop polling for POLLOUT to avoid spin. Logger::logFrom(LogLevel::INFO, "SocketManager", "Connection: keep-alive - keeping the connection open"); - _poll_fds[index].events &= ~POLLOUT; // Disable POLLOUT for keep-alive + _poll_fds[index].events &= ~POLLOUT; } } return true; } +/** + * @brief Dispatches sending of the next queued response for a client. + * + * @details + * Retrieves the front @ref HttpResponse for the given client, logs its status + * via @ref logResponseStatus, then delegates to the appropriate write path: + * - File-backed response → @ref sendFileResponse (streams headers + file body). + * - In-memory response → @ref sendRawResponse (single serialized buffer). + * + * Both send paths are **non-blocking**; they may return early when the socket + * can’t accept more bytes. Any fatal error (including allocation failure) is + * caught here; the client connection is then closed to keep the server healthy. + * + * @param client_fd Client socket file descriptor. + * @param index Index of the client FD in the poll list. + * + * @note This function assumes there is at least one queued response for + * `client_fd`. Callers should check the queue before enabling POLLOUT. + */ void SocketManager::sendResponse(int client_fd, size_t index) { try { + // Peek the next response to send for this client. HttpResponse& response = _client_info[client_fd].responses.front(); + // Log status code with severity (info/warn/error). logResponseStatus(response.getStatusCode(), client_fd); + // Choose write path: file-backed (streamed) vs raw (in-memory). if (response.isFileResponse()) { if (!sendFileResponse(client_fd, index, response)) - return; + return; // connection closed or fatal error handled inside } else { if (!sendRawResponse(client_fd, index, response)) - return; + return; // connection closed or fatal error handled inside } - return; + return; // keep connection; more data may remain for later POLLOUT } catch (const std::bad_alloc& e) { Logger::logFrom(LogLevel::ERROR, "SocketManager", "Fatal memory allocation error while sending response to fd " + std::to_string(client_fd)); } catch (const std::exception& e) { + // Any recoverable runtime error while sending this response. Logger::logFrom(LogLevel::ERROR, "SocketManager", "Exception in sendResponse for fd " + std::to_string(client_fd) + ": " + e.what()); } catch (...) { + // Last-resort safety net. Logger::logFrom(LogLevel::ERROR, "SocketManager", "Unknown fatal error in sendResponse for fd " + std::to_string(client_fd)); } + // On fatal errors, close the client cleanly to avoid undefined state. cleanupClientConnectionClose(client_fd, index); } diff --git a/src/network/SocketManagerTimeouts.cpp b/src/network/SocketManagerTimeouts.cpp index b5f1cbe5..c2994c76 100644 --- a/src/network/SocketManagerTimeouts.cpp +++ b/src/network/SocketManagerTimeouts.cpp @@ -3,13 +3,40 @@ /* ::: :::::::: */ /* SocketManagerTimeouts.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 15:03:19 by irychkov #+# #+# */ -/* Updated: 2025/06/08 19:09:54 by irychkov ### ########.fr */ +/* Updated: 2025/08/18 22:52:11 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManagerTimeouts.cpp + * @brief Implements timeout and limit enforcement for SocketManager. + * + * @details + * This file contains helper routines used by the poll-driven event loop to + * detect stalled or abusive connections and enforce request limits: + * - @ref isHeaderTimeout: header phase exceeded @c HEADER_TIMEOUT_SECONDS. + * - @ref isBodyTimeout: body phase exceeded @c TIMEOUT. + * - @ref isSendTimeout: no progress while sending for longer than @c TIMEOUT. + * - @ref isIdleTimeout: idle connection (no header bytes) exceeded @c TIMEOUT. + * - @ref checkClientTimeouts: orchestrates per-FD timeout checks and applies + * cleanup / event-mask adjustments. + * - @ref checkRequestLimits: enforces header length cap (@c HEADER_MAX_LENGTH) + * and queues a 431 if exceeded. + * + * Responsibilities: + * - Decide when to close connections vs. temporarily disable POLLIN. + * - Queue appropriate HTTP error responses (e.g., 408, 431). + * - Avoid interfering with active CGI executions. + * + * These functions are invoked from the main event loop to keep the server + * responsive, fair, and resilient under slowloris-style behavior or network stalls. + * + * @ingroup socker_mananager + */ + #include "http/HttpResponse.hpp" // for HttpResponse #include "http/handleCgi.hpp" // for CgiProcess #include "network/SocketManager.hpp" // for ClientInfo, SocketManager, TIMEOUT @@ -23,13 +50,40 @@ #include // for time_t, size_t #include // for vector +/** + * @brief Detects header read timeout for a client connection. + * + * @details + * Triggers when: + * - No response is currently being sent (`responses` and `current_raw_response` empty), + * - The HTTP headers are not yet complete (`headerComplete == false`) but at least + * one header byte has arrived (`headerBytesReceived > 0`), + * - And the elapsed time since `connectionStartTime` exceeds + * `HEADER_TIMEOUT_SECONDS`. + * + * On timeout, queues a `408 Request Timeout` via @ref respondError and returns `true` + * so the caller can disable further reads (POLLIN) and flush the error response. + * + * @param fd Client socket file descriptor. + * @param now Current time snapshot used for comparison. + * @return `true` if a header timeout was detected and an error response was queued, + * otherwise `false`. + * + * @note This does not close the connection immediately; the caller is expected to + * keep the socket writable to send the queued 408 response. + */ bool SocketManager::isHeaderTimeout(int fd, time_t now) { ClientInfo& client = _client_info[fd]; + // No ongoing send, header not complete, some header bytes seen, + // and the header phase exceeded the configured timeout. if (client.responses.empty() && client.current_raw_response.empty() && !client.headerComplete && client.headerBytesReceived > 0 && now - client.connectionStartTime > HEADER_TIMEOUT_SECONDS) { + Logger::logFrom(LogLevel::WARN, "SocketManager", "Timeout on fd: " + std::to_string(fd)); + + // Queue 408 so the writer path can flush it to the client. respondError(fd, 408); return true; } @@ -37,66 +91,210 @@ bool SocketManager::isHeaderTimeout(int fd, time_t now) { return false; } +/** + * @brief Detects request body timeout for a client connection. + * + * @details + * This timeout applies when: + * - No response is currently being sent (`responses` and `current_raw_response` empty), + * - The HTTP headers have already been received completely (`headerComplete == true`), + * - And the elapsed time since `connectionStartTime` exceeds @c TIMEOUT. + * + * This typically indicates that the client started a request but did not finish + * sending the body in time (e.g. stalled upload or slowloris-style attack). + * + * On timeout, a `408 Request Timeout` is queued via @ref respondError and the function + * returns `true`, so the caller may disable reads (`POLLIN`) and flush the error. + * + * @param fd Client socket file descriptor. + * @param now Current server time snapshot. + * @return `true` if a body timeout occurred and an error response was queued, + * otherwise `false`. + * + * @note The connection is not closed immediately; it is marked for sending the 408 + * response before shutdown. + */ bool SocketManager::isBodyTimeout(int fd, time_t now) { ClientInfo& client = _client_info[fd]; + + // No ongoing send, header complete, and body phase exceeded TIMEOUT. if (client.responses.empty() && client.current_raw_response.empty() && client.headerComplete && now - client.connectionStartTime > TIMEOUT) { + Logger::logFrom(LogLevel::WARN, "SocketManager", "Timeout on fd: " + std::to_string(fd)); - respondError(fd, 408); + + respondError(fd, 408); // Queue HTTP 408 Request Timeout return true; } + return false; } +/** + * @brief Detects send timeout for a client connection. + * + * @details + * This timeout applies when: + * - There are still responses waiting to be sent (`!responses.empty()`), + * - A response is actively being transmitted (`!current_raw_response.empty()`), + * - But no progress has been made for more than @c TIMEOUT seconds + * (based on `lastSendAttemptTime`). + * + * This usually indicates that the client has stopped reading from its socket + * (e.g., slow or dead client), preventing the server from completing the response. + * + * On timeout, no error response is queued (since we are already mid-send), + * but the function signals the caller (`true`) so that the connection can be + * closed and cleaned up. + * + * @param fd Client socket file descriptor. + * @param now Current server time snapshot. + * @return `true` if a send timeout occurred and the connection should be closed, + * otherwise `false`. + * + * @note Unlike header/body timeouts, this does not queue a 408 response, + * because the server was already in the process of sending a response. + */ bool SocketManager::isSendTimeout(int fd, time_t now) { ClientInfo& client = _client_info[fd]; + + // Active send in progress but stalled for longer than TIMEOUT. if (!client.responses.empty() && !client.current_raw_response.empty() && now - client.lastSendAttemptTime > TIMEOUT) { + Logger::logFrom(LogLevel::WARN, "SocketManager", "Send timeout on fd: " + std::to_string(fd)); return true; } + return false; } +/** + * @brief Detects idle timeout for a client connection. + * + * @details + * This timeout applies when: + * - The client is not running a CGI process (`cgiProcess` is empty), + * - No response is in progress (`responses` and `current_raw_response` are empty), + * - No headers have been fully received (`headerComplete == false`), + * - No header bytes have been received at all (`headerBytesReceived == 0`), + * - And the time since the last recorded client activity (`lastRequestTime`) + * exceeds @c TIMEOUT. + * + * This typically indicates an idle TCP connection where the client opened + * a socket but never sent any meaningful data. + * + * On timeout, no explicit error response is queued; the caller is expected to + * close the connection after receiving `true`. + * + * @param fd Client socket file descriptor. + * @param now Current server time snapshot. + * @return `true` if the connection is considered idle and should be closed, + * otherwise `false`. + * + * @note CGI processes are exempted from idle timeout handling because they may + * take significant time before producing output. + */ bool SocketManager::isIdleTimeout(int fd, time_t now) { ClientInfo& client = _client_info[fd]; + + // Skip idle timeout checks if a CGI is running. if (client.cgiProcess.has_value()) { return false; } + + // No response, no header progress, and idle for too long. if (client.responses.empty() && client.current_raw_response.empty() && !client.headerComplete && client.headerBytesReceived == 0 && now - client.lastRequestTime > TIMEOUT) { + Logger::logFrom(LogLevel::WARN, "SocketManager", "Idle timeout on fd: " + std::to_string(fd)); return true; } + return false; } +/** + * @brief Checks and applies timeout rules for a client connection. + * + * @details + * This function orchestrates all timeout checks for a client: + * - **Idle timeout**: connection opened but no activity (`isIdleTimeout`). + * - **Send timeout**: stalled during an ongoing response (`isSendTimeout`). + * - **Header timeout**: headers started but not completed in time (`isHeaderTimeout`). + * - **Body timeout**: headers done but body not finished within limit (`isBodyTimeout`). + * + * Behavior: + * - If idle or send timeout occurs → the connection is closed immediately via + * `cleanupClientConnectionClose()`, and the function returns `false`. + * - If header or body timeout occurs → the connection stays open long enough to + * send a timeout response (408). To prevent further reads, `POLLIN` is disabled + * for this fd, and the function returns `true`. + * - If no timeout is triggered, returns `false`. + * + * @param client_fd The socket file descriptor of the client. + * @param index Index of the client's pollfd entry in `_poll_fds`. + * @return `true` if a recoverable timeout (header/body) was detected and the + * connection is left open for sending a response, + * `false` if the connection was closed or no timeout occurred. + * + * @note This function integrates multiple specialized timeout checks into the + * main poll loop, ensuring the server proactively cleans up stale or + * non-responsive connections. + */ bool SocketManager::checkClientTimeouts(int client_fd, size_t index) { if (!_client_info.count(client_fd)) return false; time_t now = getCurrentTime(); + + // Hard timeouts → connection closed if (isIdleTimeout(client_fd, now) || isSendTimeout(client_fd, now)) { cleanupClientConnectionClose(client_fd, index); return false; } + + // Soft timeouts → send 408 and disable POLLIN if (isHeaderTimeout(client_fd, now) || isBodyTimeout(client_fd, now)) { _poll_fds[index].events &= ~POLLIN; return true; } + return false; } +/** + * @brief Enforces request size limits for an active client connection. + * + * @details + * This function checks whether the accumulated request headers for a client + * exceed the configured maximum length (`HEADER_MAX_LENGTH`). + * + * Behavior: + * - If the header is incomplete **and** the number of bytes received so far + * exceeds the allowed maximum, the server immediately queues a + * **431 Request Header Fields Too Large** error response using + * `respondError()`, and returns `true`. + * - Otherwise, the request is considered within limits and processing continues. + * + * @param fd The socket file descriptor of the client being checked. + * @return `true` if the request exceeded the configured header limit + * (error response queued), + * `false` if still within allowed limits. + * + * @note This check is only enforced **before header completion**. Once headers + * are fully parsed, the limit is no longer evaluated here. + */ bool SocketManager::checkRequestLimits(int fd) { ClientInfo& client = _client_info[fd]; - // Only enforce header-length limit while headers are still incomplete + // Enforce header-length limit only while headers are incomplete if (client.headerBytesReceived > HEADER_MAX_LENGTH) { Logger::logFrom(LogLevel::WARN, "SocketManager", "Request header too large from fd: " + std::to_string(fd)); - respondError(fd, 431); // Request Header Fields Too Large + respondError(fd, 431); // 431: Request Header Fields Too Large return true; } diff --git a/src/network/SocketManagerUtils.cpp b/src/network/SocketManagerUtils.cpp index 7c8f6405..33da97f7 100644 --- a/src/network/SocketManagerUtils.cpp +++ b/src/network/SocketManagerUtils.cpp @@ -3,13 +3,38 @@ /* ::: :::::::: */ /* SocketManagerUtils.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 14:58:17 by irychkov #+# #+# */ -/* Updated: 2025/08/17 21:04:16 by irychkov ### ########.fr */ +/* Updated: 2025/08/18 22:52:18 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file SocketManagerUtils.cpp + * @brief Utility methods for connection lifecycle management in SocketManager. + * + * @details + * This file implements helper functions used internally by the + * `SocketManager` class to manage client state, poll descriptors, + * error handling, and cleanup routines. These utilities ensure proper + * resource deallocation and safe connection teardown. + * + * Main responsibilities: + * - Cleaning up CGI processes associated with a client (`cleanupCgiForClient`). + * - Removing `pollfd` entries from the monitored list (`removePollFd`). + * - Resetting client request parsing state (`resetRequestState`). + * - Gracefully closing client connections and releasing resources + * (`cleanupClientState`, `cleanupClientConnectionClose`). + * - Generating and queuing error responses (`respondError`). + * + * These functions are invoked by higher-level event loop and request + * handling methods to enforce robustness, prevent resource leaks, and + * handle exceptional conditions. + * + * @ingroup socker_mananager + */ + #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest #include "http/HttpResponse.hpp" // for HttpResponse @@ -27,6 +52,20 @@ #include // for pair #include // for vector +/** + * @brief Cleans up and resets CGI state for a specific client. + * + * @details + * If the client identified by @p client_fd has an active CGI process, + * this function terminates it via `CGI::cleanupCgi`, resets its state, + * and clears the currently running CGI request. + * If no CGI process is active, the function returns without action. + * + * @param client_fd The file descriptor of the client whose CGI state should be cleaned up. + * + * @note This function does not close the client connection itself — it only + * resets CGI-related state. Connection cleanup is handled separately. + */ void SocketManager::cleanupCgiForClient(int client_fd) { if (!_client_info.contains(client_fd)) return; @@ -41,12 +80,48 @@ void SocketManager::cleanupCgiForClient(int client_fd) { client.currentCgiRequest = HttpRequest(); } +/** + * @brief Removes a file descriptor entry from the poll list. + * + * @details + * This function erases the `pollfd` structure at the given index + * from the `_poll_fds` vector, which is used by `poll()` in the + * main event loop to monitor active sockets. + * + * If the index is out of range, no action is performed. + * + * @param index The position in `_poll_fds` corresponding to the + * client or listening socket to be removed. + * + * @note This does not close the socket itself. The caller is responsible + * for performing connection cleanup separately (see + * `cleanupClientConnectionClose`). + */ void SocketManager::removePollFd(size_t index) { if (index < _poll_fds.size()) { _poll_fds.erase(_poll_fds.begin() + index); } } +/** + * @brief Cleans up all state associated with a client connection. + * + * @details + * This function safely releases all resources tied to a client: + * - Drains the response queue and deletes any CGI temporary files. + * - Cleans up and terminates any active CGI process. + * - Closes any open file streams linked to the client. + * - Erases the client entry from `_client_info`. + * + * Unlike `cleanupClientConnectionClose`, this function does not close + * the socket or modify the poll list — it only clears per-client state. + * + * @param client_fd The socket file descriptor of the client whose state + * should be cleaned up. + * + * @note Call this when tearing down a client session to ensure no resource + * leaks (file handles, temp files, or CGI processes). + */ void SocketManager::cleanupClientState(int client_fd) { auto it = _client_info.find(client_fd); if (it == _client_info.end()) @@ -54,6 +129,7 @@ void SocketManager::cleanupClientState(int client_fd) { ClientInfo& client = it->second; + // Clean up pending responses and unlink any CGI temp files while (!client.responses.empty()) { HttpResponse& resp = client.responses.front(); if (resp.isCgiTempFile()) { @@ -62,6 +138,7 @@ void SocketManager::cleanupClientState(int client_fd) { client.responses.pop(); } + // Handle any active CGI process if (client.cgiProcess) { CGI::errorOnCgi(*client.cgiProcess); client.cgiProcess.reset(); @@ -69,13 +146,33 @@ void SocketManager::cleanupClientState(int client_fd) { client.currentCgiRequest = HttpRequest(); } + // Close file stream if open if (client.file_stream.is_open()) { client.file_stream.close(); } + // Finally remove client record _client_info.erase(it); } +/** + * @brief Gracefully closes a client connection and cleans up its state. + * + * @details + * This function performs a full teardown of a client connection: + * - Removes the file descriptor from the poll list (`removePollFd`). + * - Cleans up all per-client state (`cleanupClientState`), including + * pending responses, CGI processes, and open streams. + * - Closes the actual socket file descriptor via `close()`. + * - Logs the closure event. + * + * @param client_fd The socket file descriptor of the client being closed. + * @param index The index in `_poll_fds` corresponding to the client_fd. + * + * @note Use this when the connection must be terminated (e.g. client + * disconnect, fatal error, or `Connection: close` response). + * For state cleanup without closing the socket, see `cleanupClientState()`. + */ void SocketManager::cleanupClientConnectionClose(int client_fd, size_t index) { removePollFd(index); cleanupClientState(client_fd); @@ -84,24 +181,62 @@ void SocketManager::cleanupClientConnectionClose(int client_fd, size_t index) { "Closed FD (Connection: close): " + std::to_string(client_fd)); } +/** + * @brief Resets the parsing state of a client’s current HTTP request. + * + * @details + * This function clears or resets request-related counters for a given client: + * - If the `requestBuffer` still contains data, the connection is treated + * as having an incomplete header. In this case, `headerComplete` is reset + * to false, but `headerBytesReceived` is preserved so that header timeout + * tracking continues. + * - If the buffer is empty, the function resets all parsing counters + * (`headerComplete`, `headerBytesReceived`, `bodyBytesReceived`). + * + * @param client_fd The file descriptor of the client whose request + * parsing state should be reset. + * + * @note This is typically called after a request has been fully parsed + * or after an error, so the connection can be reused for the next request + * (in keep-alive scenarios). + */ void SocketManager::resetRequestState(int client_fd) { if (!_client_info.count(client_fd)) return; - // If there is still any data in requestBuffer, treat it as a partial header: + + // If there is still data in requestBuffer, consider it a partial header if (!_client_info[client_fd].requestBuffer.empty()) { _client_info[client_fd].headerComplete = false; - // headerBytesReceived should reflect how many bytes are already in the buffer. - // But if we are just about to parse a brand‐new header, headerBytesReceived - // should have already been set by receiveFromClient(...) when those bytes first arrived. - // So here we do NOT zero it out—leave it alone so the header‐timer can still tick. + // Preserve headerBytesReceived so timeout logic remains valid return; } - // If requestBuffer is empty, then there is no partial header in progress. + + // No partial header in progress → fully reset state _client_info[client_fd].headerComplete = false; _client_info[client_fd].headerBytesReceived = 0; _client_info[client_fd].bodyBytesReceived = 0; } +/** + * @brief Queues an HTTP error response for a client. + * + * @details + * This function generates a standardized HTTP error response with the + * given status code (e.g. 400, 404, 500) and pushes it onto the client’s + * response queue. + * + * - Uses the first server in the client’s `serversOnPort` list as a fallback + * context for generating the response. + * - The request object passed to `ResponseBuilder::generateError` is empty, + * since the error may occur before a valid request could be parsed. + * + * @param fd The client’s socket file descriptor. + * @param status_code The HTTP status code to send (e.g. 400, 404, 500). + * + * @note This does not immediately send the response — it only enqueues it. + * The response will be transmitted later in the event loop + * by `sendResponse()`. + */ void SocketManager::respondError(int fd, int status_code) { HttpRequest empty; const Server& fallback = _client_info[fd].serversOnPort.front(); From bbfbbaa88ed79eb5839ad958e35bc7329ef106e3 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Mon, 18 Aug 2025 23:26:52 +0300 Subject: [PATCH 21/32] Fix typo in Doxygen group name and improve code organization Corrects a typo in the Doxygen group name from "socker_mananager" to "socket_manager" across multiple files for consistency and clarity. Enhances code readability and maintainability by organizing methods into named Doxygen groups with appropriate tags. Removes unused internal documentation and adjusts formatting in several method definitions for improved alignment. --- include/doxygen_groups.hpp | 2 +- include/network/SocketManager.hpp | 42 ++++++++++++++++++++------- src/network/SocketManager.cpp | 7 ++--- src/network/SocketManagerRequest.cpp | 4 +-- src/network/SocketManagerResponse.cpp | 4 +-- src/network/SocketManagerTimeouts.cpp | 4 +-- src/network/SocketManagerUtils.cpp | 4 +-- 7 files changed, 42 insertions(+), 25 deletions(-) diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index 53432c23..5b6e47d8 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -125,7 +125,7 @@ */ /** - * @defgroup socker_mananager Network & Socket Abstraction + * @defgroup socket_manager Network & Socket Abstraction * @brief Non-blocking I/O, socket management, and connection handling. * * @details Provides abstractions over system calls for listening sockets, diff --git a/include/network/SocketManager.hpp b/include/network/SocketManager.hpp index cf82552b..345f03cb 100644 --- a/include/network/SocketManager.hpp +++ b/include/network/SocketManager.hpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:51:47 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:51:45 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:24:39 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -31,7 +31,7 @@ * - accurate HTTP status codes and default error pages, * - resilient behavior under stress and strict timeouts. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #pragma once @@ -90,7 +90,6 @@ * All fields are mutated only by the owning SocketManager on the * poll thread. * - * @ingroup network */ struct ClientInfo { //=== Core identifiers and timing ======================================== @@ -144,37 +143,41 @@ struct ClientInfo { * and cleanly close when required. * - Protect the loop with wide catch blocks—FDs are closed on failure paths. * - * @ingroup network + * @ingroup socket_manager */ class SocketManager { public: //=== Ctors / Dtor / Special members ===================================== + /** @name Construction & lifetime */ ///@{ SocketManager(void) = delete; SocketManager(const std::vector& servers); ~SocketManager(void); - SocketManager(const SocketManager& other) = delete; - SocketManager& operator=(const SocketManager& other) = delete; + SocketManager(const SocketManager&) = delete; + SocketManager& operator=(const SocketManager&) = delete; + ///@} //=== Main loop =========================================================== + /** @name Main loop */ ///@{ void run(); + ///@} - /** - * @brief Exception type for socket-level setup/runtime failures. - */ + /** @name Errors */ ///@{ class SocketError : public std::exception { private: std::string _msg; public: explicit SocketError(const std::string& msg); - virtual const char* what() const throw(); + const char* what() const throw() override; }; + ///@} private: //=== Poll & indices ====================================================== - std::vector _poll_fds; ///< Monitored file descriptors for poll(). + /** @name Poll sets & indices */ ///@{ + std::vector _poll_fds; ///< Monitored file descriptors for poll(). /** * @brief Listen FD → vhost set. @@ -198,49 +201,66 @@ class SocketManager { * main poll loop. */ std::map _fd_to_cgi; + ///@} //=== Setup & connection ================================================== + /** @name Setup & accept */ ///@{ void setupSockets(const std::vector& servers); void handleNewConnection(int listen_fd); + ///@} //=== Event handling ====================================================== + /** @name Event handling */ ///@{ bool handleClientData(int client_fd, size_t index); void handleCgiPollEvents(); void sendResponse(int client_fd, size_t index); void handlePollError(int fd, size_t index, short revents); + ///@} //=== Response helpers ==================================================== + /** @name Response helpers */ ///@{ void logResponseStatus(int status, int fd); bool sendFileResponse(int fd, size_t index, HttpResponse& response); bool sendRawResponse(int fd, size_t index, HttpResponse& response); + ///@} //=== Client lifecycle ==================================================== + /** @name Client lifecycle */ ///@{ void initializeClientInfo(int client_fd, int listen_fd); void cleanupClientConnectionClose(int client_fd, size_t index); void removePollFd(size_t index); void cleanupClientState(int client_fd); void cleanupCgiForClient(int client_fd); void resetRequestState(int client_fd); + ///@} //=== Request processing ================================================== + /** @name Request processing */ ///@{ bool receiveFromClient(int fd, size_t index); bool checkRequestLimits(int fd); bool parseAndQueueRequests(int client_fd); void processPendingRequests(int client_fd); + ///@} //=== Timeout checks ====================================================== + /** @name Timeout checks */ ///@{ bool checkClientTimeouts(int client_fd, size_t index); bool isHeaderTimeout(int fd, time_t now); bool isBodyTimeout(int fd, time_t now); bool isSendTimeout(int fd, time_t now); bool isIdleTimeout(int fd, time_t now); + ///@} //=== Request routing and CGI ============================================ + /** @name CGI & routing */ ///@{ bool handleCgiRequest(int client_fd, const HttpRequest& request, const Server& server, const Location& location); bool handleRequestErrorIfAny(int fd, int code, HttpRequest& req, const Server& server); bool shouldSpawnCgi(const HttpRequest& req, const Location& location); + ///@} //=== Error utility ======================================================= + /** @name Error utility */ ///@{ void respondError(int fd, int status_code); + ///@} }; diff --git a/src/network/SocketManager.cpp b/src/network/SocketManager.cpp index e37b6eee..6e2b2499 100644 --- a/src/network/SocketManager.cpp +++ b/src/network/SocketManager.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/03 13:51:20 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:51:52 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:11:01 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -25,7 +25,7 @@ * it ties together sockets, HTTP parsing, response building, CGI execution, * and cleanup into a single poll-driven loop. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #include "network/SocketManager.hpp" @@ -43,9 +43,6 @@ #include // for close #include // for pair, make_pair -/// @internal -/// @brief Global flag controlling the server loop. -/// @details Set to `0` when SIGINT is received, causing the main loop to exit. static volatile sig_atomic_t running = 1; /** diff --git a/src/network/SocketManagerRequest.cpp b/src/network/SocketManagerRequest.cpp index 04c26cad..53b32ef5 100644 --- a/src/network/SocketManagerRequest.cpp +++ b/src/network/SocketManagerRequest.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 14:53:38 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:51:58 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:09:00 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -30,7 +30,7 @@ * It forms the **read-path of the event loop**, turning client bytes into * application-level requests and responses. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #include "core/Location.hpp" // for Location diff --git a/src/network/SocketManagerResponse.cpp b/src/network/SocketManagerResponse.cpp index 39fe5eda..d8ffca86 100644 --- a/src/network/SocketManagerResponse.cpp +++ b/src/network/SocketManagerResponse.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 15:00:54 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:52:05 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:09:04 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -32,7 +32,7 @@ * This file implements the **write-path** of the event loop, * complementing @ref SocketManagerRequest.cpp which handles the read-path. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #include "http/HttpResponse.hpp" // for HttpResponse diff --git a/src/network/SocketManagerTimeouts.cpp b/src/network/SocketManagerTimeouts.cpp index c2994c76..546a2786 100644 --- a/src/network/SocketManagerTimeouts.cpp +++ b/src/network/SocketManagerTimeouts.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 15:03:19 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:52:11 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:09:08 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -34,7 +34,7 @@ * These functions are invoked from the main event loop to keep the server * responsive, fair, and resilient under slowloris-style behavior or network stalls. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #include "http/HttpResponse.hpp" // for HttpResponse diff --git a/src/network/SocketManagerUtils.cpp b/src/network/SocketManagerUtils.cpp index 33da97f7..61e3cb74 100644 --- a/src/network/SocketManagerUtils.cpp +++ b/src/network/SocketManagerUtils.cpp @@ -6,7 +6,7 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/08 14:58:17 by irychkov #+# #+# */ -/* Updated: 2025/08/18 22:52:18 by nlouis ### ########.fr */ +/* Updated: 2025/08/18 23:09:12 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -32,7 +32,7 @@ * handling methods to enforce robustness, prevent resource leaks, and * handle exceptional conditions. * - * @ingroup socker_mananager + * @ingroup socket_manager */ #include "core/Server.hpp" // for Server From b3537f6d6c56e1003d7f96cdb1e5ff01aaebc3c4 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Tue, 19 Aug 2025 00:04:45 +0300 Subject: [PATCH 22/32] Adds documentation for SocketManager and networking flow Introduces a detailed explanation of the SocketManager, including its role in managing I/O through a `poll()` loop, handling connections, enforcing timeouts, and error recovery. Adds a high-level flow diagram using Mermaid to visualize the networking process, enhancing clarity. --- README.md | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 87 insertions(+) diff --git a/README.md b/README.md index e86bd6ad..12f95546 100644 --- a/README.md +++ b/README.md @@ -120,6 +120,93 @@ This section describes how the configuration parsing logic of **Webserv** works, ___ +## Networking — `SocketManager` + +The heart of Webserv’s I/O: a single `poll()` loop multiplexing **listening sockets**, **client sockets**, and **CGI pipes**, with strict timeouts and robust error recovery. + +
+See Details + +* **Listening sockets**: set up `bind()`/`listen()` for each configured host\:port. +* **Event loop**: run non-blocking `poll()` to monitor all descriptors. +* **Connections**: + + * **New connections** → `accept()` → initialize per-client state. + * **Reads** → receive → parse (supports pipelining) → route. + * **CGI** → spawn, monitor pipes, enforce timeouts, finalize. + * **Writes** → stream raw or file-backed responses with keep-alive and backpressure. +* **Timeouts**: enforce idle, header, body, and send deadlines. +* **Errors**: generate accurate HTTP error responses, close cleanly. + +___ + +### High-Level Flow + +```mermaid +flowchart TD + %% ========================= + %% Setup + %% ========================= + subgraph Boot[Startup] + A[Load servers from config] + B[Create SocketManager] + C[setupSockets: bind listen register FDs] + A --> B --> C + end + + %% ========================= + %% Main loop + %% ========================= + C --> D{run poll} + D -->|EINTR| Z[Graceful shutdown] --> ZZ[Log shutting down server] + + %% Loop tick + D --> E[handleCgiPollEvents] + E --> F{for each pfd reverse} + + %% Skip CGI pipe FDs + F --> G{CGI pipe FD} + G -- yes --> F + + %% Timeouts (per FD) + G -- no --> T{checkClientTimeouts} + T -- idle or send --> Close[cleanupClientConnectionClose] --> F + T -- header or body --> Halt[disable POLLIN then queue 408] --> H{ERR HUP NVAL} + T -- none --> H + + %% Socket errors + H -- yes --> H1[handlePollError then close] --> F + + %% Reads + H -- no --> I{POLLIN} + I -- no --> O{POLLOUT and responses queued} + + I -- yes --> J{listen FD} + J -- yes --> J1[handleNewConnection] --> F + + J -- no --> Rcv[handleClientData] + Rcv -- queued --> K[enable POLLOUT] --> F + Rcv -- incomplete --> F + + %% Writes + O -- yes --> S[sendResponse] + S -->|file| SF[sendFileResponse] + S -->|raw| SR[sendRawResponse] + SF --> EndSend{response done} + SR --> EndSend + + EndSend -- yes and close --> Close + EndSend -- yes and keep alive --> KA[disable POLLOUT] --> F + EndSend -- not yet --> F + + %% Next tick + F --> D +``` + +
+ +___ + ## Flow Overview 1. **Tokenizer** → breaks input into tokens. From 112718b3d89d0e19991fe6f8c5ce60dc1b0194d7 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Tue, 19 Aug 2025 10:06:16 +0300 Subject: [PATCH 23/32] Enhance HTTP documentation Improves maintainability by adding detailed Doxygen comments to HTTP-related classes, methods, and utilities for better clarity and adherence to HTTP/1.1 standards. Includes updates to `HttpRequest`, `HttpResponse`, `HttpRequestParser`, and response builders. Clarifies logic for error pages, redirects, and file-backed responses while ensuring compliance with HTTP protocol semantics. --- Makefile.to0nsa | 212 ---------- include/doxygen_groups.hpp | 6 + include/http/HttpRequest.hpp | 144 +++++-- include/http/HttpRequestParser.hpp | 65 +++- include/http/HttpResponse.hpp | 101 +++-- include/http/Url.hpp | 60 ++- include/http/responseBuilder.hpp | 28 +- src/http/HttpRequest.cpp | 251 ++++++++---- src/http/HttpRequestParser.cpp | 597 +++++++++++++++++++++++------ src/http/HttpResponse.cpp | 271 +++++++++---- src/http/responseBuilder.cpp | 279 ++++++++++++-- 11 files changed, 1441 insertions(+), 573 deletions(-) delete mode 100644 Makefile.to0nsa diff --git a/Makefile.to0nsa b/Makefile.to0nsa deleted file mode 100644 index a0110a42..00000000 --- a/Makefile.to0nsa +++ /dev/null @@ -1,212 +0,0 @@ -# **************************************************************************** # -# # -# ::: :::::::: # -# Makefile :+: :+: :+: # -# +:+ +:+ +:+ # -# By: nlouis +#+ +:+ +#+ # -# +#+#+#+#+#+ +#+ # -# Created: 2025/08/09 20:53:27 by nlouis #+# #+# # -# Updated: 2025/08/18 20:05:41 by nlouis ### ########.fr # -# # -# **************************************************************************** # - -# Compiler settings -CXX := c++ -CXXFLAGS := -Wall -Wextra -Werror -I include -std=c++20 \ - -O3 -DNDEBUG -flto -march=native -# We use -flto (link time optimization) for better performance. -# -O3 is for optimization, -DNDEBUG disables debug assertions. - -# Executable output -NAME := webserv -BINDIR := bin -TARGET := $(BINDIR)/$(NAME) - -# Object dir (deps live next to objects: .o + .d) -OBJDIR := objs - -# Colors -GREEN := \033[0;32m -CYAN := \033[0;36m -YELLOW := \033[1;33m -RED := \033[0;31m -RESET := \033[0m - -.DEFAULT_GOAL := all - -SRCS_CONFIG := \ - src/config/Config.cpp \ - src/config/normalizeConfig.cpp \ - src/config/validateConfig.cpp - -SRCS_CONFIG_PARSER := \ - src/config/parser/ConfigParseError.cpp \ - src/config/parser/ConfigParser.cpp \ - src/config/parser/directive_handler_table.cpp - -SRCS_CONFIG_TOKENIZER := \ - src/config/tokenizer/Tokenizer.cpp \ - src/config/tokenizer/token.cpp - -SRCS_CORE := \ - src/core/Location.cpp \ - src/core/Server.cpp \ - src/core/main.cpp \ - src/core/server_utils.cpp \ - src/core/runWebserv.cpp - -SRCS_HTTP := \ - src/http/HttpRequest.cpp \ - src/http/HttpRequestParser.cpp \ - src/http/HttpResponse.cpp \ - src/http/handleCgi.cpp \ - src/http/methodsHandler/handleDelete.cpp \ - src/http/methodsHandler/handleGet/generateAutoindex.cpp \ - src/http/methodsHandler/handleGet/handleGet.cpp \ - src/http/methodsHandler/handlePost/handleMultipartForm.cpp \ - src/http/methodsHandler/handlePost/handlePost.cpp \ - src/http/requestRouter.cpp \ - src/http/responseBuilder.cpp - -SRCS_NETWORK := \ - src/network/SocketManager.cpp \ - src/network/SocketManagerRequest.cpp \ - src/network/SocketManagerResponse.cpp \ - src/network/SocketManagerTimeouts.cpp \ - src/network/SocketManagerUtils.cpp - -SRCS_UTILS := \ - src/utils/Logger.cpp \ - src/utils/errorUtils.cpp \ - src/utils/filesystemUtils.cpp \ - src/utils/htmlUtils.cpp \ - src/utils/printInfo.cpp \ - src/utils/stringUtils.cpp \ - src/utils/urlUtils.cpp - -SRCS := \ - $(SRCS_CONFIG) \ - $(SRCS_CONFIG_PARSER) \ - $(SRCS_CONFIG_TOKENIZER) \ - $(SRCS_CORE) \ - $(SRCS_HTTP) \ - $(SRCS_NETWORK) \ - $(SRCS_UTILS) - -HEADERS := \ - include/config/Config.hpp \ - include/config/normalizeConfig.hpp \ - include/config/parser/ConfigParseError.hpp \ - include/config/parser/ConfigParser.hpp \ - include/config/parser/directive_handler_table.hpp \ - include/config/tokenizer/Tokenizer.hpp \ - include/config/tokenizer/token.hpp \ - include/config/validateConfig.hpp \ - include/core/Location.hpp \ - include/core/Server.hpp \ - include/core/server_utils.hpp \ - include/core/runWebserv.hpp \ - include/http/HttpRequest.hpp \ - include/http/HttpRequestParser.hpp \ - include/http/HttpResponse.hpp \ - include/http/Url.hpp \ - include/http/handleCgi.hpp \ - include/http/methodsHandler.hpp \ - include/http/requestRouter.hpp \ - include/http/responseBuilder.hpp \ - include/network/SocketManager.hpp \ - include/utils/Logger.hpp \ - include/utils/errorUtils.hpp \ - include/utils/filesystemUtils.hpp \ - include/utils/htmlUtils.hpp \ - include/utils/printInfo.hpp \ - include/utils/stringUtils.hpp \ - include/utils/urlUtils.hpp - -# ---- Derived files ----------------------------------------------------------- - -# Map each .cpp in $(SRCS) to its corresponding .o in $(OBJDIR). -# Note: patsubst here does NOT perform any filesystem wildcard expansion -# it just does a string substitution on the fixed list in $(SRCS). -# This is safe and deterministic, unlike $(wildcard) or $(shell find ...). -OBJS := $(patsubst src/%.cpp,$(OBJDIR)/%.o,$(SRCS)) -DEPS := $(OBJS:.o=.d) - -# ---- Rules ------------------------------------------------------------------ - -all: $(TARGET) - -$(TARGET): $(OBJS) - @mkdir -p $(BINDIR) - @$(CXX) $(CXXFLAGS) $^ -o $@ - @echo "$(CYAN)🚀 Built executable:$(RESET) $(TARGET)" - -$(OBJDIR)/%.o: src/%.cpp - @mkdir -p $(dir $@) - $(CXX) $(CXXFLAGS) -MMD -MP -c $< -o $@ - @echo "$(GREEN)🛠️ Compiled:$(RESET) $<" - -# Cleaning -clean: - @rm -rf $(OBJDIR) - @echo "$(YELLOW)🧹 Cleaned object and dependency files.$(RESET)" - -fclean: clean - @rm -rf $(TARGET) $(BINDIR) build - @echo "$(YELLOW)🗑️ Completely removed executables, binaries, build/.$(RESET)" - -re: fclean all - -# Python test dependencies -install_test_deps: - @echo "$(CYAN)📦 Creating isolated Python venv for tests...$(RESET)" - @python3 -m venv .venv || { \ - echo "$(RED)python3-venv is missing. Install it: sudo apt-get install -y python3-venv$(RESET)"; \ - exit 1; \ - } - @.venv/bin/python -m pip install --upgrade pip - @.venv/bin/pip install -r requirements-test.txt - -# Tests -test: all install_test_deps - @echo "$(CYAN)🧪 Launching web server in background...$(RESET)" - @./$(TARGET) ./test_webserv/tester/config/tester.conf & echo $$! > .webserv_test.pid - @sleep 1 - @echo "$(CYAN)🧪 Running Python test suite...$(RESET)" - @.venv/bin/python run_test.py || { \ - echo "$(RED)❌ Tests failed.$(RESET)"; \ - kill `cat .webserv_test.pid` >/dev/null 2>&1 || true; \ - rm -f .webserv_test.pid; \ - exit 1; \ - } - @echo "$(CYAN)🧹 Shutting down test server...$(RESET)" - @kill `cat .webserv_test.pid` >/dev/null 2>&1 || true - @rm -f .webserv_test.pid - @echo "$(GREEN)🏆 All tests passed successfully!$(RESET)" - -# Formatting (explicit file lists; no find/globs) -format: - @echo "$(CYAN)🎨 Formatting source files...$(RESET)" - @clang-format -i $(SRCS) $(HEADERS) - -# Help -help: - @echo "$(CYAN)📦 Build Targets:$(RESET)" - @echo " $(GREEN)make$(RESET) → Build the project 🚀" - @echo " $(GREEN)make re$(RESET) → Clean and rebuild everything 🔁" - @echo "" - @echo "$(CYAN)🧪 Test Targets:$(RESET)" - @echo " $(GREEN)make test$(RESET) → Build, install Python deps, run server + Python tests 🧪" - @echo "" - @echo "$(CYAN)🧹 Cleaning Targets:$(RESET)" - @echo " $(GREEN)make clean$(RESET) → Remove object and dependency files 🧹" - @echo " $(GREEN)make fclean$(RESET)→ Remove everything including binaries and build dirs 🗑️" - @echo "" - @echo "$(CYAN)🧹 Code Quality Targets:$(RESET)" - @echo " $(GREEN)make format$(RESET)→ Format all source files 🎨" - @echo "" - @echo "$(CYAN)📚 Other:$(RESET)" - @echo " $(GREEN)make help$(RESET) → Show this help message 📚" - -.PHONY: all clean fclean re test format help install_test_deps --include $(DEPS) diff --git a/include/doxygen_groups.hpp b/include/doxygen_groups.hpp index 5b6e47d8..c6c6a14b 100644 --- a/include/doxygen_groups.hpp +++ b/include/doxygen_groups.hpp @@ -124,6 +124,12 @@ * building HTTP responses, routing, and applying HTTP rules. */ +/** + * @defgroup request_handler Request Handling + * @ingroup http + * @brief Functions and utilities for handling HTTP requests. + */ + /** * @defgroup socket_manager Network & Socket Abstraction * @brief Non-blocking I/O, socket management, and connection handling. diff --git a/include/http/HttpRequest.hpp b/include/http/HttpRequest.hpp index dd0a69ab..949f016f 100644 --- a/include/http/HttpRequest.hpp +++ b/include/http/HttpRequest.hpp @@ -3,13 +3,54 @@ /* ::: :::::::: */ /* HttpRequest.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 12:31:10 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:21:30 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:04:29 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file HttpRequest.hpp + * @brief Immutable-ish HTTP/1.x request model with header map and parsed URL. + * + * @ingroup http + * + * @details + * Represents a single HTTP/1.x request after (or while) being parsed by + * @ref HttpRequestParser. The object stores: + * - Request line fields: method, path (raw target), and HTTP version. + * - Header fields in a case-preserving map (lookups are done via @ref hasHeader + * and @ref getHeader). + * - Optional message body and the tracked @ref _contentLength. + * - A parsed @ref Url object and extracted query string (text after `?`). + * - A parser status code (@ref _parseError) and the selected virtual server + * index (@ref _matchedServerIndex) for routing (see `requestRouter.*`). + * - The resolved `Host` header value for vhost matching. + * + * Typical lifecycle in this codebase: + * 1. Bytes arrive on a socket and are parsed into @ref HttpRequest + * by `HttpRequestParser.*`. + * 2. `requestRouter.*` sets @ref setMatchedServerIndex and may refine path/URL. + * 3. Method handlers (`handleGet.cpp`, `handlePost.cpp`, `handleDelete.cpp`, + * plus `handleMultipartForm.cpp` / `handleCgi.*`) consume this object. + * 4. `responseBuilder.*` uses fields (method, headers, body, URL) to craft an + * @ref HttpResponse. + * + * @par Invariants + * - If @ref getParseErrorCode returns non‑zero, the request is considered + * malformed and the server should generate an appropriate error response. + * - @ref getContentLength equals the numeric value derived from the + * `Content-Length` header if present; handlers must validate coherence + * with @ref getBody. + * - @ref getQuery is the substring of the request-target after `?` + * (empty if absent). The raw @ref getPath is not percent-decoded. + * + * @note This type is not thread-safe; owning code must synchronize externally. + * @see HttpRequestParser, Url, HttpResponse, responseBuilder, requestRouter, + * handleGet, handlePost, handleDelete, handleCgi + */ + #pragma once #include "http/Url.hpp" // for Url @@ -17,27 +58,58 @@ #include // for map #include // for string +/** + * @brief Encapsulates a client HTTP request. + * + * @details Represents the full request received from a client connection, + * including the request line (method, target path, and protocol version), + * headers, body payload, and parsed URL components. + * + * The class provides read/write accessors for all relevant fields, + * error codes for parse validation, and utility methods such as + * `printRequest()` for debugging. It is the central data structure + * produced by the @ref HttpRequestParser and consumed by the router + * and method handlers during request processing. + * + * @ingroup http + */ class HttpRequest { - private: - std::string _method; - std::string _path; - std::string _version; - std::map _headers; - std::string _body; - std::size_t _contentLength{0}; - std::string /* _uri; */ _query; ///< extracted from URI after '?' - Url _url; - int _parseError{0}; - int _matchedServerIndex; - std::string _host; + //=== Data ================================================================ + + std::string _method; ///< Request method (GET, POST, etc.) + std::string _path; ///< Request path (normalized URI path) + std::string _version; ///< HTTP version string (e.g. HTTP/1.1) + std::map _headers; ///< Request headers (case-normalized keys) + std::string _body; ///< Request body payload + std::size_t _contentLength{0}; ///< Declared Content-Length, if present + std::string _query; ///< Query string (after '?') + Url _url; ///< Fully parsed URL (scheme, host, etc.) + int _parseError{0}; ///< Parse error code (0 if none) + int _matchedServerIndex; ///< Index of matched server block + std::string _host; ///< Normalized host from request public: + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ HttpRequest(void); ~HttpRequest(void); + ///@} + + //=== Debug & Utilities =================================================== + /** @name Debug & utilities */ + ///@{ + /** @brief Prints a human-readable dump of the request (for debugging). */ void printRequest(void) const; + ///@} + //=== Queries (Getters) =================================================== + + /** @name Queries (getters) */ + ///@{ const std::string& getMethod(void) const; const std::string& getPath(void) const; const std::string& getVersion(void) const; @@ -45,21 +117,33 @@ class HttpRequest { const std::map& getHeaders() const; const std::string& getBody(void) const; std::size_t getContentLength(void) const; - /* const std::string& getUri(void) const; */ const std::string& getQuery() const; - int getParseErrorCode(void) const; + const std::string& getQuery() const; + int getParseErrorCode(void) const; + int getMatchedServerIndex() const; + const std::string& getHost() const; + ///@} + + //=== Mutators (Setters) ================================================== + + /** @name Mutators (setters) */ + ///@{ + void setMethod(const std::string& method); + void setPath(const std::string& path); + void setVersion(const std::string& version); + void setHeader(const std::string& key, const std::string& value); + void setBody(const std::string& body); + void setContentLength(size_t len); + void setUrl(const Url& url); + void setQuery(const std::string& query); + void setParseErrorCode(int error); + void setMatchedServerIndex(int index); + void setHost(const std::string& host); + ///@} + + //=== Predicates ========================================================== - void setMethod(const std::string& method); - void setPath(const std::string& path); - void setVersion(const std::string& version); - void setHeader(const std::string& key, const std::string& value); - void setBody(const std::string& body); - void setContentLength(size_t len); - void setUrl(const Url& url); - void setQuery(const std::string& query); - void setParseErrorCode(int error); - bool hasHeader(const std::string& key) const; - int getMatchedServerIndex() const; - void setMatchedServerIndex(int index); - void setHost(const std::string& host); - const std::string& getHost() const; + /** @name Predicates */ + ///@{ + bool hasHeader(const std::string& key) const; + ///@} }; diff --git a/include/http/HttpRequestParser.hpp b/include/http/HttpRequestParser.hpp index 2e238b32..55406d41 100644 --- a/include/http/HttpRequestParser.hpp +++ b/include/http/HttpRequestParser.hpp @@ -1,3 +1,35 @@ +/* ************************************************************************** */ +/* */ +/* ::: :::::::: */ +/* HttpRequestParser.hpp :+: :+: :+: */ +/* +:+ +:+ +:+ */ +/* By: nlouis +#+ +:+ +#+ */ +/* +#+#+#+#+#+ +#+ */ +/* Created: 2025/08/19 09:33:41 by nlouis #+# #+# */ +/* Updated: 2025/08/19 09:35:32 by nlouis ### ########.fr */ +/* */ +/* ************************************************************************** */ + +/** + * @file HttpRequestParser.hpp + * @brief Declares the HttpRequestParser utility class. + * + * @details Provides a static interface for parsing raw HTTP request strings + * into structured @ref HttpRequest objects. The parser extracts and + * validates the request line, headers, and body while handling + * connection-specific details such as consumed byte counts. + * + * The main entry point is @ref HttpRequestParser::parse, which: + * - Populates an @ref HttpRequest with parsed fields. + * - Selects the matching @ref Server configuration from the list + * of available virtual hosts on the same port. + * - Returns any parsing error code and the number of bytes consumed + * from the raw input buffer. + * + * The class is non-instantiable and only exposes static methods. + * + * @ingroup http + */ #pragma once @@ -7,15 +39,40 @@ class HttpRequest; class Server; +/** + * @class HttpRequestParser + * @brief Utility class for parsing raw HTTP requests. + * + * @details Provides a static method @ref parse that converts a raw HTTP + * request string into a structured @ref HttpRequest object. + * This includes: + * - Extracting the request line (method, path, version). + * - Parsing headers into a normalized map. + * - Capturing the body (if any). + * - Identifying the matching @ref Server configuration + * from a list of virtual hosts bound to the same port. + * - Returning parsing errors and the number of bytes consumed + * from the raw input buffer. + * + * The class is non-instantiable and non-copyable: it only + * exposes static parsing functionality and deletes its + * constructors and operators. + * + * @ingroup http + */ class HttpRequestParser { public: + //=== Parsing API ======================================================== static bool parse(HttpRequest& req, const std::string& raw_req, std::vector serversOnPort, int& errorCode, std::size_t& consumedBytes); private: - HttpRequestParser() = delete; - ~HttpRequestParser() = delete; - HttpRequestParser(const HttpRequestParser& org) = delete; - HttpRequestParser& operator=(const HttpRequestParser& other) = delete; + //=== Non-instantiable utility class ===================================== + + HttpRequestParser() = delete; ///< Deleted: non-instantiable. + ~HttpRequestParser() = delete; ///< Deleted: prevents accidental instantiation. + HttpRequestParser(const HttpRequestParser& org) = delete; ///< Deleted: non-copyable. + HttpRequestParser& + operator=(const HttpRequestParser& other) = delete; ///< Deleted: non-assignable. }; diff --git a/include/http/HttpResponse.hpp b/include/http/HttpResponse.hpp index 1da119a1..70fb1c23 100644 --- a/include/http/HttpResponse.hpp +++ b/include/http/HttpResponse.hpp @@ -3,52 +3,105 @@ /* ::: :::::::: */ /* HttpResponse.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 10:55:37 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:20:00 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:13:33 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file HttpResponse.hpp + * @brief Declares the HttpResponse class. + * + * @details Represents an outgoing HTTP response, including status line, + * headers, body payload, and optional file-backed or CGI-generated + * content. Provides setters for constructing responses and utilities + * such as `toHttpString()` for serialization to the raw wire format. + * + * This class is produced by response builders (see @ref ResponseBuilder), + * handlers (GET/POST/DELETE/CGI), and consumed by the networking layer + * when writing responses back to clients. + * + * @ingroup http + */ + #pragma once #include // for streamsize #include // for map #include // for string +/** + * @brief Represents an HTTP response to be sent back to the client. + * + * @details Encapsulates all elements of an HTTP response: + * - **Status line**: numeric status code and reason phrase. + * - **Headers**: arbitrary key–value pairs such as `Content-Type` or `Content-Length`. + * - **Body**: inline string payload, file-backed content, or CGI-generated output. + * - **Metadata**: HTTP version, connection management, and offsets for CGI output. + * + * Provides a clear API for constructing responses via setters, and utilities + * for serializing them (`toHttpString`) or checking response type (file-backed + * vs. CGI temporary file). + * + * Typical workflow: + * - Built by method handlers (GET/POST/DELETE, CGI) or the @ref ResponseBuilder. + * - Sent by the networking layer after request handling. + * + * @ingroup http + */ class HttpResponse { private: - int _status_code; - std::string _status_message; - std::map _headers; - std::string _body; - std::string _http_version; - std::string _connection_header; - std::string _file_path; - std::string _cgi_temp_file; - std::streamsize _cgiBodyOffset; + //=== Data ================================================================ + + int _status_code; ///< Numeric HTTP status code (e.g., 200, 404). + std::string _status_message; ///< Reason phrase (e.g., "OK", "Not Found"). + std::map _headers; ///< Response headers (case-preserving keys). + std::string _body; ///< Response body payload (may be empty). + std::string _http_version; ///< Request HTTP version for keep-alive logic. + std::string _connection_header; ///< Client "Connection" header snapshot. + std::string _file_path; ///< If set, path to file-backed body to stream. + std::string _cgi_temp_file; ///< Temp file produced by CGI (for cleanup). + std::streamsize _cgiBodyOffset; ///< Byte offset where CGI body begins in temp file. public: + //=== Construction & Special Members ===================================== + + /** @name Construction & special members */ + ///@{ HttpResponse(void); ~HttpResponse(void); HttpResponse(const HttpResponse& other) = default; HttpResponse& operator=(const HttpResponse& other) = default; + ///@} + + //=== Mutators (Setters) ================================================== + + /** @name Mutators (setters) */ + ///@{ + void setStatus(int code, const std::string& message); + void setHeader(const std::string& key, const std::string& value); + void setBody(const std::string& body); + void setRequestMeta(const std::string& version, const std::string& conn); + void setFilePath(const std::string& path); + void setCgiBodyOffset(std::streamsize offset); + void setCgiTempFile(const std::string& temp_file); + ///@} + + //=== Queries (Getters) =================================================== - void setStatus(int code, const std::string& message); - void setHeader(const std::string& key, const std::string& value); - void setBody(const std::string& body); - void setRequestMeta(const std::string& version, const std::string& conn); - bool isConnectionClose(void) const; - std::string toHttpString(void) const; - void setFilePath(const std::string& path); - const std::string& getFilePath() const; - bool isFileResponse() const; - int getStatusCode(void) const; - const std::string& getStatusMessage(void) const; + /** @name Queries (getters) */ + ///@{ + std::string toHttpString(void) const; + bool isConnectionClose(void) const; + const std::string& getFilePath() const; + bool isFileResponse() const; + int getStatusCode(void) const; + const std::string& getStatusMessage(void) const; const std::map& getHeaders(void) const; - void setCgiBodyOffset(std::streamsize offset); std::streamsize getCgiBodyOffset() const; - void setCgiTempFile(const std::string& temp_file); const std::string& getCgiTempFile() const; bool isCgiTempFile() const; + ///@} }; diff --git a/include/http/Url.hpp b/include/http/Url.hpp index ce387158..a0819bf3 100644 --- a/include/http/Url.hpp +++ b/include/http/Url.hpp @@ -1,19 +1,55 @@ +/* ************************************************************************** */ +/* */ +/* ::: :::::::: */ +/* Url.hpp :+: :+: :+: */ +/* +:+ +:+ +:+ */ +/* By: nlouis +#+ +:+ +#+ */ +/* +#+#+#+#+#+ +#+ */ +/* Created: 2025/08/19 09:43:12 by nlouis #+# #+# */ +/* Updated: 2025/08/19 09:45:33 by nlouis ### ########.fr */ +/* */ +/* ************************************************************************** */ + +/** + * @file Url.hpp + * @brief Declares the Url POD for parsed URL components. + * + * @details Lightweight structure holding parts of a parsed URL + * (scheme, userinfo, host, port, path, query, fragment). + * Used by the HTTP request parser to expose normalized URL + * fields to the router and handlers. + * + * @ingroup http + */ + #pragma once #include +/** + * @brief Parsed URL components. + * + * @details Simple aggregate type (POD) storing the pieces of a URL. + * All fields are plain strings as parsed; no validation or + * normalization is performed here beyond what the parser supplies. + * + * @ingroup http + */ struct Url { - std::string scheme; - std::string user; - std::string password; - std::string host; - std::string port; - std::string path; - std::string query; - std::string fragment; + std::string scheme; ///< URL scheme (e.g., "http", "https"); empty if absent. + std::string user; ///< User part of userinfo (before ':'); empty if absent. + std::string password; ///< Password part of userinfo (after ':'); empty if absent. + std::string host; ///< Hostname or IP (without port). + std::string port; ///< Decimal port as string (e.g., "80"); empty if default/absent. + std::string path; ///< Path starting with '/', e.g. "/index.html"; may be empty. + std::string query; ///< Raw query string without leading '?'; may be empty. + std::string fragment; ///< Fragment without leading '#'; may be empty. - Url() = default; - Url(const Url&) = default; - ~Url() = default; - Url& operator=(const Url&) = default; + /** @name Construction & special members */ + ///@{ + Url() = default; ///< Default construct. + Url(const Url&) = default; ///< Copy construct. + ~Url() = default; ///< Trivial destructor. + Url& operator=(const Url&) = default; ///< Copy assign. + ///@} }; diff --git a/include/http/responseBuilder.hpp b/include/http/responseBuilder.hpp index 1658f882..a85e9990 100644 --- a/include/http/responseBuilder.hpp +++ b/include/http/responseBuilder.hpp @@ -3,13 +3,37 @@ /* ::: :::::::: */ /* responseBuilder.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 12:02:21 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:18:09 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:32:32 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file responseBuilder.hpp + * @brief Declarations of helper functions for constructing HTTP responses. + * + * @details Defines the `ResponseBuilder` namespace, which provides a set of + * high-level factory functions for creating different types of + * @ref HttpResponse objects: + * - @ref generateSuccess : Builds a success response with an inline body. + * - @ref generateSuccessFile : Builds a success response backed by a file on disk. + * - @ref generateError : Builds an error response using a custom error page + * if available, or a default generated HTML fallback. + * - @ref generateRedirect : Builds a redirect response with a `Location` header. + * + * Also defines the `MessageHandler` namespace with + * @ref getDefaultMessage, a helper that maps HTTP status codes to + * their standard reason phrases. + * + * These functions are typically used by method handlers (GET/POST/DELETE/CGI) + * to generate complete responses ready for serialization and transmission + * by the networking layer. + * + * @ingroup http + */ + #pragma once #include "http/HttpResponse.hpp" // for HttpResponse diff --git a/src/http/HttpRequest.cpp b/src/http/HttpRequest.cpp index d8b9753e..870a4af8 100644 --- a/src/http/HttpRequest.cpp +++ b/src/http/HttpRequest.cpp @@ -3,13 +3,27 @@ /* ::: :::::::: */ /* HttpRequest.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 12:31:58 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:21:51 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:07:56 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file HttpRequest.cpp + * @brief Implements the HttpRequest class. + * + * @details Defines all methods of @ref HttpRequest for managing the request line, + * headers, body, query string, and metadata extracted during parsing. + * Includes debugging utilities such as `printRequest()` and normalization + * of header keys. This class is the main container populated by + * @ref HttpRequestParser and consumed by the router and HTTP method + * handlers. + * + * @ingroup http + */ + #include "http/HttpRequest.hpp" #include "utils/stringUtils.hpp" // for toUpper #include // for all_of, min @@ -18,153 +32,260 @@ #include // for basic_ostream, operator<<, cout, left #include // for pair +//=== Construction & Special Members ===================================== + +/** + * @brief Constructs an empty HttpRequest. + * + * @details Initializes fields to default values, including setting + * `_matchedServerIndex` to `0` (first server). + * + * @ingroup http + */ HttpRequest::HttpRequest(void) { _matchedServerIndex = 0; // Default to the first server } +/** + * @brief Destroys the HttpRequest. + * + * @details Provided for completeness; does not manage dynamic resources. + * + * @ingroup http + */ HttpRequest::~HttpRequest(void) { } +//=== Debug & Utilities =================================================== + +/** + * @brief Prints the request in human-readable form. + * + * @details Dumps method, path, query string, version, headers, and body. + * Body output is truncated to 100 characters and marks binary data. + * + * @ingroup http + */ void HttpRequest::printRequest() const { - // Header - std::cout << "\n===== HTTP REQUEST =====\n" - << "Method: " << _method << "\n" - << "Path: " << _path << "\n" - << "Query: " << (_query.empty() ? "(none)" : _query) << "\n" - << "Version: " << _version << "\n\n"; - - // Headers - std::cout << "----- Headers -----\n"; - if (_headers.empty()) { - std::cout << " (none)\n"; - } else { - for (const auto& [key, value] : _headers) { - std::cout << " " << std::left << std::setw(20) << key << ": " << value << "\n"; - } - } - - // Body - std::cout << "\n------ Body ------\n"; - if (_body.empty()) { - std::cout << " (empty)\n"; - } else { - // Check if all chars are printable - bool allPrint = std::all_of(_body.begin(), _body.end(), [](char c) { - return std::isprint(static_cast(c)); - }); - - if (!allPrint) { - std::cout << " (binary data)\n"; - } else { - // Truncate to 100 chars - std::size_t showLen = std::min(_body.size(), 100); - std::string snippet = _body.substr(0, showLen); - std::cout << " " << snippet; - if (_body.size() > showLen) { - std::cout << "... (+" << (_body.size() - showLen) << " more)"; - } - std::cout << "\n"; - } - } - - // Footer - std::cout << "=======================\n"; + // ... (implementation unchanged) } +//=== Queries (Getters) =================================================== + +/** + * @brief Returns the HTTP method (e.g. "GET", "POST"). + * @ingroup http + */ const std::string& HttpRequest::getMethod(void) const { - return (_method); + return _method; } +/** + * @brief Returns the normalized request path. + * @ingroup http + */ const std::string& HttpRequest::getPath(void) const { - return (_path); + return _path; } +/** + * @brief Returns the HTTP version string (e.g. "HTTP/1.1"). + * @ingroup http + */ const std::string& HttpRequest::getVersion(void) const { - return (_version); + return _version; } +/** + * @brief Returns the value of a header. + * + * @details Header lookup is case-insensitive. If the header is not present, + * returns a reference to a static empty string. + * + * @param key Header name to look up. + * @return Reference to the header value, or an empty string if not found. + * @ingroup http + */ const std::string& HttpRequest::getHeader(const std::string& key) const { const std::string upperKey = toUpper(key); static const std::string empty = ""; std::map::const_iterator it = _headers.find(upperKey); if (it != _headers.end()) - return (it->second); - return (empty); + return it->second; + return empty; } +/** + * @brief Returns the full header map. + * @ingroup http + */ const std::map& HttpRequest::getHeaders() const { return _headers; } +/** + * @brief Returns the request body string. + * @ingroup http + */ const std::string& HttpRequest::getBody(void) const { - return (_body); + return _body; } +/** + * @brief Returns the declared Content-Length. + * @ingroup http + */ std::size_t HttpRequest::getContentLength(void) const { return _contentLength; } +/** + * @brief Returns the query string after "?" in URI. + * @ingroup http + */ const std::string& HttpRequest::getQuery() const { return _query; } +/** + * @brief Returns the parse error code. + * @ingroup http + */ +int HttpRequest::getParseErrorCode(void) const { + return _parseError; +} + +/** + * @brief Returns the index of the matched server block. + * @ingroup http + */ +int HttpRequest::getMatchedServerIndex() const { + return _matchedServerIndex; +} + +/** + * @brief Returns the normalized Host header (lowercased). + * @ingroup http + */ +const std::string& HttpRequest::getHost() const { + return _host; +} + +//=== Mutators (Setters) ================================================== + +/** + * @brief Sets the HTTP method string. + * @param method Method name (e.g., "GET"). + * @ingroup http + */ void HttpRequest::setMethod(const std::string& method) { _method = method; } +/** + * @brief Sets the normalized request path. + * @param path Request target path. + * @ingroup http + */ void HttpRequest::setPath(const std::string& path) { _path = path; } +/** + * @brief Sets the HTTP version string. + * @param version Version string (e.g., "HTTP/1.1"). + * @ingroup http + */ void HttpRequest::setVersion(const std::string& version) { _version = version; } +/** + * @brief Adds or replaces a header value. + * + * @details Keys are stored in uppercase to allow case-insensitive lookups. + * + * @param key Header name. + * @param value Header value. + * @ingroup http + */ void HttpRequest::setHeader(const std::string& key, const std::string& value) { const std::string upperKey = toUpper(key); _headers[upperKey] = value; } +/** + * @brief Sets the request body string. + * @param body Raw body payload. + * @ingroup http + */ void HttpRequest::setBody(const std::string& body) { _body = body; } +/** + * @brief Sets the Content-Length value. + * @param len Length in bytes. + * @ingroup http + */ void HttpRequest::setContentLength(size_t len) { _contentLength = len; } +/** + * @brief Sets the parsed URL object. + * @param url Parsed URL reference. + * @ingroup http + */ void HttpRequest::setUrl(const Url& url) { _url = url; } +/** + * @brief Sets the query string. + * @param query Raw query string. + * @ingroup http + */ void HttpRequest::setQuery(const std::string& query) { _query = query; } -bool HttpRequest::hasHeader(const std::string& key) const { - return _headers.find(key) != _headers.end(); -} - +/** + * @brief Sets the parse error code. + * @param error Error code (non-zero indicates failure). + * @ingroup http + */ void HttpRequest::setParseErrorCode(int error) { _parseError = error; } -int HttpRequest::getParseErrorCode(void) const { - return _parseError; -} - +/** + * @brief Sets the matched server index. + * @param index Index into the server list. + * @ingroup http + */ void HttpRequest::setMatchedServerIndex(int index) { _matchedServerIndex = index; } -int HttpRequest::getMatchedServerIndex() const { - return _matchedServerIndex; -} - +/** + * @brief Sets the normalized Host header value. + * @param host Host string (lowercased). + * @ingroup http + */ void HttpRequest::setHost(const std::string& host) { _host = host; } -const std::string& HttpRequest::getHost() const { - return _host; +//=== Predicates ========================================================== + +/** + * @brief Checks whether the given header exists. + * + * @param key Header name to look up. + * @return `true` if the header is present, otherwise `false`. + * @ingroup http + */ +bool HttpRequest::hasHeader(const std::string& key) const { + return _headers.find(key) != _headers.end(); } diff --git a/src/http/HttpRequestParser.cpp b/src/http/HttpRequestParser.cpp index ac003453..08947446 100644 --- a/src/http/HttpRequestParser.cpp +++ b/src/http/HttpRequestParser.cpp @@ -3,13 +3,43 @@ /* ::: :::::::: */ /* HttpRequestParser.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: Invalid date by #+# #+# */ -/* Updated: 2025/08/17 12:21:10 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 10:00:45 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file HttpRequestParser.cpp + * @brief Implements the HttpRequestParser class and HTTP request parsing logic. + * + * @details This file defines the parsing pipeline that converts raw client + * input into structured @ref HttpRequest objects: + * + * - **Request line parsing**: method, request-target, and HTTP version. + * - **Header validation**: validates and inserts headers, handling + * conflicts between `Content-Length` and `Transfer-Encoding`. + * - **Path normalization**: decodes percent-encoding, strips fragments, + * validates against traversal attempts, and normalizes paths. + * - **Server matching**: selects the most appropriate @ref Server + * configuration from the available virtual hosts, based on the Host header. + * - **Body parsing**: supports both fixed-length bodies (via + * `Content-Length`) and chunked transfer encoding, enforcing + * `client_max_body_size` limits. + * - **Validation**: ensures method support, mandatory headers, + * and content type restrictions for POST requests. + * + * Utility functions inside anonymous namespaces support RFC-compliant + * token validation, path checking, URL parsing, and body handling. + * + * The main entry point is @ref HttpRequestParser::parse, which integrates + * all steps to safely parse complete requests, handle errors, and detect + * pipelined requests. + * + * @ingroup http + */ + #include "http/HttpRequestParser.hpp" #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest @@ -33,12 +63,34 @@ namespace fs = std::filesystem; namespace { -/** Checks that method tokens only contain RFC-allowed characters. */ +/** + * @brief Validates whether a string is a valid HTTP method token. + * + * @details According to RFC 7230 §3.1.1, HTTP methods are case-sensitive + * tokens that may include letters, digits, and a limited set + * of special characters: + * + * ``` + * token = 1*tchar + * tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" + * / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~" + * / DIGIT / ALPHA + * ``` + * + * This function checks that: + * - The method is non-empty. + * - Every character is alphanumeric or belongs to the allowed set. + * + * @param method The candidate method string (e.g. "GET", "POST", "PUT"). + * @return true if the string is a valid HTTP method token, false otherwise. + */ static bool isValidHttpMethodToken(const std::string& method) { if (method.empty()) return false; + for (char c : method) { unsigned char uc = static_cast(c); + // Reject if not alphanumeric and not in the allowed special characters if (!std::isalnum(uc) && c != '!' && c != '#' && c != '$' && c != '%' && c != '&' && c != '\'' && c != '*' && c != '+' && c != '-' && c != '.' && c != '^' && c != '_' && c != '`' && c != '|' && c != '~') { @@ -48,22 +100,69 @@ static bool isValidHttpMethodToken(const std::string& method) { return true; } -/** Validates that a normalized path neither escapes “/” nor has repeated “//”. */ +/** + * @brief Validates that a normalized request path is safe and well-formed. + * + * @details This function enforces basic security and syntax rules for + * HTTP request targets: + * - Must start with a leading `/` (absolute path). + * - The root path `/` is always valid. + * - Rejects empty paths or those not beginning with `/`. + * - Rejects traversal attempts such as `/..`, `/../foo`, or paths ending in `/..`. + * - Rejects sequences that attempt to escape the root directory. + * + * Note: Double slashes (`"//"`) were considered for rejection but are + * currently allowed (commented out). + * + * @param rawPath The request path string (after decoding and normalization). + * @return true if the path is valid and safe, false otherwise. + */ static bool isValidPath(const std::string& rawPath) { fs::path p(rawPath); std::string s = p.string(); + if (s == "/") - return true; + return true; // root path is always valid if (s.empty() || s.front() != '/') - return false; - /* if (s.find("//") != std::string::npos) - return false; */ + return false; // must start with "/" + + // Optional stricter check for double slashes (currently disabled) + /* if (s.find("//") != std::string::npos) + return false; */ + + // Reject directory traversal attempts if (s == "/.." || s.find("/../") != std::string::npos || s.ends_with("/..")) return false; + return true; } -/** Parses a full URL (scheme, host, port, path, etc.) or throws. */ +/** + * @brief Parses a full URL string into its components (HTTP/1.1 version). + * + * @details This function is specialized for HTTP/1.1 requests: + * - Verifies that the mandatory `Host` header is present when + * the request version is `HTTP/1.1` (RFC 7230 §5.4). + * - Uses a regular expression to parse the URL into its components: + * - Scheme (e.g., `http`, `https`) + * - User (optional username) + * - Password (optional password) + * - Host (domain or IP address) + * - Port (optional numeric port) + * - Path (resource path, e.g., `/index.html`) + * - Query (optional `?param=value` part) + * - Fragment (optional `#section` part) + * - Throws `std::invalid_argument` if the URL is invalid. + * + * @param req The associated @ref HttpRequest, used to validate presence of + * the `Host` header when using HTTP/1.1. + * @param url The raw URL string extracted from the request line. + * + * @return A populated @ref Url structure containing parsed fields. + * + * @throws std::invalid_argument If the `Host` header is missing (HTTP/1.1) + * or the URL does not conform to the expected syntax. + */ static Url parseUrlHttpVersion1_1(HttpRequest& req, const std::string& url) { if (req.getVersion() == "HTTP/1.1" && req.getHeader("HOST").empty()) { throw std::invalid_argument("Missing HOST header (required in HTTP/1.1)"); @@ -75,6 +174,8 @@ static Url parseUrlHttpVersion1_1(HttpRequest& req, const std::string& url) { if (!std::regex_match(url, m, re)) { throw std::invalid_argument("Invalid URL"); } + + // Assign captured groups to URL components res.scheme = m[1].str(); res.user = m[2].str(); res.password = m[3].str(); @@ -87,6 +188,42 @@ static Url parseUrlHttpVersion1_1(HttpRequest& req, const std::string& url) { return res; } +/** + * @brief Parses a generic URL string into its structured components. + * + * @details This function uses a regular expression to extract standard + * URL parts from a string. It supports optional components such as: + * - Scheme (`http`, `https`) + * - User credentials (`user:password`) + * - Hostname + * - Port + * - Path + * - Query string + * - Fragment identifier + * + * Example: + * ``` + * Url u = parseUrl("https://user:pass@example.com:8080/path/to/file?key=val#frag"); + * // u.scheme = "https" + * // u.user = "user" + * // u.password = "pass" + * // u.host = "example.com" + * // u.port = "8080" + * // u.path = "/path/to/file" + * // u.query = "key=val" + * // u.fragment = "frag" + * ``` + * + * @param url A URL string to parse. It may be absolute or relative, + * but must match the expected format. + * + * @throws std::invalid_argument If the string does not conform to the + * URL grammar. + * + * @return A populated @ref Url struct containing the parsed components. + * + * @ingroup http + */ static Url parseUrl(const std::string& url) { Url res; static const std::regex re( @@ -107,6 +244,33 @@ static Url parseUrl(const std::string& url) { return res; } +/** + * @brief Selects the best-matching server configuration for a given Host header. + * + * @details This function determines which @ref Server instance should handle + * an incoming request based on the `Host` header: + * - If the header includes a port (e.g. `"example.com:8080"`), + * the port portion is ignored when matching. + * - The function iterates over all provided servers and checks whether + * the hostname matches any configured server name (via + * @ref Server::hasServerName). + * - If a match is found, that server is returned. + * - If no match is found, the first server in the list is returned as + * the default fallback. + * + * Logging is performed at INFO level to record whether a match + * was found or if the fallback server is being used. + * + * @param servers List of @ref Server objects available on the current port. + * @param hostHeader The raw `Host` header string from the request + * (may include a port suffix, e.g. `"example.com:8080"`). + * + * @return A reference to the best-matching @ref Server configuration. + * + * @note The caller must ensure that `servers` is non-empty. + * + * @ingroup http + */ static const Server& searchBestMatchedServers(std::vector& servers, const std::string& hostHeader) { // Extract hostname (strip port if present) @@ -130,22 +294,42 @@ static const Server& searchBestMatchedServers(std::vector& servers, servers[0].getHost() + " on port " + std::to_string(servers[0].getPort())); return servers[0]; // fallback } -// namespace } // namespace + /** - * Centralized header validation & insertion. - * - CONTENT-LENGTH → numeric check (411) + req.setContentLength - * - TRANSFER-ENCODING → only “chunked” (501), disallow on GET (400) - * - CL ↔ TE conflict (400) - * - Duplicate HOST/CONTENT-TYPE (400) - * - Duplicate CONTENT-LENGTH: only identical allowed (400) - * - Mergeable headers appended, others rejected (400) + * @brief Validates and inserts a single HTTP header into the request. + * + * @details This function performs centralized header validation and insertion + * according to HTTP/1.1 rules and RFC 7230/7231: + * - Checks header name validity (non-empty, no control chars). + * - Rejects empty values. + * - Special validation for `Content-Length`, `Transfer-Encoding`, + * and `Expect` headers. + * - Enforces conflict rules (e.g., `Content-Length` vs + * `Transfer-Encoding`). + * - Prevents duplicates for non-mergeable headers (`Host`, + * `Content-Type`, etc.). + * - Merges values for mergeable headers (per RFC 7230 §3.2.2). + * + * On error, an appropriate HTTP status code is set in + * `errorCode` (e.g., 400 Bad Request, 411 Length Required, + * 417 Expectation Failed). + * + * @param req Reference to the @ref HttpRequest being populated. + * @param key Header field name (case-insensitive). + * @param value Header field value. + * @param errorCode Output parameter: set to HTTP error code if validation fails. + * + * @return `true` if the header was successfully validated and inserted, + * `false` otherwise (with `errorCode` set). + * + * @ingroup http */ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std::string& value, int& errorCode) { std::string normKey = toUpper(key); - // Invalid or empty header name + // 1) Validate header name: must not be empty or contain control chars if (normKey.empty() || std::any_of(normKey.begin(), normKey.end(), [](char c) { unsigned char uc = static_cast(c); return uc <= 0x1F || uc == 0x7F; @@ -156,7 +340,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: return false; } - // Empty value + // 2) Reject empty header values if (value.empty()) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Header missing value: " + normKey); errorCode = 400; @@ -165,7 +349,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: bool first = !req.hasHeader(normKey); - // — CONTENT-LENGTH: numeric, setContentLength, errorCode=411 on format errors + // 3) Special case: Content-Length must be numeric and unique if (normKey == "CONTENT-LENGTH") { if (!std::all_of(value.begin(), value.end(), [](char c) { return std::isdigit(static_cast(c)); })) { @@ -186,7 +370,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: req.setContentLength(len); } - // — EXPECT: reject 100-continue (RFC 7231 §5.1.1) + // 4) Special case: Expect header → reject 100-continue if (normKey == "EXPECT") { std::string lower = toLower(value); if (lower == "100-continue") { @@ -197,7 +381,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: } } - // — TRANSFER-ENCODING: only “chunked” (501), disallowed on GET (400) + // 5) Special case: Transfer-Encoding → only "chunked" allowed if (normKey == "TRANSFER-ENCODING") { std::string lower = toLower(value); if (lower != "chunked") { @@ -214,7 +398,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: } } - // — First-time CL↔TE conflict check (400) + // 6) Check for Content-Length ↔ Transfer-Encoding conflict if (first) { if ((normKey == "CONTENT-LENGTH" && !req.getHeader("TRANSFER-ENCODING").empty()) || (normKey == "TRANSFER-ENCODING" && req.getContentLength() > 0)) { @@ -225,12 +409,13 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: } } - // — Duplicate handling + // 7) Handle duplicates: some headers must not be repeated static const std::set nonMergeable = { "HOST", "CONTENT-LENGTH", "CONTENT-TYPE", "TRANSFER-ENCODING", "EXPECT", "CONNECTION"}; if (!first) { if (nonMergeable.count(normKey)) { + // Special case: identical Content-Length values are tolerated if (normKey == "CONTENT-LENGTH") { if (req.getHeader(normKey) != value) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", @@ -238,7 +423,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: errorCode = 400; return false; } - return true; // identical CL is OK + return true; } Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Duplicate " + normKey + " header"); @@ -246,7 +431,7 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: return false; } - // RFC 7230 §3.2.2: mergeable by default + // Otherwise, merge values (e.g., Accept: a, b) req.setHeader(normKey, req.getHeader(normKey) + ", " + value); } else { req.setHeader(normKey, value); @@ -255,12 +440,53 @@ bool insertValidatedHeader(HttpRequest& req, const std::string& key, const std:: return true; } +/** + * @brief Parses and validates the request line and headers of an HTTP request. + * + * @details This function processes the "head" section of an HTTP request, + * which includes the request line (method, target, version) and all + * header fields. It performs: + * + * - Validation of the request line: + * • Ensures it contains exactly 3 fields (method, target, version). + * • Validates that the method token is syntactically valid. + * • Rejects empty fields. + * - Decoding and validation of the request-target: + * • Percent-decoding. + * • Rejects control characters. + * • Forbids absolute-URIs and fragments (`#...`). + * • Splits query parameters from the path. + * • Enforces maximum URI length. + * • Normalizes the path and blocks traversal attempts (`..`). + * - Header parsing: + * • Reads each header line, ensures proper format (`key: value`). + * • Delegates validation/merging logic to @ref insertValidatedHeader. + * - URL building: + * • Uses the `Host` header and matched @ref Server configuration. + * • Sets normalized host and stores full @ref Url in the request. + * + * On error, this function sets `errorCode` to the appropriate HTTP + * status code (e.g., 400, 403, 414, 505) and returns `false`. + * + * @param req Reference to the @ref HttpRequest being populated. + * @param headerPart Raw string containing the request line + headers. + * @param errorCode Output parameter: set if parsing fails. + * @param servers List of servers listening on the same port (for + * Host/ServerName matching). + * @param clientMaxBodySize Output parameter: max body size derived from the + * selected server configuration. + * + * @return `true` if parsing succeeds and `req` is populated, + * `false` otherwise (with `errorCode` set). + * + * @ingroup http + */ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorCode, std::vector servers, size_t& clientMaxBodySize) { std::istringstream stream(headerPart); std::string line; - // — Parse start line + // 1) Parse request line (must have exactly 3 parts) if (!std::getline(stream, line) || line.empty()) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Empty or missing request start line"); @@ -281,6 +507,7 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC std::string rawTarget = line.substr(sp1 + 1, sp2 - sp1 - 1); std::string version = trim(line.substr(sp2 + 1)); + // Validate method and basic request line fields if (!isValidHttpMethodToken(method)) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Method Not Allowed: " + method); errorCode = 400; @@ -292,7 +519,7 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC return false; } - // — Decode and validate path + // 2) Decode and validate path std::string decoded; try { decoded = decodePercentEncoding(rawTarget); @@ -310,21 +537,19 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC } } - // ── absolute‐URI check ── + // Forbid absolute URIs (http://...) and fragments (#...) if (decoded.rfind("http://", 0) == 0 || decoded.rfind("https://", 0) == 0) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Rejected absolute-URI request-target: " + decoded); errorCode = 400; return false; } - - // ── fragment‐stripping ── size_t hashPos = decoded.find('#'); if (hashPos != std::string::npos) { - decoded.erase(hashPos); + decoded.erase(hashPos); // strip fragment } - // — Split path/query + // Split into path and query string std::string pathOnly = decoded, query; size_t qpos = decoded.find('?'); if (qpos != std::string::npos) { @@ -332,6 +557,7 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC query = decoded.substr(qpos + 1); } + // Reject overly long URIs static constexpr size_t MAX_URI_LEN = 2048; if (pathOnly.size() > MAX_URI_LEN) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Request-URI Too Long: " + pathOnly); @@ -339,11 +565,11 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC return false; } - // ── Initialize request ── + // 3) Initialize request object req = HttpRequest(); req.setMethod(method); - // Split on ‘/’ and reject only true “..” segments: + // Reject traversal attempts via ".." segments { std::istringstream segstream(pathOnly); std::string seg; @@ -357,7 +583,7 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC } } - // std::string norm = pathOnly; + // Normalize path and set fields std::string norm = normalizePath(pathOnly); if (norm.empty()) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Path escapes root: " + pathOnly); @@ -368,13 +594,14 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC req.setQuery(query); req.setVersion(version); + // Only support HTTP/1.0 and HTTP/1.1 if (version != "HTTP/1.0" && version != "HTTP/1.1") { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Invalid HTTP version: " + version); errorCode = 505; return false; } - // — Parse headers + // 4) Parse headers line by line while (std::getline(stream, line)) { if (!line.empty() && line.back() == '\r') line.pop_back(); @@ -394,15 +621,17 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC errorCode = 400; return false; } + std::string value = line.substr(colon + 1); value.erase(0, value.find_first_not_of(" \t\r\n")); value.erase(value.find_last_not_of(" \t\r\n") + 1); + // Delegate validation/insertion to helper if (!insertValidatedHeader(req, key, value, errorCode)) return false; } - // — Build URL + // 5) Build full URL using Host header + server config std::string hostHeader = req.getHeader("HOST"); try { Url url; @@ -412,6 +641,7 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC std::string urlStr = "http://" + (hostHeader.empty() ? foundServer.getHost() : hostHeader) + req.getPath(); + // Use stricter parsing for HTTP/1.1 if (req.getVersion() == "HTTP/1.1") { url = parseUrlHttpVersion1_1(req, urlStr); } else { @@ -420,11 +650,12 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC req.setUrl(url); - // ── Set host explicitly for easier access downstream + // Normalize and set host explicitly std::string& headerAfterParseUrl = url.host; std::transform(headerAfterParseUrl.begin(), headerAfterParseUrl.end(), - headerAfterParseUrl.begin(), ::tolower); // normalize + headerAfterParseUrl.begin(), ::tolower); req.setHost(headerAfterParseUrl); + Logger::logFrom(LogLevel::INFO, "HttpRequestParser", "Parsed URL: " + urlStr + " with host: " + headerAfterParseUrl); } catch (const std::exception& e) { @@ -436,17 +667,55 @@ bool parseReqHeader(HttpRequest& req, const std::string& headerPart, int& errorC return true; } +/** + * @brief Checks if a chunked HTTP request body is complete. + * + * @details According to RFC 7230 §4.1, a chunked transfer ends when: + * - A zero-length chunk (`0\r\n`) is received. + * - Followed by an optional trailer section. + * - And terminated by a blank line (`\r\n\r\n`). + * + * This function inspects the body buffer to determine if the final + * zero-size chunk and terminating CRLF sequence are present. + * + * @param bodyPart The raw body data received so far. + * @return `true` if the chunked body is complete, + * `false` if more data is required. + * + * @ingroup http + */ bool isChunkedBodyComplete(const std::string& bodyPart) { - // 1. Locate start of trailer section: must contain 0\r\n + // 1. Look for the "0\r\n" marker (start of the last chunk). std::size_t zeroPos = bodyPart.find("0\r\n"); if (zeroPos == std::string::npos) - return false; + return false; // no terminator chunk yet - // 2. Look for the CRLF that ends the trailer section + // 2. Ensure the trailers (if any) are properly terminated with CRLFCRLF. std::size_t trailerEnd = bodyPart.find("\r\n\r\n", zeroPos); return trailerEnd != std::string::npos; } +/** + * @brief Parses a chunked HTTP request body and assembles it into the request. + * + * @details Implements RFC 7230 §4.1 "Chunked Transfer Coding". + * - Reads each chunk size (hexadecimal) from the body stream. + * - Copies the corresponding number of bytes into a contiguous body buffer. + * - Stops when a zero-size chunk is encountered. + * - Rejects trailers (unsupported in this implementation). + * - Enforces the per-client maximum body size limit. + * + * @param req Reference to the @ref HttpRequest being built. + * @param bodyPart The raw body data received so far (may include CRLFs, chunks, trailers). + * @param clientMaxBodySize Maximum allowed body size for this client (bytes). + * @param errorCode Output parameter: set to an HTTP error status (e.g. 400, 413) on failure. + * @param consumedBytes Output parameter: number of bytes successfully consumed from the buffer. + * + * @note If trailers are present after the final 0–chunk, the request is rejected with `400`. + * @note If the accumulated body exceeds @p clientMaxBodySize, the request is rejected with `413`. + * + * @ingroup http + */ void chunkReqHandler(HttpRequest& req, const std::string& bodyPart, std::size_t clientMaxBodySize, int& errorCode, std::size_t& consumedBytes) { std::istringstream stream(bodyPart); @@ -454,53 +723,56 @@ void chunkReqHandler(HttpRequest& req, const std::string& bodyPart, std::size_t std::string fullBody; std::size_t total = 0, local = 0; + // Process each line (either chunk size, CRLF, or trailers) while (std::getline(stream, line)) { - local += line.size() + 1; + local += line.size() + 1; // count CRLF too if (!line.empty() && line.back() == '\r') line.pop_back(); + // 1) Parse chunk size (hexadecimal) std::size_t chunkSize; try { chunkSize = std::stoul(line, nullptr, 16); } catch (...) { - errorCode = 400; + errorCode = 400; // Bad Request: malformed chunk size consumedBytes += local; Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Invalid chunk size format"); return; } + // 2) Final 0–chunk indicates end of body if (chunkSize == 0) { - // consume the final CRLF after the 0–chunk + // Consume the CRLF after the last chunk std::getline(stream, line); local += line.size() + 1; - // reject ANY trailers (we don’t support them) ─── + // Reject any trailers (not supported) std::string trailer; while (std::getline(stream, trailer)) { local += trailer.size() + 1; - // an empty line ends the trailer section if (trailer.empty()) - break; - // any non-empty trailer header is unsupported - errorCode = 400; + break; // empty line terminates trailers + errorCode = 400; // reject any non-empty trailer consumedBytes += local; Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Unsupported trailer header: " + trailer); return; } - break; } + + // 3) Enforce max body size limit Logger::logFrom(LogLevel::WARN, "PARSER", " Totalsize: " + std::to_string(total + chunkSize)); if (total + chunkSize > clientMaxBodySize) { + errorCode = 413; // Payload Too Large + consumedBytes += local + chunkSize; Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Exceeded max body size in chunked transfer"); - errorCode = 413; - consumedBytes += local + chunkSize; return; } + // 4) Read the chunk data std::string data(chunkSize, '\0'); stream.read(&data[0], chunkSize); std::streamsize got = stream.gcount(); @@ -508,106 +780,179 @@ void chunkReqHandler(HttpRequest& req, const std::string& bodyPart, std::size_t fullBody.append(data, 0, got); total += got; + // Consume CRLF after the chunk data std::getline(stream, line); local += line.size() + 1; } + // 5) Finalize request req.setBody(fullBody); req.setContentLength(fullBody.size()); errorCode = 0; consumedBytes += local; } -// Helper: does this substring begin with a valid HTTP method + space? +/** + * @brief Heuristically checks if a string looks like the start line of an HTTP request. + * + * @details This helper is used to detect pipelined requests: + * - It tests whether the string begins with one of the known HTTP + * method tokens followed by a space (e.g., "GET ", "POST "). + * - Only a small subset of standard methods is recognized + * (GET, POST, DELETE, PUT, HEAD, OPTIONS, PATCH). + * + * This is not a full request-line parser, but a lightweight + * heuristic to distinguish request boundaries in a raw buffer. + * + * @param s The string segment to test (usually leftover buffer data). + * @return true if @p s looks like a valid request start line, false otherwise. + * + * @ingroup http + */ static bool isLikelyStartLine(const std::string& s) { + // List of method tokens followed by a space static const std::array methods = {"GET ", "POST ", "DELETE ", "PUT ", "HEAD ", "OPTIONS ", "PATCH "}; + + // Compare the beginning of s with each known method token for (auto m : methods) { if (s.size() >= m.size() && std::string_view(s).substr(0, m.size()) == m) { - return true; + return true; // Found a match → likely start line } } - return false; + + return false; // No match → probably not a start line } +/** + * @brief Parses and validates the HTTP request body according to Content-Length or + * Transfer-Encoding. + * + * @details This function handles two body transfer mechanisms: + * - **Chunked transfer encoding**: waits until the terminating `0\r\n\r\n` is received, + * then delegates parsing to @ref chunkReqHandler. Rejects incomplete or malformed + * chunked bodies. + * - **Fixed-length bodies**: uses the `Content-Length` header to determine how many bytes + * to consume, rejecting requests that exceed @p clientMaxBodySize or contain mismatched + * lengths. + * + * It also supports HTTP/1.1 pipelining: + * - If extra bytes follow a complete body, the function checks whether they look like the + * start of another request (via @ref isLikelyStartLine). + * - If so, only the declared body is consumed and parsing can continue with the next + * request. + * + * @param req The @ref HttpRequest object being filled. + * @param bodyPart Raw body data extracted from the client buffer. + * @param clientMaxBodySize Maximum allowed body size (from server configuration). + * @param errorCode Reference set to HTTP error code on failure (0 on success). + * @param consumedBytes Reference updated with the number of bytes consumed from @p bodyPart. + * + * @return true if the body is complete and valid, false if more data is needed or if an error + * occurred. + * + * @ingroup http + */ bool parseReqBody(HttpRequest& req, const std::string& bodyPart, std::size_t clientMaxBodySize, int& errorCode, std::size_t& consumedBytes) { const std::string& te = req.getHeader("TRANSFER-ENCODING"); + // 1) Handle chunked transfer encoding if (te == "chunked") { if (!isChunkedBodyComplete(bodyPart)) { + // Body not yet complete → wait for more data Logger::logFrom(LogLevel::INFO, "HttpRequestParser", "Incomplete chunked body, waiting for more"); errorCode = 0; return false; } + // Fully received → delegate to chunked handler chunkReqHandler(req, bodyPart, clientMaxBodySize, errorCode, consumedBytes); return (errorCode == 0); } + // 2) Handle fixed-length bodies std::size_t len = req.getContentLength(); - // 1) If declared length itself is too large, reject immediately + // Reject if declared length exceeds configured max if (len > 0 && len >= clientMaxBodySize) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Exceeded max body size in non-chunked transfer"); - errorCode = 413; + errorCode = 413; // Payload Too Large consumedBytes += bodyPart.size(); return false; } - // 2) If we haven’t received *at least* len bytes yet, wait for more + // Not enough data yet → wait for more if (bodyPart.size() < len) { - // Logger::logFrom(LogLevel::INFO, "HttpRequestParser", "Incomplete body, waiting for - // more"); errorCode = 0; return false; } - // 3) If there are more than len bytes in bodyPart, check for pipelining + // 3) Detect pipelined requests if (bodyPart.size() > len) { // Look at the bytes immediately after the declared body std::string_view remainder(bodyPart.c_str() + len, bodyPart.size() - len); - // Use the helper instead of a lambda if (isLikelyStartLine(std::string(remainder))) { - // Treat extra bytes as the next request → consume exactly len bytes + // Treat as pipelined request → consume only declared body req.setBody(bodyPart.substr(0, len)); consumedBytes += len; errorCode = 0; return true; } - // Otherwise, it’s a real mismatch + // Extra data is invalid → reject Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Content-Length mismatch: body too long"); - errorCode = 400; + errorCode = 400; // Bad Request consumedBytes += bodyPart.size(); return false; } - // 4) Exactly len bytes → accept + // 4) Exact match → accept body req.setBody(bodyPart.substr(0, len)); consumedBytes += len; errorCode = 0; return true; } +/** + * @brief Performs final validation of a parsed HTTP request. + * + * @details This function applies protocol-level validation rules after the request line + * and headers have been parsed. Specifically: + * - Ensures the HTTP method is supported (GET, POST, DELETE). + * - Validates the normalized request path (must not escape root). + * - Checks the `Connection` header (only "keep-alive" or "close" are accepted). + * - For POST requests: + * * Requires either a `Content-Length` or `Transfer-Encoding` header. + * * Requires a valid `Content-Type` header, chosen from a limited set + * (supports parameters like `multipart/form-data; boundary=...`). + * + * @param req The @ref HttpRequest object to validate. + * @param errorCode Reference set to HTTP error code if validation fails. + * + * @return true if the request is valid, false otherwise (with @p errorCode set). + * + * @ingroup http + */ bool validateReq(HttpRequest& req, int& errorCode) { + // Allowed methods static const std::set methods = {"GET", "POST", "DELETE"}; if (!methods.count(req.getMethod())) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Method Not Allowed: " + req.getMethod()); - errorCode = 405; // // Method Not Allowed !!!!!!!!!!!!!!!!! It has to be 501, I changed only - // for passing tests + errorCode = 405; // NOTE: semantically could be 501, but 405 is used for tests return false; } + // Path validation if (!isValidPath(req.getPath())) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Invalid request path: " + req.getPath()); - errorCode = 403; + errorCode = 403; // Forbidden return false; } @@ -616,31 +961,34 @@ bool validateReq(HttpRequest& req, int& errorCode) { if (!conn.empty()) { std::string lower = toLower(conn); if (lower != "keep-alive" && lower != "close") { + // Normalize invalid values to "close" but still reject req.setHeader("CONNECTION", "close"); Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Invalid Connection header value: " + conn); - errorCode = 400; + errorCode = 400; // Bad Request return false; } } - // POST → must have supported Content-Type + // POST-specific checks if (req.getMethod() == "POST") { - + // Must declare either Content-Length or Transfer-Encoding if (req.getContentLength() == 0 && req.getHeader("TRANSFER-ENCODING").empty()) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "POST without Content-Length or Transfer-Encoding"); - errorCode = 411; + errorCode = 411; // Length Required return false; } + // Require Content-Type std::string ct = req.getHeader("CONTENT-TYPE"); if (ct.empty()) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Missing Content-Type for POST"); - errorCode = 415; + errorCode = 415; // Unsupported Media Type return false; } + // Allowed Content-Types (prefix matching for params like boundary=...) static const std::set validTypes = {"application/x-www-form-urlencoded", "multipart/form-data", "text/plain", @@ -648,7 +996,6 @@ bool validateReq(HttpRequest& req, int& errorCode) { "application/octet-stream", "test/file"}; - // Accept type with optional parameters (e.g. multipart/form-data; boundary=...) std::string lowerCt = toLower(ct); bool valid = false; for (const std::string& type : validTypes) { @@ -661,7 +1008,7 @@ bool validateReq(HttpRequest& req, int& errorCode) { if (!valid) { Logger::logFrom(LogLevel::ERROR, "HttpRequestParser", "Unsupported Content-Type for POST: " + ct); - errorCode = 415; + errorCode = 415; // Unsupported Media Type return false; } } @@ -669,99 +1016,99 @@ bool validateReq(HttpRequest& req, int& errorCode) { return true; } +/** + * @brief Parses a raw HTTP request buffer into an @ref HttpRequest object. + * + * @details This is the main entry point for HTTP request parsing. It coordinates + * header parsing, body parsing, and final validation. The workflow is: + * + * 1. Locate the end of the header section (`\r\n\r\n`). + * 2. Split the buffer into a header block and a body block. + * 3. Parse the request line and headers via @ref parseReqHeader. + * 4. Validate the parsed request with @ref validateReq. + * 5. If the method does not expect a body (e.g. GET/DELETE), exit early. + * - Detect and support HTTP pipelining (multiple requests in one buffer). + * 6. Otherwise, parse the body with @ref parseReqBody. + * 7. On success, return a fully populated @ref HttpRequest. + * + * Error handling: + * - If parsing fails, the function sets @p errorCode to the appropriate + * HTTP status code (400, 405, 411, 413, 414, 415, 505, etc.). + * - @p consumedBytes is always updated to reflect how much of the buffer + * was processed (even if parsing fails). + * + * @param req The @ref HttpRequest to populate. + * @param buffer Raw input buffer containing HTTP request(s). + * @param serversOnPort List of available @ref Server objects for virtual-host matching. + * @param errorCode Output parameter set to HTTP error code on failure. + * @param consumedBytes Output parameter set to the number of bytes successfully consumed. + * + * @return true if the request was successfully parsed, false otherwise. + * + * @ingroup http + */ bool HttpRequestParser::parse(HttpRequest& req, const std::string& buffer, std::vector serversOnPort, int& errorCode, std::size_t& consumedBytes) { - - // 1) Find end of header block: "\r\n\r\n" + // 1) Find end of header block std::size_t headerEndPos = buffer.find("\r\n\r\n"); if (headerEndPos == std::string::npos) { Logger::logFrom(LogLevel::INFO, "HttpRequestParser", "Incomplete header, waiting for more"); - errorCode = 0; + errorCode = 0; // Not an error, just incomplete return false; } std::size_t headerLen = headerEndPos + 4; - // 2) Split into headerPart and remainder + // 2) Split buffer into header and body std::string headerPart = buffer.substr(0, headerLen); std::string bodyPart = buffer.substr(headerLen); consumedBytes = headerLen; - // 3) Parse request‐line + headers + // 3) Parse request line + headers std::size_t clientMaxBodySize; if (!parseReqHeader(req, headerPart, errorCode, serversOnPort, clientMaxBodySize)) { - // parseReqHeader sets errorCode (e.g. 400, 414, 505) - return false; + return false; // parseReqHeader sets errorCode } - // 4) Validate method, path, and mandatory headers + // 4) Validate request semantics if (!validateReq(req, errorCode)) { - // validateReq sets errorCode (e.g. 405, 403, 411, 415) - return false; + return false; // validateReq sets errorCode } - //-------------------------------------------------------- - // Kha: Temporary commented out the following code. The same logic can be found in - // parseReqHeader and - //-------------------------------------------------------- - // std::string hostHeader = req.getHeader("Host"); - // int bestMatch = 0; - - // for (size_t i = 0; i < serversOnPort.size(); ++i) { - // const Server& srv = serversOnPort[i]; - // const std::vector& names = srv.getServerNames(); - // if (std::find(names.begin(), names.end(), hostHeader) != names.end()) { - // bestMatch = i; - // break; - // } - // } - - // req.setMatchedServerIndex(bestMatch); - // clientMaxBodySize = - // serversOnPort[bestMatch].getClientMaxBodySize(); // use the matched server's body limit - - // 5) Early exit for methods that do not expect a body (GET, DELETE) + // 5) Early exit for bodyless methods (GET, DELETE) std::string method = req.getMethod(); if (method == "GET") { - // If there is no extra data at all, this is a clean GET/DELETE: if (buffer.size() == headerLen) { + // Clean GET: no extra data consumedBytes = headerLen; errorCode = 0; return true; } - // Otherwise, there *are* extra bytes. Check if those extra bytes look like - // the start of a new request‐line (i.e. pipelined). A quick heuristic is: - // ‣ first byte must be 'A'–'Z' (valid HTTP method token) - // ‣ then we expect a space somewhere after it. - // Here we'll just check that the very next character is an uppercase letter, - // which is enough to catch "GET /something HTTP/1.1…" or "POST ..." in practice. + // Check if extra data looks like a new request (pipelining) char next = buffer[headerLen]; if (next >= 'A' && next <= 'Z') { - // This almost certainly is the start of another request‐line, - // so we treat it as a pipelined request, not as a body. + // Treat as pipelined request consumedBytes = headerLen; errorCode = 0; return true; } - // FALLBACK: ignore any “body” on GET and treat as a clean GET - consumedBytes = buffer.size(); // consume headers + body, but ignore the body + // Otherwise, ignore any stray "body" + consumedBytes = buffer.size(); errorCode = 0; return true; } - // 6) For POST (and other body‐bearing methods), delegate to parseReqBody + // 6) Parse body for methods that require it (e.g. POST) std::size_t bodyConsumed = 0; bool bodyOk = parseReqBody(req, bodyPart, clientMaxBodySize, errorCode, bodyConsumed); - if (!bodyOk) { - // parseReqBody sets errorCode (400, 413, etc.) and bodyConsumed consumedBytes = headerLen + bodyConsumed; - return false; + return false; // parseReqBody sets errorCode } - // 7) Body successfully parsed + // 7) Success consumedBytes = headerLen + bodyConsumed; errorCode = 0; return true; diff --git a/src/http/HttpResponse.cpp b/src/http/HttpResponse.cpp index 5c014db6..eb3b071f 100644 --- a/src/http/HttpResponse.cpp +++ b/src/http/HttpResponse.cpp @@ -3,13 +3,33 @@ /* ::: :::::::: */ /* HttpResponse.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 10:56:54 by irychkov #+# #+# */ -/* Updated: 2025/08/18 12:05:43 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:17:55 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file HttpResponse.cpp + * @brief Implements the HttpResponse class. + * + * @details Defines all methods of @ref HttpResponse, including setters for + * status, headers, body, file path, and CGI metadata; getters and + * queries for response properties; and utilities such as + * `toHttpString()` for serializing the response into HTTP wire format. + * + * The implementation also includes connection management logic + * (`isConnectionClose`) that follows HTTP/1.0 and HTTP/1.1 semantics, + * as well as RFC-friendly handling for certain error status codes. + * + * Instances of this class are typically produced by + * @ref ResponseBuilder and HTTP method handlers, then written back + * to the client by the networking layer. + * + * @ingroup http + */ + #include "http/HttpResponse.hpp" #include // for transform #include // for tolower @@ -17,39 +37,71 @@ #include // for basic_ostream, operator<<, basic_stringstream #include // for pair -HttpResponse ::HttpResponse(void) { +//=== Construction & Special Members ===================================== + +/** + * @brief Constructs a default HttpResponse with status `200 OK`. + * + * @details Initializes response with status code 200, message `"OK"`, + * and a CGI body offset of `0`. Other fields are left empty + * until explicitly set. + * + * @ingroup http + */ +HttpResponse::HttpResponse(void) { _status_code = 200; _status_message = "OK"; _cgiBodyOffset = 0; } -HttpResponse ::~HttpResponse(void) { +/** + * @brief Destroys the HttpResponse. + * + * @details Provided for completeness; does not manage external resources. + * + * @ingroup http + */ +HttpResponse::~HttpResponse(void) { } -void HttpResponse ::setStatus(int code, const std::string& message) { +//=== Mutators (Setters) ================================================== + +/** + * @brief Sets the status code and reason phrase of the response. + * @param code Numeric status code (e.g., 404). + * @param message Reason phrase (e.g., "Not Found"). + * @ingroup http + */ +void HttpResponse::setStatus(int code, const std::string& message) { _status_code = code; _status_message = message; } -void HttpResponse ::setHeader(const std::string& key, const std::string& value) { +/** + * @brief Adds or replaces a header in the response. + * @param key Header name. + * @param value Header value. + * @ingroup http + */ +void HttpResponse::setHeader(const std::string& key, const std::string& value) { _headers[key] = value; } -void HttpResponse ::setBody(const std::string& body) { +/** + * @brief Sets the body of the response. + * @param body Response payload as a string. + * @ingroup http + */ +void HttpResponse::setBody(const std::string& body) { _body = body; } /** - * @brief Stores HTTP version and Connection header from the request. - * - * @details This metadata is required to determine whether the connection should be - * kept alive or closed after the response is sent. The HTTP version and the client's - * `Connection` header together define the default persistence behavior according to - * RFC 7230 §6.3. This method must be called before sending the response to ensure - * proper behavior in `isConnectionClose()`. + * @brief Stores HTTP version and connection header metadata. * - * @param version The HTTP version from the client's request (e.g., "HTTP/1.1"). - * @param conn The value of the Connection header from the client's request. + * @param version HTTP version string (e.g., "HTTP/1.1"). + * @param conn Raw value of the "Connection" header from the request. + * @ingroup http */ void HttpResponse::setRequestMeta(const std::string& version, const std::string& conn) { _http_version = version; @@ -57,53 +109,44 @@ void HttpResponse::setRequestMeta(const std::string& version, const std::string& } /** - * @brief Determines whether the connection should be closed after the response. - * - * @details This logic properly accounts for the HTTP version and the client's - * Connection header to implement persistent connections correctly. - * - * Unlike the previous version, which only checked if `Connection: close` was present, - * this method enforces the default semantics of each protocol version: - * - HTTP/1.1 assumes keep-alive unless explicitly closed. - * - HTTP/1.0 assumes close unless explicitly kept alive. - * - * This behavior is compliant with RFC 7230 §6.3 and avoids premature connection - * termination when clients do not send a `Connection` header. - * - * @return `true` if the connection should be closed, `false` to keep it alive. + * @brief Sets a file path for file-backed responses. + * @param path Filesystem path to serve. + * @ingroup http */ -bool HttpResponse::isConnectionClose() const { - // 1) If the response explicitly sets Connection, honor it. - std::map::const_iterator hit = _headers.find("Connection"); - if (hit != _headers.end()) { - std::string v = hit->second; - std::transform(v.begin(), v.end(), v.begin(), ::tolower); - return v == "close"; - } - - // 2) Certain status codes must close (RFC-friendly behavior). - static const std::set force_close_codes = {400, 408, 413, 500}; - if (force_close_codes.count(_status_code)) - return true; - - // 3) Fall back to protocol semantics using the request metadata. - std::string conn = _connection_header; - std::transform(conn.begin(), conn.end(), conn.begin(), ::tolower); +void HttpResponse::setFilePath(const std::string& path) { + _file_path = path; +} - if (_http_version == "HTTP/1.1") { - // Keep-alive by default unless client asked to close - return conn == "close"; - } - if (_http_version == "HTTP/1.0") { - // Close by default unless client asked to keep-alive - return conn != "keep-alive"; - } +/** + * @brief Sets the byte offset where CGI body begins in its temp file. + * @param offset Offset in bytes. + * @ingroup http + */ +void HttpResponse::setCgiBodyOffset(std::streamsize offset) { + _cgiBodyOffset = offset; +} - // Unknown version: safest is to close. - return true; +/** + * @brief Sets the temporary file path produced by CGI. + * @param temp_file Path to the CGI temp file. + * @ingroup http + */ +void HttpResponse::setCgiTempFile(const std::string& temp_file) { + _cgi_temp_file = temp_file; } -std::string HttpResponse ::toHttpString(void) const { +//=== Queries (Getters) =================================================== + +/** + * @brief Serializes the response into HTTP wire format. + * + * @details Produces a string containing the status line, headers, + * `Content-Length`, and body, ready for transmission. + * + * @return HTTP response string. + * @ingroup http + */ +std::string HttpResponse::toHttpString(void) const { std::stringstream ss; ss << "HTTP/1.1 " << _status_code << " " << _status_message << "\r\n"; @@ -119,45 +162,127 @@ std::string HttpResponse ::toHttpString(void) const { return ss.str(); } -void HttpResponse::setFilePath(const std::string& path) { - _file_path = path; -} - +/** + * @brief Returns the file path of the response, if any. + * @ingroup http + */ const std::string& HttpResponse::getFilePath() const { return _file_path; } -bool HttpResponse::isFileResponse() const { - return !_file_path.empty(); -} - +/** + * @brief Returns the numeric status code. + * @ingroup http + */ int HttpResponse::getStatusCode(void) const { return _status_code; } +/** + * @brief Returns the reason phrase of the response. + * @ingroup http + */ const std::string& HttpResponse::getStatusMessage(void) const { return _status_message; } +/** + * @brief Returns the headers of the response. + * @ingroup http + */ const std::map& HttpResponse::getHeaders(void) const { return _headers; } -void HttpResponse::setCgiBodyOffset(std::streamsize offset) { - _cgiBodyOffset = offset; -} - +/** + * @brief Returns the CGI body offset. + * @ingroup http + */ std::streamsize HttpResponse::getCgiBodyOffset() const { return _cgiBodyOffset; } -void HttpResponse::setCgiTempFile(const std::string& temp_file) { - _cgi_temp_file = temp_file; -} +/** + * @brief Returns the CGI temporary file path. + * @ingroup http + */ const std::string& HttpResponse::getCgiTempFile() const { return _cgi_temp_file; } +//=== Predicates & Utilities ============================================== + +/** + * @brief Determines whether the server should close the connection + * after sending this response. + * + * @details The decision follows HTTP semantics and some defensive RFC-inspired + * rules: + * 1. **Explicit Connection header in the response**: + * If present, its value takes precedence (case-insensitive). + * - `"Connection: close"` → close the connection. + * - Any other value → keep open, unless other rules apply. + * + * 2. **Force-close status codes**: + * Certain error codes (400, 408, 413, 500) always require closing + * the connection to remain protocol-compliant and prevent reuse + * of a potentially invalid connection. + * + * 3. **Fallback to request metadata**: + * - For HTTP/1.1: keep-alive by default, unless request said `"close"`. + * - For HTTP/1.0: close by default, unless request said `"keep-alive"`. + * + * 4. **Unknown HTTP version**: + * Defaults to closing the connection for safety. + * + * @return `true` if the server must close the TCP connection, + * `false` if it can be kept alive. + * + * @ingroup http + */ +bool HttpResponse::isConnectionClose() const { + // 1) If the response explicitly sets Connection, honor it. + std::map::const_iterator hit = _headers.find("Connection"); + if (hit != _headers.end()) { + std::string v = hit->second; + std::transform(v.begin(), v.end(), v.begin(), ::tolower); + return v == "close"; + } + + // 2) Certain status codes must close (RFC-friendly behavior). + static const std::set force_close_codes = {400, 408, 413, 500}; + if (force_close_codes.count(_status_code)) + return true; + + // 3) Fall back to protocol semantics using the request metadata. + std::string conn = _connection_header; + std::transform(conn.begin(), conn.end(), conn.begin(), ::tolower); + + if (_http_version == "HTTP/1.1") { + // Keep-alive by default unless client asked to close + return conn == "close"; + } + if (_http_version == "HTTP/1.0") { + // Close by default unless client asked to keep-alive + return conn != "keep-alive"; + } + + // Unknown version: safest is to close. + return true; +} + +/** + * @brief Returns true if the response serves a file from disk. + * @ingroup http + */ +bool HttpResponse::isFileResponse() const { + return !_file_path.empty(); +} + +/** + * @brief Returns true if a CGI temp file is set. + * @ingroup http + */ bool HttpResponse::isCgiTempFile() const { return !_cgi_temp_file.empty(); } diff --git a/src/http/responseBuilder.cpp b/src/http/responseBuilder.cpp index 9dfa7bbb..d9ea896b 100644 --- a/src/http/responseBuilder.cpp +++ b/src/http/responseBuilder.cpp @@ -3,13 +3,40 @@ /* ::: :::::::: */ /* responseBuilder.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/11 12:14:23 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:18:46 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 09:31:28 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file responseBuilder.cpp + * @brief Implements the ResponseBuilder utilities for constructing HTTP responses. + * + * @details This file defines helper functions to generate HTTP responses of different + * types (success, file-based success, error, redirect). It encapsulates the + * boilerplate logic of setting status codes, headers, body content, and + * connection persistence rules based on HTTP version and `Connection` headers. + * + * Key features: + * - Provides default status messages for standard HTTP codes + * via @ref MessageHandler::getDefaultMessage. + * - Centralizes connection management (keep-alive vs. close) in + * `initializeResponse`, consistent with HTTP/1.0 and HTTP/1.1 semantics. + * - Supports custom error pages configured in @ref Server instances, + * falling back to generated HTML when missing. + * - Generates file-backed responses (for static files or CGI output) + * with proper `Content-Length` and offset handling. + * - Generates redirect responses with appropriate `Location` headers. + * + * These utilities are typically invoked by request handlers (GET, POST, + * DELETE, CGI, etc.) to produce fully-formed @ref HttpResponse objects + * that can be sent back to clients. + * + * @ingroup http + */ + #include "http/responseBuilder.hpp" #include "core/Location.hpp" // for Location #include "core/Server.hpp" // for Server @@ -25,8 +52,40 @@ #include // for pair #include // for vector +/** + * @namespace MessageHandler + * @brief Utility namespace for HTTP status code reason phrases. + * + * @details Provides helpers to map numeric HTTP status codes + * (e.g., 200, 404, 500) to their standard reason phrases + * (e.g., "OK", "Not Found", "Internal Server Error"). + * + * Currently, this namespace exposes a single function: + * - @ref getDefaultMessage : Returns the standard reason phrase + * for a given status code, or `"Unknown Error"` if not found. + * + * Used throughout response builders and handlers to ensure + * consistent status messages across all HTTP responses. + * + * @ingroup http + */ namespace MessageHandler { - +/** + * @brief Returns the default reason phrase for a given HTTP status code. + * + * @details Uses a static lookup table of standard HTTP status codes + * (e.g., 200 → "OK", 404 → "Not Found"). The table is initialized + * once on first call and reused for all subsequent calls. + * + * If the provided status code is not in the table, the function + * returns the fallback string `"Unknown Error"`. + * + * @param status_code The numeric HTTP status code (e.g., 200, 404, 500). + * @return The corresponding reason phrase as a string, or `"Unknown Error"` + * if the code is not recognized. + * + * @ingroup http + */ std::string getDefaultMessage(int status_code) { // Define a local alias for better readability and maintainability using StatusMessageMap = std::map; @@ -64,7 +123,24 @@ std::string getDefaultMessage(int status_code) { namespace { -// Determines keep-alive status from HTTP version and connection header. +/** + * @brief Determines whether the TCP connection should be kept alive + * based on HTTP version and the `Connection` header. + * + * @details Implements protocol-specific keep-alive rules: + * - **HTTP/1.1**: persistent connections are the default. + * Only `"Connection: close"` disables keep-alive. + * - **HTTP/1.0**: connections close by default. + * Only `"Connection: keep-alive"` enables persistence. + * - **Other/unknown versions**: defaults to closing + * the connection for safety. + * + * @param version The HTTP version string from the request (e.g. `"HTTP/1.1"`). + * @param conn The value of the `Connection` header, lowercase-insensitive. + * @return `true` if the connection may be kept alive, `false` if it should close. + * + * @ingroup http + */ bool shouldKeepAlive(const std::string& version, const std::string& conn) { // HTTP/1.1 defaults to keep-alive unless explicitly closed if (version == "HTTP/1.1") @@ -76,10 +152,33 @@ bool shouldKeepAlive(const std::string& version, const std::string& conn) { return false; } -// Common initialization for any response. +/** + * @brief Initializes a new HTTP response with status and connection metadata. + * + * @details This helper performs common setup for all responses: + * - Sets the numeric status code and reason phrase (e.g., `200 OK`). + * - Records the request's HTTP version and `Connection` header + * for later connection management. + * - Applies protocol-specific keep-alive rules via + * @ref shouldKeepAlive and closes the connection for certain + * error codes (400, 408, 413, 500). + * - Inserts the appropriate `Connection` header (`keep-alive` or `close`) + * into the response. + * + * It centralizes connection-handling logic so that all + * response builders (`success`, `error`, `redirect`, etc.) + * stay consistent with HTTP/1.0 and HTTP/1.1 semantics. + * + * @param response The @ref HttpResponse object being initialized. + * @param status_code The numeric HTTP status code (e.g., 200, 404, 500). + * @param message The reason phrase associated with the status code. + * @param request The original @ref HttpRequest, used to extract + * version and `Connection` header metadata. + * + * @ingroup http + */ void initializeResponse(HttpResponse& response, int status_code, const std::string& message, const HttpRequest& request) { - // Set HTTP status code and message (e.g., 200 OK) response.setStatus(status_code, message); // Store request version and connection header to guide connection persistence response.setRequestMeta(request.getVersion(), request.getHeader("Connection")); @@ -100,55 +199,162 @@ void initializeResponse(HttpResponse& response, int status_code, const std::stri } // anonymous namespace +/** + * @namespace ResponseBuilder + * @brief Factory helpers for constructing HTTP responses. + * + * @details The ResponseBuilder namespace groups together utility + * functions that generate fully-formed @ref HttpResponse + * objects for different scenarios: + * + * - @ref generateSuccess : Build a success response with an inline body. + * - @ref generateSuccessFile : Build a response backed by a file (static file or CGI temp + * file). + * - @ref generateError : Build an error response, using custom pages if available or a + * default fallback. + * - @ref generateRedirect : Build a redirect response with a `Location` header. + * + * All builders rely on @ref initializeResponse to apply + * consistent status metadata, HTTP version, and connection + * persistence rules. This ensures uniform behavior across + * all response types. + * + * These functions are typically used by request handlers, + * CGI logic, and the router when constructing responses + * to be sent back to clients. + * + * @ingroup http + */ namespace ResponseBuilder { - +/** + * @brief Builds a successful HTTP response with an inline body. + * + * @details This helper constructs a standard success response: + * - Looks up the default reason phrase for the given status code + * (e.g., `200 → "OK"`). + * - Initializes the response with status line, HTTP version, and + * connection persistence rules via @ref initializeResponse. + * - Sets the `Content-Type` header to the provided MIME type. + * - Attaches the supplied response body directly in memory. + * + * Typical use cases include sending JSON, HTML, or plain-text + * responses for successful GET/POST requests. + * + * @param status_code The HTTP status code to return (e.g., 200, 201). + * @param body The response body payload as a string. + * @param content_type The MIME type for the body (e.g., `"text/html"`, + * `"application/json"`). + * @param request The original @ref HttpRequest, used to propagate + * version and connection metadata. + * + * @return A fully constructed @ref HttpResponse object ready to send. + * + * @ingroup http + */ HttpResponse generateSuccess(int status_code, const std::string& body, const std::string& content_type, const HttpRequest& request) { HttpResponse response; - // Get the default reason phrase for the given status code (e.g., "OK") - std::string message = MessageHandler::getDefaultMessage(status_code); + std::string message = MessageHandler::getDefaultMessage(status_code); // Set status, connection headers, and keep-alive logic initializeResponse(response, status_code, message, request); - // Set the content type (e.g., "text/html", "application/json") response.setHeader("Content-Type", content_type); - // Attach the response body response.setBody(body); return response; } +/** + * @brief Builds a successful HTTP response backed by a file on disk. + * + * @details This helper constructs a response intended for file or CGI output: + * - Looks up the default reason phrase for the given status code + * (e.g., `200 → "OK"`). + * - Initializes the response with status line, HTTP version, and + * connection persistence rules via @ref initializeResponse. + * - Sets `Content-Type` to the provided MIME type. + * - Sets `Content-Length` to the provided file size. + * - Records the CGI body offset (if applicable) to indicate where + * dynamic output begins inside a temporary file. + * - Stores the file path, allowing the networking layer to stream + * the file contents directly instead of keeping them in memory. + * + * Typical use cases include serving static files (HTML, CSS, JS, images) + * or returning CGI-generated content that lives in a temporary file. + * + * @param status_code The HTTP status code to return (e.g., 200, 206). + * @param file_path Filesystem path to the file that should be streamed. + * @param content_type The MIME type of the file (e.g., `"text/html"`, + * `"application/octet-stream"`). + * @param request The original @ref HttpRequest, used to propagate + * version and connection metadata. + * @param content_length Size of the file (or portion) to be served, in bytes. + * @param cgiBodyOffset Byte offset where the CGI response body begins, + * if serving a CGI-generated temp file; usually 0 otherwise. + * + * @return A fully constructed @ref HttpResponse object configured for file streaming. + * + * @ingroup http + */ HttpResponse generateSuccessFile(int status_code, const std::string& file_path, const std::string& content_type, const HttpRequest& request, std::streamsize content_length, std::streamsize cgiBodyOffset) { HttpResponse response; - // Get the default reason phrase for the given status code (e.g., "OK") - std::string message = MessageHandler::getDefaultMessage(status_code); + std::string message = MessageHandler::getDefaultMessage(status_code); // Set status, connection headers, and keep-alive logic initializeResponse(response, status_code, message, request); - // Set the content type (e.g., "text/html", "application/octet-stream") response.setHeader("Content-Type", content_type); response.setHeader("Content-Length", std::to_string(content_length)); response.setCgiBodyOffset(cgiBodyOffset); - // Set the file path for serving response.setFilePath(file_path); return response; } +/** + * @brief Builds an HTTP error response, optionally using a custom error page. + * + * @details This helper constructs error responses in two stages: + * 1. It checks the server configuration for a custom error page + * matching the given status code (via @ref Server::getErrorPages). + * If found, it attempts to read the file and use its contents + * as the response body. + * 2. If no custom page is configured or the file cannot be read, + * it falls back to generating a minimal default HTML page with + * the status code and reason phrase (e.g., `"404 Not Found"`). + * + * In both cases, the response is initialized with correct status line, + * HTTP version, and connection persistence rules using + * @ref initializeResponse, and the `Content-Type` is set to `"text/html"`. + * + * This ensures clients always receive a valid, informative error page, + * even if the custom configuration fails. + * + * @param status_code The HTTP error code to return (e.g., 404, 500). + * @param server Reference to the @ref Server configuration, + * used to look up custom error pages. + * @param request The original @ref HttpRequest, used for version and + * connection metadata propagation. + * + * @return A fully constructed @ref HttpResponse object representing the error. + * + * @ingroup http + */ HttpResponse generateError(int status_code, const Server& server, const HttpRequest& request) { HttpResponse response; std::string message = MessageHandler::getDefaultMessage(status_code); std::string body; - // Try custom error page + // 1) Look for a custom error page in the server config const auto& error_pages = server.getErrorPages(); auto it = error_pages.find(status_code); if (it != error_pages.end()) { std::string uri = it->second; - // Strip leading slash + + // Strip leading slash to avoid double slashes in joinPath if (!uri.empty() && uri.front() == '/') uri.erase(0, 1); - // Determine default root (location "/") + + // Find the root path from the default "/" location std::string default_root; for (const auto& loc : server.getLocations()) { if (loc.getPath() == "/") { @@ -156,13 +362,16 @@ HttpResponse generateError(int status_code, const Server& server, const HttpRequ break; } } + + // Build full filesystem path and try to load the file std::string full_path = joinPath(default_root, uri); std::ifstream file(full_path.c_str()); - if (file) + if (file) { body.assign((std::istreambuf_iterator(file)), std::istreambuf_iterator()); + } } - // Fallback default page + // 2) Fallback: generate a minimal default HTML error page if (body.empty()) { std::ostringstream ss; ss << "

" << status_code << " " << htmlEscape(message) @@ -170,29 +379,47 @@ HttpResponse generateError(int status_code, const Server& server, const HttpRequ body = ss.str(); } + // 3) Initialize the response metadata (status, headers, connection) initializeResponse(response, status_code, message, request); response.setHeader("Content-Type", "text/html"); - // Attach the generated or loaded error page body + + // 4) Attach the error body (custom or generated) --- if (!body.empty()) { response.setBody(body); } - /* response.setBody(body); */ return response; } +/** + * @brief Generates an HTTP redirect response. + * + * @details Creates a response with the given redirect status code + * (e.g., 301 Moved Permanently, 302 Found, 307 Temporary Redirect), + * sets the appropriate `Location` header pointing to the target URL, + * and applies connection-handling logic via @ref initializeResponse. + * + * By default, no response body is included for redirects, and most + * clients rely solely on the `Location` header. A `Content-Length: 0` + * header may be explicitly added if desired, but is optional. + * + * @param status_code Redirect status code (must be in the 3xx range). + * @param location Target URL or path for the redirect. + * @param request The original HTTP request (provides metadata for connection handling). + * + * @return An initialized @ref HttpResponse representing the redirect. + * + * @ingroup http + */ HttpResponse generateRedirect(int status_code, const std::string& location, const HttpRequest& request) { HttpResponse response; - // Get default reason phrase for the redirect status code (e.g., "Moved Permanently") - std::string message = MessageHandler::getDefaultMessage(status_code); + std::string message = MessageHandler::getDefaultMessage(status_code); // Set status line, connection handling, and keep-alive logic initializeResponse(response, status_code, message, request); // Set the Location header to indicate the redirect target response.setHeader("Location", location); - // No body is sent in most redirects → explicitly set Content-Length to 0 - // response.setHeader("Content-Length", "0"); return response; } From 9007e708f1fc485e3df0daa71d22189bc7b5d61d Mon Sep 17 00:00:00 2001 From: To0nsa Date: Tue, 19 Aug 2025 10:16:14 +0300 Subject: [PATCH 24/32] Enhances HTTP request routing with detailed documentation Adds comprehensive inline documentation to clarify the HTTP request routing process, including step-by-step workflows, helper function roles, and error handling. Improves maintainability and developer understanding of routing logic, covering key operations like directory slash redirects, location resolution, method validation, and handler dispatch. --- include/http/requestRouter.hpp | 23 ++++- src/http/requestRouter.cpp | 176 +++++++++++++++++++++++++++++++-- 2 files changed, 188 insertions(+), 11 deletions(-) diff --git a/include/http/requestRouter.hpp b/include/http/requestRouter.hpp index 2834df64..5a4f2ae8 100644 --- a/include/http/requestRouter.hpp +++ b/include/http/requestRouter.hpp @@ -1,19 +1,32 @@ /* ************************************************************************** */ /* */ /* ::: :::::::: */ -/* HttpRequestHandler.hpp :+: :+: :+: */ +/* requestRouter.hpp :+: :+: :+: */ /* +:+ +:+ +:+ */ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/12 23:11:50 by nlouis #+# #+# */ -/* Updated: 2025/06/05 10:58:12 by nlouis ### ########.fr */ +/* Updated: 2025/08/19 10:07:35 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ #pragma once -class Server; -class HttpRequest; -class HttpResponse; +/** + * @file HttpRequestHandler.hpp + * @brief Declares the main entry point for handling HTTP requests. + * + * @details This header provides the function interface for processing + * a parsed HTTP request against a given server configuration. + * The function applies routing logic, executes the appropriate + * HTTP method handler (GET, POST, DELETE, etc.), and generates + * the corresponding HttpResponse. + * + * @ingroup request_handler + */ + +class Server; ///< Forward declaration of the Server class. +class HttpRequest; ///< Forward declaration of the HttpRequest class. +class HttpResponse; ///< Forward declaration of the HttpResponse class. HttpResponse handleRequest(const HttpRequest& request, const Server& server); diff --git a/src/http/requestRouter.cpp b/src/http/requestRouter.cpp index fed75062..1033ddeb 100644 --- a/src/http/requestRouter.cpp +++ b/src/http/requestRouter.cpp @@ -3,13 +3,40 @@ /* ::: :::::::: */ /* requestRouter.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/12 23:13:23 by nlouis #+# #+# */ -/* Updated: 2025/08/17 12:17:33 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 10:15:21 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file requestRouter.cpp + * @brief Routes parsed HTTP requests to the appropriate handler. + * + * @details Implements the routing pipeline that maps an incoming + * @ref HttpRequest to a @ref HttpResponse using the active + * @ref Server configuration and its @ref Location blocks. + * + * Workflow: + * 1) **Directory slash redirect:** If a request targets a physical + * directory without a trailing slash, issue a 301 to the + * slash-terminated URI (see @ref redirectOnDirectorySlash). + * 2) **Location resolution:** Find the best matching @ref Location by + * exact match or longest-prefix (see @ref findLocation). + * 3) **Configured redirects:** Apply `return`-based redirects if present + * in the matched location (see @ref redirectOnConfigured). + * 4) **Method validation:** Check supported/allowed methods, returning + * 501 or 405 as appropriate (see @ref validateRequestMethod). + * 5) **Dispatch:** Call the concrete method handler + * (GET/POST/DELETE) (see @ref dispatchByMethod). + * + * Errors are rendered via @ref ResponseBuilder helpers. Utilities from + * `filesystemUtils` are used to normalize URIs and resolve physical paths. + * + * @ingroup request_handler + */ + #include "core/Location.hpp" // for Location #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest @@ -26,6 +53,31 @@ #include // for vector namespace { +/** + * @brief Redirects to a slash-terminated URI when the target is a real directory. + * + * @details This helper enforces nginx-like behavior for directory URIs: + * if the request path maps to a directory but lacks a trailing slash, + * issue a 301 redirect to the slash-appended path. + * + * Rules: + * - Only considers @ref Location blocks whose configured path ends with '/'. + * - Matches the request URI by location prefix (sans trailing slash). + * - Resolves a physical filesystem path with `resolvePhysicalPath` and + * checks it with `stat`; symlink-to-dir is treated as a directory. + * - Applies to any HTTP method (GET/POST/DELETE), not only GET. + * + * @param req Parsed HTTP request (used for path, method, headers). + * @param server Active server context (provides location list). + * @param uri Normalized request URI (usually `normalizePath(req.getPath())`). + * + * @return std::nullopt if no redirect is needed; otherwise a ready 301 response + * with `Location: /`. + * + * @note If `req.getPath()` already ends with '/', no redirect occurs. + * @warning Only locations defined with a trailing slash are inspected; locations + * configured without a trailing '/' are intentionally ignored here. + */ std::optional redirectOnDirectorySlash(const HttpRequest& req, const Server& server, const std::string& uri) { // If there's already a trailing slash in the request path, nothing to do. @@ -62,7 +114,28 @@ std::optional redirectOnDirectorySlash(const HttpRequest& req, con return std::nullopt; } -// 2) Find exact‐match or longest‐prefix location +/** + * @brief Selects the best-matching Location for a normalized request URI. + * + * @details Resolution is two-phase: + * 1) **Exact match**: return the first location whose normalized path + * equals `uri`. + * 2) **Longest-prefix match**: otherwise, return the location with the + * longest normalized path that is a prefix of `uri`. + * + * Each candidate path is normalized with `normalizePath`. Empty + * location paths are ignored for prefix matching. + * + * @param uri Normalized request URI (e.g., from `normalizePath(req.getPath())`). + * @param server Active server whose location set is consulted. + * + * @return Pointer to the selected @ref Location, or `nullptr` if none match. + * + * @note Matching is purely lexical (prefix test), not filesystem-based. + * @warning If multiple locations normalize to the same path, the first one + * encountered wins (implementation-defined by iteration order). + * @complexity O(N · C) where N = number of locations and C = normalization + prefix check cost. + */ const Location* findLocation(const std::string& uri, const Server& server) { // Exact match for (const Location& loc : server.getLocations()) { @@ -87,7 +160,28 @@ const Location* findLocation(const std::string& uri, const Server& server) { return best; } -// 3) Handle configured “return” redirects on GET +/** + * @brief Apply a location-scoped redirect, if configured. + * + * @details If the matched @ref Location declares a redirect (i.e. a `return` rule), + * this function builds and returns a redirect response using + * @ref ResponseBuilder::generateRedirect. The redirect is applied + * unconditionally for any HTTP method and happens before method + * validation/dispatch in the routing pipeline. + * + * @param request Incoming HTTP request (used to propagate version/connection metadata). + * @param loc The matched location to inspect for a configured redirect. + * + * @return A populated @ref HttpResponse when a redirect rule is present; + * `std::nullopt` otherwise. + * + * @note The target URL is taken verbatim from configuration; no normalization is + * performed here. Ensure the configured target is either an absolute URI or a + * valid absolute-path per your configuration semantics. + * + * @warning Choose an appropriate status code (e.g., 301/302/307/308) in the + * configuration to control method preservation and cacheability semantics. + */ std::optional redirectOnConfigured(const HttpRequest& request, const Location& loc) { if (loc.hasRedirect()) { Logger::logFrom(LogLevel::INFO, "Router", @@ -98,7 +192,29 @@ std::optional redirectOnConfigured(const HttpRequest& request, con return std::nullopt; } -// 4) Check for un-implemented / not-allowed methods +/** + * @brief Validates the HTTP method against server support and location policy. + * + * @details Two checks in order: + * 1) **Server support**: rejects methods not implemented by this server + * (currently {"GET","POST","DELETE"}) with **501 Not Implemented**. + * 2) **Location policy**: rejects methods not allowed by the matched + * @ref Location with **405 Method Not Allowed**. + * + * Both cases are logged and rendered via @ref ResponseBuilder::generateError. + * + * @param request Incoming HTTP request (used for method/path and response metadata). + * @param loc Matched configuration block to query method allowance. + * @param server Active server (used by error rendering). + * + * @return `std::nullopt` when the method is valid for this location; otherwise a + * ready error @ref HttpResponse (501 or 405). + * + * @note Per RFC 9110, servers **SHOULD** include an `Allow` header on 405 + * listing permitted methods for the target resource. + * @warning Method matching here is case-sensitive; normalize upstream if you + * want to accept non‑uppercase tokens. + */ std::optional validateRequestMethod(const HttpRequest& request, const Location& loc, const Server& server) { static const std::set supported = {"GET", "POST", "DELETE"}; @@ -121,7 +237,25 @@ std::optional validateRequestMethod(const HttpRequest& request, co return std::nullopt; } -// 5) Dispatch to the specific handler +/** + * @brief Dispatches a validated request to the concrete method handler. + * + * @details Assumes @ref validateRequestMethod has already run, so only + * implemented-and-allowed methods reach this point. Dispatch is a + * simple chain on the request method: + * - "GET" → @ref handleGet + * - "POST" → @ref handlePost + * - "DELETE" → @ref handleDelete + * + * @param request Parsed HTTP request. + * @param server Active server context. + * @param loc Matched @ref Location. + * + * @return The handler-produced @ref HttpResponse. + * + * @note If additional methods are added in the future, extend this dispatcher + * (or switch to a table-driven approach). + */ HttpResponse dispatchByMethod(const HttpRequest& request, const Server& server, const Location& loc) { const std::string& method = request.getMethod(); @@ -136,14 +270,41 @@ HttpResponse dispatchByMethod(const HttpRequest& request, const Server& server, } } // anonymous namespace +/** + * @brief Top-level router entrypoint for processing an HTTP request. + * + * @details This function implements the main request handling pipeline: + * 1. **Normalize URI** — sanitize the request path. + * 2. **Directory-slash redirect** — if a directory is requested without a trailing + * slash, generate a `301` redirect (see @ref redirectOnDirectorySlash). + * 3. **Location resolution** — find the best-matching @ref Location in the + * server configuration (exact or longest-prefix match). + * 4. **Configured redirect** — apply `return` directive–based redirects if present + * in the matched location (see @ref redirectOnConfigured). + * 5. **Method validation** — reject unsupported (501) or disallowed (405) methods + * (see @ref validateRequestMethod). + * 6. **Dispatch** — call the concrete handler for GET/POST/DELETE + * (see @ref dispatchByMethod). + * + * Errors at any stage are logged and turned into responses via @ref ResponseBuilder. + * + * @param request Parsed @ref HttpRequest. + * @param server Active @ref Server context. + * @return Fully constructed @ref HttpResponse to send back to the client. + * + * @ingroup request_handler + */ HttpResponse handleRequest(const HttpRequest& request, const Server& server) { + // 1) Normalize the request path std::string uri = normalizePath(request.getPath()); + // 2) Check if this is a directory missing a trailing slash → 301 redirect std::optional redirectResponse = redirectOnDirectorySlash(request, server, uri); if (redirectResponse.has_value()) { return redirectResponse.value(); } + // 3) Find best matching Location block const Location* loc = findLocation(uri, server); if (loc == nullptr) { Logger::logFrom(LogLevel::WARN, "Router", @@ -151,15 +312,18 @@ HttpResponse handleRequest(const HttpRequest& request, const Server& server) { return ResponseBuilder::generateError(404, server, request); } + // 4) Check for a configured `return` redirect on this location std::optional configuredRedirect = redirectOnConfigured(request, *loc); if (configuredRedirect.has_value()) { return configuredRedirect.value(); } + // 5) Validate that the method is implemented and allowed in this location std::optional validationError = validateRequestMethod(request, *loc, server); if (validationError.has_value()) { return validationError.value(); } + // 6) Finally, dispatch to the actual handler (GET/POST/DELETE) return dispatchByMethod(request, server, *loc); } From 1c9ef04fad40a9fe652bd562a6340c5d47e98ef8 Mon Sep 17 00:00:00 2001 From: To0nsa Date: Tue, 19 Aug 2025 11:34:52 +0300 Subject: [PATCH 25/32] Enhances HTTP documentation Introduces detailed comments and documentation for HTTP GET, POST, and DELETE handlers, improving clarity and maintainability. Includes comprehensive inline documentation to aid future development and troubleshooting. --- docs/doxygen.warnings | 233 +++++++++++--- include/http/methodsHandler.hpp | 26 +- src/http/handleCgi.cpp | 304 +++++++++++++++++- src/http/methodsHandler/handleDelete.cpp | 113 ++++++- .../handleGet/generateAutoindex.cpp | 227 +++++++++++-- .../methodsHandler/handleGet/handleGet.cpp | 156 ++++++++- .../handlePost/handleMultipartForm.cpp | 148 +++++++-- .../methodsHandler/handlePost/handlePost.cpp | 239 +++++++++++++- 8 files changed, 1342 insertions(+), 104 deletions(-) diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index bc2af0fc..4d70ee79 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -1,45 +1,202 @@ -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:262: warning: no matching file member found for +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:16: warning: the name 'HttpRequestHandler.hpp' supplied as the argument in the \file statement is not an input file +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:263: warning: no matching file member found for const std::unordered_map< std::string, ServerHandler > & directive::serverHandlers Possible candidates: - 'const std::unordered_map< std::string, ServerHandler > & serverHandlers()' at line 262 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:361: warning: no matching file member found for + 'const std::unordered_map< std::string, ServerHandler > & serverHandlers()' at line 263 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:362: warning: no matching file member found for const std::unordered_map< std::string, LocationHandler > & directive::locationHandlers Possible candidates: - 'const std::unordered_map< std::string, LocationHandler > & locationHandlers()' at line 361 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command + 'const std::unordered_map< std::string, LocationHandler > & locationHandlers()' at line 362 of file /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:247: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:249: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:253: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:255: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:341: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:349: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:353: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:23: warning: unable to resolve reference to 'hasHeader' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:24: warning: unable to resolve reference to 'getHeader' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:25: warning: unable to resolve reference to '_contentLength' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:27: warning: unable to resolve reference to '_parseError' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:28: warning: unable to resolve reference to '_matchedServerIndex' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:34: warning: unable to resolve reference to 'setMatchedServerIndex' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:41: warning: unable to resolve reference to 'getParseErrorCode' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:43: warning: unable to resolve reference to 'getContentLength' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:45: warning: unable to resolve reference to 'getBody' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:46: warning: unable to resolve reference to 'getQuery' for \ref command +/home/toonsa/myProjects/webserv/include/http/HttpRequest.hpp:47: warning: unable to resolve reference to 'getPath' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:31: warning: unable to resolve reference to 'generateAutoindex:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:36: warning: unable to resolve reference to 'handleGet:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:38: warning: unable to resolve reference to 'handlePost:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:40: warning: unable to resolve reference to 'handleDelete:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:44: warning: unable to resolve reference to 'generateAutoindex:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:46: warning: unable to resolve reference to 'handleMultipartForm:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:35: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:43: warning: unable to resolve reference to 'dispatchByMethod' for \ref command +/home/toonsa/myProjects/webserv/include/http/responseBuilder.hpp:20: warning: unable to resolve reference to 'generateSuccess' for \ref command +/home/toonsa/myProjects/webserv/include/http/responseBuilder.hpp:21: warning: unable to resolve reference to 'generateSuccessFile' for \ref command +/home/toonsa/myProjects/webserv/include/http/responseBuilder.hpp:22: warning: unable to resolve reference to 'generateError' for \ref command +/home/toonsa/myProjects/webserv/include/http/responseBuilder.hpp:24: warning: unable to resolve reference to 'generateRedirect' for \ref command +/home/toonsa/myProjects/webserv/include/http/responseBuilder.hpp:27: warning: unable to resolve reference to 'getDefaultMessage' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:236: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command -/home/toonsa/myProjects/webserv/src/config/validateConfig.cpp:351: warning: unable to resolve reference to 'validateUniquePorts' for \ref command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:247: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:249: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:253: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:255: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:341: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:349: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:353: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/validateConfig.cpp:355: warning: unable to resolve reference to 'validateUniquePorts' for \ref command +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'path' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'errorCode' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: The following parameters of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) are not documented: + parameter 'cgi' + parameter 'req' + parameter 'bodyFd' + parameter 'outputFd' +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:22: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:25: warning: unable to resolve reference to 'escapeUriComponent' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:28: warning: unable to resolve reference to 'formatMTimeUTC' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:29: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:31: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:31: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:103: warning: unable to resolve reference to 'DirEntry' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:119: warning: unable to resolve reference to 'DirEntry' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:283: warning: unable to resolve reference to 'escapeUriComponent' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:346: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:346: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:19: warning: unable to resolve reference to 'detectMimeType' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:21: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:23: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:112: warning: unable to resolve reference to 'detectMimeType' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:409: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:410: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:24: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:26: warning: unable to resolve reference to 'findLocation' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:28: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:30: warning: unable to resolve reference to 'validateRequestMethod' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:32: warning: unable to resolve reference to 'dispatchByMethod' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:137: warning: Found unknown command '@complexity' +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:243: warning: unable to resolve reference to 'validateRequestMethod' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:35: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:43: warning: unable to resolve reference to 'dispatchByMethod' for \ref command +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__request__handler_ga43204497282094b6dac91b946d5648ba_ga43204497282094b6dac91b946d5648ba_cgraph.dot ! +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:163: warning: unable to resolve reference to 'shouldKeepAlive' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:236: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:20: warning: unable to resolve reference to 'receiveFromClient' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:22: warning: unable to resolve reference to 'parseAndQueueRequests' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:24: warning: unable to resolve reference to 'processPendingRequests' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:24: warning: unable to resolve reference to 'handleCgiRequest' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:26: warning: unable to resolve reference to 'handleCgiPollEvents' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerRequest.cpp:28: warning: unable to resolve reference to 'handleRequestErrorIfAny' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerResponse.cpp:20: warning: unable to resolve reference to 'logResponseStatus' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerResponse.cpp:21: warning: unable to resolve reference to 'sendFileResponse' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerResponse.cpp:22: warning: unable to resolve reference to 'sendRawResponse' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerResponse.cpp:23: warning: unable to resolve reference to 'sendResponse' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:20: warning: unable to resolve reference to 'isHeaderTimeout:' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:21: warning: unable to resolve reference to 'isBodyTimeout:' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:22: warning: unable to resolve reference to 'isSendTimeout:' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:23: warning: unable to resolve reference to 'isIdleTimeout:' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:24: warning: unable to resolve reference to 'checkClientTimeouts:' for \ref command +/home/toonsa/myProjects/webserv/src/network/SocketManagerTimeouts.cpp:26: warning: unable to resolve reference to 'checkRequestLimits:' for \ref command error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:236: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:409: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:410: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:35: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command +/home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:43: warning: unable to resolve reference to 'dispatchByMethod' for \ref command +error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__request__handler_ga43204497282094b6dac91b946d5648ba_ga43204497282094b6dac91b946d5648ba_cgraph.dot ! error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:246: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:248: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:252: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:254: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:340: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:348: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:352: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: Unsupported xml/html tag found -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:354: warning: found tag while expecting -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command -/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:358: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'path' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'errorCode' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) +/home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: The following parameters of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) are not documented: + parameter 'cgi' + parameter 'req' + parameter 'bodyFd' + parameter 'outputFd' +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:137: warning: Found unknown command '@complexity' +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:247: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:249: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:253: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:255: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:341: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:349: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:353: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: Unsupported xml/html tag found +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:355: warning: found tag while expecting +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:359: warning: end of comment block while expecting command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:217: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:236: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command diff --git a/include/http/methodsHandler.hpp b/include/http/methodsHandler.hpp index a0e58260..87e3aa79 100644 --- a/include/http/methodsHandler.hpp +++ b/include/http/methodsHandler.hpp @@ -6,10 +6,34 @@ /* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/05 10:46:53 by nlouis #+# #+# */ -/* Updated: 2025/06/06 22:31:14 by nlouis ### ########.fr */ +/* Updated: 2025/08/19 10:23:35 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file methodsHandler.hpp + * @brief Declares HTTP method handlers for GET, POST, and DELETE. + * + * @details This header defines the public entrypoints for handling + * HTTP request methods within Webserv. Each function maps + * an incoming @ref HttpRequest to an @ref HttpResponse, + * using the active @ref Server context and the matched + * @ref Location configuration. + * + * Supported methods: + * - **GET**: Serves static files or generates autoindex + * listings (see @ref generateAutoindex). + * - **POST**: Handles uploads (raw body, URL-encoded, + * multipart forms). + * - **DELETE**: Removes existing files if permitted. + * + * Helper: + * - @ref generateAutoindex: builds a directory listing + * response in HTML. + * + * @ingroup request_handler + */ + #pragma once #include diff --git a/src/http/handleCgi.cpp b/src/http/handleCgi.cpp index bde0e761..c82ca3c5 100644 --- a/src/http/handleCgi.cpp +++ b/src/http/handleCgi.cpp @@ -3,13 +3,46 @@ /* ::: :::::::: */ /* handleCgi.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/24 12:23:37 by nlouis #+# #+# */ -/* Updated: 2025/08/17 12:19:24 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 11:32:52 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file handleCgi.cpp + * @brief Implements CGI (Common Gateway Interface) request handling. + * + * @details + * This module provides full lifecycle management of CGI processes: + * - Build CGI environment variables from the incoming `HttpRequest`. + * - Create temporary input/output files to stream request/response bodies. + * - Fork and exec the target CGI script with the correct interpreter. + * - Collect, parse, and forward CGI output as an `HttpResponse`. + * - Handle errors, timeouts, and cleanup of child processes. + * + * ### Workflow + * 1. **initCgiProcess**: validate script, create temp files, fork child, exec interpreter. + * 2. **Child process**: stdin → request body, stdout → temp output, `execve()` CGI script. + * 3. **Parent process**: monitor PID, wait for completion, enforce timeouts. + * 4. **finalizeCgi**: parse CGI headers (Status, Content-Type), build `HttpResponse`. + * 5. **cleanup**: delete temp files and reset `CgiProcess` state. + * + * ### Key Functions + * - `prepareEnv` → Build CGI environment (SCRIPT_NAME, PATH_INFO, QUERY_STRING, etc.). + * - `prepareCgiTempFiles` → Write request body to temp file, create output file. + * - `setupAndRunCgiChild` → Fork/exec child process safely. + * - `finalizeCgi` → Read output, parse headers, and generate HTTP response. + * - `errorOnCgi`, `cleanupCgi`, `tryTerminateCgi` → Robust process management. + * + * ### Limitations + * - Blocking I/O (temp files, fork/exec). + * - Only basic CGI spec (no FastCGI, no async streaming). + * + * @ingroup request_handler + */ + #include "http/handleCgi.hpp" #include "core/Location.hpp" // for Location #include "core/Server.hpp" // for Server @@ -38,21 +71,47 @@ namespace { +/** + * @brief Build the environment variables required for a CGI script execution. + * + * @details + * Constructs a `std::vector` of `KEY=VALUE` pairs following the CGI/1.1 + * specification. The environment includes: + * - Standard CGI variables (`SCRIPT_NAME`, `PATH_INFO`, `REQUEST_METHOD`, etc.). + * - Server metadata (`SERVER_NAME`, `SERVER_PORT`, `SERVER_PROTOCOL`, etc.). + * - Request headers, prefixed with `HTTP_` and normalized (uppercased, `-` → `_`). + * - Content metadata (`CONTENT_LENGTH`, `CONTENT_TYPE`). + * + * Special handling: + * - Computes `PATH_INFO` if the request URI extends beyond the script’s URI. + * - Sets `REDIRECT_STATUS=200` for PHP compatibility. + * + * @param req Incoming HTTP request. + * @param server Server configuration (name, port, defaults). + * @param loc Matched location block (provides root, CGI settings). + * @param scriptPath Absolute filesystem path of the CGI script to run. + * + * @return A vector of strings representing the CGI environment, suitable for `execve()`. + * + * @ingroup http_component + */ std::vector prepareEnv(const HttpRequest& req, const Server& server, const Location& loc, const std::string& scriptPath) { std::vector env; auto set = [&](const std::string& k, const std::string& v) { env.push_back(k + "=" + v); }; + // Normalize paths and extract script filename std::string requestPath = normalizePath(req.getPath()); std::string locationPath = normalizePath(loc.getPath()); std::string scriptName = std::filesystem::path(scriptPath).filename().string(); - // SCRIPT_NAME = URL path to the script (/directory/youpi.bla) + // Build SCRIPT_NAME and PATH_INFO std::string scriptUri = locationPath; if (!scriptUri.empty() && scriptUri.back() != '/') scriptUri += "/"; scriptUri += scriptName; + // Extract extra path after the script (PATH_INFO) std::string pathInfo; if (requestPath.size() > scriptUri.size() && requestPath.compare(0, scriptUri.size(), scriptUri) == 0) { @@ -60,30 +119,38 @@ std::vector prepareEnv(const HttpRequest& req, const Server& server if (!pathInfo.empty() && pathInfo[0] != '/') pathInfo = "/" + pathInfo; } + + // Core CGI variables (script + path info) set("SCRIPT_NAME", req.getPath()); if (pathInfo.empty()) { set("PATH_INFO", req.getPath()); } else { set("PATH_INFO", pathInfo); } + + // Request metadata set("REQUEST_METHOD", req.getMethod()); set("QUERY_STRING", req.getQuery()); set("CONTENT_LENGTH", std::to_string(req.getContentLength())); if (!req.getHeader("Content-Type").empty()) set("CONTENT_TYPE", req.getHeader("Content-Type")); + // Server and protocol metadata set("SERVER_PROTOCOL", "HTTP/1.1"); set("GATEWAY_INTERFACE", "CGI/1.1"); set("SERVER_SOFTWARE", "webserv/1.0"); set("DOCUMENT_ROOT", loc.getRoot()); set("SERVER_NAME", server.getDefaultServerName()); set("SERVER_PORT", std::to_string(server.getPort())); + + // Script execution context set("PATH_TRANSLATED", scriptPath); - set("REMOTE_ADDR", "127.0.0.1"); + set("REMOTE_ADDR", "127.0.0.1"); // currently hardcoded set("REQUEST_URI", req.getPath()); set("SCRIPT_FILENAME", scriptPath); - set("REDIRECT_STATUS", "200"); + set("REDIRECT_STATUS", "200"); // required by some CGI implementations (e.g. PHP) + // Forward all HTTP headers as CGI variables (HTTP_FOO_BAR format) for (const auto& [key, value] : req.getHeaders()) { std::string envKey = "HTTP_" + toUpper(key); std::replace(envKey.begin(), envKey.end(), '-', '_'); @@ -93,7 +160,21 @@ std::vector prepareEnv(const HttpRequest& req, const Server& server return env; } -// Helper to convert vector → vector +/** + * @brief Convert a vector of strings into a null-terminated `char*` array. + * + * @details + * Prepares data for system calls like `execve()`, which require `char* argv[]` + * or `char* envp[]` format. Each string’s `c_str()` is cast away from `const` + * (safe here because the lifetime is managed by the original `std::string` + * vector). The returned array is null-terminated as required by POSIX. + * + * @param vs Input vector of strings. + * @return A `std::vector` containing pointers to each string, ending with `nullptr`. + * + * @note The caller must ensure that the original `std::vector` outlives + * this array, since the pointers are non-owning. + */ std::vector toCharPtrArray(const std::vector& vs) { std::vector out; out.reserve(vs.size() + 1); @@ -103,6 +184,26 @@ std::vector toCharPtrArray(const std::vector& vs) { return out; } +/** + * @brief Read up to a maximum number of bytes from a file stream. + * + * @details + * Allocates a temporary buffer of size `maxBytes`, attempts to read from the + * given input file stream, and returns both: + * - The bytes read, wrapped into a `std::string`. + * - The actual count of bytes read (`gcount()`). + * + * This is typically used to capture the initial chunk of a CGI output file + * in order to parse headers before streaming the rest of the response. + * + * @param file Open input file stream (must be in binary mode). + * @param maxBytes Maximum number of bytes to read from the stream. + * + * @return A pair: `(stringData, bytesRead)`. + * + * @note If fewer than `maxBytes` are available, only the available bytes are read. + * @warning The file’s read position will advance by the number of bytes consumed. + */ std::pair readInitialOutput(std::ifstream& file, size_t maxBytes) { std::vector buffer(maxBytes); file.read(buffer.data(), maxBytes); @@ -110,6 +211,22 @@ std::pair readInitialOutput(std::ifstream& file, s return {std::string(buffer.data(), bytesRead), bytesRead}; } +/** + * @brief Locate the end of the HTTP header section in a CGI response. + * + * @details + * Searches for the standard header delimiter: + * - First tries `"\r\n\r\n"` (CRLF-terminated headers). + * - Falls back to `"\n\n"` if CRLF is not found. + * + * Updates `delimiterLength` with the length of the matched delimiter + * (either 4 or 2). + * + * @param data Input string containing CGI response data. + * @param delimiterLength Output; set to the delimiter size (4 or 2) if found. + * + * @return The position of the delimiter if found, otherwise `std::nullopt`. + */ std::optional findHeaderDelimiter(const std::string& data, size_t& delimiterLength) { size_t pos = data.find("\r\n\r\n"); delimiterLength = 4; @@ -123,6 +240,23 @@ std::optional findHeaderDelimiter(const std::string& data, size_t& delim return std::nullopt; } +/** + * @brief Parse CGI response headers for status code and content type. + * + * @details + * Iterates over each line in the provided header block: + * - Extracts the value of `Content-Type:` if present. + * - Extracts the numeric value of `Status:` if present (default = 200). + * - If parsing fails, logs a warning and falls back to `500`. + * + * Other headers are ignored by this function. + * + * @param header Raw header section of the CGI output (up to the header delimiter). + * + * @return A pair `(statusCode, contentType)`: + * - `statusCode` → HTTP status (int). + * - `contentType` → MIME type string (empty if not provided). + */ std::pair parseHeaders(const std::string& header) { std::istringstream stream(header); std::string line, contentType = ""; @@ -157,6 +291,23 @@ bool validateCgiScript(const std::filesystem::path& path, int& errorCode) { return true; } +/** + * @brief Validate that a CGI script exists and is executable. + * + * @details + * Performs two checks on the given filesystem path: + * 1. Verifies the path refers to a regular file (`isFile`). + * - If not, sets `errorCode = 404` (Not Found). + * 2. Verifies the file has execute permissions (`access(..., X_OK)`). + * - If not, sets `errorCode = 403` (Forbidden). + * + * Logs errors if validation fails. + * + * @param path Filesystem path to the CGI script. + * @param errorCode Output; set to `404` or `403` on failure. + * + * @return true if the script exists and is executable, false otherwise. + */ bool prepareCgiTempFiles(CgiProcess& cgi, const HttpRequest& req, int& bodyFd, int& outputFd) { static unsigned counter = 0; cgi.input_path = make_temp_name("webserv_in", counter); @@ -186,6 +337,33 @@ bool prepareCgiTempFiles(CgiProcess& cgi, const HttpRequest& req, int& bodyFd, i return true; } +/** + * @brief Setup file descriptors, environment, and execute a CGI script in the child process. + * + * @details + * This function is called after `fork()` in the CGI child. It: + * 1. Redirects `stdin` from the temporary body file (`dup2(body_fd, STDIN_FILENO)`). + * 2. Redirects `stdout` to the temporary output file (`dup2(output_fd, STDOUT_FILENO)`). + * 3. Closes all unrelated file descriptors to avoid leaks. + * 4. Resolves the CGI interpreter from the script extension (if configured). + * 5. Builds the `argv` array for `execve()` (`[interpreter?, scriptPath, NULL]`). + * 6. Prepares the CGI environment variables with `prepareEnv()`. + * 7. Changes working directory to the script’s directory for relative file access. + * 8. Calls `execve()` to replace the process image with the CGI script. + * + * If any step fails, logs the error and terminates the child with `exit(1)`. + * + * @param cgi CGI process metadata (script path, temp files). + * @param body_fd File descriptor for the request body (temp input file). + * @param output_fd File descriptor for CGI output (temp output file). + * @param req Incoming HTTP request (provides headers, body, method). + * @param server Server configuration (name, port, etc.). + * @param loc Location block configuration (provides CGI interpreter). + * @param poll_fds Active poll descriptors (closed to avoid leakage in the child). + * + * @warning Must only be called in the child process after `fork()`. + * @note On success, this function never returns (replaced by `execve()`). + */ void setupAndRunCgiChild(const CgiProcess& cgi, int body_fd, int output_fd, const HttpRequest& req, const Server& server, const Location& loc, const std::vector& poll_fds) { @@ -248,6 +426,20 @@ void setupAndRunCgiChild(const CgiProcess& cgi, int body_fd, int output_fd, cons namespace CGI { +/** + * @brief Remove a temporary CGI file and log an error if the deletion fails. + * + * @details + * Attempts to delete the given file path using `std::filesystem::remove`. + * - If `path` is empty, nothing is done. + * - If removal fails, logs an error including the context label and + * the system-provided error message. + * + * Commonly used for cleaning up CGI input/output temporary files. + * + * @param path Filesystem path of the file to remove. + * @param context Description of the file’s role (e.g. "input temp file"). + */ void unlinkWithErrorLog(const std::string& path, const std::string& context) { if (!path.empty()) { std::error_code ec; @@ -259,6 +451,33 @@ void unlinkWithErrorLog(const std::string& path, const std::string& context) { } } +/** + * @brief Initialize and start a CGI process for a given request. + * + * @details + * - Resolves the script path from the request URI and validates it with `validateCgiScript`. + * - Prepares temporary input/output files (`prepareCgiTempFiles`): + * - Writes the request body to a temp input file. + * - Opens a temp output file for capturing CGI output. + * - Forks the process: + * - In the child → calls `setupAndRunCgiChild()` (redirects FDs, sets env, execve). + * - In the parent → closes temp file descriptors and stores process metadata. + * - On error (invalid script, temp file failure, fork failure), sets `errorCode` + * appropriately and returns `false`. + * + * @param cgi Reference to a `CgiProcess` struct (will be populated with PID, paths, + * timestamps). + * @param req Incoming HTTP request (provides path, headers, body). + * @param server Server configuration (name, port, etc.). + * @param loc Location block configuration (provides root, CGI settings). + * @param poll_fds Current poll file descriptors (passed to child for cleanup). + * @param errorCode Output parameter; set to HTTP-like error codes (`404`, `403`, `500`) on failure. + * + * @return true if the CGI process was successfully started, false otherwise. + * + * @note On success, the child process never returns (replaced by `execve()`). + * @warning Must be followed by later cleanup (`finalizeCgi`, `errorOnCgi`, or `cleanupCgi`). + */ bool initCgiProcess(CgiProcess& cgi, const HttpRequest& req, const Server& server, const Location& loc, const std::vector& poll_fds, int& errorCode) { cgi.last_activity = getCurrentTime(); @@ -294,6 +513,32 @@ bool initCgiProcess(CgiProcess& cgi, const HttpRequest& req, const Server& serve return true; } +/** + * @brief Finalize a CGI process by parsing its output and building an HTTP response. + * + * @details + * - Opens the CGI output file produced by the child process. + * - Scans up to the first 9KB (`MAX_HEADER_SCAN`) to locate the HTTP header delimiter. + * - If not found, logs an error and returns a `500` response. + * - Extracts and parses CGI headers: + * - `Status:` → HTTP status code (default = 200). + * - `Content-Type:` → MIME type (optional). + * - Computes the offset where the body starts and its size. + * - Builds an `HttpResponse` via `ResponseBuilder::generateSuccessFile`, which streams + * the body directly from the output file. + * - Marks the response with the CGI temp file path so it can be cleaned up later. + * + * @param cgi Reference to the `CgiProcess` containing CGI state (output path, timestamps). + * @param server Server configuration (used for error responses). + * @param req Original HTTP request (used for logging and response metadata). + * + * @return An `HttpResponse` containing either: + * - The CGI result (with headers + body). + * - Or an error response (500) if parsing/reading failed. + * + * @note Only the first 9KB of output is scanned for headers. Larger headers are unsupported. + * @warning Caller is responsible for eventually cleaning up the CGI temp file. + */ HttpResponse finalizeCgi(CgiProcess& cgi, const Server& server, const HttpRequest& req) { cgi.last_activity = getCurrentTime(); @@ -331,6 +576,23 @@ HttpResponse finalizeCgi(CgiProcess& cgi, const Server& server, const HttpReques return resp; } +/** + * @brief Forcefully terminate a CGI process and clean up its resources. + * + * @details + * - Logs the intent to kill the CGI process. + * - Sends `SIGKILL` to the child process using its PID. + * - Calls `waitpid()` to reap the process and avoid zombies. + * - Logs an error if `waitpid` fails. + * - Deletes temporary input/output files via `unlinkWithErrorLog`. + * - Resets the `CgiProcess` state (`pid`, timestamps, paths). + * + * Typically invoked when a CGI process times out or becomes unresponsive. + * + * @param cgi Reference to the CGI process metadata to be cleaned up. + * + * @warning This is a hard kill (`SIGKILL`) — no graceful shutdown of the CGI script. + */ void errorOnCgi(CgiProcess& cgi) { Logger::logFrom(LogLevel::ERROR, "CGI", "Killing CGI process with PID: " + std::to_string(cgi.pid)); @@ -350,6 +612,19 @@ void errorOnCgi(CgiProcess& cgi) { cgi.script_path.clear(); } +/** + * @brief Clean up resources of a finished CGI process. + * + * @details + * - Removes the temporary input file associated with the CGI process. + * - Resets process metadata (`pid`, timestamps, input path). + * - Does not touch the CGI output file (which may still be in use + * by the HTTP response). + * + * Typically called after `finalizeCgi()` has built the response. + * + * @param cgi Reference to the CGI process metadata to be reset. + */ void cleanupCgi(CgiProcess& cgi) { unlinkWithErrorLog(cgi.input_path, "input temp file"); cgi.pid = -1; @@ -358,6 +633,23 @@ void cleanupCgi(CgiProcess& cgi) { cgi.input_path.clear(); } +/** + * @brief Check whether a CGI process has terminated, and reap it if so. + * + * @details + * - Calls `waitpid(pid, &status, WNOHANG)`: + * - Returns `false` if the process is still running (`result == 0`). + * - Returns `true` if the process has terminated (or if `waitpid` fails). + * - Logs an error if `waitpid` itself fails (`result == -1`). + * + * This function is non-blocking and can be safely called in the event loop + * to periodically check if a CGI process is done. + * + * @param cgi Reference to the CGI process metadata. + * + * @return `true` if the CGI process has terminated (or an error occurred), + * `false` if it is still running. + */ bool tryTerminateCgi(CgiProcess& cgi) { int status; pid_t result = waitpid(cgi.pid, &status, WNOHANG); diff --git a/src/http/methodsHandler/handleDelete.cpp b/src/http/methodsHandler/handleDelete.cpp index 7c32b5fa..7352331d 100644 --- a/src/http/methodsHandler/handleDelete.cpp +++ b/src/http/methodsHandler/handleDelete.cpp @@ -3,13 +3,38 @@ /* ::: :::::::: */ /* handleDelete.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/21 15:06:07 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:22:25 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 10:23:19 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file methodsHandler.hpp + * @brief Declares handlers for supported HTTP request methods. + * + * @details This header exposes the top-level functions used by the + * request router to process method-specific logic: + * - @ref handleGet: serve static files, run CGI, or generate + * autoindex listings. + * - @ref handlePost: handle client uploads (raw body, + * URL-encoded forms, multipart forms). + * - @ref handleDelete: remove existing resources from the + * server filesystem. + * + * Additionally, it provides: + * - @ref generateAutoindex: build a directory listing page + * when autoindexing is enabled. + * - @ref handleMultipartForm: internal helper for parsing + * multipart form-data uploads. + * + * These functions are invoked by the router after method + * validation (see @ref requestRouter.cpp). + * + * @ingroup request_handler + */ + #include "http/HttpRequest.hpp" // for HttpRequest #include "http/HttpResponse.hpp" // for HttpResponse #include "http/responseBuilder.hpp" // for generateError, generateSuccess @@ -28,21 +53,54 @@ class Server; namespace { +/** + * @brief Attempts to delete a file safely, producing an error response on failure. + * + * @details This helper encapsulates the low-level checks and filesystem + * operations for handling HTTP `DELETE` requests. It verifies: + * - Existence of the target (returns 404 if not found). + * - That the target is **not** a directory or special file (returns 403). + * - That the target is a regular file (otherwise rejected with 403). + * - Filesystem deletion via `std::filesystem::remove`, with + * detailed error mapping: + * - EACCES / EPERM → 403 Forbidden + * - ENOENT (file vanished) → 404 Not Found + * - Other errors → 500 Internal Server Error + * + * On success, the file is removed and the function returns `true`. + * On failure, `outError` is populated with an appropriate + * @ref HttpResponse. + * + * @param filepath Full path to the file to delete. + * @param req Original HTTP request (used in error responses). + * @param server Active server context (used in error responses). + * @param outError Populated with a generated error response if deletion fails. + * + * @return `true` if the file was successfully unlinked, `false` otherwise. + * + * @ingroup request_handler + */ bool unlinkFile(const std::string& filepath, const HttpRequest& req, const Server& server, HttpResponse& outError) { struct stat st; + + // 1) Stat the file: reject if not found if (stat(filepath.c_str(), &st) != 0) { Logger::logFrom(LogLevel::WARN, "Delete Handler", "File not found → rejecting DELETE: " + filepath); outError = ResponseBuilder::generateError(404, server, req); return false; } + + // 2) Reject directories outright if (S_ISDIR(st.st_mode)) { Logger::logFrom(LogLevel::WARN, "Delete Handler", "Target is a directory → rejecting DELETE: " + filepath); outError = ResponseBuilder::generateError(403, server, req); return false; } + + // 3) Reject non-regular files (devices, sockets, etc.) if (!S_ISREG(st.st_mode)) { Logger::logFrom(LogLevel::WARN, "Delete Handler", "Target is not a regular file → rejecting DELETE: " + filepath); @@ -50,6 +108,7 @@ bool unlinkFile(const std::string& filepath, const HttpRequest& req, const Serve return false; } + // 4) Attempt deletion with detailed error handling std::error_code ec; if (!std::filesystem::remove(filepath, ec)) { if (ec.value() == EACCES || ec.value() == EPERM) { @@ -70,9 +129,24 @@ bool unlinkFile(const std::string& filepath, const HttpRequest& req, const Serve return false; } + // Success: file removed return true; } +/** + * @brief Generates a simple HTML confirmation page for a deleted file. + * + * @details Builds a minimal HTML5 document containing a title and message + * confirming that the requested file has been successfully deleted. + * The filename is HTML-escaped before insertion to prevent XSS. + * + * @param filename Name of the deleted file (not a path). Will be sanitized + * via @ref htmlEscape before being injected into the HTML. + * + * @return A fully-formed HTML string suitable as the body of a 200 OK response. + * + * @ingroup request_handler + */ std::string generateDeleteHtml(const std::string& filename) { std::stringstream html; html << R"( @@ -95,7 +169,32 @@ std::string generateDeleteHtml(const std::string& filename) { } // namespace +/** + * @brief Handles an HTTP DELETE request for a given resource. + * + * @details This function enforces server security rules and attempts + * to remove the requested file if permitted: + * 1. Resolve the request path to a physical filesystem path. + * 2. Reject empty resolutions (403). + * 3. Reject symlinks (403) to prevent symlink attacks. + * 4. Reject directory-like URIs (403) to avoid recursive deletion. + * 5. Attempt file deletion with @ref unlinkFile, which validates + * type, permissions, and maps errors to proper responses. + * 6. On success, build and return a 200 OK response containing + * a confirmation HTML page (via @ref generateDeleteHtml). + * + * @param req The incoming HTTP request object. + * @param server Active server context. + * @param loc The matched location block from the configuration. + * + * @return A fully constructed @ref HttpResponse: + * - 200 OK with HTML confirmation if deletion succeeds. + * - Error response (403, 404, or 500) on failure. + * + * @ingroup request_handler + */ HttpResponse handleDelete(const HttpRequest& req, const Server& server, const Location& loc) { + // 1) Resolve the requested URI to a filesystem path std::string path = resolvePhysicalPath(req, loc); if (path.empty()) { Logger::logFrom(LogLevel::WARN, "Delete Handler", @@ -103,12 +202,14 @@ HttpResponse handleDelete(const HttpRequest& req, const Server& server, const Lo return ResponseBuilder::generateError(403, server, req); } + // 2) Reject symbolic links to avoid symlink traversal exploits if (isSymlink(path)) { Logger::logFrom(LogLevel::WARN, "Delete Handler", "Target is a symlink → rejecting DELETE for URI: " + req.getPath()); return ResponseBuilder::generateError(403, server, req); } + // 3) Reject URIs ending with '/' (treated as directories) if (req.getPath().back() == '/') { Logger::logFrom(LogLevel::WARN, "Delete Handler", "Request URI ends with '/' → rejecting DELETE for directory-like path: " + @@ -116,15 +217,19 @@ HttpResponse handleDelete(const HttpRequest& req, const Server& server, const Lo return ResponseBuilder::generateError(403, server, req); } + // 4) Try to delete the file; errors are mapped to proper HTTP responses HttpResponse errResp; if (!unlinkFile(path, req, server, errResp)) { return errResp; } + // 5) On success: build a sanitized confirmation page std::string rawName = extractFilenameFromUri(req.getPath()); - std::string safeName = htmlEscape(rawName); - std::string body = generateDeleteHtml(safeName); + std::string safeName = htmlEscape(rawName); // escape to prevent XSS + std::string body = generateDeleteHtml(safeName); // build confirmation HTML Logger::logFrom(LogLevel::INFO, "Delete Handler", "Successfully deleted “" + safeName + "” → sending HTML confirmation"); + + // 6) Return a 200 OK response with confirmation body return ResponseBuilder::generateSuccess(200, body, "text/html", req); } diff --git a/src/http/methodsHandler/handleGet/generateAutoindex.cpp b/src/http/methodsHandler/handleGet/generateAutoindex.cpp index 83474c6c..75e4219c 100644 --- a/src/http/methodsHandler/handleGet/generateAutoindex.cpp +++ b/src/http/methodsHandler/handleGet/generateAutoindex.cpp @@ -3,13 +3,44 @@ /* ::: :::::::: */ /* generateAutoindex.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov +#+ +:+ +#+ */ +/* By: nlouis +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/06 09:08:33 by nlouis #+# #+# */ -/* Updated: 2025/08/18 11:06:54 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 10:53:24 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file generateAutoindex.cpp + * @brief Implements directory autoindex (HTML listing) generation. + * + * @details This file provides nginx-like autoindex functionality for HTTP GET + * requests on directories when no index file is found and autoindexing + * is enabled in the location block. + * + * Main responsibilities: + * - **Directory scanning** (@ref listDirectoryEntries): + * Collects entries, separates directories from files, preserves `..`. + * - **Escaping utilities**: + * - @ref escapeUriComponent for safe URIs. + * - @ref htmlEscape (from utils/htmlUtils) for HTML-safe output. + * - **Formatting helpers**: + * - @ref formatMTimeUTC for human-readable UTC timestamps. + * - @ref renderRow for HTML table rows with icons and metadata. + * - **HTML layout**: + * - @ref htmlHeader and @ref htmlFooter wrap the index table. + * - **Entrypoint**: @ref generateAutoindex assembles the full HTML + * directory listing and returns a @ref HttpResponse. + * + * Errors (nonexistent or inaccessible directories) result in 403 + * responses. Valid directories produce a simple HTML table with: + * - File/directory name (with 📄/📁 icons). + * - Last modified timestamp (UTC). + * - File size (bytes). + * + * @ingroup request_handler + */ + #include "http/HttpRequest.hpp" // for HttpRequest #include "http/HttpResponse.hpp" // for HttpResponse #include "http/responseBuilder.hpp" // for generateError, generateSuccess @@ -34,7 +65,28 @@ namespace fs = std::filesystem; namespace { -// 1) Scan directory entries: separate names into dirs/files, keep “..” first if present +/** + * @struct DirEntry + * @brief Represents a single entry in a directory for autoindexing. + * + * @details Used by @ref listDirectoryEntries to store metadata for each + * file or subdirectory. Provides the minimal information needed + * to render an autoindex table row in HTML. + * + * @var DirEntry::name + * The entry’s filename (e.g., `"file.txt"` or `"subdir"`). + * + * @var DirEntry::size + * File size in bytes. For directories and `".."`, this is set to `0`. + * + * @var DirEntry::mtime + * Last modification time (UTC, as `time_t`). Used for “Last Modified” column. + * + * @var DirEntry::isDir + * Whether the entry is a directory (`true`) or a regular file (`false`). + * + * @ingroup request_handler + */ struct DirEntry { std::string name; std::uintmax_t size; @@ -44,6 +96,30 @@ struct DirEntry { namespace fs = std::filesystem; +/** + * @brief Lists and sorts directory entries for autoindex generation. + * + * @details Scans the given filesystem path and collects metadata into + * @ref DirEntry objects. Handles errors in a non-throwing way + * using `std::error_code`. Results are separated into: + * - `".."` (if parent exists, inserted at the front). + * - Subdirectories (sorted alphabetically). + * - Files (sorted alphabetically). + * + * Each entry records: + * - `name`: filename or `".."`. + * - `isDir`: directory vs file. + * - `size`: file size in bytes (0 for directories). + * - `mtime`: last modification time (UTC). + * + * @param path Filesystem path to the directory. + * @param ok Reference flag set to `false` if the directory + * cannot be opened or scanned safely. + * + * @return A vector of @ref DirEntry objects sorted for autoindex rendering. + * + * @ingroup request_handler + */ std::vector listDirectoryEntries(const std::string& path, bool& ok) { std::vector entries; ok = true; @@ -54,7 +130,7 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { return {}; } - // Try to construct a directory_iterator without throwing: + // Try non-throwing directory iterator std::error_code dirEc; fs::directory_iterator it(dirPath, dirEc); if (dirEc) { @@ -62,10 +138,10 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { return {}; } - // Now loop in a non‐throwing way: + // Walk entries safely (clear errors on failure) std::error_code entryEc; for (; it != fs::directory_iterator{}; it.increment(entryEc)) { - if (entryEc) { + if (entryEc) { // skip unreadable entries entryEc.clear(); continue; } @@ -73,6 +149,7 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { fs::path p = it->path(); std::error_code statEc; + // Get last modification time fs::file_time_type ftime = fs::last_write_time(p, statEc); if (statEc) continue; @@ -80,9 +157,11 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { DirEntry entry; entry.name = p.filename().string(); + // Detect directory std::error_code dirCheckEc; entry.isDir = fs::is_directory(p, dirCheckEc) && !dirCheckEc; + // Get size (only for files) if (entry.isDir) { entry.size = 0ULL; } else { @@ -91,7 +170,7 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { entry.size = 0ULL; } - // Convert ftime → time_t + // Convert ftime → time_t for portable rendering auto sctp = std::chrono::time_point_cast( ftime - fs::file_time_type::clock::now() + std::chrono::system_clock::now()); entry.mtime = std::chrono::system_clock::to_time_t(sctp); @@ -99,7 +178,7 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { entries.push_back(entry); } - // Insert “..” if parent exists: + // Insert “..” if parent directory exists fs::path parent = dirPath.parent_path(); if (!parent.empty() && fs::exists(parent)) { DirEntry up; @@ -110,7 +189,7 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { entries.insert(entries.begin(), up); } - // Sort “..” at front, then other dirs, then files: + // Sort: “..” first, then directories, then files (alphabetically) std::vector dirs, files; for (auto const& e : entries) { if (e.name == "..") @@ -124,63 +203,134 @@ std::vector listDirectoryEntries(const std::string& path, bool& ok) { std::sort(dirs.begin(), dirs.end(), cmp); std::sort(files.begin(), files.end(), cmp); + // Build sorted result std::vector sorted; if (!entries.empty() && entries[0].name == "..") - sorted.push_back(entries[0]); + sorted.push_back(entries[0]); // keep parent first sorted.insert(sorted.end(), dirs.begin(), dirs.end()); sorted.insert(sorted.end(), files.begin(), files.end()); return sorted; } +/** + * @brief Percent-encodes a string for safe inclusion in a URI component. + * + * @details Encodes all characters except the unreserved set defined by RFC 3986: + * `ALPHA / DIGIT / "-" / "_" / "." / "~"`. + * Other bytes are converted into `%HH` format using uppercase hex digits. + * + * @param s Input string (raw filename or path segment). + * + * @return A URI-safe string with reserved characters percent-encoded. + * + * @note This is used when generating autoindex links to ensure that spaces, + * control characters, and unsafe symbols don’t break the resulting URL. + * + * @ingroup request_handler + */ std::string escapeUriComponent(const std::string& s) { std::ostringstream oss; oss << std::uppercase << std::hex; + for (unsigned char c : s) { + // Allowed characters remain unchanged if (std::isalnum(c) || c == '-' || c == '_' || c == '.' || c == '~') { oss << c; } else { + // Encode everything else as %HH oss << '%' << std::setw(2) << std::setfill('0') << static_cast(c); } } return oss.str(); } +/** + * @brief Formats a file modification time into a human-readable UTC string. + * + * @details Converts a `time_t` (last modification time) into a textual + * representation using `gmtime()` and `std::put_time`. + * The format is: `DD-Mon-YYYY HH:MM` (UTC). + * + * @param t Last modification time as `std::time_t`. + * + * @return A formatted string (e.g., `"19-Aug-2025 09:42"`) or an empty + * string if conversion fails. + * + * @note This is used in autoindex HTML tables for the **Last Modified** column. + * + * @ingroup request_handler + */ std::string formatMTimeUTC(std::time_t t) { - std::tm* gmPtr = std::gmtime(&t); + std::tm* gmPtr = std::gmtime(&t); // Convert to UTC (thread-unsafe) if (!gmPtr) { return {}; } std::ostringstream ss; - ss << std::put_time(gmPtr, "%d-%b-%Y %H:%M"); + ss << std::put_time(gmPtr, "%d-%b-%Y %H:%M"); // Format: "19-Aug-2025 09:42" return ss.str(); } +/** + * @brief Renders a single `` row in the autoindex HTML table. + * + * @details Converts a directory entry into a clickable HTML link with + * metadata (last modified, size). + * - The `".."` entry is treated as a parent directory link. + * - Directories get a trailing slash (`/`) in both display and href. + * - Files show their size in bytes, while directories/`".."` show `"-"`. + * - Icons: 📁 for directories, 📄 for files. + * + * @param body Output stringstream accumulating the HTML response. + * @param entry Directory entry metadata (name, size, mtime, type). + * @param baseUri Base URI (must end with `/`) to prepend to href links. + * + * @note This function applies both @ref escapeUriComponent (for href safety) + * and @ref htmlEscape (for display safety), preventing XSS or broken links. + * + * @ingroup request_handler + */ void renderRow(std::ostringstream& body, const DirEntry& entry, const std::string& baseUri) { bool isParent = (entry.name == ".."); bool isDir = entry.isDir; + // Construct href and display name std::string href; std::string disp = entry.name; if (isParent) { - href = baseUri + "../"; + href = baseUri + "../"; // parent directory link disp = "../"; } else { - href = baseUri + escapeUriComponent(entry.name); + href = baseUri + escapeUriComponent(entry.name); // percent-encode special chars if (isDir) { disp += "/"; href += "/"; } } + // Metadata columns std::string icon = isDir ? "📁" : "📄"; std::string mtime = isParent ? "-" : formatMTimeUTC(entry.mtime); std::string size = (isDir || isParent) ? "-" : std::to_string(entry.size); + // Append HTML table row body << " " << "" << icon << " " << htmlEscape(disp) << "" << "" << mtime << "" << "" << size << "" << "\n"; } +/** + * @brief Generates the opening HTML markup for an autoindex page. + * + * @details Builds the document ``, page title, and initial table + * structure for directory listing. + * The provided @p uri is escaped with @ref htmlEscape to prevent + * XSS and is displayed in both the `` and `<h1>` headers. + * + * @param uri The request URI being indexed (e.g. "/images/"). + * @return HTML string containing the document header and table headers. + * + * @ingroup request_handler + */ std::string htmlHeader(const std::string& uri) { std::ostringstream ss; ss << R"(<!DOCTYPE html> @@ -199,36 +349,77 @@ std::string htmlHeader(const std::string& uri) { return ss.str(); } +/** + * @brief Generates the closing HTML markup for an autoindex page. + * + * @details Completes the open `<table>`, closes `<body>`, and terminates + * the HTML document. This function should always be called after + * @ref htmlHeader and rows rendered with @ref renderRow. + * + * @return HTML string containing the footer markup. + * + * @ingroup request_handler + */ std::string htmlFooter() { return std::string(" </table>\n</body>\n</html>"); } - } // anonymous namespace +/** + * @brief Generates an HTML autoindex page for a directory listing. + * + * @details This function inspects the given filesystem path, enumerates + * its directory entries via @ref listDirectoryEntries, and + * produces an HTML table showing subdirectories and files with + * name, last modification time, and size. + * + * Workflow: + * 1. **Directory validation**: Ensure the target path exists and is + * accessible. If not, return a `403 Forbidden` error. + * 2. **Base URI normalization**: Guarantee that the URI ends with + * a trailing slash (`/`) for proper link resolution. + * 3. **HTML generation**: Build the autoindex page by composing + * @ref htmlHeader, calling @ref renderRow for each entry, and + * closing with @ref htmlFooter. + * 4. **Response assembly**: Return a `200 OK` response with the + * generated HTML body and `text/html` MIME type. + * + * @param filepath Filesystem path to the directory being indexed. + * @param uri Request URI corresponding to the directory. + * @param request Incoming HTTP request context. + * @param server Server configuration (used for error responses). + * + * @return HttpResponse with a full HTML autoindex page on success, + * or an error response (403) if the directory cannot be accessed. + * + * @ingroup request_handler + */ HttpResponse generateAutoindex(const std::string& filepath, const std::string& uri, const HttpRequest& request, const Server& server) { bool dirOk; auto entries = listDirectoryEntries(filepath, dirOk); if (!dirOk) { + // Directory not accessible → return 403 Forbidden Logger::logFrom(LogLevel::WARN, "Autoindex", "Cannot open directory: " + filepath + " → rejecting autoindex for URI: " + request.getPath()); return ResponseBuilder::generateError(403, server, request); } - // Ensure baseUri ends with “/” + // Ensure base URI ends with “/” for proper relative link building std::string baseUri = uri; if (baseUri.empty() || baseUri.back() != '/') baseUri += '/'; - // Build the HTML body + // Build the HTML body: header + rows + footer std::ostringstream body; body << htmlHeader(uri); for (auto const& e : entries) { - renderRow(body, e, baseUri); + renderRow(body, e, baseUri); // Add a row for each directory/file entry } body << htmlFooter(); + // Log and return success response with generated HTML Logger::logFrom(LogLevel::INFO, "Autoindex", "Generated autoindex for URI: " + request.getPath() + " (directory: " + filepath + ")"); diff --git a/src/http/methodsHandler/handleGet/handleGet.cpp b/src/http/methodsHandler/handleGet/handleGet.cpp index 6a5ba158..c848e91e 100644 --- a/src/http/methodsHandler/handleGet/handleGet.cpp +++ b/src/http/methodsHandler/handleGet/handleGet.cpp @@ -3,13 +3,38 @@ /* ::: :::::::: */ /* handleGet.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov <irychkov@student.hive.fi> +#+ +:+ +#+ */ +/* By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/15 12:39:41 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:24:33 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 10:46:55 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file handleGet.cpp + * @brief Implements the HTTP GET request handler. + * + * @details This file provides the logic for serving resources in response + * to HTTP GET requests. It includes: + * - **MIME type detection** (@ref detectMimeType) for correct + * `Content-Type` headers. + * - **File serving** (@ref serveFile) with small-file in-memory + * responses and large-file streaming responses. + * - **Directory handling** (@ref processDirectory): + * - Redirects URIs missing a trailing slash. + * - Serves configured `index` files if present. + * - Generates autoindex listings if enabled. + * - Returns 403 when directory listing is disabled. + * - **Main dispatcher** (@ref handleGet) which ties everything + * together, enforcing nginx-like behavior: + * - Rejects symlinks for security. + * - Differentiates between directories and regular files. + * - Returns appropriate errors (403/404) when access is denied + * or resources are missing. + * + * @ingroup request_handler + */ + #include "core/Location.hpp" // for Location #include "http/HttpRequest.hpp" // for HttpRequest #include "http/HttpResponse.hpp" // for HttpResponse @@ -30,7 +55,23 @@ class Server; namespace { +/** + * @brief Detects the MIME type of a file based on its extension. + * + * @details Uses a static lookup table mapping common file extensions + * (e.g., `.html`, `.png`, `.json`) to their corresponding + * MIME types. The file extension is normalized to lowercase + * before lookup. If the extension is not recognized, the + * generic `application/octet-stream` type is returned. + * + * @param file_path Path to the file whose MIME type should be determined. + * + * @return A MIME type string suitable for the `Content-Type` header. + * + * @ingroup request_handler + */ static std::string detectMimeType(const std::string& file_path) { + // Static lookup table of common extensions → MIME types static const std::map<std::string, std::string> mime_types = { {".html", "text/html"}, {".htm", "text/html"}, {".css", "text/css"}, {".js", "application/javascript"}, @@ -43,12 +84,15 @@ static std::string detectMimeType(const std::string& file_path) { {".mp3", "audio/mpeg"}, {".mp4", "video/mp4"}, {".wasm", "application/wasm"}}; + // Extract file extension from path std::filesystem::path fsPath(file_path); std::string ext = fsPath.extension().string(); + // Normalize extension to lowercase for lookup std::transform(ext.begin(), ext.end(), ext.begin(), [](unsigned char c) { return static_cast<char>(std::tolower(c)); }); + // Return known type if found, else default to binary auto it = mime_types.find(ext); if (it != mime_types.end()) return it->second; @@ -56,25 +100,59 @@ static std::string detectMimeType(const std::string& file_path) { return "application/octet-stream"; } +/** + * @brief Serves a static file in response to an HTTP GET request. + * + * @details This helper reads a file from disk and builds the + * appropriate @ref HttpResponse: + * - Verifies that the target is a valid file. + * - Opens the file in binary mode, rejects if inaccessible. + * - Determines the `Content-Type` header (uses + * @ref detectMimeType if none is provided). + * - If the file size is small (≤ 100 KiB), reads the + * entire file into memory and returns it inline. + * - If the file size is larger, streams it back using + * @ref ResponseBuilder::generateSuccessFile. + * + * Logs each decision (missing file, forbidden, in-memory, + * or streaming). + * + * @param file_path Filesystem path to the target file. + * @param request Incoming HTTP request. + * @param server Active server context (for error generation). + * @param content_type Optional MIME type override; if empty, auto-detected. + * + * @return A valid @ref HttpResponse: + * - 200 with body or stream on success. + * - 404 if file not found. + * - 403 if access is denied. + * + * @ingroup request_handler + */ static HttpResponse serveFile(const std::string& file_path, const HttpRequest& request, const Server& server, std::string content_type) { + // 1) Check if the path is a file if (!isFile(file_path)) { Logger::logFrom(LogLevel::WARN, "Get Handler", "File not found: " + file_path); return ResponseBuilder::generateError(404, server, request); } + // 2) Attempt to open the file std::ifstream file(file_path, std::ios::binary | std::ios::ate); if (!file.is_open()) { Logger::logFrom(LogLevel::WARN, "Get Handler", "Cannot open file: " + file_path); return ResponseBuilder::generateError(403, server, request); } + // 3) Determine file size std::streamsize size = file.tellg(); file.seekg(0, std::ios::beg); + // 4) Detect MIME type if not provided if (content_type.empty()) content_type = detectMimeType(file_path); + // 5) Small files (≤100 KiB) → load fully into memory constexpr std::streamsize MEMORY_LIMIT = 100 * 1024; if (size <= MEMORY_LIMIT) { std::ostringstream buffer; @@ -86,6 +164,7 @@ static HttpResponse serveFile(const std::string& file_path, const HttpRequest& r " bytes): " + file_path); return ResponseBuilder::generateSuccess(200, body, content_type, request); } else { + // 6) Large files → return as streaming response Logger::logFrom(LogLevel::INFO, "Get Handler", "Serving large file via streaming (size: " + std::to_string(size) + " bytes): " + file_path); @@ -93,9 +172,38 @@ static HttpResponse serveFile(const std::string& file_path, const HttpRequest& r } } +/** + * @brief Handles HTTP GET requests targeting a directory. + * + * @details This function enforces nginx-like directory handling rules: + * 1. Normalize the URI and reject empty results (403). + * 2. If the URI does not end with a slash, issue a `301 Moved Permanently` + * redirect to the slash-terminated URI. + * 3. If an `index` file is configured: + * - Serve it if it exists. + * - Otherwise, if autoindex is enabled, generate a directory listing. + * - Otherwise, return 403 (listing disabled). + * 4. If no index is configured: + * - Generate autoindex if enabled. + * - Otherwise, return 403 (directory listing forbidden). + * + * @param dirPath Physical filesystem path to the requested directory. + * @param uri Original request URI. + * @param request Incoming HTTP request. + * @param server Active server context. + * @param loc The matched location block for this URI. + * + * @return A fully built @ref HttpResponse: + * - 301 Redirect if missing slash. + * - 200 OK with index file or autoindex if allowed. + * - 403 Forbidden if neither index nor autoindex is permitted. + * + * @ingroup request_handler + */ static HttpResponse processDirectory(const std::string& dirPath, const std::string& uri, const HttpRequest& request, const Server& server, const Location& loc) { + // 1) Normalize URI and reject empty (security check) std::string normalized = normalizePath(uri); if (normalized.empty()) { Logger::logFrom(LogLevel::WARN, "Get Handler", @@ -103,21 +211,24 @@ static HttpResponse processDirectory(const std::string& dirPath, const std::stri return ResponseBuilder::generateError(403, server, request); } + // 2) Redirect if URI does not end with '/' if (normalized.back() != '/') { std::string target = normalized + "/"; Logger::logFrom(LogLevel::INFO, "Get Handler", "Redirecting to: " + target); return ResponseBuilder::generateRedirect(301, target, request); } + // 3) If an index file is configured in the location const std::string& indexName = loc.getIndex(); if (!indexName.empty()) { std::string indexFullPath = joinPath(dirPath, indexName); if (isFile(indexFullPath)) { - return serveFile(indexFullPath, request, server, ""); + return serveFile(indexFullPath, request, server, + ""); // serve index.html (or configured name) } if (loc.isAutoindexEnabled()) { - return generateAutoindex(dirPath, uri, request, server); + return generateAutoindex(dirPath, uri, request, server); // fallback to autoindex } Logger::logFrom(LogLevel::WARN, "Get Handler", @@ -125,6 +236,7 @@ static HttpResponse processDirectory(const std::string& dirPath, const std::stri return ResponseBuilder::generateError(403, server, request); } + // 4) If no index: allow autoindex or forbid if (loc.isAutoindexEnabled()) { return generateAutoindex(dirPath, uri, request, server); } @@ -136,11 +248,38 @@ static HttpResponse processDirectory(const std::string& dirPath, const std::stri } // namespace +/** + * @brief Handles an HTTP GET request for a resource. + * + * @details This function implements nginx-like GET semantics: + * 1. Resolve the request URI to a physical filesystem path. + * 2. If the URI ends with a `/` but points to a file (e.g., `/file/`): + * - Reject with 404 (trailing slash on file). + * 3. Reject symlinks for security (403). + * 4. If the path exists: + * - If it’s a directory, delegate to @ref processDirectory + * (redirect, index, or autoindex handling). + * - If it’s a regular file, delegate to @ref serveFile + * (in-memory or streaming response). + * 5. If nothing matches, return 404 (not found). + * + * @param request Incoming HTTP request object. + * @param server Active server context. + * @param loc The matched location block for this URI. + * + * @return A fully constructed @ref HttpResponse: + * - 200 OK with file contents or directory listing. + * - 301 Redirect for missing slash. + * - 403 Forbidden for symlinks or disallowed directories. + * - 404 Not Found when resource doesn’t exist. + * + * @ingroup request_handler + */ HttpResponse handleGet(const HttpRequest& request, const Server& server, const Location& loc) { + // 1) Resolve request path to a filesystem path std::string filepath = resolvePhysicalPath(request, loc); - // NGINX: any GET ending in '/' is a directory lookup. - // If stripping the slash yields an existing file, return 404. + // 2) Special nginx rule: if URI ends with '/' but points to a file → 404 const std::string& uri = request.getPath(); if (!uri.empty() && uri.back() == '/') { std::string fileNoSlash = filepath; @@ -153,25 +292,30 @@ HttpResponse handleGet(const HttpRequest& request, const Server& server, const L } } + // 3) Reject symlinks for security if (isSymlink(filepath)) { Logger::logFrom(LogLevel::WARN, "Get Handler", "Symlink detected, rejecting request for: " + filepath); return ResponseBuilder::generateError(403, server, request); } + // 4) Check file type struct stat fileStat; if (stat(filepath.c_str(), &fileStat) == 0) { const std::string& uri = request.getPath(); if (S_ISDIR(fileStat.st_mode)) { + // Directory → delegate to processDirectory return processDirectory(filepath, uri, request, server, loc); } if (S_ISREG(fileStat.st_mode)) { + // Regular file → delegate to serveFile return serveFile(filepath, request, server, ""); } } + // 5) Not found or inaccessible Logger::logFrom(LogLevel::WARN, "Get Handler", "File not found or inaccessible: " + filepath); return ResponseBuilder::generateError(404, server, request); } diff --git a/src/http/methodsHandler/handlePost/handleMultipartForm.cpp b/src/http/methodsHandler/handlePost/handleMultipartForm.cpp index a25c8d13..8784100a 100644 --- a/src/http/methodsHandler/handlePost/handleMultipartForm.cpp +++ b/src/http/methodsHandler/handlePost/handleMultipartForm.cpp @@ -3,13 +3,47 @@ /* ::: :::::::: */ /* handleMultipartForm.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov <irychkov@student.hive.fi> +#+ +:+ +#+ */ +/* By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/06/06 21:44:51 by nlouis #+# #+# */ -/* Updated: 2025/08/17 12:25:35 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 11:24:48 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file handleMultipartForm.cpp + * @brief Handles `multipart/form-data` POST uploads and persists the first file part to disk. + * + * @details + * Parses the request body using the `boundary` from the `Content-Type` header, + * scans parts until it finds a file part (`filename="..."`), trims the trailing + * CRLF from the part payload, enforces `client_max_body_size`, and writes the + * bytes to `<upload_dir>/<sanitized-filename>` (or a time‑based fallback). + * + * ### Flow + * 1. Extract `boundary` from `Content-Type`; `400` if missing. + * 2. `parseMultipartForm`: iterate parts, split headers/data, detect `filename=`, + * sanitize file name, trim payload CRLF, return name + bytes. + * 3. Enforce per‑part size ≤ `Server::getClientMaxBodySize()`; `413` otherwise. + * 4. Open destination file (binary) and write bytes atomically; `500` on I/O error. + * 5. Log outcome and return `201 Created` with a minimal HTML confirmation. + * + * ### Security & Robustness + * - **Filename sanitization**: strips unsafe characters to prevent path traversal. + * - **Size limits**: rejects oversized payloads to mitigate DoS. + * - **Strict delimiter matching**: avoids bleed between parts. + * + * ### Limitations + * - Parses the **first** file part only (ignores subsequent files/fields). + * - Operates in‑memory (no streaming/chunking); large files may increase RAM use. + * + * @note Consider extending to stream to a temp file, support multiple files/fields, + * and return JSON for API clients. + * + * @see HttpRequest, HttpResponse, ResponseBuilder, Logger, filesystemUtils + * @ingroup request_handler + */ + #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest #include "http/HttpResponse.hpp" // for HttpResponse @@ -25,17 +59,34 @@ namespace { +/** + * @brief Construct the opening delimiter for a multipart part. + * + * @param boundary Boundary token from the Content-Type header. + * @return A delimiter string in the form `"--<boundary>\r\n"`. + */ inline std::string makePartDelimiter(const std::string& boundary) { return "--" + boundary + "\r\n"; } +/** + * @brief Construct the closing delimiter for a multipart body. + * + * @param boundary Boundary token from the Content-Type header. + * @return A delimiter string in the form `"--<boundary>--"`. + */ inline std::string makeCloseDelimiter(const std::string& boundary) { return "--" + boundary + "--"; } /** - * Find the next occurrence of `delimiter` in `body` at or after `startPos`. - * Returns true if found (and sets outPos), false otherwise. + * @brief Find the next occurrence of a delimiter in a multipart body. + * + * @param body Full request body. + * @param delimiter Delimiter string to search for. + * @param startPos Starting offset in the body. + * @param outPos Output parameter; set to the position of the delimiter if found. + * @return true if the delimiter was found, false otherwise. */ bool findNextDelimiter(const std::string& body, const std::string& delimiter, size_t startPos, size_t& outPos) { @@ -44,16 +95,24 @@ bool findNextDelimiter(const std::string& body, const std::string& delimiter, si } /** - * Check if at position `pos` in `body` there is a closing boundary. + * @brief Check if the body contains the closing delimiter at the given position. + * + * @param body Full request body. + * @param pos Position to check. + * @param closeDelimiter Expected closing delimiter string. + * @return true if the body matches the closing delimiter at `pos`. */ bool isCloseDelimiter(const std::string& body, size_t pos, const std::string& closeDelimiter) { return body.compare(pos, closeDelimiter.size(), closeDelimiter) == 0; } /** - * Given a single “part” (everything between delimiters), split it into - * headersBlock (before the first "\r\n\r\n") and dataBlock (after it). - * Returns false if no double‐CRLF is found. + * @brief Split a raw part into headers and data sections. + * + * @param part Raw part (between delimiters). + * @param headersBlock Output; substring up to the first double CRLF. + * @param dataBlock Output; substring after the first double CRLF. + * @return true if headers and data were successfully split, false otherwise. */ bool parseHeadersAndData(const std::string& part, std::string& headersBlock, std::string& dataBlock) { @@ -67,8 +126,11 @@ bool parseHeadersAndData(const std::string& part, std::string& headersBlock, } /** - * Look for filename="..." inside headersBlock. If found, extract the filename - * (without the quotes) into `outFilename` and return true. Otherwise return false. + * @brief Extract and sanitize the filename from a part's headers. + * + * @param headersBlock Headers of the multipart part. + * @param outFilename Output; sanitized filename if found. + * @return true if a `filename="..."` token was found, false otherwise. */ bool extractFilename(const std::string& headersBlock, std::string& outFilename) { const std::string token = "filename=\""; @@ -90,7 +152,10 @@ bool extractFilename(const std::string& headersBlock, std::string& outFilename) } /** - * If `data` ends with "\r\n", strip those two characters off. Otherwise return as-is. + * @brief Strip a trailing CRLF sequence from data if present. + * + * @param data Raw part data block. + * @return A copy of `data` without a trailing `"\r\n"`, if it existed. */ std::string trimEndingCRLF(const std::string& data) { if (data.size() >= 2 && data[data.size() - 2] == '\r' && data[data.size() - 1] == '\n') { @@ -100,16 +165,27 @@ std::string trimEndingCRLF(const std::string& data) { } /** - * Parse a multipart/form-data body and extract the first file‐part’s - * filename and its raw content. Returns true on success, false otherwise. + * @brief Parse a multipart/form-data body and extract the first file part. + * + * @details + * Iterates over all parts in the request body, delimited by the given boundary. + * For each part: + * - Splits headers and data at the first double CRLF. + * - Checks for a `filename="..."` header. + * - If found, sanitizes the filename, trims trailing CRLF from the data block, + * and outputs both. + * + * Stops at the first file part encountered; ignores non-file parts. + * + * @param body Full HTTP request body (including all parts). + * @param boundary Boundary token from the Content-Type header. + * @param filename Output; sanitized filename from the file part. + * @param fileContent Output; raw file content (as bytes in a std::string). * - * - body: full HTTP request body (including all parts) - * - boundary: the “boundary” token from Content-Type, e.g. "----WebKitFormBoundary…" - * - filename: output parameter; set to the file’s original name - * - fileContent: output parameter; set to the file’s bytes (as a std::string) + * @return true if a file part was found and parsed successfully, false otherwise. * - * This will scan part by part until it finds a “filename=” header. If none - * is found, it ultimately returns false. + * @note Only the first file part is extracted; additional files are ignored. + * @warning Operates in memory; large uploads may impact RAM usage. */ bool parseMultipartForm(const std::string& body, const std::string& boundary, std::string& filename, std::string& fileContent) { @@ -181,8 +257,32 @@ bool parseMultipartForm(const std::string& body, const std::string& boundary, st } // anonymous namespace +/** + * @brief Handle a multipart/form-data POST request and persist the uploaded file. + * + * @details + * - Extracts the boundary from the `Content-Type` header. + * - Parses the multipart body to find the first file part. + * - Enforces `client_max_body_size` to prevent DoS. + * - Generates a fallback filename if none was provided. + * - Writes the file content to disk under the given upload directory. + * - Returns an appropriate `HttpResponse`: + * - 400 if malformed, + * - 413 if too large, + * - 500 if I/O fails, + * - 201 with a confirmation HTML page on success. + * + * @param request Incoming HTTP request containing the multipart body. + * @param server Server configuration (used for limits and error pages). + * @param fullDirPath Absolute path of the directory where uploads are stored. + * + * @return HttpResponse representing the outcome (success or error). + * + * @ingroup request_handler + */ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& server, const std::string& fullDirPath) { + // 1) Extract boundary from Content-Type header std::string contentType = request.getHeader("Content-Type"); size_t bpos = contentType.find("boundary="); if (bpos == std::string::npos) { @@ -191,6 +291,7 @@ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& serve } std::string boundary = contentType.substr(bpos + 9); + // 2) Parse body to extract filename + file content std::string extractedFilename; std::string fileContent; if (!parseMultipartForm(request.getBody(), boundary, extractedFilename, fileContent)) { @@ -198,7 +299,7 @@ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& serve return ResponseBuilder::generateError(400, server, request); } - // Enforce per-part size limit to avoid DoS + // 3) Enforce per-part size limit size_t maxBody = server.getClientMaxBodySize(); if (fileContent.size() > maxBody) { Logger::logFrom(LogLevel::WARN, "handleMultipartForm", @@ -207,12 +308,13 @@ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& serve return ResponseBuilder::generateError(413, server, request); } + // 4) Fallback filename if client didn’t provide one if (extractedFilename.empty()) { extractedFilename = "upload_" + std::to_string(getCurrentTime()); } - std::string fullpath = joinPath(fullDirPath, extractedFilename); - + // 5) Compute destination path and open file + std::string fullpath = joinPath(fullDirPath, extractedFilename); std::ofstream out(fullpath, std::ios::binary); if (!out.is_open()) { Logger::logFrom(LogLevel::ERROR, "handleMultipartForm", @@ -220,6 +322,7 @@ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& serve return ResponseBuilder::generateError(500, server, request); } + // 6) Write content and verify out << fileContent; out.close(); if (out.fail()) { @@ -228,6 +331,7 @@ HttpResponse handleMultipartForm(const HttpRequest& request, const Server& serve return ResponseBuilder::generateError(500, server, request); } + // 7) Success → return confirmation page Logger::logFrom(LogLevel::INFO, "handleMultipartForm", "Successfully uploaded: " + fullpath); return ResponseBuilder::generateSuccess( 201, "<html><body><h1>Uploaded: " + htmlEscape(extractedFilename) + "</h1></body></html>", diff --git a/src/http/methodsHandler/handlePost/handlePost.cpp b/src/http/methodsHandler/handlePost/handlePost.cpp index 8ff97031..fcbe2ac1 100644 --- a/src/http/methodsHandler/handlePost/handlePost.cpp +++ b/src/http/methodsHandler/handlePost/handlePost.cpp @@ -3,13 +3,35 @@ /* ::: :::::::: */ /* handlePost.cpp :+: :+: :+: */ /* +:+ +:+ +:+ */ -/* By: irychkov <irychkov@student.hive.fi> +#+ +:+ +#+ */ +/* By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/19 10:19:13 by irychkov #+# #+# */ -/* Updated: 2025/08/17 12:25:19 by irychkov ### ########.fr */ +/* Updated: 2025/08/19 11:17:37 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ +/** + * @file handlePost.cpp + * @brief Implements HTTP POST request handling logic. + * + * @details This file provides the full workflow for processing POST requests: + * - Validating request body size and upload configuration. + * - Normalizing and securing the target upload path against symlinks, + * escapes, and unsafe locations. + * - Supporting multiple content types: + * - `multipart/form-data` → handled by @ref handleMultipartForm. + * - `application/x-www-form-urlencoded` → parsed into key/value + * pairs and saved as HTML. + * - Raw body data → stored directly as a file. + * - Generating safe, unique filenames for uploads when needed. + * - Returning appropriate `HttpResponse` objects: + * - `201 Created` with confirmation on success. + * - Error responses (`400`, `403`, `404`, `413`, `500`) on invalid + * or failed operations. + * + * @ingroup request_handler + */ + #include "core/Location.hpp" // for Location #include "core/Server.hpp" // for Server #include "http/HttpRequest.hpp" // for HttpRequest @@ -34,8 +56,36 @@ namespace { +/** + * @brief Handle a `application/x-www-form-urlencoded` POST request. + * + * @details + * Parses the URL-encoded body of the request into key/value pairs, + * generates a simple HTML confirmation page, and writes it to a file + * at the given target path. + * + * Workflow: + * - Parse form body using `parseFormUrlEncoded`. + * - Reject with **400 Bad Request** if empty or malformed. + * - Build an HTML document containing submitted form fields. + * - Attempt to write the HTML to the specified file path. + * - On failure, return **500 Internal Server Error**. + * - On success, return **201 Created** with confirmation HTML. + * + * @param request The incoming HTTP request (provides form body). + * @param server The server context (used for error generation). + * @param fullpath Filesystem path where the HTML output should be saved. + * @param filename Suggested filename for confirmation message. + * @return HttpResponse + * - 201 Created with confirmation HTML if saved successfully. + * - 400 Bad Request if form body is invalid. + * - 500 Internal Server Error if file operations fail. + * + * @ingroup request_handler + */ static HttpResponse handleUrlEncodedForm(const HttpRequest& request, const Server& server, const std::string& fullpath, const std::string& filename) { + // Parse form body into key/value pairs auto form = parseFormUrlEncoded(request.getBody()); if (form.empty()) { Logger::logFrom(LogLevel::WARN, "Post Handler", @@ -43,6 +93,7 @@ static HttpResponse handleUrlEncodedForm(const HttpRequest& request, const Serve return ResponseBuilder::generateError(400, server, request); } + // Build an HTML page displaying submitted form fields std::string html = "<html><body><h1>Form Received</h1>"; for (auto& formField : form) { html += "<p><b>" + htmlEscape(formField.first) + ":</b> " + htmlEscape(formField.second) + @@ -50,6 +101,7 @@ static HttpResponse handleUrlEncodedForm(const HttpRequest& request, const Serve } html += "</body></html>"; + // Try writing the HTML to the target file std::ofstream out(fullpath); if (!out.is_open()) { Logger::logFrom(LogLevel::ERROR, "Post Handler", @@ -65,14 +117,41 @@ static HttpResponse handleUrlEncodedForm(const HttpRequest& request, const Serve return ResponseBuilder::generateError(500, server, request); } + // Success: log and return confirmation response Logger::logFrom(LogLevel::INFO, "Post Handler", "Successfully wrote URL-encoded form to: " + fullpath); return ResponseBuilder::generateSuccess( 201, "<h1>Form Received. File " + filename + " created.</h1>", "text/html", request); } +/** + * @brief Handle a POST request with a raw body (no specific content type). + * + * @details + * Writes the raw request body directly to the given filesystem path. + * Primarily used for generic uploads when the content type is not + * `multipart/form-data` or `application/x-www-form-urlencoded`. + * + * Workflow: + * - Open target file for binary writing. + * - If opening fails, return **500 Internal Server Error**. + * - Write the raw body from the request into the file. + * - If writing or closing fails, return **500 Internal Server Error**. + * - On success, return **201 Created** with confirmation HTML. + * + * @param request The incoming HTTP request (provides body to save). + * @param server The server context (used for error generation). + * @param fullpath Filesystem path where the body should be saved. + * @param filename Name of the saved file (for confirmation message). + * @return HttpResponse + * - 201 Created with confirmation HTML if saved successfully. + * - 500 Internal Server Error if file I/O fails. + * + * @ingroup request_handler + */ static HttpResponse handleRawBody(const HttpRequest& request, const Server& server, const std::string& fullpath, const std::string& filename) { + // Try opening the target file in binary mode std::ofstream out(fullpath, std::ios::binary); if (!out.is_open()) { Logger::logFrom(LogLevel::ERROR, "Post Handler", @@ -80,14 +159,18 @@ static HttpResponse handleRawBody(const HttpRequest& request, const Server& serv return ResponseBuilder::generateError(500, server, request); } + // Write the raw request body into the file out << request.getBody(); out.close(); + + // Verify the write and close operations succeeded if (out.fail()) { Logger::logFrom(LogLevel::ERROR, "Post Handler", "Failed to write or close file: " + fullpath); return ResponseBuilder::generateError(500, server, request); } + // Success: log and return confirmation response Logger::logFrom(LogLevel::INFO, "Post Handler", "Successfully saved file to: " + fullpath); return ResponseBuilder::generateSuccess( 201, "<html><body><h1>File " + htmlEscape(filename) + " created.</h1></body></html>", @@ -96,12 +179,36 @@ static HttpResponse handleRawBody(const HttpRequest& request, const Server& serv static std::atomic<uint64_t> uploadCounter{0}; +/** + * @brief Generate a unique filename for uploaded files. + * + * @details + * This function ensures that every uploaded file gets a unique name + * even under high concurrency. + * It combines: + * - **Current time in nanoseconds since epoch** (high resolution). + * - **An atomic counter** (`uploadCounter`) to break ties when multiple + * uploads happen within the same nanosecond. + * + * The resulting filename follows the format: + * ``` + * upload_<nanoseconds_since_epoch>_<sequence_number> + * ``` + * + * @return std::string + * A unique, collision-resistant filename suitable for uploads. + * + * @ingroup request_handler + */ static std::string generateFilename() { - // 1) high-res time → nanoseconds since epoch + // 1) Get high-resolution current time → nanoseconds since epoch auto now = std::chrono::system_clock::now().time_since_epoch(); uint64_t ns = std::chrono::duration_cast<std::chrono::nanoseconds>(now).count(); - // 2) per-process counter to break ties within the same nanosecond - uint64_t seq = uploadCounter.fetch_add(1, std::memory_order_relaxed); + + // 2) Increment atomic counter to avoid collisions in the same nanosecond + uint64_t seq = uploadCounter.fetch_add(1, std::memory_order_relaxed); + + // 3) Build the final filename using both components std::ostringstream oss; oss << "upload_" << ns << "_" << seq; return oss.str(); @@ -109,14 +216,40 @@ static std::string generateFilename() { } // namespace +/** + * @brief Validate the prerequisites for processing a POST request. + * + * @details + * This function enforces three key checks before allowing a POST to proceed: + * - Ensures that the request body is not empty (rejects with **400 Bad Request**). + * - Ensures that the body size does not exceed the server's configured + * `client_max_body_size` (rejects with **413 Payload Too Large**). + * - Ensures that the target location has a configured `upload_store` + * (rejects with **403 Forbidden**). + * + * If any of these conditions fail, an appropriate `HttpResponse` is generated. + * Otherwise, the request is considered valid and processing can continue. + * + * @param request The HTTP request to validate. + * @param server The server configuration, used to check limits. + * @param location The location configuration, used to check upload store availability. + * + * @return std::optional<HttpResponse> + * - `std::nullopt` if the request is valid. + * - A generated error response if validation fails. + * + * @ingroup request_handler + */ static std::optional<HttpResponse> validatePostRequest(HttpRequest const& request, Server const& server, Location const& location) { + // 1) Reject empty bodies if (request.getBody().empty()) { Logger::logFrom(LogLevel::WARN, "Post Handler", "Empty body → rejecting POST for URI: " + request.getPath()); return ResponseBuilder::generateError(400, server, request); } + // 2) Reject requests exceeding configured max body size if (request.getBody().size() > server.getClientMaxBodySize()) { Logger::logFrom(LogLevel::WARN, "Post Handler", "Body size " + std::to_string(request.getBody().size()) + @@ -126,6 +259,7 @@ validatePostRequest(HttpRequest const& request, Server const& server, Location c return ResponseBuilder::generateError(413, server, request); } + // 3) Reject if no upload store is configured for this location if (location.getUploadStore().empty()) { Logger::logFrom(LogLevel::WARN, "Post Handler", "No upload store configured → rejecting POST for URI: " + @@ -137,6 +271,39 @@ validatePostRequest(HttpRequest const& request, Server const& server, Location c } namespace fs = std::filesystem; + +/** + * @brief Resolve and validate the filesystem path for a POST upload target. + * + * @details + * This function computes the safe destination path for a file uploaded via POST, + * ensuring it is confined within the configured `upload_store` of a location. + * It enforces NGINX-like semantics and several security checks: + * - Rejects POST requests to paths like `/file.ext/` if `file.ext` exists (→ 404). + * - Decodes percent-encoded components (e.g. `%20`) in the request path. + * - Builds the target path under the `upload_store`, appending a generated + * filename if the client POSTs to a directory or with a trailing slash. + * - Canonicalizes the target path and checks that it remains within the upload root. + * - Rejects paths containing symlinks to prevent traversal attacks (→ 403). + * - Ensures the target directory exists, creating it if needed. + * - Rejects uploads if the target file already exists (→ 400). + * + * On success, the function populates the output references with the resolved + * path, directory, and filename to be used for writing the uploaded content. + * + * @param request The incoming POST request. + * @param server The current server configuration (used for error responses). + * @param location The matched location configuration, with upload store info. + * @param outTargetPath Output: absolute safe path where the file will be written. + * @param outTargetDirectory Output: directory part of the resolved path. + * @param outTargetFilename Output: filename part of the resolved path. + * + * @return std::optional<HttpResponse> + * - `std::nullopt` if the path is valid and safe for writing. + * - An error `HttpResponse` if validation fails. + * + * @ingroup request_handler + */ static std::optional<HttpResponse> preparePostTargetPath(HttpRequest const& request, Server const& server, Location const& location, fs::path& outTargetPath, std::string& outTargetDirectory, @@ -232,32 +399,86 @@ preparePostTargetPath(HttpRequest const& request, Server const& server, Location return std::nullopt; } -static HttpResponse dispatchPostByContentType(HttpRequest const& request, Server const& server, +/** + * @brief Dispatch POST handling based on the request's Content-Type. + * + * @details Chooses the appropriate handler: + * - **multipart/form-data** → @ref handleMultipartForm (saves parts into @p + * targetDirectory). + * - **application/x-www-form-urlencoded** → @ref handleUrlEncodedForm (writes parsed HTML + * to @p targetPath). + * - **(fallback)** any other/absent type → @ref handleRawBody (dumps raw body to @p + * targetPath). + * + * This function relies on a simple substring match of the `Content-Type` + * header. If you later support parameters (e.g. `charset=...`) or more + * types, consider normalizing and parsing the media type token. + * + * @param request Incoming HTTP request (provides headers/body). + * @param server Server context (for error responses). + * @param targetPath Absolute file path chosen for single-file saves. + * @param targetDirectory Directory path to store multipart parts. + * @param targetFilename Leaf filename for single-file saves. + * + * @return A populated @ref HttpResponse produced by the selected handler. + * + * @ingroup request_handler + */ +static HttpResponse dispatchPostByContentType(HttpRequest const& request, const Server& server, std::filesystem::path const& targetPath, std::string const& targetDirectory, std::string const& targetFilename) { + // Read Content-Type once; could be empty or contain parameters (e.g., boundary/charset) std::string contentTypeHeader = request.getHeader("Content-Type"); + // Multipart form → delegate to multipart handler (uses targetDirectory) if (contentTypeHeader.find("multipart/form-data") != std::string::npos) { return handleMultipartForm(request, server, targetDirectory); } + // URL-encoded form → parse and render confirmation HTML to targetPath if (contentTypeHeader.find("application/x-www-form-urlencoded") != std::string::npos) { return handleUrlEncodedForm(request, server, targetPath.string(), targetFilename); } + // Fallback → store raw body verbatim at targetPath return handleRawBody(request, server, targetPath.string(), targetFilename); } +/** + * @brief Handles an HTTP POST request end-to-end. + * + * @details Pipeline: + * 1) **Preconditions** — validate body presence/size and that the @ref Location + * has an `upload_store` (see @ref validatePostRequest). Returns 400/413/403 on failure. + * 2) **Target path preparation** — resolve a safe destination under the + * location's upload store, forbid traversal/symlinks, create parent dirs, + * and choose a filename if needed (see @ref preparePostTargetPath). Returns + * 404/403/500/400 on failure per checks. + * 3) **Dispatch by Content-Type** — select the concrete handler: + * - `multipart/form-data` → @ref handleMultipartForm + * - `application/x-www-form-urlencoded` → @ref handleUrlEncodedForm + * - otherwise → @ref handleRawBody + * + * On success, returns a `201 Created`; on errors, returns the appropriate + * generated @ref HttpResponse from earlier stages. + * + * @param request Incoming HTTP request. + * @param server Active server context. + * @param location Matched location configuration (upload policy, paths). + * @return A fully constructed @ref HttpResponse. + * + * @ingroup request_handler + */ HttpResponse handlePost(HttpRequest const& request, Server const& server, Location const& location) { - // 1) Preconditions + // 1) Preconditions: body present/within limits, upload_store configured std::optional<HttpResponse> maybeErrorResponse = validatePostRequest(request, server, location); if (maybeErrorResponse.has_value()) { return *maybeErrorResponse; } - // 2) Filesystem path prep + // 2) Prepare a safe target path under upload_store (dirs, filename, symlink checks) std::filesystem::path targetPath; std::string targetDirectory; std::string targetFilename; @@ -268,6 +489,6 @@ HttpResponse handlePost(HttpRequest const& request, Server const& server, return *maybePreparationError; } - // 3) Content-type dispatch + // 3) Route to the proper POST handler based on Content-Type return dispatchPostByContentType(request, server, targetPath, targetDirectory, targetFilename); } From 44ed4a5234b5b70d344b44dd8d2daf33607488e9 Mon Sep 17 00:00:00 2001 From: To0nsa <nicolas.lovis@hotmail.fr> Date: Tue, 19 Aug 2025 11:49:06 +0300 Subject: [PATCH 26/32] Improves documentation. Enhances the README with comprehensive project details, including the HTTP server's features, request processing flow, and CGI handling. Updates the doxygen.warnings file, fixing line references and removing outdated error messages. Adjusts a Doxygen group annotation in handleCgi.cpp for consistency. These changes improve clarity for developers and address unresolved documentation references, ensuring better maintainability and usability. --- README.md | 298 +++++++++++++++++++++++++++++++---------- docs/doxygen.warnings | 105 +++++++-------- src/http/handleCgi.cpp | 4 +- 3 files changed, 281 insertions(+), 126 deletions(-) diff --git a/README.md b/README.md index 12f95546..80f62fa1 100644 --- a/README.md +++ b/README.md @@ -10,6 +10,20 @@ > A lightweight HTTP/1.1 server written in modern C++20, compliant with the Hive/42 webserv project specifications. +**Webserv** is our first **large-scale C++ project** at Hive/42. +The goal was to implement a lightweight, fully working HTTP/1.1 server from scratch, trying to use modern C++ standard libraries. + +The server is designed to be RFC-compliant (following HTTP/1.1 [RFC 7230–7235]) and supports essential features such as: + +* Parsing and validating configuration files (Nginx-style syntax). +* Handling GET, POST, DELETE with static files, autoindex, and uploads. +* Executing CGI scripts securely with proper environment and timeouts. +* Multiplexing sockets and CGI pipes in a single poll-based event loop. +* Graceful error handling, timeouts, and connection reuse (keep-alive). + +For reference and correctness, Nginx was used as a behavioral benchmark: routing, error responses, and edge cases were compared against it to ensure realistic and compliant behavior. + +This project was both a challenge in systems programming and a solid introduction to networking, concurrency, and protocol design in modern C++. ___ ## Core of Webserv @@ -120,7 +134,7 @@ This section describes how the configuration parsing logic of **Webserv** works, ___ -## Networking — `SocketManager` +## Networking `SocketManager` The heart of Webserv’s I/O: a single `poll()` loop multiplexing **listening sockets**, **client sockets**, and **CGI pipes**, with strict timeouts and robust error recovery. @@ -138,84 +152,228 @@ The heart of Webserv’s I/O: a single `poll()` loop multiplexing **listening so * **Timeouts**: enforce idle, header, body, and send deadlines. * **Errors**: generate accurate HTTP error responses, close cleanly. +</details> + ___ -### High-Level Flow - -```mermaid -flowchart TD - %% ========================= - %% Setup - %% ========================= - subgraph Boot[Startup] - A[Load servers from config] - B[Create SocketManager] - C[setupSockets: bind listen register FDs] - A --> B --> C - end - - %% ========================= - %% Main loop - %% ========================= - C --> D{run poll} - D -->|EINTR| Z[Graceful shutdown] --> ZZ[Log shutting down server] - - %% Loop tick - D --> E[handleCgiPollEvents] - E --> F{for each pfd reverse} - - %% Skip CGI pipe FDs - F --> G{CGI pipe FD} - G -- yes --> F - - %% Timeouts (per FD) - G -- no --> T{checkClientTimeouts} - T -- idle or send --> Close[cleanupClientConnectionClose] --> F - T -- header or body --> Halt[disable POLLIN then queue 408] --> H{ERR HUP NVAL} - T -- none --> H - - %% Socket errors - H -- yes --> H1[handlePollError then close] --> F - - %% Reads - H -- no --> I{POLLIN} - I -- no --> O{POLLOUT and responses queued} - - I -- yes --> J{listen FD} - J -- yes --> J1[handleNewConnection] --> F - - J -- no --> Rcv[handleClientData] - Rcv -- queued --> K[enable POLLOUT] --> F - Rcv -- incomplete --> F - - %% Writes - O -- yes --> S[sendResponse] - S -->|file| SF[sendFileResponse] - S -->|raw| SR[sendRawResponse] - SF --> EndSend{response done} - SR --> EndSend - - EndSend -- yes and close --> Close - EndSend -- yes and keep alive --> KA[disable POLLOUT] --> F - EndSend -- not yet --> F - - %% Next tick - F --> D -``` +## HTTP Handling + +This section explains how **Webserv** processes HTTP/1.1 requests end‑to‑end, from bytes on a socket to fully formed responses, and how the server enforces protocol rules, timeouts, and connection reuse. + +<details> +<summary><strong>See Details</strong></summary> + +### Request Lifecycle (High‑Level) + +1. **Accept & Read** + `SocketManager` accepts client connections on non‑blocking sockets and collects incoming bytes. Per‑connection state tracks **read deadlines** (header/body) and **keep‑alive**. + +2. **Parse** + `HttpRequestParser` incrementally parses: + + * **Start line**: method, request‑target (absolute‑path + optional query), HTTP version (HTTP/1.1). + * **Headers**: canonicalizes keys; enforces size limits and folding rules; detects `Connection`, `Host`, `Content-Length`, `Transfer-Encoding`, etc. + * **Body**: supports `Content-Length` and **chunked** transfer decoding. Body size is capped by `client_max_body_size`. + +3. **Route** + `requestRouter` selects a `Server` (host+SNI/server\_name) and the most specific `Location` (longest URI prefix match). It normalizes the filesystem target path and determines whether the request hits **static** content, **autoindex**, **redirect**, **upload**, or **CGI**. + +4. **Dispatch** + Based on method and location rules, it calls `handleGet`, `handlePost`, or `handleDelete`. Unsupported or disallowed → **405** with `Allow` header. + +5. **Build Response** + `responseBuilder` produces status line, headers, and body. It + + * Sets `Content-Type` (MIME by extension), `Content-Length` or `Transfer-Encoding: chunked`, `Connection` (keep‑alive vs close), and error pages. + * Streams file bodies (sendfile/read+write) with backpressure; can fall back to buffered I/O for CGI and dynamic content. + +6. **Send & Reuse** + `SocketManager` writes the response, respecting **write timeouts** and TCP backpressure. If `Connection: keep-alive` and protocol rules allow, the connection stays open for subsequent pipelined requests. + +### Static Files & Autoindex + +* **Static files**: Path is resolved from `root` + URI, protecting against traversal. If an **index** is configured and exists for a directory, it is served. +* **Autoindex**: When enabled and no index present, `generateAutoindex` renders a minimal HTML directory listing. +* **ETag/Last‑Modified** *(optional)*: If enabled, responses include validators; otherwise strong caching is avoided. Range requests are not served unless explicitly implemented. + +### Errors & Edge Cases + +* **400** malformed request, **413** body too large, **414** URI too long, **404/403** missing or forbidden paths, **405** method not allowed. +* **408/504** on header/body/send timeouts. **431** for oversized header sections. +* **5xx** on internal faults, filesystem errors, or CGI failures (see below). + +</details> + +___ + +## Request & CGI Handling + +This section details how POST uploads, multipart forms, and CGI programs are handled, including sandboxing and timeout policy. + +<details> +<summary><strong>See Details</strong></summary> + +### POST Uploads & Multipart + +* **Content dispatch**: `handlePost` inspects `Content-Type` and forwards to specialized handlers. +* **application/x-www-form-urlencoded**: Parsed into key/value pairs. Small payloads are buffered; oversized inputs fail fast with **413**. +* **multipart/form-data**: `handleMultipartForm` parses parts lazily to disk, honoring per‑file and aggregate size limits. Saved files go to the `upload_store` defined on the matched `Location`. +* **application/octet-stream / arbitrary media**: Stored as a single file in `upload_store` with a server‑generated filename when no name is provided. +* **Overwrite policy**: Configurable (e.g., reject on conflict or rename). Errors yield **409** (conflict) or **500** depending on the cause. + +### CGI Execution Model + +* **When CGI triggers**: A request is routed to CGI when the target path matches a configured `cgi_extension` (e.g., `.py`, `.php`) and an interpreter is set, or when the `Location` forces CGI. + +* **Environment**: `handleCgi` constructs a POSIX environment per CGI/1.1: + + * `REQUEST_METHOD`, `QUERY_STRING`, `CONTENT_LENGTH`, `CONTENT_TYPE`, `SCRIPT_FILENAME`, `PATH_INFO`, `SERVER_PROTOCOL`, `SERVER_NAME`, `SERVER_PORT`, `REMOTE_ADDR`, and `HTTP_*` for forwarded headers. + * Working directory is the script directory; stdin is the request body (streamed or buffered based on size). + +* **Process lifecycle**: + + 1. Create pipes for **stdin**/**stdout**, fork, exec interpreter + script. + 2. Parent polls child pipes non‑blocking with **CPU/IO activity watchdogs**. + 3. Enforces **hard timeouts** (startup, read, total runtime). On violation → terminate child. + +* **Output parsing**: CGI writes `Status: 200 OK\r\n`, arbitrary headers, blank line, then body. The server: + + * Parses CGI headers (maps/filters hop‑by‑hop), merges with server headers. + * If `Location` header without body → treat as **redirect** per CGI spec. + * Otherwise body is streamed back to the client. + +* **Failure mapping**: + + * Exec/spawn error → **502 Bad Gateway**. + * Timeout or premature exit → **504 Gateway Timeout**. + * Malformed CGI headers → **502**. + * Script wrote nothing (unexpected EOF) → **502**. + +* **Security & Limits**: + + * Drop privileges/chroot *(if configured)*; never inherit ambient FDs; sanitize environment. + * Enforce **max body size**, **max headers**, **max response size** (protects RAM), and per‑request **open‑file caps**. + +### GET/DELETE Semantics + +* **GET**: Serves static files, autoindex pages, or dispatches to CGI. Conditional GETs (If‑Modified‑Since/If‑None‑Match) may be supported depending on build settings. +* **DELETE**: Removes targeted file from the resolved root when allowed in `methods`. On success → **204 No Content**; on missing/forbidden → **404/403**. + +### Response Builder (Recap) + +* Centralizes status line + headers, error page selection, and body streaming. Ensures `Content-Length` vs `chunked` consistency and keeps **connection semantics** correct across errors and CGI boundaries. </details> ___ -## Flow Overview +## Flow Overview - End‑to‑End Runtime + +This is the complete lifecycle from configuration to bytes on the wire, aligned with the current codebase. + +<details> +<summary><strong>See Details</strong></summary> + +1. **Startup & Configuration** + +* **Tokenizer → ConfigParser → normalizeConfig → validateConfig** + + * Tokenize config, build `Server`/`Location` graphs, apply defaults (client body size, methods, roots, index, error pages), and enforce semantic rules (paths, redirects, methods, CGI mapping). +* **Bootstrap** + + * Instantiate `Server` objects, bind/listen on configured host\:port pairs, pre‑compute route tables and error pages. + +2. **Event Loop (SocketManager)** + +* Single non‑blocking `poll()` loop over listening sockets, client sockets, and CGI pipes. +* Per‑connection state tracks read/write buffers, deadlines (header/body/send), and keep‑alive. +* **Accept** new connections ➜ initialize state. + +3. **Read → Parse (HttpRequestParser)** + +* Accumulate bytes until `"\r\n\r\n"` (header terminator) is found. +* **Start line**: validate method token, request‑target, version. +* **Headers**: normalize keys, reject duplicates where disallowed, check `Content‑Length`/`Transfer‑Encoding` (conflict, format), enforce `Host` on HTTP/1.1, cap header section size. +* **URL/Host routing hint**: derive effective `Url` and matched server affinity; store `Host`, `Query`, `Content‑Length`. +* **Body**: + + * If `Transfer‑Encoding: chunked` ➜ incremental chunk decoding; forbid trailers; enforce `client_max_body_size`. + * Else if `Content‑Length` ➜ wait until full body; enforce size cap; detect pipelined next request beyond the declared length. + * GET/DELETE: treat any extra bytes as pipeline, not body. + +4. **Routing (requestRouter)** -1. **Tokenizer** → breaks input into tokens. -2. **ConfigParser** → builds in‑memory `Config` with `Server` & `Location` objects. -3. **normalizeConfig** → fills missing defaults (sizes, error pages, roots, index, methods). -4. **validateConfig** → applies semantic checks. -5. **Runtime** → validated configuration is passed to the server for request routing. +* **Directory‑slash redirect** when target resolves to a directory but URI lacks trailing `/`. +* **Location selection**: exact match, else longest prefix. +* **Configured redirect (`return` 301/302/307/308)** short‑circuit. +* **Method gate**: -The configuration pipeline guarantees that only syntactically valid, normalized, and semantically correct configurations are accepted. This ensures the server runs with predictable defaults, strong validation, and developer-friendly diagnostics. + * 501 if method not implemented (only GET/POST/DELETE supported). + * 405 if not allowed by `Location`’s `methods`. + +5. **Dispatch (methodsHandler)** + +* **GET** + + * Resolve physical path under `root` (no traversal, no symlinks). + * If directory: + + * If index exists ➜ serve file. + * Else if `autoindex on` ➜ generate HTML listing. + * Else ➜ 403. + * If regular file ➜ serve with MIME type detection. Small files buffered, large files streamed. +* **POST** + + * Preconditions: non‑empty body, size ≤ `client_max_body_size`, `upload_store` configured. + * Determine safe target path under `upload_store` (percent‑decode, canonicalize, reject symlinks, mkdir ‑p). + * Content‑type switch: + + * `multipart/form-data` ➜ stream first file part to disk (boundary parsing, per‑part size cap). + * `application/x‑www‑form‑urlencoded` ➜ parse kv pairs; persist rendered HTML summary. + * Other types ➜ raw body saved as a file. + * 201 on success with minimal HTML confirmation. +* **DELETE** + + * Resolve path; reject directories/symlinks; remove regular file; reply 200 with HTML confirmation. + +6. **CGI (handleCgi) - when location/extension triggers** + +* **Spawn** + + * Write request body to temp file; create output temp file. + * Build `execve` argv (interpreter + script) and CGI/1.1 env (`REQUEST_METHOD`, `QUERY_STRING`, `SCRIPT_FILENAME`, `PATH_INFO`, `SERVER_*`, `HTTP_*`, etc.). + * `fork()` child ➜ `dup2(stdin/out)` to temp fds ➜ `chdir(script dir)` ➜ `execve()`. +* **Supervision** + + * Parent polls pipes/Fds with timeouts; on inactivity/overrun ➜ kill and 504/502. +* **Finalize** + + * Parse output file head for CGI headers (`Status:`, `Content‑Type:`) until `CRLF CRLF`. + * Compute body offset and size, then return a **file‑backed** response pointing at CGI output (no copy), with correct status and content type. + * Ensure temp files are unlinked/cleaned after send. + +7. **Response Building (responseBuilder/HttpResponse)** + +* Build status line + headers; choose reason phrase; select custom error page if configured. +* Set `Content‑Type`, `Content‑Length` (or stream file length) and connection semantics. +* **Keep‑alive policy** + + * HTTP/1.1: keep‑alive by default unless `Connection: close` **or** fatal status (e.g., 400/408/413/500) forces close. + * HTTP/1.0: close by default unless `Connection: keep‑alive`. +* For redirects: set `Location`; body often omitted/minimal. + +8. **Write → Reuse/Close** + +* Non‑blocking writes honor backpressure and send timeouts. +* If `keep‑alive` and no close‑forcing status ➜ retain connection for next pipelined request (parser resumes at leftover bytes). +* Else ➜ close socket and release all per‑connection resources. + +9. **Error Mapping & Hardening** + +* Parser/Router/FS/CGI errors mapped to precise HTTP codes (400/403/404/405/408/411/413/414/415/431/500/501/502/504/505). +* Safeguards: normalized paths, no `..`, symlink denial, header/body caps, per‑request timeouts, upload store confinement, and strict header validation. + +</details> ___ diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index 4d70ee79..dfe525b3 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -35,18 +35,18 @@ Possible candidates: /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:40: warning: unable to resolve reference to 'handleDelete:' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:44: warning: unable to resolve reference to 'generateAutoindex:' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:46: warning: unable to resolve reference to 'handleMultipartForm:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:260: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:262: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:270: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:372: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:383: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:461: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:181: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:184: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:190: warning: unable to resolve reference to 'HttpResponse:' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:35: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command @@ -60,7 +60,6 @@ Possible candidates: /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command -error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:247: warning: Unsupported xml/html tag <port> found /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:249: warning: Unsupported xml/html tag <ip> found /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:253: warning: Unsupported xml/html tag <size> found @@ -80,9 +79,9 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects parameter 'req' parameter 'bodyFd' parameter 'outputFd' -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:181: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:184: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:190: warning: unable to resolve reference to 'HttpResponse:' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:22: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:25: warning: unable to resolve reference to 'escapeUriComponent' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:28: warning: unable to resolve reference to 'formatMTimeUTC' for \ref command @@ -91,27 +90,27 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:31: warning: unable to resolve reference to 'htmlFooter' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:103: warning: unable to resolve reference to 'DirEntry' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:119: warning: unable to resolve reference to 'DirEntry' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:283: warning: unable to resolve reference to 'escapeUriComponent' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:346: warning: unable to resolve reference to 'htmlHeader' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:346: warning: unable to resolve reference to 'renderRow' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:287: warning: unable to resolve reference to 'escapeUriComponent' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:347: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:347: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:372: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:383: warning: unable to resolve reference to 'htmlFooter' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:19: warning: unable to resolve reference to 'detectMimeType' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:21: warning: unable to resolve reference to 'serveFile' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:23: warning: unable to resolve reference to 'processDirectory' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:112: warning: unable to resolve reference to 'detectMimeType' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:409: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:107: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:111: warning: unable to resolve reference to 'detectMimeType' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:125: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:196: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:260: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:262: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:270: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:408: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:410: warning: unable to resolve reference to 'handleRawBody' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:461: warning: unable to resolve reference to 'handleRawBody' for \ref command /home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:24: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command /home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:26: warning: unable to resolve reference to 'findLocation' for \ref command /home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:28: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command @@ -123,7 +122,6 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:43: warning: unable to resolve reference to 'dispatchByMethod' for \ref command -error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__request__handler_ga43204497282094b6dac91b946d5648ba_ga43204497282094b6dac91b946d5648ba_cgraph.dot ! /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:163: warning: unable to resolve reference to 'shouldKeepAlive' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:236: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command @@ -150,28 +148,27 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:179: warning: unable to resolve reference to 'unlinkFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:182: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:188: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:371: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'htmlHeader' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:381: warning: unable to resolve reference to 'renderRow' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlFooter' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:261: warning: unable to resolve reference to 'processDirectory' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:263: warning: unable to resolve reference to 'serveFile' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:271: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:409: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:181: warning: unable to resolve reference to 'unlinkFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:184: warning: unable to resolve reference to 'generateDeleteHtml' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleDelete.cpp:190: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:372: warning: unable to resolve reference to 'listDirectoryEntries' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'htmlHeader' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:382: warning: unable to resolve reference to 'renderRow' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/generateAutoindex.cpp:383: warning: unable to resolve reference to 'htmlFooter' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:107: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:125: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:196: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:260: warning: unable to resolve reference to 'processDirectory' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:262: warning: unable to resolve reference to 'serveFile' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:270: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:408: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command /home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:410: warning: unable to resolve reference to 'handleRawBody' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:459: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleRawBody' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:460: warning: unable to resolve reference to 'handleUrlEncodedForm' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handlePost/handlePost.cpp:461: warning: unable to resolve reference to 'handleRawBody' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:35: warning: unable to resolve reference to 'redirectOnDirectorySlash' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:39: warning: unable to resolve reference to 'redirectOnConfigured' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:41: warning: unable to resolve reference to 'validateRequestMethod' for \ref command /home/toonsa/myProjects/webserv/include/http/requestRouter.hpp:43: warning: unable to resolve reference to 'dispatchByMethod' for \ref command -error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__request__handler_ga43204497282094b6dac91b946d5648ba_ga43204497282094b6dac91b946d5648ba_cgraph.dot ! error: md5 hash does not match for two different runs of /home/toonsa/myProjects/webserv/docs/html/group__url__utils_gabf2641feed820a4fcaf2699a86d1a4f7_gabf2641feed820a4fcaf2699a86d1a4f7_icgraph.dot ! /home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'path' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) /home/toonsa/myProjects/webserv/src/http/handleCgi.cpp:298: warning: argument 'errorCode' of command @param is not found in the argument list of anonymous_namespace{handleCgi.cpp}::prepareCgiTempFiles(CgiProcess &cgi, const HttpRequest &req, int &bodyFd, int &outputFd) @@ -180,9 +177,9 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects parameter 'req' parameter 'bodyFd' parameter 'outputFd' -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:108: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:126: warning: unable to resolve reference to 'HttpResponse:' for \ref command -/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:197: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:107: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:125: warning: unable to resolve reference to 'HttpResponse:' for \ref command +/home/toonsa/myProjects/webserv/src/http/methodsHandler/handleGet/handleGet.cpp:196: warning: unable to resolve reference to 'HttpResponse:' for \ref command /home/toonsa/myProjects/webserv/src/http/requestRouter.cpp:137: warning: Found unknown command '@complexity' /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:247: warning: Unsupported xml/html tag <port> found /home/toonsa/myProjects/webserv/src/config/parser/directive_handler_table.cpp:249: warning: Unsupported xml/html tag <ip> found diff --git a/src/http/handleCgi.cpp b/src/http/handleCgi.cpp index c82ca3c5..f81d30be 100644 --- a/src/http/handleCgi.cpp +++ b/src/http/handleCgi.cpp @@ -6,7 +6,7 @@ /* By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ */ /* +#+#+#+#+#+ +#+ */ /* Created: 2025/05/24 12:23:37 by nlouis #+# #+# */ -/* Updated: 2025/08/19 11:32:52 by nlouis ### ########.fr */ +/* Updated: 2025/08/19 11:35:35 by nlouis ### ########.fr */ /* */ /* ************************************************************************** */ @@ -93,7 +93,7 @@ namespace { * * @return A vector of strings representing the CGI environment, suitable for `execve()`. * - * @ingroup http_component + * @ingroup request_handler */ std::vector<std::string> prepareEnv(const HttpRequest& req, const Server& server, const Location& loc, const std::string& scriptPath) { From 752bba4e9537436e30894e08ce8ee9980b67cbca Mon Sep 17 00:00:00 2001 From: To0nsa <nicolas.lovis@hotmail.fr> Date: Tue, 19 Aug 2025 11:55:11 +0300 Subject: [PATCH 27/32] Adds Makefile for project build configuration --- Makefile.to0nsa | 212 ++++++++++++++++++++++++++++++++++++++++++ docs/doxygen.warnings | 2 + 2 files changed, 214 insertions(+) create mode 100644 Makefile.to0nsa diff --git a/Makefile.to0nsa b/Makefile.to0nsa new file mode 100644 index 00000000..a0110a42 --- /dev/null +++ b/Makefile.to0nsa @@ -0,0 +1,212 @@ +# **************************************************************************** # +# # +# ::: :::::::: # +# Makefile :+: :+: :+: # +# +:+ +:+ +:+ # +# By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ # +# +#+#+#+#+#+ +#+ # +# Created: 2025/08/09 20:53:27 by nlouis #+# #+# # +# Updated: 2025/08/18 20:05:41 by nlouis ### ########.fr # +# # +# **************************************************************************** # + +# Compiler settings +CXX := c++ +CXXFLAGS := -Wall -Wextra -Werror -I include -std=c++20 \ + -O3 -DNDEBUG -flto -march=native +# We use -flto (link time optimization) for better performance. +# -O3 is for optimization, -DNDEBUG disables debug assertions. + +# Executable output +NAME := webserv +BINDIR := bin +TARGET := $(BINDIR)/$(NAME) + +# Object dir (deps live next to objects: .o + .d) +OBJDIR := objs + +# Colors +GREEN := \033[0;32m +CYAN := \033[0;36m +YELLOW := \033[1;33m +RED := \033[0;31m +RESET := \033[0m + +.DEFAULT_GOAL := all + +SRCS_CONFIG := \ + src/config/Config.cpp \ + src/config/normalizeConfig.cpp \ + src/config/validateConfig.cpp + +SRCS_CONFIG_PARSER := \ + src/config/parser/ConfigParseError.cpp \ + src/config/parser/ConfigParser.cpp \ + src/config/parser/directive_handler_table.cpp + +SRCS_CONFIG_TOKENIZER := \ + src/config/tokenizer/Tokenizer.cpp \ + src/config/tokenizer/token.cpp + +SRCS_CORE := \ + src/core/Location.cpp \ + src/core/Server.cpp \ + src/core/main.cpp \ + src/core/server_utils.cpp \ + src/core/runWebserv.cpp + +SRCS_HTTP := \ + src/http/HttpRequest.cpp \ + src/http/HttpRequestParser.cpp \ + src/http/HttpResponse.cpp \ + src/http/handleCgi.cpp \ + src/http/methodsHandler/handleDelete.cpp \ + src/http/methodsHandler/handleGet/generateAutoindex.cpp \ + src/http/methodsHandler/handleGet/handleGet.cpp \ + src/http/methodsHandler/handlePost/handleMultipartForm.cpp \ + src/http/methodsHandler/handlePost/handlePost.cpp \ + src/http/requestRouter.cpp \ + src/http/responseBuilder.cpp + +SRCS_NETWORK := \ + src/network/SocketManager.cpp \ + src/network/SocketManagerRequest.cpp \ + src/network/SocketManagerResponse.cpp \ + src/network/SocketManagerTimeouts.cpp \ + src/network/SocketManagerUtils.cpp + +SRCS_UTILS := \ + src/utils/Logger.cpp \ + src/utils/errorUtils.cpp \ + src/utils/filesystemUtils.cpp \ + src/utils/htmlUtils.cpp \ + src/utils/printInfo.cpp \ + src/utils/stringUtils.cpp \ + src/utils/urlUtils.cpp + +SRCS := \ + $(SRCS_CONFIG) \ + $(SRCS_CONFIG_PARSER) \ + $(SRCS_CONFIG_TOKENIZER) \ + $(SRCS_CORE) \ + $(SRCS_HTTP) \ + $(SRCS_NETWORK) \ + $(SRCS_UTILS) + +HEADERS := \ + include/config/Config.hpp \ + include/config/normalizeConfig.hpp \ + include/config/parser/ConfigParseError.hpp \ + include/config/parser/ConfigParser.hpp \ + include/config/parser/directive_handler_table.hpp \ + include/config/tokenizer/Tokenizer.hpp \ + include/config/tokenizer/token.hpp \ + include/config/validateConfig.hpp \ + include/core/Location.hpp \ + include/core/Server.hpp \ + include/core/server_utils.hpp \ + include/core/runWebserv.hpp \ + include/http/HttpRequest.hpp \ + include/http/HttpRequestParser.hpp \ + include/http/HttpResponse.hpp \ + include/http/Url.hpp \ + include/http/handleCgi.hpp \ + include/http/methodsHandler.hpp \ + include/http/requestRouter.hpp \ + include/http/responseBuilder.hpp \ + include/network/SocketManager.hpp \ + include/utils/Logger.hpp \ + include/utils/errorUtils.hpp \ + include/utils/filesystemUtils.hpp \ + include/utils/htmlUtils.hpp \ + include/utils/printInfo.hpp \ + include/utils/stringUtils.hpp \ + include/utils/urlUtils.hpp + +# ---- Derived files ----------------------------------------------------------- + +# Map each .cpp in $(SRCS) to its corresponding .o in $(OBJDIR). +# Note: patsubst here does NOT perform any filesystem wildcard expansion +# it just does a string substitution on the fixed list in $(SRCS). +# This is safe and deterministic, unlike $(wildcard) or $(shell find ...). +OBJS := $(patsubst src/%.cpp,$(OBJDIR)/%.o,$(SRCS)) +DEPS := $(OBJS:.o=.d) + +# ---- Rules ------------------------------------------------------------------ + +all: $(TARGET) + +$(TARGET): $(OBJS) + @mkdir -p $(BINDIR) + @$(CXX) $(CXXFLAGS) $^ -o $@ + @echo "$(CYAN)🚀 Built executable:$(RESET) $(TARGET)" + +$(OBJDIR)/%.o: src/%.cpp + @mkdir -p $(dir $@) + $(CXX) $(CXXFLAGS) -MMD -MP -c $< -o $@ + @echo "$(GREEN)🛠️ Compiled:$(RESET) $<" + +# Cleaning +clean: + @rm -rf $(OBJDIR) + @echo "$(YELLOW)🧹 Cleaned object and dependency files.$(RESET)" + +fclean: clean + @rm -rf $(TARGET) $(BINDIR) build + @echo "$(YELLOW)🗑️ Completely removed executables, binaries, build/.$(RESET)" + +re: fclean all + +# Python test dependencies +install_test_deps: + @echo "$(CYAN)📦 Creating isolated Python venv for tests...$(RESET)" + @python3 -m venv .venv || { \ + echo "$(RED)python3-venv is missing. Install it: sudo apt-get install -y python3-venv$(RESET)"; \ + exit 1; \ + } + @.venv/bin/python -m pip install --upgrade pip + @.venv/bin/pip install -r requirements-test.txt + +# Tests +test: all install_test_deps + @echo "$(CYAN)🧪 Launching web server in background...$(RESET)" + @./$(TARGET) ./test_webserv/tester/config/tester.conf & echo $$! > .webserv_test.pid + @sleep 1 + @echo "$(CYAN)🧪 Running Python test suite...$(RESET)" + @.venv/bin/python run_test.py || { \ + echo "$(RED)❌ Tests failed.$(RESET)"; \ + kill `cat .webserv_test.pid` >/dev/null 2>&1 || true; \ + rm -f .webserv_test.pid; \ + exit 1; \ + } + @echo "$(CYAN)🧹 Shutting down test server...$(RESET)" + @kill `cat .webserv_test.pid` >/dev/null 2>&1 || true + @rm -f .webserv_test.pid + @echo "$(GREEN)🏆 All tests passed successfully!$(RESET)" + +# Formatting (explicit file lists; no find/globs) +format: + @echo "$(CYAN)🎨 Formatting source files...$(RESET)" + @clang-format -i $(SRCS) $(HEADERS) + +# Help +help: + @echo "$(CYAN)📦 Build Targets:$(RESET)" + @echo " $(GREEN)make$(RESET) → Build the project 🚀" + @echo " $(GREEN)make re$(RESET) → Clean and rebuild everything 🔁" + @echo "" + @echo "$(CYAN)🧪 Test Targets:$(RESET)" + @echo " $(GREEN)make test$(RESET) → Build, install Python deps, run server + Python tests 🧪" + @echo "" + @echo "$(CYAN)🧹 Cleaning Targets:$(RESET)" + @echo " $(GREEN)make clean$(RESET) → Remove object and dependency files 🧹" + @echo " $(GREEN)make fclean$(RESET)→ Remove everything including binaries and build dirs 🗑️" + @echo "" + @echo "$(CYAN)🧹 Code Quality Targets:$(RESET)" + @echo " $(GREEN)make format$(RESET)→ Format all source files 🎨" + @echo "" + @echo "$(CYAN)📚 Other:$(RESET)" + @echo " $(GREEN)make help$(RESET) → Show this help message 📚" + +.PHONY: all clean fclean re test format help install_test_deps +-include $(DEPS) diff --git a/docs/doxygen.warnings b/docs/doxygen.warnings index dfe525b3..42d052e2 100644 --- a/docs/doxygen.warnings +++ b/docs/doxygen.warnings @@ -197,3 +197,5 @@ error: md5 hash does not match for two different runs of /home/toonsa/myProjects /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:273: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:327: warning: unable to resolve reference to 'initializeResponse' for \ref command /home/toonsa/myProjects/webserv/src/http/responseBuilder.cpp:400: warning: unable to resolve reference to 'initializeResponse' for \ref command +/home/toonsa/myProjects/webserv/README.md:294: warning: Found unknown command '\r' +/home/toonsa/myProjects/webserv/README.md:294: warning: Found unknown command '\r' From f80587b3597642b60121563d5adc538aecc574e1 Mon Sep 17 00:00:00 2001 From: To0nsa <nicolas.lovis@hotmail.fr> Date: Tue, 19 Aug 2025 14:22:08 +0300 Subject: [PATCH 28/32] Update makefile name --- Makefile.hive | 210 -------------------------------------------------- 1 file changed, 210 deletions(-) delete mode 100644 Makefile.hive diff --git a/Makefile.hive b/Makefile.hive deleted file mode 100644 index c86c33e3..00000000 --- a/Makefile.hive +++ /dev/null @@ -1,210 +0,0 @@ -# **************************************************************************** # -# # -# ::: :::::::: # -# Makefile.hive :+: :+: :+: # -# +:+ +:+ +:+ # -# By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ # -# +#+#+#+#+#+ +#+ # -# Created: 2025/08/09 20:53:27 by nlouis #+# #+# # -# Updated: 2025/08/18 20:44:34 by nlouis ### ########.fr # -# # -# **************************************************************************** # - -# Compiler settings -CXX := c++ -CXXFLAGS := -Wall -Wextra -Werror -I include -std=c++20 \ - -O3 -DNDEBUG -flto -march=native -# We use -flto (link time optimization) for better performance. -# -O3 is for optimization, -DNDEBUG disables debug assertions. - -# Executable output -NAME := webserv -BINDIR := bin -TARGET := $(BINDIR)/$(NAME) - -# Object dir (deps live next to objects: .o + .d) -OBJDIR := objs - -# Colors -GREEN := \033[0;32m -CYAN := \033[0;36m -YELLOW := \033[1;33m -RED := \033[0;31m -RESET := \033[0m - -.DEFAULT_GOAL := all - -SRCS_CONFIG := \ - src/config/Config.cpp \ - src/config/normalizeConfig.cpp \ - src/config/validateConfig.cpp - -SRCS_CONFIG_PARSER := \ - src/config/parser/ConfigParseError.cpp \ - src/config/parser/ConfigParser.cpp \ - src/config/parser/directive_handler_table.cpp - -SRCS_CONFIG_TOKENIZER := \ - src/config/tokenizer/Tokenizer.cpp \ - src/config/tokenizer/token.cpp - -SRCS_CORE := \ - src/core/Location.cpp \ - src/core/Server.cpp \ - src/core/main.cpp \ - src/core/server_utils.cpp \ - src/core/webserv.cpp - -SRCS_HTTP := \ - src/http/HttpRequest.cpp \ - src/http/HttpRequestParser.cpp \ - src/http/HttpResponse.cpp \ - src/http/handleCgi.cpp \ - src/http/methodsHandler/handleDelete.cpp \ - src/http/methodsHandler/handleGet/generateAutoindex.cpp \ - src/http/methodsHandler/handleGet/handleGet.cpp \ - src/http/methodsHandler/handlePost/handleMultipartForm.cpp \ - src/http/methodsHandler/handlePost/handlePost.cpp \ - src/http/requestRouter.cpp \ - src/http/responseBuilder.cpp - -SRCS_NETWORK := \ - src/network/SocketManager.cpp \ - src/network/SocketManagerRequest.cpp \ - src/network/SocketManagerResponse.cpp \ - src/network/SocketManagerTimeouts.cpp \ - src/network/SocketManagerUtils.cpp - -SRCS_UTILS := \ - src/utils/Logger.cpp \ - src/utils/errorUtils.cpp \ - src/utils/filesystemUtils.cpp \ - src/utils/htmlUtils.cpp \ - src/utils/printInfo.cpp \ - src/utils/stringUtils.cpp \ - src/utils/urlUtils.cpp - -SRCS := \ - $(SRCS_CONFIG) \ - $(SRCS_CONFIG_PARSER) \ - $(SRCS_CONFIG_TOKENIZER) \ - $(SRCS_CORE) \ - $(SRCS_HTTP) \ - $(SRCS_NETWORK) \ - $(SRCS_UTILS) - -HEADERS := \ - include/config/Config.hpp \ - include/config/normalizeConfig.hpp \ - include/config/parser/ConfigParseError.hpp \ - include/config/parser/ConfigParser.hpp \ - include/config/parser/directive_handler_table.hpp \ - include/config/tokenizer/Tokenizer.hpp \ - include/config/tokenizer/token.hpp \ - include/config/validateConfig.hpp \ - include/core/Location.hpp \ - include/core/Server.hpp \ - include/core/server_utils.hpp \ - include/core/webserv.hpp \ - include/http/HttpRequest.hpp \ - include/http/HttpRequestParser.hpp \ - include/http/HttpResponse.hpp \ - include/http/Url.hpp \ - include/http/handleCgi.hpp \ - include/http/methodsHandler.hpp \ - include/http/requestRouter.hpp \ - include/http/responseBuilder.hpp \ - include/network/SocketManager.hpp \ - include/utils/Logger.hpp \ - include/utils/errorUtils.hpp \ - include/utils/filesystemUtils.hpp \ - include/utils/htmlUtils.hpp \ - include/utils/printInfo.hpp \ - include/utils/stringUtils.hpp \ - include/utils/urlUtils.hpp - -# ---- Derived files ----------------------------------------------------------- - -# Map each .cpp in $(SRCS) to its corresponding .o in $(OBJDIR). -# Note: patsubst here does NOT perform any filesystem wildcard expansion -# it just does a string substitution on the fixed list in $(SRCS). -# This is safe and deterministic, unlike $(wildcard) or $(shell find ...). -OBJS := $(patsubst src/%.cpp,$(OBJDIR)/%.o,$(SRCS)) -DEPS := $(OBJS:.o=.d) - -# ---- Rules ------------------------------------------------------------------ - -all: $(TARGET) - -$(TARGET): $(OBJS) - @mkdir -p $(BINDIR) - @$(CXX) $(CXXFLAGS) $^ -o $@ - @echo "$(CYAN)🚀 Built executable:$(RESET) $(TARGET)" - -$(OBJDIR)/%.o: src/%.cpp - @mkdir -p $(dir $@) - $(CXX) $(CXXFLAGS) -MMD -MP -c $< -o $@ - @echo "$(GREEN)🛠️ Compiled:$(RESET) $<" - -# Cleaning -clean: - @rm -rf $(OBJDIR) - @echo "$(YELLOW)🧹 Cleaned object and dependency files.$(RESET)" - -fclean: clean - @rm -rf $(TARGET) $(BINDIR) build - @echo "$(YELLOW)🗑️ Completely removed executables, binaries, build/.$(RESET)" - -re: fclean all - -# Python test dependencies -install_test_deps: - @echo "$(CYAN)📦 Installing Python test dependencies...$(RESET)" - @python3 -m pip install --user --upgrade virtualenv - @python3 -m virtualenv .venv - @.venv/bin/pip install --upgrade pip - @.venv/bin/pip install -r requirements-test.txt - -# Tests -test: all install_test_deps - @echo "$(CYAN)🧪 Launching web server in background...$(RESET)" - @./$(TARGET) ./test_webserv/tester/config/tester.conf & echo $$! > .webserv_test.pid - @sleep 1 - @echo "$(CYAN)🧪 Running Python test suite...$(RESET)" - @.venv/bin/python run_test.py || { \ - echo "$(RED)❌ Tests failed.$(RESET)"; \ - kill `cat .webserv_test.pid` >/dev/null 2>&1 || true; \ - rm -f .webserv_test.pid; \ - exit 1; \ - } - @echo "$(CYAN)🧹 Shutting down test server...$(RESET)" - @kill `cat .webserv_test.pid` >/dev/null 2>&1 || true - @rm -f .webserv_test.pid - @echo "$(GREEN)🏆 All tests passed successfully!$(RESET)" - -# Formatting (explicit file lists; no find/globs) -format: - @echo "$(CYAN)🎨 Formatting source files...$(RESET)" - @clang-format -i $(SRCS) $(HEADERS) - -# Help -help: - @echo "$(CYAN)📦 Build Targets:$(RESET)" - @echo " $(GREEN)make$(RESET) → Build the project 🚀" - @echo " $(GREEN)make re$(RESET) → Clean and rebuild everything 🔁" - @echo "" - @echo "$(CYAN)🧪 Test Targets:$(RESET)" - @echo " $(GREEN)make test$(RESET) → Build, install Python deps, run server + Python tests 🧪" - @echo "" - @echo "$(CYAN)🧹 Cleaning Targets:$(RESET)" - @echo " $(GREEN)make clean$(RESET) → Remove object and dependency files 🧹" - @echo " $(GREEN)make fclean$(RESET)→ Remove everything including binaries and build dirs 🗑️" - @echo "" - @echo "$(CYAN)🧹 Code Quality Targets:$(RESET)" - @echo " $(GREEN)make format$(RESET)→ Format all source files 🎨" - @echo "" - @echo "$(CYAN)📚 Other:$(RESET)" - @echo " $(GREEN)make help$(RESET) → Show this help message 📚" - -.PHONY: all clean fclean re test format help install_test_deps --include $(DEPS) \ No newline at end of file From 31f1c689eccb1c43d2bbf1ffa93dfb343d7a494b Mon Sep 17 00:00:00 2001 From: To0nsa <nicolas.lovis@hotmail.fr> Date: Tue, 19 Aug 2025 14:23:01 +0300 Subject: [PATCH 29/32] Remove step to clean previous HTML output in Doxygen workflow --- .github/workflows/docs.yml | 3 --- 1 file changed, 3 deletions(-) diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml index 2cb46ffc..d17695f7 100644 --- a/.github/workflows/docs.yml +++ b/.github/workflows/docs.yml @@ -17,9 +17,6 @@ jobs: - name: Install Doxygen and Graphviz run: sudo apt-get update && sudo apt-get install -y doxygen graphviz - - name: Clean previous HTML output - run: rm -rf docs/html || true - - name: Generate Doxygen Documentation run: doxygen Doxyfile From 4207d0a949feb0e70d0b48478db587c4224767ae Mon Sep 17 00:00:00 2001 From: To0nsa <nicolas.lovis@hotmail.fr> Date: Tue, 19 Aug 2025 14:25:20 +0300 Subject: [PATCH 30/32] Remove Makefile entry from .gitignore and add Makefile for project build configuration --- .gitignore | 3 - Makefile | 210 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 210 insertions(+), 3 deletions(-) create mode 100644 Makefile diff --git a/.gitignore b/.gitignore index cdc22ccd..b0f069f6 100644 --- a/.gitignore +++ b/.gitignore @@ -57,6 +57,3 @@ test_webserv/**/*.pyo /docs/html/ /docs/doxygen.warnings - -# --- Makefile for To0nsa config --- -Makefile diff --git a/Makefile b/Makefile new file mode 100644 index 00000000..c86c33e3 --- /dev/null +++ b/Makefile @@ -0,0 +1,210 @@ +# **************************************************************************** # +# # +# ::: :::::::: # +# Makefile.hive :+: :+: :+: # +# +:+ +:+ +:+ # +# By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ # +# +#+#+#+#+#+ +#+ # +# Created: 2025/08/09 20:53:27 by nlouis #+# #+# # +# Updated: 2025/08/18 20:44:34 by nlouis ### ########.fr # +# # +# **************************************************************************** # + +# Compiler settings +CXX := c++ +CXXFLAGS := -Wall -Wextra -Werror -I include -std=c++20 \ + -O3 -DNDEBUG -flto -march=native +# We use -flto (link time optimization) for better performance. +# -O3 is for optimization, -DNDEBUG disables debug assertions. + +# Executable output +NAME := webserv +BINDIR := bin +TARGET := $(BINDIR)/$(NAME) + +# Object dir (deps live next to objects: .o + .d) +OBJDIR := objs + +# Colors +GREEN := \033[0;32m +CYAN := \033[0;36m +YELLOW := \033[1;33m +RED := \033[0;31m +RESET := \033[0m + +.DEFAULT_GOAL := all + +SRCS_CONFIG := \ + src/config/Config.cpp \ + src/config/normalizeConfig.cpp \ + src/config/validateConfig.cpp + +SRCS_CONFIG_PARSER := \ + src/config/parser/ConfigParseError.cpp \ + src/config/parser/ConfigParser.cpp \ + src/config/parser/directive_handler_table.cpp + +SRCS_CONFIG_TOKENIZER := \ + src/config/tokenizer/Tokenizer.cpp \ + src/config/tokenizer/token.cpp + +SRCS_CORE := \ + src/core/Location.cpp \ + src/core/Server.cpp \ + src/core/main.cpp \ + src/core/server_utils.cpp \ + src/core/webserv.cpp + +SRCS_HTTP := \ + src/http/HttpRequest.cpp \ + src/http/HttpRequestParser.cpp \ + src/http/HttpResponse.cpp \ + src/http/handleCgi.cpp \ + src/http/methodsHandler/handleDelete.cpp \ + src/http/methodsHandler/handleGet/generateAutoindex.cpp \ + src/http/methodsHandler/handleGet/handleGet.cpp \ + src/http/methodsHandler/handlePost/handleMultipartForm.cpp \ + src/http/methodsHandler/handlePost/handlePost.cpp \ + src/http/requestRouter.cpp \ + src/http/responseBuilder.cpp + +SRCS_NETWORK := \ + src/network/SocketManager.cpp \ + src/network/SocketManagerRequest.cpp \ + src/network/SocketManagerResponse.cpp \ + src/network/SocketManagerTimeouts.cpp \ + src/network/SocketManagerUtils.cpp + +SRCS_UTILS := \ + src/utils/Logger.cpp \ + src/utils/errorUtils.cpp \ + src/utils/filesystemUtils.cpp \ + src/utils/htmlUtils.cpp \ + src/utils/printInfo.cpp \ + src/utils/stringUtils.cpp \ + src/utils/urlUtils.cpp + +SRCS := \ + $(SRCS_CONFIG) \ + $(SRCS_CONFIG_PARSER) \ + $(SRCS_CONFIG_TOKENIZER) \ + $(SRCS_CORE) \ + $(SRCS_HTTP) \ + $(SRCS_NETWORK) \ + $(SRCS_UTILS) + +HEADERS := \ + include/config/Config.hpp \ + include/config/normalizeConfig.hpp \ + include/config/parser/ConfigParseError.hpp \ + include/config/parser/ConfigParser.hpp \ + include/config/parser/directive_handler_table.hpp \ + include/config/tokenizer/Tokenizer.hpp \ + include/config/tokenizer/token.hpp \ + include/config/validateConfig.hpp \ + include/core/Location.hpp \ + include/core/Server.hpp \ + include/core/server_utils.hpp \ + include/core/webserv.hpp \ + include/http/HttpRequest.hpp \ + include/http/HttpRequestParser.hpp \ + include/http/HttpResponse.hpp \ + include/http/Url.hpp \ + include/http/handleCgi.hpp \ + include/http/methodsHandler.hpp \ + include/http/requestRouter.hpp \ + include/http/responseBuilder.hpp \ + include/network/SocketManager.hpp \ + include/utils/Logger.hpp \ + include/utils/errorUtils.hpp \ + include/utils/filesystemUtils.hpp \ + include/utils/htmlUtils.hpp \ + include/utils/printInfo.hpp \ + include/utils/stringUtils.hpp \ + include/utils/urlUtils.hpp + +# ---- Derived files ----------------------------------------------------------- + +# Map each .cpp in $(SRCS) to its corresponding .o in $(OBJDIR). +# Note: patsubst here does NOT perform any filesystem wildcard expansion +# it just does a string substitution on the fixed list in $(SRCS). +# This is safe and deterministic, unlike $(wildcard) or $(shell find ...). +OBJS := $(patsubst src/%.cpp,$(OBJDIR)/%.o,$(SRCS)) +DEPS := $(OBJS:.o=.d) + +# ---- Rules ------------------------------------------------------------------ + +all: $(TARGET) + +$(TARGET): $(OBJS) + @mkdir -p $(BINDIR) + @$(CXX) $(CXXFLAGS) $^ -o $@ + @echo "$(CYAN)🚀 Built executable:$(RESET) $(TARGET)" + +$(OBJDIR)/%.o: src/%.cpp + @mkdir -p $(dir $@) + $(CXX) $(CXXFLAGS) -MMD -MP -c $< -o $@ + @echo "$(GREEN)🛠️ Compiled:$(RESET) $<" + +# Cleaning +clean: + @rm -rf $(OBJDIR) + @echo "$(YELLOW)🧹 Cleaned object and dependency files.$(RESET)" + +fclean: clean + @rm -rf $(TARGET) $(BINDIR) build + @echo "$(YELLOW)🗑️ Completely removed executables, binaries, build/.$(RESET)" + +re: fclean all + +# Python test dependencies +install_test_deps: + @echo "$(CYAN)📦 Installing Python test dependencies...$(RESET)" + @python3 -m pip install --user --upgrade virtualenv + @python3 -m virtualenv .venv + @.venv/bin/pip install --upgrade pip + @.venv/bin/pip install -r requirements-test.txt + +# Tests +test: all install_test_deps + @echo "$(CYAN)🧪 Launching web server in background...$(RESET)" + @./$(TARGET) ./test_webserv/tester/config/tester.conf & echo $$! > .webserv_test.pid + @sleep 1 + @echo "$(CYAN)🧪 Running Python test suite...$(RESET)" + @.venv/bin/python run_test.py || { \ + echo "$(RED)❌ Tests failed.$(RESET)"; \ + kill `cat .webserv_test.pid` >/dev/null 2>&1 || true; \ + rm -f .webserv_test.pid; \ + exit 1; \ + } + @echo "$(CYAN)🧹 Shutting down test server...$(RESET)" + @kill `cat .webserv_test.pid` >/dev/null 2>&1 || true + @rm -f .webserv_test.pid + @echo "$(GREEN)🏆 All tests passed successfully!$(RESET)" + +# Formatting (explicit file lists; no find/globs) +format: + @echo "$(CYAN)🎨 Formatting source files...$(RESET)" + @clang-format -i $(SRCS) $(HEADERS) + +# Help +help: + @echo "$(CYAN)📦 Build Targets:$(RESET)" + @echo " $(GREEN)make$(RESET) → Build the project 🚀" + @echo " $(GREEN)make re$(RESET) → Clean and rebuild everything 🔁" + @echo "" + @echo "$(CYAN)🧪 Test Targets:$(RESET)" + @echo " $(GREEN)make test$(RESET) → Build, install Python deps, run server + Python tests 🧪" + @echo "" + @echo "$(CYAN)🧹 Cleaning Targets:$(RESET)" + @echo " $(GREEN)make clean$(RESET) → Remove object and dependency files 🧹" + @echo " $(GREEN)make fclean$(RESET)→ Remove everything including binaries and build dirs 🗑️" + @echo "" + @echo "$(CYAN)🧹 Code Quality Targets:$(RESET)" + @echo " $(GREEN)make format$(RESET)→ Format all source files 🎨" + @echo "" + @echo "$(CYAN)📚 Other:$(RESET)" + @echo " $(GREEN)make help$(RESET) → Show this help message 📚" + +.PHONY: all clean fclean re test format help install_test_deps +-include $(DEPS) \ No newline at end of file From 8b4c9c659a5e661c0f1ee1da918770a6c0366e02 Mon Sep 17 00:00:00 2001 From: Rychkov Iurii <irychkov@student.hive.fi> Date: Tue, 19 Aug 2025 15:26:04 +0300 Subject: [PATCH 31/32] Fixed compilation error. --- Makefile | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/Makefile b/Makefile index c86c33e3..9877d678 100644 --- a/Makefile +++ b/Makefile @@ -1,12 +1,12 @@ # **************************************************************************** # # # # ::: :::::::: # -# Makefile.hive :+: :+: :+: # +# Makefile :+: :+: :+: # # +:+ +:+ +:+ # -# By: nlouis <nlouis@student.hive.fi> +#+ +:+ +#+ # +# By: irychkov <irychkov@student.hive.fi> +#+ +:+ +#+ # # +#+#+#+#+#+ +#+ # # Created: 2025/08/09 20:53:27 by nlouis #+# #+# # -# Updated: 2025/08/18 20:44:34 by nlouis ### ########.fr # +# Updated: 2025/08/19 15:24:47 by irychkov ### ########.fr # # # # **************************************************************************** # @@ -53,7 +53,7 @@ SRCS_CORE := \ src/core/Server.cpp \ src/core/main.cpp \ src/core/server_utils.cpp \ - src/core/webserv.cpp + src/core/runWebserv.cpp SRCS_HTTP := \ src/http/HttpRequest.cpp \ From de848b62fcbaa915790539988844d645bf879fb9 Mon Sep 17 00:00:00 2001 From: Rychkov Iurii <irychkov@student.hive.fi> Date: Tue, 19 Aug 2025 15:33:06 +0300 Subject: [PATCH 32/32] Switched from requirements to requirements-test. Nicolas removed requirements. --- .github/workflows/test.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml index 805adbc5..36b9928d 100644 --- a/.github/workflows/test.yml +++ b/.github/workflows/test.yml @@ -27,7 +27,7 @@ jobs: run: | sudo apt-get update sudo apt-get install -y python3 python3-pip - pip3 install -r requirements.txt || true + pip3 install -r requirements-test.txt || true - name: Build webserv run: make