Test

.gitattributes

@@ -1,25 +1,35 @@
 # .gitattributes for CLMS_documents
+
 # Always keep the target branch's version of .llm_cache during merges
-.llm_cache merge=ours
+.llm_cache/** merge=ours linguist-generated
 
-# Treat all shell scripts as text with LF endings
+# Text files with LF line endings
 *.sh text eol=lf
+*.py text eol=lf
+*.js text eol=lf
+*.qmd text eol=lf
+*.md text eol=lf
+*.yml text eol=lf
+*.yaml text eol=lf
+*.css text eol=lf
+*.html text eol=lf
+*.json text eol=lf
+*.lua text eol=lf
 
-# Treat images as binary (no diff)
+# Images as binary (no diff, case-insensitive)
 *.png binary
 *.jpg binary
 *.jpeg binary
 *.svg binary
 *.emf binary
 *.wmf binary
-*.JPG binary
-*.PNG binary
-*.SVG binary
+*.gif binary
 
-# Optional: treat docx/pdf as binary
+# Documents as binary
 *.docx binary
 *.pdf binary
 
-# Exclude secrets or temp files from export/archive
+# Exclude from export/archive
 *.env export-ignore
 *.bak export-ignore
+.llm_cache/ export-ignore

.github/non_browsable_doc_map.json

@@ -10,6 +10,11 @@
       "source": "products/CLDS_survey_evaluation_final.qmd",
       "base": "ygwvotluybqwbma6gy8sgaqrfd9ndd8iu078aqn6q10boj016xkhjwegnirwigxb",
       "url": "/ygwvotluybqwbma6gy8sgaqrfd9ndd8iu078aqn6q10boj016xkhjwegnirwigxb.html"
+    },
+    {
+      "source": "Copernicus_Land_Data_Store_CLDS/Survey_Evaluation_v1.qmd",
+      "base": "dr3yejtzxollhygq0lwfgjxf755jvzyly2lemmvpgiqrzf5own2074qpdivug54c",
+      "url": "/dr3yejtzxollhygq0lwfgjxf755jvzyly2lemmvpgiqrzf5own2074qpdivug54c.html"
     }
   ]
 }

.github/scripts/build-docs.sh

@@ -10,12 +10,17 @@ python3 .github/scripts/update_url_mappings.py
 echo "🔄 Grouping documents by category..."
 python3 .github/scripts/group_docs_by_category.py
 
+
 # Change to DOCS directory as it'll be the root of rendered content
 cd DOCS
+
+# Copy quarto configuration and meta files from _meta to DOCS
+cp ../_meta/_quarto*.yml .
+cp -r ../_meta .
+
 # Link assets to origin_DOCS as these files need to be served from rendered content
 ln -s ../assets assets
 
-
 echo "🖼 Render all documents to HTML"
 if [[ -n "$SKIP_DOCX" ]]; then
   echo "   (DOCX generation will be skipped)"
@@ -33,7 +38,9 @@ else
   # _quarto.yml is already the default with headers - no copying needed
 fi
 
-QUARTO_CHROMIUM_HEADLESS_MODE=new quarto render --to html --no-clean
+# Render all files together (changelog filter uses original-filename from YAML headers)
+echo "🔄 Rendering all documents to HTML..."
+QUARTO_CHROMIUM_HEADLESS_MODE=new quarto render --to html --no-clean 
 
 # Backup the correct sitemap as it may be overwritten by next operations
 sleep 5

.github/scripts/filters/inject_contact_info.lua

@@ -0,0 +1,62 @@
+-- inject_contact_info.lua
+-- Quarto filter to automatically inject standard contact information
+-- after the YAML header in documents
+
+local filter_dir = debug.getinfo(1, "S").source:match("@?(.*/)") or ""
+local template_path = filter_dir .. "../../templates/contact_template.md"
+
+-- Function to read the contact template
+function read_contact_template()
+  local file = io.open(template_path, "r")
+  if not file then
+    return nil
+  end
+  local content = file:read("*all")
+  file:close()
+  return content
+end
+
+-- Function to check if contact should be injected
+function should_inject_contact(meta)
+  -- Check if contact injection is explicitly disabled
+  if meta.contact and pandoc.utils.stringify(meta.contact) == "false" then
+    return false
+  end
+
+  -- Default: inject contact unless explicitly disabled
+  return true
+end
+
+-- Main filter function
+function Pandoc(doc)
+  -- Check if we should inject contact info
+  if not should_inject_contact(doc.meta) then
+    return doc
+  end
+
+  -- Read the contact template
+  local contact_content = read_contact_template()
+  if not contact_content then
+    -- If template doesn't exist, skip injection
+    return doc
+  end
+
+  -- Parse the contact template as markdown
+  local contact_doc = pandoc.read(contact_content, "markdown")
+
+  -- Insert contact blocks at the beginning of the document
+  local new_blocks = {}
+
+  -- Add contact blocks first
+  for i, block in pairs(contact_doc.blocks) do
+    table.insert(new_blocks, block)
+  end
+
+  -- Add original document blocks
+  for i, block in pairs(doc.blocks) do
+    table.insert(new_blocks, block)
+  end
+
+  -- Return document with injected contact info
+  return pandoc.Pandoc(new_blocks, doc.meta)
+end

.github/scripts/generate_commit_message.sh

@@ -0,0 +1,49 @@
+#!/bin/bash
+# Generate detailed commit message body from PR changed files
+# Usage: generate_commit_message.sh <pr_number>
+
+set -e
+
+PR_NUMBER=$1
+
+if [ -z "$PR_NUMBER" ]; then
+  echo "Usage: $0 <pr_number>"
+  exit 1
+fi
+
+# Get changed .qmd files only
+changed_files=$(gh pr view "$PR_NUMBER" --json files -q '.files[].path' | grep '^DOCS/.*\.qmd$' || echo "")
+
+if [ -z "$changed_files" ]; then
+  echo "No documentation files (.qmd) changed"
+  exit 0
+fi
+
+# Group files by category subdirectory
+declare -A categories
+while IFS= read -r file; do
+  [ -z "$file" ] && continue
+  # Extract category (first directory after DOCS/)
+  category=$(echo "$file" | sed 's|^DOCS/\([^/]*\)/.*|\1|')
+  # Extract filename without path and .qmd extension
+  filename=$(basename "$file" .qmd)
+
+  # Add to category array
+  if [ -n "${categories[$category]}" ]; then
+    categories[$category]="${categories[$category]}\n- ${filename}"
+  else
+    categories[$category]="- ${filename}"
+  fi
+done <<< "$changed_files"
+
+# Build message body, sorted by category name
+body=""
+for category in $(echo "${!categories[@]}" | tr ' ' '\n' | sort); do
+  # Format category name (replace underscores with spaces, title case)
+  formatted_category=$(echo "$category" | sed 's/_/ /g')
+  body="${body}\n\n${formatted_category}:"
+  body="${body}\n${categories[$category]}"
+done
+
+# Output the body
+echo -e "$body"

.github/scripts/generate_index_all.mjs

@@ -2,7 +2,7 @@ import { readdir, writeFile, access } from "fs/promises";
 import { join, basename } from "path";
 
 const DOCS_DIR = "../DOCS";
-const IGNORED_FOLDERS = new Set(["theme", "templates", "includes", "assets", "_site", ".quarto", "non-browsable"]);
+const IGNORED_FOLDERS = new Set(["assets", "_site", "_meta", ".quarto", "non-browsable"]);
 
 function formatTitle(name) {
   return name.replace(/[-_]/g, " ").replace(/\b\w/g, (c) => c.toUpperCase());

.github/scripts/generate_intros_and_keywords.py

@@ -22,7 +22,15 @@
 ROOT_DIR = (SCRIPT_DIR / "../../").resolve()
 CACHE_DIR = (SCRIPT_DIR / "../../.llm_cache").resolve()
 CACHE_DIR.mkdir(exist_ok=True)
-BLACKLISTED_DIRS = {"templates", "includes", "theme"}
+BLACKLISTED_DIRS = {
+    "templates",
+    "includes",
+    "theme",
+    "_meta",
+    "assets",
+    "_site",
+    ".quarto",
+}
 
 PROMPT = """You are an AI assistant helping to enrich technical documents for the Copernicus Land Monitoring Service (CLMS).