hum-parser

Racket expressions for parsing humdrum files

hum-parser is a repository of data definitions, data structures, and functions for parsing Humdrum files.

hum-parser is written in the Racket language.

hum-parser is able to parse a Humdrum file into records and spines, allowing the user to write these data into new files for use with other programs, or for use directly (i.e. without writing) with other parts of the hum-parser.

The ultimate goal of hum-parser is to map spines on to graph data structures. Such structures enable the user to write new functions that traverse Humdrum data without having to worry about parsing records or spines.

hum-parser also provides command-line tools.

Installing and Using Racket

Racket is a LISP-dialect. It is multi-paradigm, but hum-parser is chiefly functional. Read more on the language here. Consult the documentation here.

To install Racket, follow the Getting Started page of the documentation.

After installing Racket, it is recommended that the user add the path to the bin directory to their shell's environment file (where X\.XX is the version number):

export PATH=/Applications/Racket\ vX\.XX/bin:$PATH

Installing hum-parser

After installing Racket, the hum-parser repository should be cloned in the collects directory:

cd /Applications/Racket vX.XX/collects

git clone https://github.com/bel28kent/hum-parser

Racket automatically searches the collects directory when a module is imported. This simplifies paths:

(require hum-parser)

If one clones hum-parser somewhere else, then the path to the parser will need to be specified relative to the REPL or importing file:

(require "../PATH/hum-parser/main.rkt")

To install the binaries for the command-line tools:

cd /Applications/Racket vX.XX/collects/hum-parser

make all

Note on limitations of hum-parser

Documentation

Future directions

hum-parser may eventually be translated into Racket's sister language Typed Racket to allow for type-checking of user-defined types such as HumdrumFile and GlobalSpine.