Erlang library and packrat parser-generator for parsing expression grammars.
Erlang Makefile
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
priv review tweaks Aug 18, 2014
test use unicode code points instead of hardcoding utf8 chars in parser_test Aug 19, 2014
rebar Add a recent rebar and update the Makefile for dialyzer. Jan 1, 2014
rebar.config Add support and publish. Jul 30, 2015




Neotoma is a packrat parser-generator for Erlang for Parsing Expression Grammars (PEGs).
It consists of a parsing-combinator library with memoization routines, a parser for PEGs,
and a utility to generate parsers from PEGs. It is inspired by treetop, a Ruby library with
similar aims, and parsec, the parser-combinator library for Haskell.

Neotoma is licensed under the MIT License (see LICENSE).


  1. Simple, declarative parsers generated from even simpler grammars.
  2. Fully integrated, single-pass lexical and syntactic analysis (a feature of PEGs).
  3. Packrat-style memoization, boasting parse-time bound linearly to the input size (at the expense of memory usage).
  4. In-place semantic analysis/transformation, supporting single-pass end-to-end in some applications.
  5. Erlang code-generation for the lexical/syntactic analysis piece, with the option of semantic analysis/transformation inline, or in a separate module.
  6. Line/column number tracking for easy resolution of parsing errors.


  1. Clone the repository from github:
    $ git clone git://
  2. Symlink or copy the cloned repository to somewhere in your Erlang code path. $ERLANG_HOME/lib is best.
  3. Build the source:
    $ make


  1. After making sure the library is in your code path, fire up an Erlang shell.
  2. To generate a parser from a PEG, use neotoma:file/1,2 (more detailed documentation pending). For PEG examples, see the extra/ directory in the repository.
1> neotoma:file("extra/arithmetic.peg").
This will place arithmetic.erl in the same directory as the .peg file by default.


Please send pull-requests to ‘seancribbs’ on github. When submitting a patch, eunit tests are strongly encouraged.