tree-sitter-bison

Bison grammar for tree-sitter. Created from the official documentation and from examples.

It also supports embedded languages in grammar files (C, C++, D and Java).

Known limitations

This parser requires a semicolon between different grammar rules. While not in official documentation (as far as I know), it seems to be possible in practice to omit these semicolons.

For instance, while theoretically valid, this parser cannot parse this:

first_rule: token1 token1

second_rule: token3 token4

If you want to use this parser, you will need to add semicolons, like this:

first_rule: token1 token1
;

second_rule: token3 token4
;

License

• LGPL v3 or later: grammar.js and the generated parser artifacts (for example the files under src/ such as parser.c, scanner.c, grammar.json, and node-types.json).
• GPL v3 or later: Test corpora derived from GNU projects, specifically the files in test/corpus/ such as complete.txt and bash.txt.
• MIT: The remaining tree-sitter support code and bindings (the contents of bindings/, language-specific build helpers, and other infrastructure files that are not generated from grammar.js). These parts follow the license fields declared in the various package manifests.