PyPI, #4392 most downloaded on PyPI

parsley

Parsing and pattern matching made easy.

1.0M downloads/moNew releases in your daily digest, changelog included.

Latest stable

1.3

Releases on this page

Drag across the ruler to compare, or tap a release to open it.

Your coding agent can read these notes before it upgrades. Set up the MCP server →

4 releases since 2012

1.18 Mar 2013Deprecated

The 'spaces' rule has been provided under the alias 'ws'; 'spaces' is now the deprecated name.

The 'spaces' rule has been provided under the alias 'ws'; 'spaces' is now the deprecated name.
Single quotes can now be used around multiple characters to match them consecutively. For example, 'abc' is equivalent 'a' 'b' 'c'.
The 'token' rule is thus deprecated as well, since "token('foo')" is more typing than "ws 'foo'".
Hex escapes can now be used in literals for matching bytes.
'makeGrammar' now takes an 'unwrap' argument, to get a grammar class suitable for subclassing.
'makeGrammar' also takes an 'extends' argument that can be either a regular python class or a wrapped Parsley grammar, for simple subclassing.
'repeat' expressions now work for 0 repetitions properly.
ometa.runtime.TIMING flag added for enabling debug output of grammar load times.
OMeta version 1 parser moved to ometa.compat.
Twine removed.
Various bugfixes around error reporting.
Parsley no longer loads its internal grammars at import time. The 'stage' script in bin/ is used to regenerate the compiled versions after modification.
Experimental tree transformer grammars, with special syntax for destructuring terms and creating string templates, have been added.

1.02 Oct 2012

A new public API has been added in the 'parsley' module.

PyMeta is now Parsley.
A new public API has been added in the 'parsley' module.
Grammars now use the OMeta 2 syntax by default. Compatibility is available via ometa.grammar.OMeta1.
A generic AST-handling library has been added, 'terml'.
Grammars now parse to term trees instead of nested lists. The code generator has accordingly been greatly simplified.
A grammar interpreter has been added, which parses by evaluating the grammar term tree directly instead of generating a class.
A push parser has been added which accepts data in arbitrary chunks, maintaining state while waiting for more input.

20122014201620182020202220242026

20122026

One column per quarter.