SimpleParse is a parser generator, it takes a modified EBNF
grammar and customises a text-processing engine mxTextTools
to produce a relatively fast table-driven parser.
As of version 3.x, Simpleparse has moved to being unicode-native, with direct support for each of the 3 internal unicode formats that are supported by Python3 (as well as bytes format). However, the effect of the unicode changes is that the parser generator will assume unicode when generating the parser, and you will need to specify the content encoding when parsing/tagging.
So for parsing a latin-1 document provided as raw encoded bytes:
- tag(bytestream, tagtable, 'latin-1')
Only utf-8 encoding is handled among multi-byte encodings, but single-byte encodings should be handled. Note that it is generally going to be more useful to load your document as unicode text and then tag the content, but the byte-stream parsing may be necessary if you have the content already in memory as bytes and can't decode it for some reason.
You can see a full-featured sample parser
in the PyVRML97 package.
See the docs for usage.
pip install SimpleParse
