On 2012-12-30 17:32, Roy Smith wrote:
In article <mailman.1467.1356885520.29569.python-l...@python.org>,
Vlastimil Brom <vlastimil.b...@gmail.com> wrote:
you may check the new regex implementation for python
http://pypi.python.org/pypi/regex
Wow, I wasn't aware of such an effort.
At first reading, I'm amused by the concept of "strict fuzzy matching".
Those of us with long memories will be confused by the name of the
module, however. To me, "regex" is the old version, and "re" is the new
(http://docs.python.org/release/1.5.1/lib/module-regex.html). Now, it's
going to be "regex" is either the really old, or the really new version,
and "re" is what came in between. I suppose the next iteration after
this will be called "re" again :-)
I'm not sure I like the fuzzy matching stuff. On the one hand, it can
be useful. On the other hand, people already complain about how
difficult it can be to read regexes. Once we add things like
{i<=1,d<=1,s<=1,2i+2d+1s<=4} in the mix, it's going to be total line
noise.
The request for fuzzy matching came after someone saw the TRE library:
http://laurikari.net/tre/
The difference is that the syntax is a bit clearer, IMHO. :-)
--
http://mail.python.org/mailman/listinfo/python-list
