[Python-Dev] New re failures on Windows (original) (raw)

Gustavo Niemeyer niemeyer@conectiva.com
Sat, 19 Apr 2003 23:27:23 -0300

I have backed out some changes introduced in _sre.c:2.84 so that its behavior was compliant with the old behavior. More information in bug #672491.

[...]

Running retests test suite === grouping error ('^((a)c)?(ab)$', 'ab', 0, 'g1+"-"+g2+"-"+g3', 'None-None-ab' ) 'None-a-ab' should be 'None-None-ab' """

Hummm.. my changes shouldn't affect this. I'll check that out as well.

testsre is dying with a segfault:

This shouldn't happen with my changes either. I've just backed out some changes, returning to the original code.

""" C:\Code\python\PCbuild>python ../lib/test/testsre.py Running tests on character literals Running tests on sre.search and sre.match sre.match(r'(a)?a','a').lastindex FAILED expected None got result 1 sre.match(r'(a)(b)?b','ab').lastindex FAILED expected 1 got result 2 sre.match(r'(?Pa)(?Pb)?b','ab').lastgroup FAILED expected 'a' got result 'b'

These were the tests I've implemented when the patch was introduced. Unfortunately, the documentation wasn't clear about the expected behavior, and it was implemented wrongly the first time. Now I backed out the changes, and it returned to the original behavior. OTOH, it looks like part of the original problem is still there.

I'll work on it. Greg Chapman also has some ideas about this in the patch #712900.

-- Gustavo Niemeyer

[ 2AAC 7928 0FBF 0299 5EB5 60E2 2253 B29A 6664 3A0C ]