This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
PATCH: [perl 127537] /\W/ regression with UTF-8
authorKarl Williamson <khw@cpan.org>
Sat, 13 Feb 2016 22:35:11 +0000 (15:35 -0700)
committerKarl Williamson <khw@cpan.org>
Fri, 19 Feb 2016 03:44:02 +0000 (20:44 -0700)
commitac33c516140ee213a8a20ada506f97b3a7776ae4
treea86c0a90eb822f61408403d9bd47d9e21e9d0927
parentce54a8b9b1353b2e7e84528e499a996fb0697a95
PATCH: [perl 127537] /\W/ regression with UTF-8

This bug is apparently uncommon in the field, as I was the one who
discovered it.  It requires a UTF-8 pattern containing a complemented
posix class, like \W or \S, in an inverted character class, like
[^\Wfoo] in a pattern that also has a synthetic start class generated by
the regex optimizer for it .

The fix is trivial.
pod/perldelta.pod
regcomp.c
t/re/re_tests