Bug#690975: perl: Regression with /i, latin1 chars

[Niko Tyni]

Package: perl
Version: 5.14.2-14
Severity: important
Tags: patch fixed-upstream
Forwarded: http://rt.perl.org/rt3/Public/Bug/Display.html?id=101710

 perl -E 'my $x="br\x{f8}ndby"; $x =~ s/b/X/gi; say $x;' | od -c

gives the correct
0000000   X   r 370   n   d   X   y  \n
on squeeze, but the incorrect
0000000   X   r 370   n   d   b   y  \n
on wheezy.

This was fixed in 5.14.3 with the attached patch. As a regression from
squeeze, I think this is a candidate for wheezy if it's not too late
for that.
-- 
Niko Tyni   ntyni at debian.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-PATCH-perl-101710-Regression-with-i-latin1-chars.patch
Type: text/x-diff
Size: 2169 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/perl-maintainers/attachments/20121019/86ce57f4/attachment.patch>