[perl #71740] 5.11.3 breaks Unicode::Normalize

# New Ticket Created by  Claus Färber 
# Please include the string:  [perl #71740]
# in the subject line of all future correspondence about this issue. 
# <URL: http://rt.perl.org/rt3/Ticket/Display.html?id=71740 >


Hi!

perl-5.11.3 breaks Unicode::Normalize: For example, Unicode::Normalize::NFKC("\x{2000}") will produce "\x{20}\x{5}" instead of "\x{20}". (Other code points will obviously suffer from the same problem.)

lib/unicode/mktables produces lib/unicode/Decomposition.pl, containing this (the comment is supposed to indicate the number of code points in the range 2002..2006):
...
| return << 'EOF';
...
| 2000            2002
| 2001            2003
| 2002    2006    <compat> 0020 # [5]
...

cpan/Unicode-Normalize/mkheader then fails to parse this correctly, interpreting "# [5]" as a U+0005.

I would already have written a patch, however, I'm unsure whether this should be fixed in perl or in Unicode::Normalize.

Claus

NB: I've already filed a bug report against Unicode::Normalize at <http://rt.cpan.org/Public/Bug/Display.html?id=53197>.

• [perl #71740] 5.11.3 breaks Unicode::Normalize by Claus Färber