RT (3.8.7) and FCGI > 0.68 and actual utf-8 content

Hi,

perl FCGI 0.69 onwards is more picky about the strings it gets.
Thus, if you feed it perl strings with utf-8 you get a complaint about
wide characters in the string.

The following patch takes care of the problem:
— ./lib/RT/Interface/Web.pm.orig 2009-12-11 17:27:20.000000000 +0000
+++ ./lib/RT/Interface/Web.pm
@@ -88,6 +88,7 @@ sub EscapeUTF8 {
$$ref =~ s/)/)/g;
$$ref =~ s/"/"/g;
$$ref =~ s/’/’/g;

  • $$ref = Encode::encode_utf8($$ref);
    }

best regards,
spz
spz@serpens.de (S.P.Zeidler)

Thanks so much! I encountered this problem too, and your patch fixes it for
me.
JasonOn Sun, Apr 4, 2010 at 4:07 PM, S.P.Zeidler spz@serpens.de wrote:

Hi,

perl FCGI 0.69 onwards is more picky about the strings it gets.
Thus, if you feed it perl strings with utf-8 you get a complaint about
wide characters in the string.

The following patch takes care of the problem:
— ./lib/RT/Interface/Web.pm.orig 2009-12-11 17:27:20.000000000 +0000
+++ ./lib/RT/Interface/Web.pm
@@ -88,6 +88,7 @@ sub EscapeUTF8 {
$$ref =~ s/)/)/g;
$$ref =~ s/"/"/g;
$$ref =~ s/’/’/g;

  • $$ref = Encode::encode_utf8($$ref);
    }

best regards,
spz

spz@serpens.de (S.P.Zeidler)


List info:
http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-devel