| Subject: | FB_HTMLCREF and FB_XMLCREF for the UTF-8 decoder |
The following patch to Encode-2.08 seems to fix this.
There was even prepared tests for this :)
diff -ru Encode-2.08/Encode.xs Encode-2.08-hacked/Encode.xs
--- Encode-2.08/Encode.xs 2004-10-24 15:00:33.000000000 +0200
+++ Encode-2.08-hacked/Encode.xs 2004-12-01 18:32:08.818467762 +0100
@@ -331,7 +331,9 @@
break;
}
if (check & (ENCODE_PERLQQ|ENCODE_HTMLCREF|ENCODE_XMLCREF)){
- SV* subchar = newSVpvf("\\x%02" UVXf, (UV)*s);
+ SV* subchar = newSVpvf(check & ENCODE_PERLQQ ? "\\x%02" UVXf :
+ check & ENCODE_HTMLCREF ? "&#%" UVuf ";" :
+ "&#x%" UVxf ";", (UV)*s);
sv_catsv(dst, subchar);
SvREFCNT_dec(subchar);
} else {
diff -ru Encode-2.08/t/fallback.t Encode-2.08-hacked/t/fallback.t
--- Encode-2.08/t/fallback.t 2004-10-22 21:15:03.000000000 +0200
+++ Encode-2.08-hacked/t/fallback.t 2004-12-01 18:26:39.236376790 +0100
@@ -17,7 +17,7 @@
use strict;
#use Test::More qw(no_plan);
-use Test::More tests => 36;
+use Test::More tests => 40;
use Encode q(:all);
my $uo = '';
@@ -137,17 +137,17 @@
is($dst, $ah, "FB_HTMLCREF ascii");
is($src, '', "FB_HTMLCREF residue ascii");
-#$src = $ao;
-#$dst = $utf8->decode($src, FB_HTMLCREF);
-#is($dst, $uh, "FB_HTMLCREF utf8");
-#is($src, '', "FB_HTMLCREF residue utf8");
+$src = $ao;
+$dst = $utf8->decode($src, FB_HTMLCREF);
+is($dst, $uh, "FB_HTMLCREF utf8");
+is($src, '', "FB_HTMLCREF residue utf8");
$src = $uo;
$dst = $ascii->encode($src, FB_XMLCREF);
is($dst, $ax, "FB_XMLCREF ascii");
is($src, '', "FB_XMLCREF residue ascii");
-#$src = $ao;
-#$dst = $utf8->decode($src, FB_XMLCREF);
-#is($dst, $ax, "FB_XMLCREF utf8");
-#is($src, '', "FB_XMLCREF residue utf8");
+$src = $ao;
+$dst = $utf8->decode($src, FB_XMLCREF);
+is($dst, $ax, "FB_XMLCREF utf8");
+is($src, '', "FB_XMLCREF residue utf8");