This is a patch that implements this; enable by setting the
$p->backquote() attribute method. Will show up in HTML-Parser 3.57.
? HTML-Parser-3.55-threads.diff
? HTML-Parser-3.55.tar.gz
? HTML-Parser-3.56.tar.gz
? HTML-Tagset-3.10
? xxx
Index: Parser.pm
===================================================================
RCS file: /cvsroot/libwww-perl/html-parser/Parser.pm,v
retrieving revision 2.218
diff -u -p -u -r2.218 Parser.pm
--- Parser.pm 6 Feb 2007 18:09:54 -0000 2.218
+++ Parser.pm 6 Feb 2007 19:43:45 -0000
@@ -329,6 +329,14 @@ By default, the C<attr> and C<@attr> arg
entities for attribute values decoded. Enabling this attribute leaves
entities alone.
+=item $p->backquote
+
+=item $p->backquote( $bool )
+
+By default, only ' and " are recognized as quote characters around
+attribute values. MSIE also recognize backquotes for some reason.
+Enabling this attribute provide compatiblity with this behaviour.
+
=item $p->boolean_attribute_value( $val )
This method sets the value reported for boolean attributes inside HTML
Index: Parser.xs
===================================================================
RCS file: /cvsroot/libwww-perl/html-parser/Parser.xs,v
retrieving revision 2.137
diff -u -p -u -r2.137 Parser.xs
--- Parser.xs 12 Jan 2007 10:18:39 -0000 2.137
+++ Parser.xs 6 Feb 2007 19:43:45 -0000
@@ -265,6 +265,7 @@ dup_pstate(pTHX_ PSTATE *pstate, CLONE_P
pstate2->utf8_mode = pstate->utf8_mode;
pstate2->empty_element_tags = pstate->empty_element_tags;
pstate2->xml_pic = pstate->xml_pic;
+ pstate2->backquote = pstate->backquote;
pstate2->bool_attr_val =
SvREFCNT_inc(sv_dup(pstate->bool_attr_val, params));
@@ -447,6 +448,7 @@ strict_comment(pstate,...)
HTML::Parser::utf8_mode = 10
HTML::Parser::empty_element_tags = 11
HTML::Parser::xml_pic = 12
+ HTML::Parser::backquote = 13
PREINIT:
bool *attr;
CODE:
@@ -470,8 +472,9 @@ strict_comment(pstate,...)
#else
case 10: croak("The utf8_mode does not work with this perl; perl-5.8 or better required");
#endif
- case 11: attr = &pstate->empty_element_tags; break;
+ case 11: attr = &pstate->empty_element_tags; break;
case 12: attr = &pstate->xml_pic; break;
+ case 13: attr = &pstate->backquote; break;
default:
croak("Unknown boolean attribute (%d)", ix);
}
Index: hparser.c
===================================================================
RCS file: /cvsroot/libwww-perl/html-parser/hparser.c,v
retrieving revision 2.134
diff -u -p -u -r2.134 hparser.c
--- hparser.c 12 Jan 2007 10:54:06 -0000 2.134
+++ hparser.c 6 Feb 2007 19:43:46 -0000
@@ -455,7 +455,7 @@ report_event(PSTATE* p_state,
if (tokens[i+1].beg) {
char *beg = tokens[i+1].beg;
STRLEN len = tokens[i+1].end - beg;
- if (*beg == '"' || *beg == '\'') {
+ if (*beg == '"' || *beg == '\'' || (*beg == '`' && p_state->backquote)) {
assert(len >= 2 && *beg == beg[len-1]);
beg++; len -= 2;
}
@@ -1166,7 +1166,7 @@ parse_decl(PSTATE* p_state, char *beg, c
if (s == end)
goto PREMATURE;
- if (*s == '"' || *s == '\'') {
+ if (*s == '"' || *s == '\'' || (*s == '`' && p_state->backquote)) {
char *str_beg = s;
s++;
while (s < end && *s != *str_beg)
@@ -1337,7 +1337,7 @@ parse_start(PSTATE* p_state, char *beg,
PUSH_TOKEN(s, s);
break;
}
- if (*s == '"' || *s == '\'') {
+ if (*s == '"' || *s == '\'' || (*s == '`' && p_state->backquote)) {
char *str_beg = s;
s++;
while (s < end && *s != *str_beg)
Index: hparser.h
===================================================================
RCS file: /cvsroot/libwww-perl/html-parser/hparser.h,v
retrieving revision 2.34
diff -u -p -u -r2.34 hparser.h
--- hparser.h 26 Apr 2006 07:01:10 -0000 2.34
+++ hparser.h 6 Feb 2007 19:43:46 -0000
@@ -109,6 +109,7 @@ struct p_state {
bool utf8_mode;
bool empty_element_tags;
bool xml_pic;
+ bool backquote;
/* other configuration stuff */
SV* bool_attr_val;
Index: t/msie-compat.t
===================================================================
RCS file: /cvsroot/libwww-perl/html-parser/t/msie-compat.t,v
retrieving revision 1.7
diff -u -p -u -r1.7 msie-compat.t
--- t/msie-compat.t 9 Jun 2006 07:59:38 -0000 1.7
+++ t/msie-compat.t 6 Feb 2007 19:43:46 -0000
@@ -3,13 +3,13 @@
use strict;
use HTML::Parser;
-use Test::More tests => 2;
+use Test::More tests => 4;
my $TEXT = "";
sub h
{
- my($event, $tagname, $text) = @_;
- for ($event, $tagname, $text) {
+ my($event, $tagname, $text, @attr) = @_;
+ for ($event, $tagname, $text, @attr) {
if (defined) {
s/([\n\r\t])/sprintf "\\%03o", ord($1)/ge;
}
@@ -18,10 +18,10 @@ sub h
}
}
- $TEXT .= "[$event,$tagname,$text]\n";
+ $TEXT .= "[$event,$tagname,$text," . join(":", @attr) . "]\n";
}
-my $p = HTML::Parser->new(default_h => [\&h, "event,tagname,text"]);
+my $p = HTML::Parser->new(default_h => [\&h, "event,tagname,text,[EMAIL PROTECTED]"]);
$p->parse("<a>");
$p->parse("</a f>");
$p->parse("</a 'foo<>' 'bar>' x>");
@@ -33,18 +33,18 @@ $p->parse("<!--comment>text<!--comment><
$p->eof;
is($TEXT, <<'EOT');
-[start_document,<undef>,]
-[start,a,<a>]
-[end,a,</a f>]
-[end,a,</a 'foo<>' 'bar>' x>]
-[end,a,</a "foo<>" "bar>" x>]
-[comment, foo bar,</ foo bar>]
-[comment, "<>" ,</ "<>" >]
-[comment,comment,<!--comment>]
-[text,<undef>,text]
-[comment,comment,<!--comment>]
-[comment,p,<p]
-[end_document,<undef>,]
+[start_document,<undef>,,]
+[start,a,<a>,]
+[end,a,</a f>,]
+[end,a,</a 'foo<>' 'bar>' x>,]
+[end,a,</a "foo<>" "bar>" x>,]
+[comment, foo bar,</ foo bar>,]
+[comment, "<>" ,</ "<>" >,]
+[comment,comment,<!--comment>,]
+[text,<undef>,text,]
+[comment,comment,<!--comment>,]
+[comment,p,<p,]
+[end_document,<undef>,,]
EOT
$TEXT = "";
@@ -52,7 +52,28 @@ $p->parse("<!comment>");
$p->eof;
is($TEXT, <<'EOT');
-[start_document,<undef>,]
-[comment,comment,<!comment>]
-[end_document,<undef>,]
+[start_document,<undef>,,]
+[comment,comment,<!comment>,]
+[end_document,<undef>,,]
+EOT
+
+$TEXT = "";
+$p->parse(q(<a name=`foo bar`>));
+$p->eof;
+
+is($TEXT, <<'EOT');
+[start_document,<undef>,,]
+[start,a,<a name=`foo bar`>,name:`foo:bar`:bar`]
+[end_document,<undef>,,]
+EOT
+
+$p->backquote(1);
+$TEXT = "";
+$p->parse(q(<a name=`foo bar`>));
+$p->eof;
+
+is($TEXT, <<'EOT');
+[start_document,<undef>,,]
+[start,a,<a name=`foo bar`>,name:foo bar]
+[end_document,<undef>,,]
EOT
