- From: Ville Skytta via cvs-syncmail <cvsmail@w3.org>
- Date: Thu, 04 Feb 2010 19:05:24 +0000
- To: www-validator-cvs@w3.org
Update of /sources/public/perl/modules/W3C/LinkChecker/bin
In directory hutz:/tmp/cvs-serv11613
Modified Files:
checklink
Log Message:
Let LWP decode response charsets.
BOM based decoding works with LWP >= 5.827.
Index: checklink
===================================================================
RCS file: /sources/public/perl/modules/W3C/LinkChecker/bin/checklink,v
retrieving revision 4.177
retrieving revision 4.178
diff -u -d -r4.177 -r4.178
--- checklink 14 Jan 2010 16:55:18 -0000 4.177
+++ checklink 4 Feb 2010 19:05:22 -0000 4.178
@@ -1244,9 +1244,9 @@
my $response = shift;
my $error = undef;
- # @@@TODO: maybe also decode charsets?
- my $docref = $response->decoded_content(ref => 1, charset => 'none');
+ my $docref = $response->decoded_content(ref => 1);
if (defined($docref)) {
+ utf8::encode($$docref);
$response->content_ref($docref);
# Remove Content-Encoding so it won't be decoded again later.
$response->remove_header('Content-Encoding')
@@ -1256,7 +1256,12 @@
my $ct = $response->header('Content-Type');
$ct = defined($ct) ? "'$ct'" : 'undefined';
my $request_uri = $response->request->url;
- $error = "Error decoding document at <$request_uri>, Content-Type $ct, Content-Encoding $ce: '$@'";
+ # content_charset() is available in LWP >= 5.827
+ my $cs = $response->can('content_charset') ?
+ $response->content_charset() : undef;
+ $cs = defined($cs) ? "'$cs'" : 'unknown';
+ $error = "Error decoding document at <$request_uri>, Content-Type $ct, " .
+ "Content-Encoding $ce, content charset $cs: '$@'";
}
return $error;
}
Received on Thursday, 4 February 2010 19:05:25 UTC