W3C home > Mailing lists > Public > www-validator-cvs@w3.org > December 2009

perl/modules/W3C/LinkChecker/bin checklink,4.168,4.169

From: Ville Skytta via cvs-syncmail <cvsmail@w3.org>
Date: Thu, 03 Dec 2009 20:30:22 +0000
To: www-validator-cvs@w3.org
Message-Id: <E1NGIJi-0004Sb-MB@lionel-hutz.w3.org>
Update of /sources/public/perl/modules/W3C/LinkChecker/bin
In directory hutz:/tmp/cvs-serv17123

Modified Files:
	checklink 
Log Message:
Decode content encodings also when parsing for anchors, thanks to Mark Olson.

Index: checklink
===================================================================
RCS file: /sources/public/perl/modules/W3C/LinkChecker/bin/checklink,v
retrieving revision 4.168
retrieving revision 4.169
diff -u -d -r4.168 -r4.169
--- checklink	4 Nov 2009 18:44:17 -0000	4.168
+++ checklink	3 Dec 2009 20:30:20 -0000	4.169
@@ -1210,6 +1210,32 @@
   return;
 }
 
+##########################################
+# Decode Content-Encodings in a response #
+##########################################
+
+sub decode_content ($)
+{
+  my $response = shift;
+  my $error = undef;
+
+  # @@@TODO: maybe also decode charsets?
+  my $docref = $response->decoded_content(ref => 1, charset => 'none');
+  if (defined($docref)) {
+    $response->content_ref($docref);
+    # Remove Content-Encoding so it won't be decoded again later.
+    $response->remove_header('Content-Encoding')
+  } else {
+    my $ce = $response->header('Content-Encoding');
+    $ce = defined($ce) ? "'$ce'" : 'undefined';
+    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 $ce: '$@'";
+  }
+  return $error;
+}
+
 #######################################
 # Get and parse a resource to process #
 #######################################
@@ -1270,19 +1296,7 @@
     $failed_reason = "Content-Type for <$request_uri> is " .
       (defined($ct) ? "'$ct'" : 'undefined');
   } else {
-    # Pre-decode Content-Encoding.
-    # @@@TODO: maybe also decode charsets?
-    my $docref = $response->decoded_content(ref => 1, charset => 'none');
-    if (defined($docref)) {
-      $response->content_ref($docref);
-      # Remove Content-Encoding so it won't be decoded again later.
-      $response->remove_header('Content-Encoding')
-    } else {
-      my $ce = $response->header('Content-Encoding');
-      $ce = defined($ce) ? "'$ce'" : 'undefined';
-      $ct = defined($ct) ? "'$ct'" : 'undefined';
-      $failed_reason = "Error decoding document at <$request_uri>, Content-Type $ct, Content-Encoding $ce: '$@'";
-    }
+    $failed_reason = decode_content($response);
   }
   if ($failed_reason) {
     # No, there is a problem...
@@ -1850,6 +1864,10 @@
       return;
     }
     # Do it then
+    if (my $error = decode_content($response)) {
+      &hprintf("%s\n.", $error);
+    }
+    # @@@TODO: this isn't the best thing to do if a decode error occurred
     $p = &parse_document($uri, $response->base(), $response, 0, $want_links);
   } else {
     # We already had the information
Received on Thursday, 3 December 2009 20:30:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:55:16 GMT