--- test/html-webhacc/cc.cgi	2007/11/05 09:33:52	1.23
+++ test/html-webhacc/cc.cgi	2007/11/18 11:05:12	1.26
@@ -52,7 +52,6 @@
 
   $| = 0;
   my $input = get_input_document ($http, $dom);
-  my $inner_html_element = $http->get_parameter ('e');
   my $char_length = 0;
   my %time;
 
@@ -62,7 +61,12 @@
 <dt>Request URI</dt>
     <dd><code class="URI" lang="">&lt;<a href="@{[htescape $input->{request_uri}]}">@{[htescape $input->{request_uri}]}</a>&gt;</code></dd>
 <dt>Document URI</dt>
-    <dd><code class="URI" lang="">&lt;<a href="@{[htescape $input->{uri}]}">@{[htescape $input->{uri}]}</a>&gt;</code></dd>
+    <dd><code class="URI" lang="">&lt;<a href="@{[htescape $input->{uri}]}" id=anchor-document-uri>@{[htescape $input->{uri}]}</a>&gt;</code>
+    <script>
+      document.title = '<'
+          + document.getElementById ('anchor-document-uri').href + '> \\u2014 '
+          + document.title;
+    </script></dd>
 ]; # no </dl> yet
   push @nav, ['#document-info' => 'Information'];
 
@@ -74,7 +78,7 @@
     <dd><code class="URI" lang="">&lt;<a href="@{[htescape $input->{base_uri}]}">@{[htescape $input->{base_uri}]}</a>&gt;</code></dd>
 <dt>Internet Media Type</dt>
     <dd><code class="MIME" lang="en">@{[htescape $input->{media_type}]}</code>
-    @{[$input->{media_type_overridden} ? '<em>(overridden)</em>' : '']}</dd>
+    @{[$input->{media_type_overridden} ? '<em>(overridden)</em>' : defined $input->{official_type} ? $input->{media_type} eq $input->{official_type} ? '' : '<em>(sniffed; official type is: <code class=MIME lang=en>'.htescape ($input->{official_type}).'</code>)' : '<em>(sniffed)</em>']}</dd>
 <dt>Character Encoding</dt>
     <dd>@{[defined $input->{charset} ? '<code class="charset" lang="en">'.htescape ($input->{charset}).'</code>' : '(none)']}
     @{[$input->{charset_overridden} ? '<em>(overridden)</em>' : '']}</dd>
@@ -93,7 +97,8 @@
 
   if ($input->{media_type} eq 'text/html') {
     ($doc, $el) = print_syntax_error_html_section ($input, $result);
-    print_source_string_section (\($input->{s}), $input->{charset});
+    print_source_string_section
+        (\($input->{s}), $input->{charset} || $doc->input_encoding);
   } elsif ({
             'text/xml' => 1,
             'application/atom+xml' => 1,
@@ -110,7 +115,7 @@
     print_source_string_section (\($input->{s}), 'utf-8');
   } else {
     ## TODO: Change HTTP status code??
-    print_result_unknown_type_section ($input);
+    print_result_unknown_type_section ($input, $result);
   }
 
   if (defined $doc or defined $el) {
@@ -143,7 +148,8 @@
 </html>
 ];
 
-  for (qw/decode parse parse_xml parse_manifest check check_manifest/) {
+  for (qw/decode parse parse_html parse_xml parse_manifest
+          check check_manifest/) {
     next unless defined $time{$_};
     open my $file, '>>', ".cc-$_.txt" or die ".cc-$_.txt: $!";
     print $file $char_length, "\t", $time{$_}, "\n";
@@ -218,13 +224,7 @@
   
   require Encode;
   require Whatpm::HTML;
-
-  $input->{charset} ||= 'ISO-8859-1'; ## TODO: for now.
   
-  my $time1 = time;
-  my $t = Encode::decode ($input->{charset}, $input->{s});
-  $time{decode} = time - $time1;
-
   print STDOUT qq[
 <div id="parse-errors" class="section">
 <h2>Parse Errors</h2>
@@ -252,16 +252,27 @@
 
   my $doc = $dom->create_document;
   my $el;
-  $time1 = time;
+  my $inner_html_element = $http->get_parameter ('e');
   if (defined $inner_html_element and length $inner_html_element) {
+    $input->{charset} ||= 'windows-1252'; ## TODO: for now.
+    my $time1 = time;
+    my $t = Encode::decode ($input->{charset}, $input->{s});
+    $time{decode} = time - $time1;
+    
     $el = $doc->create_element_ns
         ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
+    $time1 = time;
     Whatpm::HTML->set_inner_html ($el, $t, $onerror);
+    $time{parse} = time - $time1;
   } else {
-    Whatpm::HTML->parse_string ($t => $doc, $onerror);
+    my $time1 = time;
+    Whatpm::HTML->parse_byte_string
+        ($input->{charset}, $input->{s} => $doc, $onerror);
+    $time{parse_html} = time - $time1;
   }
-  $time{parse} = time - $time1;
-
+  $doc->manakai_charset ($input->{official_charset})
+      if defined $input->{official_charset};
+  
   print STDOUT qq[</dl></div>];
 
   return ($doc, $el);
@@ -301,6 +312,8 @@
   my $doc = Message::DOM::XMLParserTemp->parse_byte_stream
       ($fh => $dom, $onerror, charset => $input->{charset});
   $time{parse_xml} = time - $time1;
+  $doc->manakai_charset ($input->{official_charset})
+      if defined $input->{official_charset};
 
   print STDOUT qq[</dl></div>];
 
@@ -759,15 +772,28 @@
   push @nav, ['#result-summary' => 'Result'];
 } # print_result_section
 
-sub print_result_unknown_type_section ($) {
-  my $input = shift;
+sub print_result_unknown_type_section ($$) {
+  my ($input, $result) = @_;
 
+  my $euri = htescape ($input->{uri});
   print STDOUT qq[
-<div id="result-summary" class="section">
-<p><em>Media type <code class="MIME" lang="en">@{[htescape $input->{media_type}]}</code> is not supported!</em></p>
+<div id="parse-errors" class="section">
+<h2>Errors</h2>
+
+<dl>
+<dt class=unsupported><code>&lt;<a href="$euri">$euri</a>&gt;</code></dt>
+    <dd class=unsupported><strong><a href="../error-description#level-u">Not
+        supported</a></strong>:
+    Media type
+    <code class="MIME" lang="en">@{[htescape $input->{media_type}]}</code>
+    is not supported.</dd>
+</dl>
 </div>
 ];
-  push @nav, ['#result-summary' => 'Result'];
+  push @nav, ['#parse-errors' => 'Errors'];
+  add_error (char => {level => 'unsupported'} => $result);
+  add_error (syntax => {level => 'unsupported'} => $result);
+  add_error (structure => {level => 'unsupported'} => $result);
 } # print_result_unknown_type_section
 
 sub print_result_input_error_section ($) {
@@ -873,7 +899,8 @@
 
 sub load_text_catalog ($) {
   my $lang = shift; # MUST be a canonical lang name
-  open my $file, '<', "cc-msg.$lang.txt" or die "$0: cc-msg.$lang.txt: $!";
+  open my $file, '<:utf8', "cc-msg.$lang.txt"
+      or die "$0: cc-msg.$lang.txt: $!";
   while (<$file>) {
     if (s/^([^;]+);([^;]*);//) {
       my ($type, $cls, $msg) = ($1, $2, $_);
@@ -981,12 +1008,10 @@
 
       ## TODO: More strict parsing...
       my $ct = $res->header ('Content-Type');
-      if (defined $ct and $ct =~ m#^([0-9A-Za-z._+-]+/[0-9A-Za-z._+-]+)#) {
-        $r->{media_type} = lc $1;
-      }
       if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) {
         $r->{charset} = lc $1;
         $r->{charset} =~ tr/\\//d;
+        $r->{official_charset} = $r->{charset};
       }
 
       my $input_charset = $http->get_parameter ('charset');
@@ -994,9 +1019,22 @@
         $r->{charset_overridden}
             = (not defined $r->{charset} or $r->{charset} ne $input_charset);
         $r->{charset} = $input_charset;
-      } 
+      }
+
+      ## TODO: Support for HTTP Content-Encoding
 
       $r->{s} = ''.$res->content;
+
+      require Whatpm::ContentType;
+      ($r->{official_type}, $r->{media_type})
+          = Whatpm::ContentType->get_sniffed_type
+              (get_file_head => sub {
+                 return substr $r->{s}, 0, shift;
+               },
+               http_content_type_byte => $ct,
+               has_http_content_encoding =>
+                   defined $res->header ('Content-Encoding'),
+               supported_image_types => {});
     } else {
       $r->{uri} = $res->request->uri;
       $r->{request_uri} = $request_uri;
@@ -1017,7 +1055,18 @@
     $r->{charset} = ''.$http->get_parameter ('_charset_');
     $r->{charset} =~ s/\s+//g;
     $r->{charset} = 'utf-8' if $r->{charset} eq '';
+    $r->{official_charset} = $r->{charset};
     $r->{header_field} = [];
+
+    require Whatpm::ContentType;
+    ($r->{official_type}, $r->{media_type})
+        = Whatpm::ContentType->get_sniffed_type
+            (get_file_head => sub {
+               return substr $r->{s}, 0, shift;
+             },
+             http_content_type_byte => undef,
+             has_http_content_encoding => 0,
+             supported_image_types => {});
   }
 
   my $input_format = $http->get_parameter ('i');
@@ -1034,6 +1083,7 @@
   if ($r->{media_type} eq 'text/xml') {
     unless (defined $r->{charset}) {
       $r->{charset} = 'us-ascii';
+      $r->{official_charset} = $r->{charset};
     } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') {
       $r->{charset_overridden} = 0;
     }
@@ -1084,4 +1134,4 @@
 
 =cut
 
-## $Date: 2007/11/05 09:33:52 $
+## $Date: 2007/11/18 11:05:12 $