/[pub]/test/html-webhacc/cc.cgi
Suika

Diff of /test/html-webhacc/cc.cgi

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.19 by wakaba, Mon Sep 10 11:51:09 2007 UTC revision 1.56 by wakaba, Mon Jul 21 08:39:12 2008 UTC
# Line 1  Line 1 
1  #!/usr/bin/perl  #!/usr/bin/perl
2  use strict;  use strict;
3    use utf8;
4    
5  use lib qw[/home/httpd/html/www/markup/html/whatpm  use lib qw[/home/httpd/html/www/markup/html/whatpm
6             /home/wakaba/work/manakai2/lib];             /home/wakaba/work/manakai2/lib];
7  use CGI::Carp qw[fatalsToBrowser];  use CGI::Carp qw[fatalsToBrowser];
8  use Scalar::Util qw[refaddr];  use Scalar::Util qw[refaddr];
 use Time::HiRes qw/time/;  
9    
10  sub htescape ($) {    require WebHACC::Input;
11    my $s = $_[0];    require WebHACC::Result;
12    $s =~ s/&/&/g;    require WebHACC::Output;
   $s =~ s/</&lt;/g;  
   $s =~ s/>/&gt;/g;  
   $s =~ s/"/&quot;/g;  
   $s =~ s{([\x00-\x09\x0B-\x1F\x7F-\xA0\x{FEFF}\x{FFFC}-\x{FFFF}])}{  
     sprintf '<var>U+%04X</var>', ord $1;  
   }ge;  
   return $s;  
 } # htescape  
13    
14    my $out;
15    
16      require Message::DOM::DOMImplementation;
17      my $dom = Message::DOM::DOMImplementation->new;
18    {
19    use Message::CGI::HTTP;    use Message::CGI::HTTP;
20    my $http = Message::CGI::HTTP->new;    my $http = Message::CGI::HTTP->new;
21    
# Line 26  sub htescape ($) { Line 23  sub htescape ($) {
23      print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400";      print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400";
24      exit;      exit;
25    }    }
26      
   binmode STDOUT, ':utf8';  
   $| = 1;  
   
   require Message::DOM::DOMImplementation;  
   my $dom = Message::DOM::DOMImplementation->new;  
   
27    load_text_catalog ('en'); ## TODO: conneg    load_text_catalog ('en'); ## TODO: conneg
28    
29    my @nav;    $out = WebHACC::Output->new;
30    print STDOUT qq[Content-Type: text/html; charset=utf-8    $out->handle (*STDOUT);
31      $out->set_utf8;
32      $out->set_flush;
33      $out->html (qq[Content-Type: text/html; charset=utf-8
34    
35  <!DOCTYPE html>  <!DOCTYPE html>
36  <html lang="en">  <html lang="en">
# Line 47  sub htescape ($) { Line 41  sub htescape ($) {
41  <body>  <body>
42  <h1><a href="../cc-interface">Web Document Conformance Checker</a>  <h1><a href="../cc-interface">Web Document Conformance Checker</a>
43  (<em>beta</em>)</h1>  (<em>beta</em>)</h1>
44  ];  ]);
45    
   $| = 0;  
46    my $input = get_input_document ($http, $dom);    my $input = get_input_document ($http, $dom);
   my $inner_html_element = $http->get_parameter ('e');  
   my $char_length = 0;  
   my %time;  
   
   print qq[  
 <div id="document-info" class="section">  
 <dl>  
 <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>  
 ]; # no </dl> yet  
   push @nav, ['#document-info' => 'Information'];  
   
 if (defined $input->{s}) {  
   $char_length = length $input->{s};  
   
   print STDOUT qq[  
 <dt>Base URI</dt>  
     <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>  
 <dt>Character Encoding</dt>  
     <dd>@{[defined $input->{charset} ? '<code class="charset" lang="en">'.htescape ($input->{charset}).'</code>' : '(none)']}  
     @{[$input->{charset_overridden} ? '<em>(overridden)</em>' : '']}</dd>  
 <dt>Length</dt>  
     <dd>$char_length byte@{[$char_length == 1 ? '' : 's']}</dd>  
 </dl>  
 </div>  
 ];  
   
   my $result = {};  
   print_http_header_section ($input, $result);  
   
   my $doc;  
   my $el;  
   
   if ($input->{media_type} eq 'text/html') {  
     ($doc, $el) = print_syntax_error_html_section ($input, $result);  
     print_source_string_section (\($input->{s}), $input->{charset});  
   } elsif ({  
             'text/xml' => 1,  
             'application/atom+xml' => 1,  
             'application/rss+xml' => 1,  
             'application/svg+xml' => 1,  
             'application/xhtml+xml' => 1,  
             'application/xml' => 1,  
            }->{$input->{media_type}}) {  
     ($doc, $el) = print_syntax_error_xml_section ($input, $result);  
     print_source_string_section (\($input->{s}), $doc->input_encoding);  
   } else {  
     ## TODO: Change HTTP status code??  
     print_result_unknown_type_section ($input);  
   }  
   
   if (defined $doc or defined $el) {  
     print_structure_dump_section ($doc, $el);  
     my $elements = print_structure_error_section ($doc, $el, $result);  
     print_table_section ($elements->{table}) if @{$elements->{table}};  
     print_id_section ($elements->{id}) if keys %{$elements->{id}};  
     print_term_section ($elements->{term}) if keys %{$elements->{term}};  
     print_class_section ($elements->{class}) if keys %{$elements->{class}};  
   }  
47    
48    print_result_section ($result);    $out->input ($input);
49  } else {    $out->unset_flush;
   print STDOUT qq[</dl></div>];  
   print_result_input_error_section ($input);  
 }  
50    
51    print STDOUT qq[    my $result = WebHACC::Result->new;
52  <ul class="navigation" id="nav-items">    $result->output ($out);
53  ];    $result->{conforming_min} = 1;
54    for (@nav) {    $result->{conforming_max} = 1;
     print STDOUT qq[<li><a href="$_->[0]">$_->[1]</a></li>];  
   }  
   print STDOUT qq[  
 </ul>  
 </body>  
 </html>  
 ];  
   
   for (qw/decode parse parse_xml check/) {  
     next unless defined $time{$_};  
     open my $file, '>>', ".cc-$_.txt" or die ".cc-$_.txt: $!";  
     print $file $char_length, "\t", $time{$_}, "\n";  
   }  
   
 exit;  
   
 sub add_error ($$$) {  
   my ($layer, $err, $result) = @_;  
   if (defined $err->{level}) {  
     if ($err->{level} eq 's') {  
       $result->{$layer}->{should}++;  
       $result->{$layer}->{score_min} -= 2;  
       $result->{conforming_min} = 0;  
     } elsif ($err->{level} eq 'w' or $err->{level} eq 'g') {  
       $result->{$layer}->{warning}++;  
     } elsif ($err->{level} eq 'unsupported') {  
       $result->{$layer}->{unsupported}++;  
       $result->{unsupported} = 1;  
     } else {  
       $result->{$layer}->{must}++;  
       $result->{$layer}->{score_max} -= 2;  
       $result->{$layer}->{score_min} -= 2;  
       $result->{conforming_min} = 0;  
       $result->{conforming_max} = 0;  
     }  
   } else {  
     $result->{$layer}->{must}++;  
     $result->{$layer}->{score_max} -= 2;  
     $result->{$layer}->{score_min} -= 2;  
     $result->{conforming_min} = 0;  
     $result->{conforming_max} = 0;  
   }  
 } # add_error  
   
 sub print_http_header_section ($$) {  
   my ($input, $result) = @_;  
   return unless defined $input->{header_status_code} or  
       defined $input->{header_status_text} or  
       @{$input->{header_field}};  
     
   push @nav, ['#source-header' => 'HTTP Header'];  
   print STDOUT qq[<div id="source-header" class="section">  
 <h2>HTTP Header</h2>  
   
 <p><strong>Note</strong>: Due to the limitation of the  
 network library in use, the content of this section might  
 not be the real header.</p>  
   
 <table><tbody>  
 ];  
   
   if (defined $input->{header_status_code}) {  
     print STDOUT qq[<tr><th scope="row">Status code</th>];  
     print STDOUT qq[<td><code>@{[htescape ($input->{header_status_code})]}</code></td></tr>];  
   }  
   if (defined $input->{header_status_text}) {  
     print STDOUT qq[<tr><th scope="row">Status text</th>];  
     print STDOUT qq[<td><code>@{[htescape ($input->{header_status_text})]}</code></td></tr>];  
   }  
     
   for (@{$input->{header_field}}) {  
     print STDOUT qq[<tr><th scope="row"><code>@{[htescape ($_->[0])]}</code></th>];  
     print STDOUT qq[<td><code>@{[htescape ($_->[1])]}</code></td></tr>];  
   }  
55    
56    print STDOUT qq[</tbody></table></div>];    $out->html ('<script src="../cc-script.js"></script>');
 } # print_http_header_section  
57    
58  sub print_syntax_error_html_section ($$) {    check_and_print ($input => $result => $out);
   my ($input, $result) = @_;  
59        
60    require Encode;    $result->generate_result_section;
   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>  
   
 <dl>];  
   push @nav, ['#parse-errors' => 'Parse Error'];  
   
   my $onerror = sub {  
     my (%opt) = @_;  
     my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level});  
     if ($opt{column} > 0) {  
       print STDOUT qq[<dt class="$cls"><a href="#line-$opt{line}">Line $opt{line}</a> column $opt{column}</dt>\n];  
     } else {  
       $opt{line} = $opt{line} - 1 || 1;  
       print STDOUT qq[<dt class="$cls"><a href="#line-$opt{line}">Line $opt{line}</a></dt>\n];  
     }  
     $type =~ tr/ /-/;  
     $type =~ s/\|/%7C/g;  
     $msg .= qq[ [<a href="../error-description#@{[htescape ($type)]}">Description</a>]];  
     print STDOUT qq[<dd class="$cls">$msg</dd>\n];  
   
     add_error ('syntax', \%opt => $result);  
   };  
   
   my $doc = $dom->create_document;  
   my $el;  
   $time1 = time;  
   if (defined $inner_html_element and length $inner_html_element) {  
     $el = $doc->create_element_ns  
         ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);  
     Whatpm::HTML->set_inner_html ($el, $t, $onerror);  
   } else {  
     Whatpm::HTML->parse_string ($t => $doc, $onerror);  
   }  
   $time{parse} = time - $time1;  
   
   print STDOUT qq[</dl></div>];  
   
   return ($doc, $el);  
 } # print_syntax_error_html_section  
   
 sub print_syntax_error_xml_section ($$) {  
   my ($input, $result) = @_;  
     
   require Message::DOM::XMLParserTemp;  
     
   print STDOUT qq[  
 <div id="parse-errors" class="section">  
 <h2>Parse Errors</h2>  
   
 <dl>];  
   push @nav, ['#parse-errors' => 'Parse Error'];  
   
   my $onerror = sub {  
     my $err = shift;  
     my $line = $err->location->line_number;  
     print STDOUT qq[<dt><a href="#line-$line">Line $line</a> column ];  
     print STDOUT $err->location->column_number, "</dt><dd>";  
     print STDOUT htescape $err->text, "</dd>\n";  
   
     add_error ('syntax', {type => $err->text,  
                 level => [  
                           $err->SEVERITY_FATAL_ERROR => 'm',  
                           $err->SEVERITY_ERROR => 'm',  
                           $err->SEVERITY_WARNING => 's',  
                          ]->[$err->severity]} => $result);  
   
     return 1;  
   };  
   
   my $time1 = time;  
   open my $fh, '<', \($input->{s});  
   my $doc = Message::DOM::XMLParserTemp->parse_byte_stream  
       ($fh => $dom, $onerror, charset => $input->{charset});  
   $time{parse_xml} = time - $time1;  
   
   print STDOUT qq[</dl></div>];  
   
   return ($doc, undef);  
 } # print_syntax_error_xml_section  
   
 sub print_source_string_section ($$) {  
   require Encode;  
   my $enc = Encode::find_encoding ($_[1]); ## TODO: charset name -> Perl name  
   return unless $enc;  
   
   my $s = \($enc->decode (${$_[0]}));  
   my $i = 1;                              
   push @nav, ['#source-string' => 'Source'];  
   print STDOUT qq[<div id="source-string" class="section">  
 <h2>Document Source</h2>  
 <ol lang="">\n];  
   if (length $$s) {  
     while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) {  
       print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n";  
       $i++;  
     }  
     if ($$s =~ /\G([^\x0A]+)/gc) {  
       print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n";  
     }  
   } else {  
     print STDOUT q[<li id="line-1"></li>];  
   }  
   print STDOUT "</ol></div>";  
 } # print_input_string_section  
   
 sub print_document_tree ($) {  
   my $node = shift;  
   my $r = '<ol class="xoxo">';  
   
   my @node = ($node);  
   while (@node) {  
     my $child = shift @node;  
     unless (ref $child) {  
       $r .= $child;  
       next;  
     }  
   
     my $node_id = 'node-'.refaddr $child;  
     my $nt = $child->node_type;  
     if ($nt == $child->ELEMENT_NODE) {  
       my $child_nsuri = $child->namespace_uri;  
       $r .= qq[<li id="$node_id" class="tree-element"><code title="@{[defined $child_nsuri ? $child_nsuri : '']}">] . htescape ($child->tag_name) .  
           '</code>'; ## ISSUE: case  
   
       if ($child->has_attributes) {  
         $r .= '<ul class="attributes">';  
         for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, $_->namespace_uri, 'node-'.refaddr $_] }  
                       @{$child->attributes}) {  
           $r .= qq[<li id="$attr->[3]" class="tree-attribute"><code title="@{[defined $_->[2] ? $_->[2] : '']}">] . htescape ($attr->[0]) . '</code> = '; ## ISSUE: case?  
           $r .= '<q>' . htescape ($attr->[1]) . '</q></li>'; ## TODO: children  
         }  
         $r .= '</ul>';  
       }  
61    
62        if ($child->has_child_nodes) {    $out->nav_list;
         $r .= '<ol class="children">';  
         unshift @node, @{$child->child_nodes}, '</ol></li>';  
       } else {  
         $r .= '</li>';  
       }  
     } elsif ($nt == $child->TEXT_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-text"><q lang="">' . htescape ($child->data) . '</q></li>';  
     } elsif ($nt == $child->CDATA_SECTION_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-cdata"><code>&lt;[CDATA[</code><q lang="">' . htescape ($child->data) . '</q><code>]]&gt;</code></li>';  
     } elsif ($nt == $child->COMMENT_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-comment"><code>&lt;!--</code><q lang="">' . htescape ($child->data) . '</q><code>--&gt;</code></li>';  
     } elsif ($nt == $child->DOCUMENT_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-document">Document';  
       $r .= qq[<ul class="attributes">];  
       $r .= qq[<li>@{[scalar get_text ('manakaiIsHTML:'.($child->manakai_is_html?1:0))]}</li>];  
       $r .= qq[<li>@{[scalar get_text ('manakaiCompatMode:'.$child->manakai_compat_mode)]}</li>];  
       unless ($child->manakai_is_html) {  
         $r .= qq[<li>XML version = <code>@{[htescape ($child->xml_version)]}</code></li>];  
         if (defined $child->xml_encoding) {  
           $r .= qq[<li>XML encoding = <code>@{[htescape ($child->xml_encoding)]}</code></li>];  
         } else {  
           $r .= qq[<li>XML encoding = (null)</li>];  
         }  
         $r .= qq[<li>XML standalone = @{[$child->xml_standalone ? 'true' : 'false']}</li>];  
       }  
       $r .= qq[</ul>];  
       if ($child->has_child_nodes) {  
         $r .= '<ol class="children">';  
         unshift @node, @{$child->child_nodes}, '</ol></li>';  
       }  
     } elsif ($nt == $child->DOCUMENT_TYPE_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-doctype"><code>&lt;!DOCTYPE&gt;</code><ul class="attributes">';  
       $r .= qq[<li class="tree-doctype-name">Name = <q>@{[htescape ($child->name)]}</q></li>];  
       $r .= qq[<li class="tree-doctype-publicid">Public identifier = <q>@{[htescape ($child->public_id)]}</q></li>];  
       $r .= qq[<li class="tree-doctype-systemid">System identifier = <q>@{[htescape ($child->system_id)]}</q></li>];  
       $r .= '</ul></li>';  
     } elsif ($nt == $child->PROCESSING_INSTRUCTION_NODE) {  
       $r .= qq'<li id="$node_id" class="tree-id"><code>&lt;?@{[htescape ($child->target)]}</code> <q>@{[htescape ($child->data)]}</q><code>?&gt;</code></li>';  
     } else {  
       $r .= qq'<li id="$node_id" class="tree-unknown">@{[$child->node_type]} @{[htescape ($child->node_name)]}</li>'; # error  
     }  
   }  
63    
64    $r .= '</ol>';    exit;
65    print STDOUT $r;  }
 } # print_document_tree  
   
 sub print_structure_dump_section ($$) {  
   my ($doc, $el) = @_;  
   
   print STDOUT qq[  
 <div id="document-tree" class="section">  
 <h2>Document Tree</h2>  
 ];  
   push @nav, ['#document-tree' => 'Tree'];  
   
   print_document_tree ($el || $doc);  
   
   print STDOUT qq[</div>];  
 } # print_structure_dump_section  
   
 sub print_structure_error_section ($$$) {  
   my ($doc, $el, $result) = @_;  
   
   print STDOUT qq[<div id="document-errors" class="section">  
 <h2>Document Errors</h2>  
   
 <dl>];  
   push @nav, ['#document-errors' => 'Document Error'];  
   
   require Whatpm::ContentChecker;  
   my $onerror = sub {  
     my %opt = @_;  
     my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}, $opt{node});  
     $type =~ tr/ /-/;  
     $type =~ s/\|/%7C/g;  
     $msg .= qq[ [<a href="../error-description#@{[htescape ($type)]}">Description</a>]];  
     print STDOUT qq[<dt class="$cls">] . get_node_link ($opt{node}) .  
         qq[</dt>\n<dd class="$cls">], $msg, "</dd>\n";  
     add_error ('structure', \%opt => $result);  
   };  
   
   my $elements;  
   my $time1 = time;  
   if ($el) {  
     $elements = Whatpm::ContentChecker->check_element ($el, $onerror);  
   } else {  
     $elements = Whatpm::ContentChecker->check_document ($doc, $onerror);  
   }  
   $time{check} = time - $time1;  
   
   print STDOUT qq[</dl></div>];  
   
   return $elements;  
 } # print_structure_error_section  
66    
67  sub print_table_section ($) {  sub check_and_print ($$$) {
68    my $tables = shift;    my ($input, $result, $out) = @_;
69        my $original_input = $out->input;
70    push @nav, ['#tables' => 'Tables'];    $out->input ($input);
   print STDOUT qq[  
 <div id="tables" class="section">  
 <h2>Tables</h2>  
   
 <!--[if IE]><script type="text/javascript" src="../excanvas.js"></script><![endif]-->  
 <script src="../table-script.js" type="text/javascript"></script>  
 <noscript>  
 <p><em>Structure of tables are visualized here if scripting is enabled.</em></p>  
 </noscript>  
 ];  
     
   require JSON;  
     
   my $i = 0;  
   for my $table_el (@$tables) {  
     $i++;  
     print STDOUT qq[<div class="section" id="table-$i"><h3>] .  
         get_node_link ($table_el) . q[</h3>];  
   
     ## TODO: Make |ContentChecker| return |form_table| result  
     ## so that this script don't have to run the algorithm twice.  
     my $table = Whatpm::HTMLTable->form_table ($table_el);  
       
     for (@{$table->{column_group}}, @{$table->{column}}, $table->{caption}) {  
       next unless $_;  
       delete $_->{element};  
     }  
       
     for (@{$table->{row_group}}) {  
       next unless $_;  
       next unless $_->{element};  
       $_->{type} = $_->{element}->manakai_local_name;  
       delete $_->{element};  
     }  
       
     for (@{$table->{cell}}) {  
       next unless $_;  
       for (@{$_}) {  
         next unless $_;  
         for (@$_) {  
           $_->{id} = refaddr $_->{element} if defined $_->{element};  
           delete $_->{element};  
           $_->{is_header} = $_->{is_header} ? 1 : 0;  
         }  
       }  
     }  
           
     print STDOUT '</div><script type="text/javascript">tableToCanvas (';  
     print STDOUT JSON::objToJson ($table);  
     print STDOUT qq[, document.getElementById ('table-$i'));</script>];  
   }  
     
   print STDOUT qq[</div>];  
 } # print_table_section  
71    
72  sub print_id_section ($) {    $input->generate_info_section ($result);
   my $ids = shift;  
     
   push @nav, ['#identifiers' => 'IDs'];  
   print STDOUT qq[  
 <div id="identifiers" class="section">  
 <h2>Identifiers</h2>  
   
 <dl>  
 ];  
   for my $id (sort {$a cmp $b} keys %$ids) {  
     print STDOUT qq[<dt><code>@{[htescape $id]}</code></dt>];  
     for (@{$ids->{$id}}) {  
       print STDOUT qq[<dd>].get_node_link ($_).qq[</dd>];  
     }  
   }  
   print STDOUT qq[</dl></div>];  
 } # print_id_section  
73    
74  sub print_term_section ($) {    $input->generate_transfer_sections ($result);
   my $terms = shift;  
     
   push @nav, ['#terms' => 'Terms'];  
   print STDOUT qq[  
 <div id="terms" class="section">  
 <h2>Terms</h2>  
   
 <dl>  
 ];  
   for my $term (sort {$a cmp $b} keys %$terms) {  
     print STDOUT qq[<dt>@{[htescape $term]}</dt>];  
     for (@{$terms->{$term}}) {  
       print STDOUT qq[<dd>].get_node_link ($_).qq[</dd>];  
     }  
   }  
   print STDOUT qq[</dl></div>];  
 } # print_term_section  
75    
76  sub print_class_section ($) {    unless (defined $input->{s}) {
77    my $classes = shift;      $result->{conforming_min} = 0;
78          return;
   push @nav, ['#classes' => 'Classes'];  
   print STDOUT qq[  
 <div id="classes" class="section">  
 <h2>Classes</h2>  
   
 <dl>  
 ];  
   for my $class (sort {$a cmp $b} keys %$classes) {  
     print STDOUT qq[<dt><code>@{[htescape $class]}</code></dt>];  
     for (@{$classes->{$class}}) {  
       print STDOUT qq[<dd>].get_node_link ($_).qq[</dd>];  
     }  
79    }    }
   print STDOUT qq[</dl></div>];  
 } # print_class_section  
80    
81  sub print_result_section ($) {    my $checker_class = {
82    my $result = shift;      'text/cache-manifest' => 'WebHACC::Language::CacheManifest',
83        'text/css' => 'WebHACC::Language::CSS',
84        'text/html' => 'WebHACC::Language::HTML',
85        'text/x-webidl' => 'WebHACC::Language::WebIDL',
86    
87        'text/xml' => 'WebHACC::Language::XML',
88        'application/atom+xml' => 'WebHACC::Language::XML',
89        'application/rss+xml' => 'WebHACC::Language::XML',
90        'image/svg+xml' => 'WebHACC::Language::XML',
91        'application/xhtml+xml' => 'WebHACC::Language::XML',
92        'application/xml' => 'WebHACC::Language::XML',
93        ## TODO: Should we make all XML MIME Types fall
94        ## into this category?
95    
96        ## NOTE: This type has different model from normal XML types.
97        'application/rdf+xml' => 'WebHACC::Language::XML',
98      }->{$input->{media_type}} || 'WebHACC::Language::Default';
99    
100      eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@";
101      my $checker = $checker_class->new;
102      $checker->input ($input);
103      $checker->output ($out);
104      $checker->result ($result);
105    
106      ## TODO: A cache manifest MUST be text/cache-manifest
107      ## TODO: WebIDL media type "text/x-webidl"
108    
109      $checker->generate_syntax_error_section;
110      $checker->generate_source_string_section;
111    
112      my @subdoc;
113      $checker->onsubdoc (sub {
114        push @subdoc, shift;
115      });
116    
117      $checker->generate_structure_dump_section;
118      $checker->generate_structure_error_section;
119      $checker->generate_additional_sections;
120    
121    print STDOUT qq[  =pod
 <div id="result-summary" class="section">  
 <h2>Result</h2>];  
   
   if ($result->{unsupported}) {    
     print STDOUT qq[<p class=uncertain id=result-para>The conformance  
         checker cannot decide whether the document is conforming or  
         not, since the document contains one or more unsupported  
         features.</p>];  
   } elsif ($result->{conforming_min}) {  
     print STDOUT qq[<p class=PASS id=result-para>No conformance-error is  
         found in this document.</p>];  
   } elsif ($result->{conforming_max}) {  
     print STDOUT qq[<p class=SEE-RESULT id=result-para>This document  
         is <strong>likely <em>non</em>-conforming</strong>, but in rare case  
         it might be conforming.</p>];  
   } else {  
     print STDOUT qq[<p class=FAIL id=result-para>This document is  
         <strong><em>non</em>-conforming</strong>.</p>];  
   }  
122    
123    print STDOUT qq[<table>    if (defined $doc or defined $el) {
 <colgroup><col><colgroup><col><col><col><colgroup><col>  
 <thead>  
 <tr><th scope=col></th><th scope=col><em class=rfc2119>MUST</em>-level  
 Errors</th><th scope=col><em class=rfc2119>SHOULD</em>-level  
 Errors</th><th scope=col>Warnings</th><th scope=col>Score</th></tr>  
 </thead><tbody>];  
   
   my $must_error = 0;  
   my $should_error = 0;  
   my $warning = 0;  
   my $score_min = 0;  
   my $score_max = 0;  
   my $score_base = 20;  
   for (  
     [Transfer => 'transfer', ''],  
     [Character => 'char', ''],  
     [Syntax => 'syntax', '#parse-errors'],  
     [Structure => 'structure', '#document-errors'],  
   ) {  
     $must_error += ($result->{$_->[1]}->{must} += 0);  
     $should_error += ($result->{$_->[1]}->{should} += 0);  
     $warning += ($result->{$_->[1]}->{warning} += 0);  
     $score_min += ($result->{$_->[1]}->{score_min} += $score_base);  
     $score_max += ($result->{$_->[1]}->{score_max} += $score_base);  
   
     my $uncertain = $result->{$_->[1]}->{unsupported} ? '?' : '';  
     my $label = $_->[0];  
     if ($result->{$_->[1]}->{must} or  
         $result->{$_->[1]}->{should} or  
         $result->{$_->[1]}->{warning} or  
         $result->{$_->[1]}->{unsupported}) {  
       $label = qq[<a href="$_->[2]">$label</a>];  
     }  
124    
125      print STDOUT qq[<tr class="@{[$uncertain ? 'uncertain' : '']}"><th scope=row>$label</th><td class="@{[$result->{$_->[1]}->{must} ? 'FAIL' : '']}">$result->{$_->[1]}->{must}$uncertain</td><td class="@{[$result->{$_->[1]}->{should} ? 'SEE-RESULT' : '']}">$result->{$_->[1]}->{should}$uncertain</td><td>$result->{$_->[1]}->{warning}$uncertain</td>];      print_listing_section ({
126      if ($uncertain) {        id => 'identifiers', label => 'IDs', heading => 'Identifiers',
127        print qq[<td class="@{[$score_max < $score_base ? $score_min < $score_max ? 'FAIL' : 'SEE-RESULT' : '']}">&#x2212;&#x221E;..$result->{$_->[1]}->{score_max}</td>];      }, $input, $elements->{id}) if keys %{$elements->{id}};
128      } elsif ($result->{$_->[1]}->{score_min} != $result->{$_->[1]}->{score_max}) {      print_listing_section ({
129        print qq[<td class="@{[$score_max < $score_base ? 'FAIL' : 'SEE-RESULT']}">$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max} + $score_base</td></tr>];        id => 'terms', label => 'Terms', heading => 'Terms',
130      } else {      }, $input, $elements->{term}) if keys %{$elements->{term}};
131        print qq[<td class="@{[$score_max < $score_base ? 'FAIL' : '']}">$result->{$_->[1]}->{score_min}</td></tr>];      print_listing_section ({
132      }        id => 'classes', label => 'Classes', heading => 'Classes',
133        }, $input, $elements->{class}) if keys %{$elements->{class}};
134      
135        print_rdf_section ($input, $elements->{rdf}) if @{$elements->{rdf}};
136      }
137    
138    =cut
139    
140      my $id_prefix = 0;
141      for my $_subinput (@subdoc) {
142        my $subinput = WebHACC::Input::Subdocument->new (++$id_prefix);
143        $subinput->{$_} = $_subinput->{$_} for keys %$_subinput;
144        $subinput->{base_uri} = $subinput->{container_node}->base_uri
145            unless defined $subinput->{base_uri};
146        $subinput->{parent_input} = $input;
147    
148        $subinput->start_section ($result);
149        check_and_print ($subinput => $result => $out);
150        $subinput->end_section ($result);
151    }    }
152    
153    $score_max += $score_base;    $out->input ($original_input);
154    } # check_and_print
   print STDOUT qq[  
 <tr class=uncertain><th scope=row>Semantics</th><td>0?</td><td>0?</td><td>0?</td><td>&#x2212;&#x221E;..$score_base</td></tr>  
 </tbody>  
 <tfoot><tr class=uncertain><th scope=row>Total</th><td>$must_error?</td><td>$should_error?</td><td>$warning?</td><td><strong>&#x2212;&#x221E;..$score_max</strong></td></tr></tfoot>  
 </table>  
   
 <p><strong>Important</strong>: This conformance checking service  
 is <em>under development</em>.  The result above might be <em>wrong</em>.</p>  
 </div>];  
   push @nav, ['#result-summary' => 'Result'];  
 } # print_result_section  
   
 sub print_result_unknown_type_section ($) {  
   my $input = shift;  
   
   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>  
 ];  
   push @nav, ['#result-summary' => 'Result'];  
 } # print_result_unknown_type_section  
   
 sub print_result_input_error_section ($) {  
   my $input = shift;  
   print STDOUT qq[<div class="section" id="result-summary">  
 <p><em><strong>Input Error</strong>: @{[htescape ($input->{error_status_text})]}</em></p>  
 </div>];  
   push @nav, ['#result-summary' => 'Result'];  
 } # print_Result_input_error_section  
   
 sub get_node_path ($) {  
   my $node = shift;  
   my @r;  
   while (defined $node) {  
     my $rs;  
     if ($node->node_type == 1) {  
       $rs = $node->manakai_local_name;  
       $node = $node->parent_node;  
     } elsif ($node->node_type == 2) {  
       $rs = '@' . $node->manakai_local_name;  
       $node = $node->owner_element;  
     } elsif ($node->node_type == 3) {  
       $rs = '"' . $node->data . '"';  
       $node = $node->parent_node;  
     } elsif ($node->node_type == 9) {  
       @r = ('') unless @r;  
       $rs = '';  
       $node = $node->parent_node;  
     } else {  
       $rs = '#' . $node->node_type;  
       $node = $node->parent_node;  
     }  
     unshift @r, $rs;  
   }  
   return join '/', @r;  
 } # get_node_path  
155    
 sub get_node_link ($) {  
   return qq[<a href="#node-@{[refaddr $_[0]]}">] .  
       htescape (get_node_path ($_[0])) . qq[</a>];  
 } # get_node_link  
156    
157  {  {
158    my $Msg = {};    my $Msg = {};
159    
160  sub load_text_catalog ($) {  sub load_text_catalog ($) {
161    #  my $self = shift;
162    my $lang = shift; # MUST be a canonical lang name    my $lang = shift; # MUST be a canonical lang name
163    open my $file, '<', "cc-msg.$lang.txt" or die "$0: cc-msg.$lang.txt: $!";    open my $file, '<:utf8', "cc-msg.$lang.txt"
164          or die "$0: cc-msg.$lang.txt: $!";
165    while (<$file>) {    while (<$file>) {
166      if (s/^([^;]+);([^;]*);//) {      if (s/^([^;]+);([^;]*);//) {
167        my ($type, $cls, $msg) = ($1, $2, $_);        my ($type, $cls, $msg) = ($1, $2, $_);
# Line 713  sub load_text_catalog ($) { Line 171  sub load_text_catalog ($) {
171    }    }
172  } # load_text_catalog  } # load_text_catalog
173    
174  sub get_text ($) {  sub get_text ($;$$) {
175    #  my $self = shift;
176    my ($type, $level, $node) = @_;    my ($type, $level, $node) = @_;
177    $type = $level . ':' . $type if defined $level;    $type = $level . ':' . $type if defined $level;
178      $level = 'm' unless defined $level;
179    my @arg;    my @arg;
180    {    {
181      if (defined $Msg->{$type}) {      if (defined $Msg->{$type}) {
182        my $msg = $Msg->{$type}->[1];        my $msg = $Msg->{$type}->[1];
183        $msg =~ s{<var>\$([0-9]+)</var>}{        $msg =~ s{<var>\$([0-9]+)</var>}{
184          defined $arg[$1] ? htescape ($arg[$1]) : '(undef)';          defined $arg[$1] ? ($arg[$1]) : '(undef)';
185        }ge;        }ge;                 ##BUG: ^ must be escaped
186        $msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{        $msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{
187          UNIVERSAL::can ($node, 'get_attribute_ns')          UNIVERSAL::can ($node, 'get_attribute_ns')
188              ? htescape ($node->get_attribute_ns (undef, $1)) : ''              ?  ($node->get_attribute_ns (undef, $1)) : ''
189        }ge;        }ge; ## BUG: ^ must be escaped
190        $msg =~ s{<var>{\@}</var>}{        $msg =~ s{<var>{\@}</var>}{        ## BUG: v must be escaped
191          UNIVERSAL::can ($node, 'value') ? htescape ($node->value) : ''          UNIVERSAL::can ($node, 'value') ? ($node->value) : ''
192        }ge;        }ge;
193        $msg =~ s{<var>{local-name}</var>}{        $msg =~ s{<var>{local-name}</var>}{
194          UNIVERSAL::can ($node, 'manakai_local_name')          UNIVERSAL::can ($node, 'manakai_local_name')
195            ? htescape ($node->manakai_local_name) : ''            ? ($node->manakai_local_name) : ''
196        }ge;        }ge;  ## BUG: ^ must be escaped
197        $msg =~ s{<var>{element-local-name}</var>}{        $msg =~ s{<var>{element-local-name}</var>}{
198          (UNIVERSAL::can ($node, 'owner_element') and          (UNIVERSAL::can ($node, 'owner_element') and
199           $node->owner_element)           $node->owner_element)
200            ? htescape ($node->owner_element->manakai_local_name)            ?  ($node->owner_element->manakai_local_name)
201            : ''            : '' ## BUG: ^ must be escaped
202        }ge;        }ge;
203        return ($type, $Msg->{$type}->[0], $msg);        return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg);
204      } elsif ($type =~ s/:([^:]*)$//) {      } elsif ($type =~ s/:([^:]*)$//) {
205        unshift @arg, $1;        unshift @arg, $1;
206        redo;        redo;
207      }      }
208    }    }
209    return ($type, '', htescape ($_[0]));    return ($type, 'level-'.$level, ($_[0]));
210                                     ## BUG: ^ must be escaped
211  } # get_text  } # get_text
212    
213  }  }
# Line 754  sub get_text ($) { Line 215  sub get_text ($) {
215  sub get_input_document ($$) {  sub get_input_document ($$) {
216    my ($http, $dom) = @_;    my ($http, $dom) = @_;
217    
218    my $request_uri = $http->get_parameter ('uri');    require Encode;
219    my $r = {};    my $request_uri = Encode::decode ('utf-8', $http->get_parameter ('uri'));
220      my $r = WebHACC::Input->new;
221    if (defined $request_uri and length $request_uri) {    if (defined $request_uri and length $request_uri) {
222      my $uri = $dom->create_uri_reference ($request_uri);      my $uri = $dom->create_uri_reference ($request_uri);
223      unless ({      unless ({
224               http => 1,               http => 1,
225              }->{lc $uri->uri_scheme}) {              }->{lc $uri->uri_scheme}) {
226        return {uri => $request_uri, request_uri => $request_uri,        $r = WebHACC::Input::Error->new;
227                error_status_text => 'URI scheme not allowed'};        $r->{uri} = $request_uri;
228          $r->{request_uri} = $request_uri;
229          $r->{error_status_text} = 'URL scheme not allowed';
230      }      }
231    
232      require Message::Util::HostPermit;      require Message::Util::HostPermit;
# Line 789  Deny ipv6=0::0/0 Line 253  Deny ipv6=0::0/0
253  Allow host=*  Allow host=*
254  EOH  EOH
255      unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) {      unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) {
256        return {uri => $request_uri, request_uri => $request_uri,        my $r = WebHACC::Input::Error->new;
257                error_status_text => 'Connection to the host is forbidden'};        $r->{uri} = $request_uri;
258          $r->{request_uri} = $request_uri;
259          $r->{error_status_text} = 'Connection to the host is forbidden';
260          return $r;
261      }      }
262    
263      require LWP::UserAgent;      require LWP::UserAgent;
# Line 802  EOH Line 269  EOH
269      $ua->protocols_allowed ([qw/http/]);      $ua->protocols_allowed ([qw/http/]);
270      $ua->max_size (1000_000);      $ua->max_size (1000_000);
271      my $req = HTTP::Request->new (GET => $request_uri);      my $req = HTTP::Request->new (GET => $request_uri);
272        $req->header ('Accept-Encoding' => 'identity, *; q=0');
273      my $res = $ua->request ($req);      my $res = $ua->request ($req);
274      ## TODO: 401 sets |is_success| true.      ## TODO: 401 sets |is_success| true.
275      if ($res->is_success or $http->get_parameter ('error-page')) {      if ($res->is_success or $http->get_parameter ('error-page')) {
# Line 811  EOH Line 279  EOH
279    
280        ## TODO: More strict parsing...        ## TODO: More strict parsing...
281        my $ct = $res->header ('Content-Type');        my $ct = $res->header ('Content-Type');
282        if (defined $ct and $ct =~ m#^([0-9A-Za-z._+-]+/[0-9A-Za-z._+-]+)#) {        if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) {
         $r->{media_type} = lc $1;  
       }  
       if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?(\S+)"?/i) {  
283          $r->{charset} = lc $1;          $r->{charset} = lc $1;
284          $r->{charset} =~ tr/\\//d;          $r->{charset} =~ tr/\\//d;
285            $r->{official_charset} = $r->{charset};
286        }        }
287    
288        my $input_charset = $http->get_parameter ('charset');        my $input_charset = $http->get_parameter ('charset');
# Line 824  EOH Line 290  EOH
290          $r->{charset_overridden}          $r->{charset_overridden}
291              = (not defined $r->{charset} or $r->{charset} ne $input_charset);              = (not defined $r->{charset} or $r->{charset} ne $input_charset);
292          $r->{charset} = $input_charset;          $r->{charset} = $input_charset;
293        }        }
294    
295          ## TODO: Support for HTTP Content-Encoding
296    
297        $r->{s} = ''.$res->content;        $r->{s} = ''.$res->content;
298    
299          require Whatpm::ContentType;
300          ($r->{official_type}, $r->{media_type})
301              = Whatpm::ContentType->get_sniffed_type
302                  (get_file_head => sub {
303                     return substr $r->{s}, 0, shift;
304                   },
305                   http_content_type_byte => $ct,
306                   has_http_content_encoding =>
307                       defined $res->header ('Content-Encoding'),
308                   supported_image_types => {});
309      } else {      } else {
310        $r->{uri} = $res->request->uri;        $r->{uri} = $res->request->uri;
311        $r->{request_uri} = $request_uri;        $r->{request_uri} = $request_uri;
# Line 847  EOH Line 326  EOH
326      $r->{charset} = ''.$http->get_parameter ('_charset_');      $r->{charset} = ''.$http->get_parameter ('_charset_');
327      $r->{charset} =~ s/\s+//g;      $r->{charset} =~ s/\s+//g;
328      $r->{charset} = 'utf-8' if $r->{charset} eq '';      $r->{charset} = 'utf-8' if $r->{charset} eq '';
329        $r->{official_charset} = $r->{charset};
330      $r->{header_field} = [];      $r->{header_field} = [];
331    
332        require Whatpm::ContentType;
333        ($r->{official_type}, $r->{media_type})
334            = Whatpm::ContentType->get_sniffed_type
335                (get_file_head => sub {
336                   return substr $r->{s}, 0, shift;
337                 },
338                 http_content_type_byte => undef,
339                 has_http_content_encoding => 0,
340                 supported_image_types => {});
341    }    }
342    
343    my $input_format = $http->get_parameter ('i');    my $input_format = $http->get_parameter ('i');
# Line 864  EOH Line 354  EOH
354    if ($r->{media_type} eq 'text/xml') {    if ($r->{media_type} eq 'text/xml') {
355      unless (defined $r->{charset}) {      unless (defined $r->{charset}) {
356        $r->{charset} = 'us-ascii';        $r->{charset} = 'us-ascii';
357          $r->{official_charset} = $r->{charset};
358      } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') {      } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') {
359        $r->{charset_overridden} = 0;        $r->{charset_overridden} = 0;
360      }      }
# Line 875  EOH Line 366  EOH
366      return $r;      return $r;
367    }    }
368    
369      $r->{inner_html_element} = $http->get_parameter ('e');
370    
371    return $r;    return $r;
372  } # get_input_document  } # get_input_document
373    
# Line 907  Wakaba <w@suika.fam.cx>. Line 400  Wakaba <w@suika.fam.cx>.
400    
401  =head1 LICENSE  =head1 LICENSE
402    
403  Copyright 2007 Wakaba <w@suika.fam.cx>  Copyright 2007-2008 Wakaba <w@suika.fam.cx>
404    
405  This library is free software; you can redistribute it  This library is free software; you can redistribute it
406  and/or modify it under the same terms as Perl itself.  and/or modify it under the same terms as Perl itself.
Legend:
Removed from v.1.19  
changed lines
  Added in v.1.56
admin@suikawiki.org
 ViewVC Help
Powered by ViewVC 1.1.24