/[suikacvs]/test/html-webhacc/cc.cgi

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

Parent Directory | Revision Log | View Patch Patch

-revision 1.1 by wakaba,
Wed Jun 27 11:08:03 2007 UTC
+revision 1.22 by wakaba,
Sun Nov  4 09:15:02 2007 UTC
 Line 2
  use strict;
  use lib qw[/home/httpd/html/www/markup/html/whatpm
-            /home/wakaba/work/manakai/lib
+            /home/wakaba/work/manakai2/lib];
-            /home/wakaba/public_html/-temp/wiki/lib];
  use CGI::Carp qw[fatalsToBrowser];
+ use Scalar::Util qw[refaddr];
  use Time::HiRes qw/time/;
- use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module
+ sub htescape ($) {
+   my $s = $_[0];
+   $s =~ s/&/&amp;/g;
+   $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
- my $http = SuikaWiki::Input::HTTP->new;
+   use Message::CGI::HTTP;
+   my $http = Message::CGI::HTTP->new;
- ## TODO: _charset_
+   if ($http->get_meta_variable ('PATH_INFO') ne '/') {
+     print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400";
- my @mode = split m#/#, scalar $http->meta_variable ('PATH_INFO'), -1;
- shift @mode if @mode and $mode[0] == '';
- ## TODO: decode unreserved characters
-   my $s = $http->parameter ('s');
-   if (length $s > 1000_000) {
-     print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long";
      exit;
    }
-   my $char_length = length $s;
-   my %time;
+   binmode STDOUT, ':utf8';
-   my $time1;
+   $| = 1;
-   my $time2;
    require Message::DOM::DOMImplementation;
-   my $dom = Message::DOM::DOMImplementation->____new;
+   my $dom = Message::DOM::DOMImplementation->new;
- #  $| = 1;
+   load_text_catalog ('en'); ## TODO: conneg
+   my @nav;
+   print STDOUT qq[Content-Type: text/html; charset=utf-8
+ <!DOCTYPE html>
+ <html lang="en">
+ <head>
+ <title>Web Document Conformance Checker (BETA)</title>
+ <link rel="stylesheet" href="../cc-style.css" type="text/css">
+ </head>
+ <body>
+ <h1><a href="../cc-interface">Web Document Conformance Checker</a>
+ (<em>beta</em>)</h1>
+ ];
+   $| = 0;
+   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 $result = {conforming_min => 1, conforming_max => 1};
+   print_http_header_section ($input, $result);
    my $doc;
    my $el;
+   my $manifest;
+   if ($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}), $doc->input_encoding);
+   } elsif ($input->{media_type} eq 'text/cache-manifest') {
+ ## TODO: MUST be text/cache-manifest
+     $manifest = print_syntax_error_manifest_section ($input, $result);
+     print_source_string_section (\($input->{s}), 'utf-8');
+   } else {
+     ## TODO: Change HTTP status code??
+     print_result_unknown_type_section ($input);
+   }
+   if (defined $doc or defined $el) {
+     print_structure_dump_dom_section ($doc, $el);
+     my $elements = print_structure_error_dom_section ($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}};
+   } elsif (defined $manifest) {
+     print_structure_dump_manifest_section ($manifest);
+     print_structure_error_manifest_section ($manifest, $result);
+   }
+   print_result_section ($result);
+ } else {
+   print STDOUT qq[</dl></div>];
+   print_result_input_error_section ($input);
+ }
+   print STDOUT qq[
+ <ul class="navigation" id="nav-items">
+ ];
+   for (@nav) {
+     print STDOUT qq[<li><a href="$_->[0]">$_->[1]</a></li>];
+   }
+   print STDOUT qq[
+ </ul>
+ </body>
+ </html>
+ ];
+   for (qw/decode parse 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";
+   }
+ 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>];
+   }
- if (@mode == 3 and $mode[0] eq 'html' and
+   print STDOUT qq[</tbody></table></div>];
-     ($mode[2] eq 'html' or $mode[2] eq 'test')) {
+ } # print_http_header_section
-   print STDOUT "Content-Type: text/plain; charset=utf-8\n\n";
+ sub print_syntax_error_html_section ($$) {
+   my ($input, $result) = @_;
    require Encode;
    require Whatpm::HTML;
-   $time1 = time;
+   $input->{charset} ||= 'ISO-8859-1'; ## TODO: for now.
-   $s = Encode::decode ('utf-8', $s);
-   $time2 = time;
-   $time{decode} = $time2 - $time1;
+   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>
-   print STDOUT "#errors\n";
+ <dl>];
+   push @nav, ['#parse-errors' => 'Parse Error'];
    my $onerror = sub {
      my (%opt) = @_;
-     print STDOUT "$opt{line},$opt{column},$opt{type}\n";
+     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);
    };
-   $doc = $dom->create_document;
+   my $doc = $dom->create_document;
+   my $el;
    $time1 = time;
-   if (length $mode[1]) {
+   if (defined $inner_html_element and length $inner_html_element) {
      $el = $doc->create_element_ns
-         ('http://www.w3.org/1999/xhtml', [undef, $mode[1]]);
+         ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
-     Whatpm::HTML->set_inner_html ($el, $s, $onerror);
+     Whatpm::HTML->set_inner_html ($el, $t, $onerror);
    } else {
-     Whatpm::HTML->parse_string ($s => $doc, $onerror);
+     Whatpm::HTML->parse_string ($t => $doc, $onerror);
    }
-   $time2 = time;
+   $time{parse} = time - $time1;
-   $time{parse} = $time2 - $time1;
+   print STDOUT qq[</dl></div>];
-   print "#document\n";
+   return ($doc, $el);
+ } # print_syntax_error_html_section
-   my $out;
-   if ($mode[2] eq 'html') {
-     $time1 = time;
-     $out = Whatpm::HTML->get_inner_html ($el || $doc);
-     $time2 = time;
-     $time{serialize_html} = $time2 - $time1;
-   } else { # test
-     $time1 = time;
-     $out = test_serialize ($el || $doc);
-     $time2 = time;
-     $time{serialize_test} = $time2 - $time1;
-   }
-   print STDOUT Encode::encode ('utf-8', $$out);
-   print STDOUT "\n";
- } elsif (@mode == 3 and $mode[0] eq 'xhtml' and
-          ($mode[2] eq 'html' or $mode[2] eq 'test')) {
-   print STDOUT "Content-Type: text/plain; charset=utf-8\n\n";
+ sub print_syntax_error_xml_section ($$) {
+   my ($input, $result) = @_;
    require Message::DOM::XMLParserTemp;
-   print STDOUT "#errors\n";
+   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;
-     print STDOUT $err->location->line_number, ",";
+     my $line = $err->location->line_number;
-     print STDOUT $err->location->column_number, ",";
+     print STDOUT qq[<dt><a href="#line-$line">Line $line</a> column ];
-     print STDOUT $err->text, "\n";
+     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;
    };
-   open my $fh, '<', \$s;
    my $time1 = time;
-   $doc = Message::DOM::XMLParserTemp->parse_byte_stream
+   open my $fh, '<', \($input->{s});
-       ($fh => $dom, $onerror, charset => 'utf-8');
+   my $doc = Message::DOM::XMLParserTemp->parse_byte_stream
-   my $time2 = time;
+       ($fh => $dom, $onerror, charset => $input->{charset});
-   $time{parse_xml} = $time2 - $time1;
+   $time{parse_xml} = time - $time1;
-   print "#document\n";
-   my $out;
-   if ($mode[2] eq 'html') {
-     ## TODO: Use XHTML serializer
-     #$out = Whatpm::HTML->get_inner_html ($doc);
-   } else { # test
-     $time1 = time;
-     $out = test_serialize ($doc);
-     $time2 = time;
-     $time{serialize_test} = $time2 - $time1;
-   }
-   print STDOUT Encode::encode ('utf-8', $$out);
-   print STDOUT "\n";
- } else {
-   print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n404";
-   exit;
- }
-   if ($http->parameter ('dom5')) {
+   print STDOUT qq[</dl></div>];
-     require Whatpm::ContentChecker;
-     my $onerror = sub {
-       my %opt = @_;
-       print STDOUT get_node_path ($opt{node}) . ';' . $opt{type} . "\n";
-     };
-     print STDOUT "#domerrors\n";
-     $time1 = time;
-     if ($el) {
-       Whatpm::ContentChecker->check_element ($el, $onerror);
-     } else {
-       Whatpm::ContentChecker->check_document ($doc, $onerror);
-     }
-     $time2 = time;
-     $time{check} = $time2 - $time1;
-   }
-   print STDOUT "#log\n";
-   for (qw/decode parse parse_xml serialize_html serialize_xml serialize_test
-           check/) {
-     next unless defined $time{$_};
-     print STDOUT {
-       decode => 'bytes->chars',
-       parse => 'html5(chars)->dom5',
-       parse_xml => 'xml1(chars)->dom5',
-       serialize_html => 'dom5->html5(char)',
-       serialize_xml => 'dom5->xml1(char)',
-       serialize_test => 'dom5->test(char)',
-       check => 'dom5 check',
-     }->{$_};
-     print STDOUT "\t", $time{$_}, "s\n";
-     open my $file, '>>', ".manakai-$_.txt" or die ".manakai-$_.txt: $!";
-     print $file $char_length, "\t", $time{$_}, "\n";
-   }
- exit;
+   return ($doc, undef);
+ } # print_syntax_error_xml_section
+ sub print_syntax_error_manifest_section ($$) {
+   my ($input, $result) = @_;
+   require Whatpm::CacheManifest;
+   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}, $opt{level});
+     print STDOUT qq[<dt class="$cls">], get_error_label (\%opt), qq[</dt>];
+     $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 $time1 = time;
+   my $manifest = Whatpm::CacheManifest->parse_byte_string
+       ($input->{s}, $input->{uri}, $input->{base_uri}, $onerror);
+   $time{parse_manifest} = time - $time1;
+   print STDOUT qq[</dl></div>];
+   return $manifest;
+ } # print_syntax_error_manifest_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 test_serialize ($) {
+ sub print_document_tree ($) {
    my $node = shift;
-   my $r = '';
+   my $r = '<ol class="xoxo">';
-   my @node = map { [$_, ''] } @{$node->child_nodes};
+   my @node = ($node);
    while (@node) {
      my $child = shift @node;
-     my $nt = $child->[0]->node_type;
+     unless (ref $child) {
-     if ($nt == $child->[0]->ELEMENT_NODE) {
+       $r .= $child;
-       $r .= '| ' . $child->[1] . '<' . $child->[0]->tag_name . ">\x0A"; ## ISSUE: case?
+       next;
+     }
-       for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value] }
-                     @{$child->[0]->attributes}) {
+     my $node_id = 'node-'.refaddr $child;
-         $r .= '| ' . $child->[1] . '  ' . $attr->[0] . '="'; ## ISSUE: case?
+     my $nt = $child->node_type;
-         $r .= $attr->[1] . '"' . "\x0A";
+     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) .
-       unshift @node,
+           '</code>'; ## ISSUE: case
-         map { [$_, $child->[1] . '  '] } @{$child->[0]->child_nodes};
-     } elsif ($nt == $child->[0]->TEXT_NODE) {
+       if ($child->has_attributes) {
-       $r .= '| ' . $child->[1] . '"' . $child->[0]->data . '"' . "\x0A";
+         $r .= '<ul class="attributes">';
-     } elsif ($nt == $child->[0]->CDATA_SECTION_NODE) {
+         for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, $_->namespace_uri, 'node-'.refaddr $_] }
-       $r .= '| ' . $child->[1] . '<![CDATA[' . $child->[0]->data . "]]>\x0A";
+                       @{$child->attributes}) {
-     } elsif ($nt == $child->[0]->COMMENT_NODE) {
+           $r .= qq[<li id="$attr->[3]" class="tree-attribute"><code title="@{[defined $_->[2] ? $_->[2] : '']}">] . htescape ($attr->[0]) . '</code> = '; ## ISSUE: case?
-       $r .= '| ' . $child->[1] . '<!-- ' . $child->[0]->data . " -->\x0A";
+           $r .= '<q>' . htescape ($attr->[1]) . '</q></li>'; ## TODO: children
-     } elsif ($nt == $child->[0]->DOCUMENT_TYPE_NODE) {
+         }
-       $r .= '| ' . $child->[1] . '<!DOCTYPE ' . $child->[0]->name . ">\x0A";
+         $r .= '</ul>';
-     } elsif ($nt == $child->[0]->PROCESSING_INSTRUCTION_NODE) {
+       }
-       $r .= '| ' . $child->[1] . '<?' . $child->[0]->target . ' ' .
-           $child->[0]->data . "?>\x0A";
+       if ($child->has_child_nodes) {
+         $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 .= '| ' . $child->[1] . $child->[0]->node_type . "\x0A"; # error
+       $r .= qq'<li id="$node_id" class="tree-unknown">@{[$child->node_type]} @{[htescape ($child->node_name)]}</li>'; # error
+     }
+   }
+   $r .= '</ol>';
+   print STDOUT $r;
+ } # print_document_tree
+ sub print_structure_dump_dom_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 || $doc);
+   print STDOUT qq[</div>];
+ } # print_structure_dump_dom_section
+ sub print_structure_dump_manifest_section ($) {
+   my $manifest = shift;
+   print STDOUT qq[
+ <div id="dump-manifest" class="section">
+ <h2>Cache Manifest</h2>
+ ];
+   push @nav, ['#dump-manifest' => 'Caceh Manifest'];
+   print STDOUT qq[<dl><dt>Explicit entries</dt>];
+   for my $uri (@{$manifest->[0]}) {
+     my $euri = htescape ($uri);
+     print STDOUT qq[<dd><code class=uri>&lt;<a href="$euri">$euri</a>></code></dd>];
+   }
+   print STDOUT qq[<dt>Fallback entries</dt><dd>
+       <table><thead><tr><th scope=row>Oppotunistic Caching Namespace</th>
+       <th scope=row>Fallback Entry</tr><tbody>];
+   for my $uri (sort {$a cmp $b} keys %{$manifest->[1]}) {
+     my $euri = htescape ($uri);
+     my $euri2 = htescape ($manifest->[1]->{$uri});
+     print STDOUT qq[<tr><td><code class=uri>&lt;<a href="$euri">$euri</a>></code></td>
+         <td><code class=uri>&lt;<a href="$euri2">$euri2</a>></code></td>];
+   }
+   print STDOUT qq[</table><dt>Online whitelist</dt>];
+   for my $uri (@{$manifest->[2]}) {
+     my $euri = htescape ($uri);
+     print STDOUT qq[<dd><code class=uri>&lt;<a href="$euri">$euri</a>></code></dd>];
+   }
+   print STDOUT qq[</dl></div>];
+ } # print_structure_dump_manifest_section
+ sub print_structure_error_dom_section ($$$) {
+   my ($doc, $el, $result) = @_;
+   print STDOUT qq[<div id="document-errors" class="section">
+ <h2>Document Errors</h2>
+ <dl>];
+   push @nav, ['#document-errors' => 'Document Error'];
+   require Whatpm::ContentChecker;
+   my $onerror = sub {
+     my %opt = @_;
+     my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}, $opt{node});
+     $type =~ tr/ /-/;
+     $type =~ s/\|/%7C/g;
+     $msg .= qq[ [<a href="../error-description#@{[htescape ($type)]}">Description</a>]];
+     print STDOUT qq[<dt class="$cls">] . get_error_label (\%opt) .
+         qq[</dt>\n<dd class="$cls">], $msg, "</dd>\n";
+     add_error ('structure', \%opt => $result);
+   };
+   my $elements;
+   my $time1 = time;
+   if ($el) {
+     $elements = Whatpm::ContentChecker->check_element ($el, $onerror);
+   } else {
+     $elements = Whatpm::ContentChecker->check_document ($doc, $onerror);
+   }
+   $time{check} = time - $time1;
+   print STDOUT qq[</dl></div>];
+   return $elements;
+ } # print_structure_error_dom_section
+ sub print_structure_error_manifest_section ($$$) {
+   my ($manifest, $result) = @_;
+   print STDOUT qq[<div id="document-errors" class="section">
+ <h2>Document Errors</h2>
+ <dl>];
+   push @nav, ['#document-errors' => 'Document Error'];
+   require Whatpm::CacheManifest;
+   Whatpm::CacheManifest->check_manifest ($manifest, 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="$cls">] . get_error_label (\%opt) .
+         qq[</dt>\n<dd class="$cls">], $msg, "</dd>\n";
+     add_error ('structure', \%opt => $result);
+   });
+   print STDOUT qq[</div>];
+ } # print_structure_error_manifest_section
+ sub print_table_section ($) {
+   my $tables = shift;
+   push @nav, ['#tables' => 'Tables'];
+   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-$i'));</script>];
+   }
+   print STDOUT qq[</div>];
+ } # print_table_section
+ sub print_id_section ($) {
+   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
+ sub print_term_section ($) {
+   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
+ sub print_class_section ($) {
+   my $classes = shift;
-   return \$r;
+   push @nav, ['#classes' => 'Classes'];
- } # test_serialize
+   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>];
+     }
+   }
+   print STDOUT qq[</dl></div>];
+ } # print_class_section
+ sub print_result_section ($) {
+   my $result = shift;
+   print STDOUT qq[
+ <div id="result-summary" class="section">
+ <h2>Result</h2>];
+   if ($result->{unsupported} and $result->{conforming_max}) {
+     print STDOUT qq[<p class=uncertain id=result-para>The conformance
+         checker cannot decide whether the document is conforming or
+         not, since the document contains one or more unsupported
+         features.  The document might or might not be conforming.</p>];
+   } elsif ($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>];
+   }
+   print STDOUT qq[<table>
+ <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;
+   my $score_unit = $score_base / 100;
+   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_unit) += $score_base);
+     $score_max += (($result->{$_->[1]}->{score_max} *= $score_unit) += $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]">$label</a>];
+     }
+     print STDOUT qq[<tr class="@{[$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>];
+     if ($uncertain) {
+       print qq[<td class="@{[$result->{$_->[1]}->{must} ? 'FAIL' : $result->{$_->[1]}->{should} ? 'SEE-RESULT' : '']}">&#x2212;&#x221E;..$result->{$_->[1]}->{score_max}</td>];
+     } elsif ($result->{$_->[1]}->{score_min} != $result->{$_->[1]}->{score_max}) {
+       print qq[<td class="@{[$result->{$_->[1]}->{must} ? 'FAIL' : 'SEE-RESULT']}">$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max}</td></tr>];
+     } else {
+       print qq[<td class="@{[$result->{$_->[1]}->{must} ? 'FAIL' : '']}">$result->{$_->[1]}->{score_min}</td></tr>];
+     }
+   }
+   $score_max += $score_base;
+   print STDOUT qq[
+ <tr class=uncertain><th scope=row>Semantics</th><td>0?</td><td>0?</td><td>0?</td><td>&#x2212;&#x221E;..$score_base</td></tr>
+ </tbody>
+ <tfoot><tr class=uncertain><th scope=row>Total</th>
+ <td class="@{[$must_error ? 'FAIL' : '']}">$must_error?</td>
+ <td class="@{[$should_error ? 'SEE-RESULT' : '']}">$should_error?</td>
+ <td>$warning?</td>
+ <td class="@{[$must_error ? 'FAIL' : $should_error ? 'SEE-RESULT' : '']}"><strong>&#x2212;&#x221E;..$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_error_label ($) {
+   my $err = shift;
+   my $r = '';
+   if (defined $err->{line}) {
+     if ($err->{column} > 0) {
+       $r = qq[<a href="#line-$err->{line}">Line $err->{line}</a> column $err->{column}];
+     } else {
+       $err->{line} = $err->{line} - 1 || 1;
+       $r = qq[<a href="#line-$err->{line}">Line $err->{line}</a>];
+     }
+   }
+   if (defined $err->{node}) {
+     $r .= ' ' if length $r;
+     $r = get_node_path ($err->{node});
+   }
+   if (defined $err->{index}) {
+     $r .= ' ' if length $r;
+     $r .= 'Index ' . (0+$err->{index});
+   }
+   if (defined $err->{value}) {
+     $r .= ' ' if length $r;
+     $r .= '<q><code>' . htescape ($err->{value}) . '</code></q>';
+   }
+   return $r;
+ } # get_error_label
  sub get_node_path ($) {
    my $node = shift;
-Line 212 
 sub get_node_path ($) {
+Line 818 
 sub get_node_path ($) {
        $rs = '"' . $node->data . '"';
        $node = $node->parent_node;
      } elsif ($node->node_type == 9) {
+       @r = ('') unless @r;
        $rs = '';
        $node = $node->parent_node;
      } else {
-Line 223 
 sub get_node_path ($) {
+Line 830 
 sub get_node_path ($) {
    return join '/', @r;
  } # get_node_path
+ sub get_node_link ($) {
+   return qq[<a href="#node-@{[refaddr $_[0]]}">] .
+       htescape (get_node_path ($_[0])) . qq[</a>];
+ } # get_node_link
+ {
+   my $Msg = {};
+ sub load_text_catalog ($) {
+   my $lang = shift; # MUST be a canonical lang name
+   open my $file, '<', "cc-msg.$lang.txt" or die "$0: cc-msg.$lang.txt: $!";
+   while (<$file>) {
+     if (s/^([^;]+);([^;]*);//) {
+       my ($type, $cls, $msg) = ($1, $2, $_);
+       $msg =~ tr/\x0D\x0A//d;
+       $Msg->{$type} = [$cls, $msg];
+     }
+   }
+ } # load_text_catalog
+ sub get_text ($) {
+   my ($type, $level, $node) = @_;
+   $type = $level . ':' . $type if defined $level;
+   my @arg;
+   {
+     if (defined $Msg->{$type}) {
+       my $msg = $Msg->{$type}->[1];
+       $msg =~ s{<var>\$([0-9]+)</var>}{
+         defined $arg[$1] ? htescape ($arg[$1]) : '(undef)';
+       }ge;
+       $msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{
+         UNIVERSAL::can ($node, 'get_attribute_ns')
+             ? htescape ($node->get_attribute_ns (undef, $1)) : ''
+       }ge;
+       $msg =~ s{<var>{\@}</var>}{
+         UNIVERSAL::can ($node, 'value') ? htescape ($node->value) : ''
+       }ge;
+       $msg =~ s{<var>{local-name}</var>}{
+         UNIVERSAL::can ($node, 'manakai_local_name')
+           ? htescape ($node->manakai_local_name) : ''
+       }ge;
+       $msg =~ s{<var>{element-local-name}</var>}{
+         (UNIVERSAL::can ($node, 'owner_element') and
+          $node->owner_element)
+           ? htescape ($node->owner_element->manakai_local_name)
+           : ''
+       }ge;
+       return ($type, $Msg->{$type}->[0], $msg);
+     } elsif ($type =~ s/:([^:]*)$//) {
+       unshift @arg, $1;
+       redo;
+     }
+   }
+   return ($type, '', htescape ($_[0]));
+ } # get_text
+ }
+ sub get_input_document ($$) {
+   my ($http, $dom) = @_;
+   my $request_uri = $http->get_parameter ('uri');
+   my $r = {};
+   if (defined $request_uri and length $request_uri) {
+     my $uri = $dom->create_uri_reference ($request_uri);
+     unless ({
+              http => 1,
+             }->{lc $uri->uri_scheme}) {
+       return {uri => $request_uri, request_uri => $request_uri,
+               error_status_text => 'URI scheme not allowed'};
+     }
+     require Message::Util::HostPermit;
+     my $host_permit = new Message::Util::HostPermit;
+     $host_permit->add_rule (<<EOH);
+ Allow host=suika port=80
+ Deny host=suika
+ Allow host=suika.fam.cx port=80
+ Deny host=suika.fam.cx
+ Deny host=localhost
+ Deny host=*.localdomain
+ Deny ipv4=0.0.0.0/8
+ Deny ipv4=10.0.0.0/8
+ Deny ipv4=127.0.0.0/8
+ Deny ipv4=169.254.0.0/16
+ Deny ipv4=172.0.0.0/11
+ Deny ipv4=192.0.2.0/24
+ Deny ipv4=192.88.99.0/24
+ Deny ipv4=192.168.0.0/16
+ Deny ipv4=198.18.0.0/15
+ Deny ipv4=224.0.0.0/4
+ Deny ipv4=255.255.255.255/32
+ Deny ipv6=0::0/0
+ Allow host=*
+ EOH
+     unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) {
+       return {uri => $request_uri, request_uri => $request_uri,
+               error_status_text => 'Connection to the host is forbidden'};
+     }
+     require LWP::UserAgent;
+     my $ua = WDCC::LWPUA->new;
+     $ua->{wdcc_dom} = $dom;
+     $ua->{wdcc_host_permit} = $host_permit;
+     $ua->agent ('Mozilla'); ## TODO: for now.
+     $ua->parse_head (0);
+     $ua->protocols_allowed ([qw/http/]);
+     $ua->max_size (1000_000);
+     my $req = HTTP::Request->new (GET => $request_uri);
+     my $res = $ua->request ($req);
+     ## TODO: 401 sets |is_success| true.
+     if ($res->is_success or $http->get_parameter ('error-page')) {
+       $r->{base_uri} = $res->base; ## NOTE: It does check |Content-Base|, |Content-Location|, and <base>. ## TODO: Use our own code!
+       $r->{uri} = $res->request->uri;
+       $r->{request_uri} = $request_uri;
+       ## 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;
+       }
+       my $input_charset = $http->get_parameter ('charset');
+       if (defined $input_charset and length $input_charset) {
+         $r->{charset_overridden}
+             = (not defined $r->{charset} or $r->{charset} ne $input_charset);
+         $r->{charset} = $input_charset;
+       }
+       $r->{s} = ''.$res->content;
+     } else {
+       $r->{uri} = $res->request->uri;
+       $r->{request_uri} = $request_uri;
+       $r->{error_status_text} = $res->status_line;
+     }
+     $r->{header_field} = [];
+     $res->scan (sub {
+       push @{$r->{header_field}}, [$_[0], $_[1]];
+     });
+     $r->{header_status_code} = $res->code;
+     $r->{header_status_text} = $res->message;
+   } else {
+     $r->{s} = ''.$http->get_parameter ('s');
+     $r->{uri} = q<thismessage:/>;
+     $r->{request_uri} = q<thismessage:/>;
+     $r->{base_uri} = q<thismessage:/>;
+     $r->{charset} = ''.$http->get_parameter ('_charset_');
+     $r->{charset} =~ s/\s+//g;
+     $r->{charset} = 'utf-8' if $r->{charset} eq '';
+     $r->{header_field} = [];
+   }
+   my $input_format = $http->get_parameter ('i');
+   if (defined $input_format and length $input_format) {
+     $r->{media_type_overridden}
+         = (not defined $r->{media_type} or $input_format ne $r->{media_type});
+     $r->{media_type} = $input_format;
+   }
+   if (defined $r->{s} and not defined $r->{media_type}) {
+     $r->{media_type} = 'text/html';
+     $r->{media_type_overridden} = 1;
+   }
+   if ($r->{media_type} eq 'text/xml') {
+     unless (defined $r->{charset}) {
+       $r->{charset} = 'us-ascii';
+     } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') {
+       $r->{charset_overridden} = 0;
+     }
+   }
+   if (length $r->{s} > 1000_000) {
+     $r->{error_status_text} = 'Entity-body too large';
+     delete $r->{s};
+     return $r;
+   }
+   return $r;
+ } # get_input_document
+ package WDCC::LWPUA;
+ BEGIN { push our @ISA, 'LWP::UserAgent'; }
+ sub redirect_ok {
+   my $ua = shift;
+   unless ($ua->SUPER::redirect_ok (@_)) {
+     return 0;
+   }
+   my $uris = $_[1]->header ('Location');
+   return 0 unless $uris;
+   my $uri = $ua->{wdcc_dom}->create_uri_reference ($uris);
+   unless ({
+            http => 1,
+           }->{lc $uri->uri_scheme}) {
+     return 0;
+   }
+   unless ($ua->{wdcc_host_permit}->check ($uri->uri_host, $uri->uri_port || 80)) {
+     return 0;
+   }
+   return 1;
+ } # redirect_ok
  =head1 AUTHOR
  Wakaba <w@suika.fam.cx>.

 Legend:



Removed from v.1.1
 


changed lines


 
Added in v.1.22
 Legend:



Removed from v.1.1
 


changed lines


 
Added in v.1.22
-Removed from v.1.1
+Added in v.1.22

admin@suikawiki.org	ViewVC Help
Powered by ViewVC 1.1.24