test/html-webhacc/cc.cgi

#!/usr/bin/perl
use strict;

use lib qw[/home/httpd/html/www/markup/html/whatpm
           /home/wakaba/work/manakai/lib
           /home/wakaba/public_html/-temp/wiki/lib];
use CGI::Carp qw[fatalsToBrowser];
use Scalar::Util qw[refaddr];

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-\x80])!sprintf '<var>U+%04X</var>', ord $1!ge;
  return $s;
} # htescape

my $http = SuikaWiki::Input::HTTP->new;

## TODO: _charset_

  my $input_format = $http->parameter ('i') || 'text/html';
  my $inner_html_element = $http->parameter ('e');
  my $input_uri = 'thismessage:/';

  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 @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>Web Document Conformance Checker (<em>beta</em>)</h1>

<div id="document-info" section="section">
<dl>
<dt>Document URI</dt>
    <dd><code class="URI" lang="">&lt;<a href="@{[htescape $input_uri]}">@{[htescape $input_uri]}</a>&gt;</code></dd>
<dt>Internet Media Type</dt>
    <dd><code class="MIME" lang="en">@{[htescape $input_format]}</code></dd>
]; # no </dl> yet
  push @nav, ['#document-info' => 'Information'];

  require Message::DOM::DOMImplementation;
  my $dom = Message::DOM::DOMImplementation->____new;
  my $doc;
  my $el;

  if ($input_format eq 'text/html') {
    require Encode;
    require Whatpm::HTML;
    
    $s = Encode::decode ('utf-8', $s);

    print STDOUT qq[
<dt>Character Encoding</dt>
    <dd>(none)</dd>
</dl>
</div>

<div id="source-string" class="section">
<h2>Document Source</h2>
];
    push @nav, ['#source-string' => 'Source'];
    print_source_string (\$s);
    print STDOUT qq[
</div>

<div id="parse-errors" class="section">
<h2>Parse Errors</h2>

<ul>
];
  push @nav, ['#parse-errors' => 'Parse Error'];

  my $onerror = sub {
    my (%opt) = @_;
    if ($opt{column} > 0) {
      print STDOUT qq[<li><a href="#line-$opt{line}">Line $opt{line}</a> column $opt{column}: ];
    } else {
      $opt{line}--;
      print STDOUT qq[<li><a href="#line-$opt{line}">Line $opt{line}</a>: ];
    }
    print STDOUT qq[@{[htescape $opt{type}]}</li>\n];
  };

  $doc = $dom->create_document;
  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, $s, $onerror);
  } else {
    Whatpm::HTML->parse_string ($s => $doc, $onerror);
  }

  print STDOUT qq[
</ul>
</div>
];
  } elsif ($input_format eq 'application/xhtml+xml') {
    require Message::DOM::XMLParserTemp;
    require Encode;
    
    my $t = Encode::decode ('utf-8', $s);

    print STDOUT qq[
<dt>Character Encoding</dt>
    <dd>(none)</dd>
</dl>
</div>

<div id="source-string" class="section">
<h2>Document Source</h2>
];
    push @nav, ['#source-string' => 'Source'];
    print_source_string (\$t);
    print STDOUT qq[
</div>

<div id="parse-errors" class="section">
<h2>Parse Errors</h2>

<ul>
];
  push @nav, ['#parse-errors' => 'Parse Error'];

  my $onerror = sub {
    my $err = shift;
    my $line = $err->location->line_number;
    print STDOUT qq[<li><a href="#line-$line">Line $line</a> column ];
    print STDOUT $err->location->column_number, ": ";
    print STDOUT htescape $err->text, "</li>\n";
    return 1;
  };

  open my $fh, '<', \$s;
  $doc = Message::DOM::XMLParserTemp->parse_byte_stream
      ($fh => $dom, $onerror, charset => 'utf-8');

    print STDOUT qq[
</ul>
</div>
];
  } else {
    print STDOUT qq[
</dl>

<div id="result-summary" class="section">
<p><em>Media type <code class="MIME" lang="en">@{[htescape $input_format]}</code> is not supported!</em></p>
</div>
];
    push @nav, ['#result-summary' => 'Result'];
  }


  if (defined $doc or defined $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>

<div id="document-errors" class="section">
<h2>Document Errors</h2>

<ul>
];
    push @nav, ['#document-errors' => 'Document Error'];

    require Whatpm::ContentChecker;
    my $onerror = sub {
      my %opt = @_;
      print STDOUT qq[<li><a href="#node-@{[refaddr $opt{node}]}">],
          htescape get_node_path ($opt{node}),
          "</a>: ", htescape $opt{type}, "</li>\n";
    };

    if ($el) {
      Whatpm::ContentChecker->check_element ($el, $onerror);
    } else {
      Whatpm::ContentChecker->check_document ($doc, $onerror);
    }

    print STDOUT qq[
</ul>
</div>
];
  }

  ## TODO: Show result

  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>
];

exit;

sub print_source_string ($) {
  my $s = $_[0];
  my $i = 1;
  print STDOUT qq[<ol lang="">\n];
  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";
  }
  print STDOUT "</ol>";
} # print_input_string

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) {
      $r .= qq'<li id="$node_id"><code>' . 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, 'node-'.refaddr $_] }
                      @{$child->attributes}) {
          $r .= qq'<li id="$attr->[2]"><code>' . htescape ($attr->[0]) . '</code> = '; ## ISSUE: case?
          $r .= '<q>' . htescape ($attr->[1]) . '</q></li>'; ## TODO: children
        }
        $r .= '</ul>';
      }

      if ($node->has_child_nodes) {
        $r .= '<ol class="children">';
        unshift @node, @{$child->child_nodes}, '</ol>';
      }
    } elsif ($nt == $child->TEXT_NODE) {
      $r .= qq'<li id="$node_id"><q>' . htescape ($child->data) . '</q></li>';
    } elsif ($nt == $child->CDATA_SECTION_NODE) {
      $r .= qq'<li id="$node_id"><code>&lt;[CDATA[</code><q>' . htescape ($child->data) . '</q><code>]]&gt;</code></li>';
    } elsif ($nt == $child->COMMENT_NODE) {
      $r .= qq'<li id="$node_id"><code>&lt;!--</code><q>' . htescape ($child->data) . '</q><code>--&gt;</code></li>';
    } elsif ($nt == $child->DOCUMENT_NODE) {
      $r .= qq'<li id="$node_id">Document</li>';
      if ($child->has_child_nodes) {
        $r .= '<ol>';
        unshift @node, @{$child->child_nodes}, '</ol>';
      }
    } elsif ($nt == $child->DOCUMENT_TYPE_NODE) {
      $r .= qq'<li id="$node_id"><code>&lt;!DOCTYPE&gt;</code><ul>';
      $r .= '<li>Name = <q>@{[htescape ($child->name)]}</q></li>';
      $r .= '<li>Public identifier = <q>@{[htescape ($child->public_id)]}</q></li>';
      $r .= '<li>System identifier = <q>@{[htescape ($child->system_id)]}</q></li>';
      $r .= '</ul></li>';
    } elsif ($nt == $child->PROCESSING_INSTRUCTION_NODE) {
      $r .= qq'<li id="$node_id"><code>&lt;?@{[htescape ($child->target)]}?&gt;</code>';
      $r .= '<ul><li>@{[htescape ($child->data)]}</li></ul></li>';
    } else {
      $r .= qq'<li id="$node_id">@{[$child->node_type]} @{[htescape ($child->node_name)]}</li>'; # error
    }
  }

  $r .= '</ol>';
  print STDOUT $r;
} # print_document_tree

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) {
      $rs = '';
      $node = $node->parent_node;
    } else {
      $rs = '#' . $node->node_type;
      $node = $node->parent_node;
    }
    unshift @r, $rs;
  }
  return join '/', @r;
} # get_node_path

=head1 AUTHOR

Wakaba <w@suika.fam.cx>.

=head1 LICENSE

Copyright 2007 Wakaba <w@suika.fam.cx>

This library is free software; you can redistribute it
and/or modify it under the same terms as Perl itself.

=cut

## $Date: 2007/06/27 12:35:24 $
1	wakaba	1.1	#!/usr/bin/perl
2			use strict;
3
4			use lib qw[/home/httpd/html/www/markup/html/whatpm
5			/home/wakaba/work/manakai/lib
6			/home/wakaba/public_html/-temp/wiki/lib];
7			use CGI::Carp qw[fatalsToBrowser];
8	wakaba	1.2	use Scalar::Util qw[refaddr];
9	wakaba	1.1
10			use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module
11
12	wakaba	1.2	sub htescape ($) {
13			my $s = $_[0];
14			$s =~ s/&/&/g;
15			$s =~ s/</</g;
16			$s =~ s/>/>/g;
17			$s =~ s/"/"/g;
18			$s =~ s!([\x00-\x09\x0B-\x1F\x7F-\x80])!sprintf '<var>U+%04X</var>', ord $1!ge;
19			return $s;
20			} # htescape
21
22	wakaba	1.1	my $http = SuikaWiki::Input::HTTP->new;
23
24			## TODO: _charset_
25
26	wakaba	1.2	my $input_format = $http->parameter ('i') \|\| 'text/html';
27			my $inner_html_element = $http->parameter ('e');
28			my $input_uri = 'thismessage:/';
29	wakaba	1.1
30			my $s = $http->parameter ('s');
31			if (length $s > 1000_000) {
32			print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long";
33			exit;
34			}
35	wakaba	1.2
36	wakaba	1.3	my @nav;
37	wakaba	1.2	print STDOUT qq[Content-Type: text/html; charset=utf-8
38
39			<!DOCTYPE html>
40			<html lang="en">
41			<head>
42			<title>Web Document Conformance Checker (BETA)</title>
43	wakaba	1.3	<link rel="stylesheet" href="../cc-style.css" type="text/css">
44	wakaba	1.2	</head>
45			<body>
46			<h1>Web Document Conformance Checker (<em>beta</em>)</h1>
47
48	wakaba	1.3	<div id="document-info" section="section">
49	wakaba	1.2	<dl>
50			<dt>Document URI</dt>
51			<dd><code class="URI" lang=""><<a href="@{[htescape $input_uri]}">@{[htescape $input_uri]}</a>></code></dd>
52			<dt>Internet Media Type</dt>
53			<dd><code class="MIME" lang="en">@{[htescape $input_format]}</code></dd>
54			]; # no </dl> yet
55	wakaba	1.3	push @nav, ['#document-info' => 'Information'];
56	wakaba	1.1
57			require Message::DOM::DOMImplementation;
58			my $dom = Message::DOM::DOMImplementation->____new;
59			my $doc;
60			my $el;
61
62	wakaba	1.2	if ($input_format eq 'text/html') {
63			require Encode;
64			require Whatpm::HTML;
65
66			$s = Encode::decode ('utf-8', $s);
67
68			print STDOUT qq[
69			<dt>Character Encoding</dt>
70			<dd>(none)</dd>
71			</dl>
72	wakaba	1.3	</div>
73	wakaba	1.2
74			<div id="source-string" class="section">
75	wakaba	1.3	<h2>Document Source</h2>
76	wakaba	1.2	];
77	wakaba	1.3	push @nav, ['#source-string' => 'Source'];
78	wakaba	1.2	print_source_string (\$s);
79			print STDOUT qq[
80			</div>
81	wakaba	1.1
82	wakaba	1.2	<div id="parse-errors" class="section">
83			<h2>Parse Errors</h2>
84
85			<ul>
86			];
87	wakaba	1.3	push @nav, ['#parse-errors' => 'Parse Error'];
88	wakaba	1.1
89			my $onerror = sub {
90			my (%opt) = @_;
91	wakaba	1.2	if ($opt{column} > 0) {
92			print STDOUT qq[<li><a href="#line-$opt{line}">Line $opt{line}</a> column $opt{column}: ];
93			} else {
94			$opt{line}--;
95			print STDOUT qq[<li><a href="#line-$opt{line}">Line $opt{line}</a>: ];
96			}
97			print STDOUT qq[@{[htescape $opt{type}]}</li>\n];
98	wakaba	1.1	};
99
100			$doc = $dom->create_document;
101	wakaba	1.2	if (defined $inner_html_element and length $inner_html_element) {
102	wakaba	1.1	$el = $doc->create_element_ns
103	wakaba	1.2	('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
104	wakaba	1.1	Whatpm::HTML->set_inner_html ($el, $s, $onerror);
105			} else {
106			Whatpm::HTML->parse_string ($s => $doc, $onerror);
107			}
108
109	wakaba	1.2	print STDOUT qq[
110			</ul>
111			</div>
112			];
113			} elsif ($input_format eq 'application/xhtml+xml') {
114			require Message::DOM::XMLParserTemp;
115			require Encode;
116
117			my $t = Encode::decode ('utf-8', $s);
118
119			print STDOUT qq[
120			<dt>Character Encoding</dt>
121			<dd>(none)</dd>
122			</dl>
123	wakaba	1.3	</div>
124	wakaba	1.2
125			<div id="source-string" class="section">
126	wakaba	1.3	<h2>Document Source</h2>
127	wakaba	1.2	];
128	wakaba	1.3	push @nav, ['#source-string' => 'Source'];
129	wakaba	1.2	print_source_string (\$t);
130			print STDOUT qq[
131			</div>
132	wakaba	1.1
133	wakaba	1.2	<div id="parse-errors" class="section">
134			<h2>Parse Errors</h2>
135	wakaba	1.1
136	wakaba	1.2	<ul>
137			];
138	wakaba	1.3	push @nav, ['#parse-errors' => 'Parse Error'];
139	wakaba	1.1
140			my $onerror = sub {
141			my $err = shift;
142	wakaba	1.2	my $line = $err->location->line_number;
143			print STDOUT qq[<li><a href="#line-$line">Line $line</a> column ];
144			print STDOUT $err->location->column_number, ": ";
145			print STDOUT htescape $err->text, "</li>\n";
146	wakaba	1.1	return 1;
147			};
148
149			open my $fh, '<', \$s;
150			$doc = Message::DOM::XMLParserTemp->parse_byte_stream
151			($fh => $dom, $onerror, charset => 'utf-8');
152
153	wakaba	1.2	print STDOUT qq[
154			</ul>
155			</div>
156			];
157			} else {
158			print STDOUT qq[
159			</dl>
160
161	wakaba	1.3	<div id="result-summary" class="section">
162	wakaba	1.2	<p><em>Media type <code class="MIME" lang="en">@{[htescape $input_format]}</code> is not supported!</em></p>
163	wakaba	1.3	</div>
164	wakaba	1.2	];
165	wakaba	1.3	push @nav, ['#result-summary' => 'Result'];
166	wakaba	1.2	}
167
168
169			if (defined $doc or defined $el) {
170			print STDOUT qq[
171			<div id="document-tree" class="section">
172			<h2>Document Tree</h2>
173			];
174	wakaba	1.3	push @nav, ['#document-tree' => 'Tree'];
175	wakaba	1.2
176			print_document_tree ($el \|\| $doc);
177
178			print STDOUT qq[
179			</div>
180	wakaba	1.1
181	wakaba	1.2	<div id="document-errors" class="section">
182			<h2>Document Errors</h2>
183
184			<ul>
185			];
186	wakaba	1.3	push @nav, ['#document-errors' => 'Document Error'];
187	wakaba	1.1
188			require Whatpm::ContentChecker;
189			my $onerror = sub {
190			my %opt = @_;
191	wakaba	1.2	print STDOUT qq[<li><a href="#node-@{[refaddr $opt{node}]}">],
192			htescape get_node_path ($opt{node}),
193			"</a>: ", htescape $opt{type}, "</li>\n";
194	wakaba	1.1	};
195	wakaba	1.2
196	wakaba	1.1	if ($el) {
197			Whatpm::ContentChecker->check_element ($el, $onerror);
198			} else {
199			Whatpm::ContentChecker->check_document ($doc, $onerror);
200			}
201	wakaba	1.2
202			print STDOUT qq[
203			</ul>
204			</div>
205			];
206	wakaba	1.1	}
207
208	wakaba	1.2	## TODO: Show result
209	wakaba	1.3
210	wakaba	1.2	print STDOUT qq[
211	wakaba	1.3	<ul class="navigation" id="nav-items">
212			];
213			for (@nav) {
214			print STDOUT qq[<li><a href="$_->[0]">$_->[1]</a></li>];
215			}
216			print STDOUT qq[
217			</ul>
218	wakaba	1.2	</body>
219			</html>
220			];
221	wakaba	1.1
222			exit;
223
224	wakaba	1.2	sub print_source_string ($) {
225			my $s = $_[0];
226			my $i = 1;
227			print STDOUT qq[<ol lang="">\n];
228			while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) {
229			print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n";
230			$i++;
231			}
232			if ($$s =~ /\G([^\x0A]+)/gc) {
233			print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n";
234			}
235			print STDOUT "</ol>";
236			} # print_input_string
237
238			sub print_document_tree ($) {
239	wakaba	1.1	my $node = shift;
240	wakaba	1.2	my $r = '<ol class="xoxo">';
241	wakaba	1.1
242	wakaba	1.2	my @node = ($node);
243	wakaba	1.1	while (@node) {
244			my $child = shift @node;
245	wakaba	1.2	unless (ref $child) {
246			$r .= $child;
247			next;
248			}
249
250			my $node_id = 'node-'.refaddr $child;
251			my $nt = $child->node_type;
252			if ($nt == $child->ELEMENT_NODE) {
253			$r .= qq'<li id="$node_id"><code>' . htescape ($child->tag_name) .
254			'</code>'; ## ISSUE: case
255
256			if ($child->has_attributes) {
257			$r .= '<ul class="attributes">';
258			for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, 'node-'.refaddr $_] }
259			@{$child->attributes}) {
260			$r .= qq'<li id="$attr->[2]"><code>' . htescape ($attr->[0]) . '</code> = '; ## ISSUE: case?
261			$r .= '<q>' . htescape ($attr->[1]) . '</q></li>'; ## TODO: children
262			}
263			$r .= '</ul>';
264			}
265
266			if ($node->has_child_nodes) {
267			$r .= '<ol class="children">';
268			unshift @node, @{$child->child_nodes}, '</ol>';
269			}
270			} elsif ($nt == $child->TEXT_NODE) {
271			$r .= qq'<li id="$node_id"><q>' . htescape ($child->data) . '</q></li>';
272			} elsif ($nt == $child->CDATA_SECTION_NODE) {
273			$r .= qq'<li id="$node_id"><code><[CDATA[</code><q>' . htescape ($child->data) . '</q><code>]]></code></li>';
274			} elsif ($nt == $child->COMMENT_NODE) {
275			$r .= qq'<li id="$node_id"><code><!--</code><q>' . htescape ($child->data) . '</q><code>--></code></li>';
276			} elsif ($nt == $child->DOCUMENT_NODE) {
277			$r .= qq'<li id="$node_id">Document</li>';
278			if ($child->has_child_nodes) {
279			$r .= '<ol>';
280			unshift @node, @{$child->child_nodes}, '</ol>';
281	wakaba	1.1	}
282	wakaba	1.2	} elsif ($nt == $child->DOCUMENT_TYPE_NODE) {
283			$r .= qq'<li id="$node_id"><code><!DOCTYPE></code><ul>';
284			$r .= '<li>Name = <q>@{[htescape ($child->name)]}</q></li>';
285			$r .= '<li>Public identifier = <q>@{[htescape ($child->public_id)]}</q></li>';
286			$r .= '<li>System identifier = <q>@{[htescape ($child->system_id)]}</q></li>';
287			$r .= '</ul></li>';
288			} elsif ($nt == $child->PROCESSING_INSTRUCTION_NODE) {
289			$r .= qq'<li id="$node_id"><code><?@{[htescape ($child->target)]}?></code>';
290			$r .= '<ul><li>@{[htescape ($child->data)]}</li></ul></li>';
291	wakaba	1.1	} else {
292	wakaba	1.2	$r .= qq'<li id="$node_id">@{[$child->node_type]} @{[htescape ($child->node_name)]}</li>'; # error
293	wakaba	1.1	}
294			}
295	wakaba	1.2
296			$r .= '</ol>';
297			print STDOUT $r;
298			} # print_document_tree
299	wakaba	1.1
300			sub get_node_path ($) {
301			my $node = shift;
302			my @r;
303			while (defined $node) {
304			my $rs;
305			if ($node->node_type == 1) {
306			$rs = $node->manakai_local_name;
307			$node = $node->parent_node;
308			} elsif ($node->node_type == 2) {
309			$rs = '@' . $node->manakai_local_name;
310			$node = $node->owner_element;
311			} elsif ($node->node_type == 3) {
312			$rs = '"' . $node->data . '"';
313			$node = $node->parent_node;
314			} elsif ($node->node_type == 9) {
315			$rs = '';
316			$node = $node->parent_node;
317			} else {
318			$rs = '#' . $node->node_type;
319			$node = $node->parent_node;
320			}
321			unshift @r, $rs;
322			}
323			return join '/', @r;
324			} # get_node_path
325
326			=head1 AUTHOR
327
328			Wakaba <w@suika.fam.cx>.
329
330			=head1 LICENSE
331
332			Copyright 2007 Wakaba <w@suika.fam.cx>
333
334			This library is free software; you can redistribute it
335			and/or modify it under the same terms as Perl itself.
336
337			=cut
338
339	wakaba	1.3	## $Date: 2007/06/27 12:35:24 $