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/manakai/lib |
/home/wakaba/work/manakai2/lib]; |
|
/home/wakaba/public_html/-temp/wiki/lib]; |
|
7 |
use CGI::Carp qw[fatalsToBrowser]; |
use CGI::Carp qw[fatalsToBrowser]; |
8 |
use Scalar::Util qw[refaddr]; |
use Scalar::Util qw[refaddr]; |
9 |
|
|
10 |
use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module |
require WebHACC::Input; |
11 |
|
require WebHACC::Result; |
12 |
|
require WebHACC::Output; |
13 |
|
|
14 |
sub htescape ($) { |
my $out; |
|
my $s = $_[0]; |
|
|
$s =~ s/&/&/g; |
|
|
$s =~ s/</</g; |
|
|
$s =~ s/>/>/g; |
|
|
$s =~ s/"/"/g; |
|
|
$s =~ s!([\x00-\x09\x0B-\x1F\x7F-\x80])!sprintf '<var>U+%04X</var>', ord $1!ge; |
|
|
return $s; |
|
|
} # htescape |
|
15 |
|
|
16 |
my $http = SuikaWiki::Input::HTTP->new; |
require Message::DOM::DOMImplementation; |
17 |
|
my $dom = Message::DOM::DOMImplementation->new; |
18 |
## TODO: _charset_ |
{ |
19 |
|
use Message::CGI::HTTP; |
20 |
|
my $http = Message::CGI::HTTP->new; |
21 |
|
|
22 |
if ($http->meta_variable ('PATH_INFO') ne '/') { |
if ($http->get_meta_variable ('PATH_INFO') ne '/') { |
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 |
|
|
|
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; |
|
|
} |
|
|
|
|
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"> |
39 |
<link rel="stylesheet" href="../cc-style.css" type="text/css"> |
<link rel="stylesheet" href="../cc-style.css" type="text/css"> |
40 |
</head> |
</head> |
41 |
<body> |
<body> |
42 |
<h1>Web Document Conformance Checker (<em>beta</em>)</h1> |
<h1><a href="../cc-interface">Web Document Conformance Checker</a> |
43 |
|
(<em>beta</em>)</h1> |
44 |
<div id="document-info" class="section"> |
]); |
45 |
<dl> |
|
46 |
<dt>Document URI</dt> |
my $input = get_input_document ($http, $dom); |
47 |
<dd><code class="URI" lang=""><<a href="@{[htescape $input_uri]}">@{[htescape $input_uri]}</a>></code></dd> |
$out->input ($input); |
48 |
<dt>Internet Media Type</dt> |
$out->unset_flush; |
49 |
<dd><code class="MIME" lang="en">@{[htescape $input_format]}</code></dd> |
|
50 |
]; # no </dl> yet |
my $char_length = 0; |
51 |
push @nav, ['#document-info' => 'Information']; |
|
52 |
|
$out->start_section (id => 'document-info', title => 'Information'); |
53 |
require Message::DOM::DOMImplementation; |
$out->html (qq[<dl> |
54 |
my $dom = Message::DOM::DOMImplementation->____new; |
<dt>Request URL</dt> |
55 |
my $doc; |
<dd>]); |
56 |
my $el; |
$out->url ($input->{request_uri}); |
57 |
|
$out->html (q[<dt>Document URL<!-- HTML5 document's address? --> |
58 |
if ($input_format eq 'text/html') { |
<dd>]); |
59 |
require Encode; |
$out->url ($input->{uri}, id => 'anchor-document-url'); |
60 |
require Whatpm::HTML; |
$out->html (q[ |
61 |
|
<script> |
62 |
$s = Encode::decode ('utf-8', $s); |
document.title = '<' |
63 |
|
+ document.getElementById ('anchor-document-url').href + '> \\u2014 ' |
64 |
print STDOUT qq[ |
+ document.title; |
65 |
<dt>Character Encoding</dt> |
</script>]); |
66 |
<dd>(none)</dd> |
## NOTE: no </dl> yet |
67 |
</dl> |
|
68 |
</div> |
if (defined $input->{s}) { |
69 |
|
$char_length = length $input->{s}; |
70 |
<div id="source-string" class="section"> |
|
71 |
<h2>Document Source</h2> |
$out->html (qq[<dt>Base URI<dd>]); |
72 |
]; |
$out->url ($input->{base_uri}); |
73 |
push @nav, ['#source-string' => 'Source']; |
$out->html (qq[<dt>Internet Media Type</dt> |
74 |
print_source_string (\$s); |
<dd><code class="MIME" lang="en">]); |
75 |
print STDOUT qq[ |
$out->text ($input->{media_type}); |
76 |
</div> |
$out->html (qq[</code> ]); |
77 |
|
if ($input->{media_type_overridden}) { |
78 |
<div id="parse-errors" class="section"> |
$out->html ('<em>(overridden)</em>'); |
79 |
<h2>Parse Errors</h2> |
} elsif (defined $input->{official_type}) { |
80 |
|
if ($input->{media_type} eq $input->{official_type}) { |
81 |
<dl> |
# |
82 |
]; |
} else { |
83 |
push @nav, ['#parse-errors' => 'Parse Error']; |
$out->html ('<em>(sniffed; official type is: <code class=MIME lang=en>'); |
84 |
|
$out->text ($input->{official_type}); |
85 |
my $onerror = sub { |
$out->html ('</code>)'); |
86 |
my (%opt) = @_; |
} |
|
my ($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]; |
|
87 |
} else { |
} else { |
88 |
$opt{line} = $opt{line} - 1 || 1; |
$out->html ('<em>(sniffed)</em>'); |
|
print STDOUT qq[<dt class="$cls"><a href="#line-$opt{line}">Line $opt{line}</a></dt>\n]; |
|
89 |
} |
} |
90 |
$opt{type} =~ tr/ /-/; |
$out->html (q[<dt>Character Encoding<dd>]); |
91 |
$opt{type} =~ s/\|/%7C/g; |
if (defined $input->{charset}) { |
92 |
$msg .= qq[ [<a href="../error-description#$opt{type}">Description</a>]]; |
$out->html ('<code class="charset" lang="en">'); |
93 |
print STDOUT qq[<dd class="$cls">$msg</dd>\n]; |
$out->text ($input->{charset}); |
94 |
}; |
$out->html ('</code>'); |
95 |
|
} else { |
96 |
$doc = $dom->create_document; |
$out->text ('(none)'); |
97 |
if (defined $inner_html_element and length $inner_html_element) { |
} |
98 |
$el = $doc->create_element_ns |
$out->html (' <em>overridden</em>') if $input->{charset_overridden}; |
99 |
('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]); |
$out->html (qq[ |
100 |
Whatpm::HTML->set_inner_html ($el, $s, $onerror); |
<dt>Length</dt> |
101 |
} else { |
<dd>$char_length byte@{[$char_length == 1 ? '' : 's']}</dd> |
|
Whatpm::HTML->parse_string ($s => $doc, $onerror); |
|
|
} |
|
|
|
|
|
print STDOUT qq[ |
|
|
</dl> |
|
|
</div> |
|
|
]; |
|
|
} elsif ($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> |
|
102 |
</dl> |
</dl> |
|
</div> |
|
103 |
|
|
104 |
<div id="source-string" class="section"> |
<script src="../cc-script.js"></script> |
105 |
<h2>Document Source</h2> |
]); |
106 |
]; |
$out->end_section; |
107 |
push @nav, ['#source-string' => 'Source']; |
|
108 |
print_source_string (\$t); |
my $result = WebHACC::Result->new; |
109 |
print STDOUT qq[ |
$result->output ($out); |
110 |
</div> |
$result->{conforming_min} = 1; |
111 |
|
$result->{conforming_max} = 1; |
112 |
|
check_and_print ($input => $result => $out); |
113 |
|
$result->generate_result_section; |
114 |
|
} else { |
115 |
|
$out->html ('</dl>'); |
116 |
|
$out->end_section; |
117 |
|
|
118 |
<div id="parse-errors" class="section"> |
my $result = WebHACC::Result->new; |
119 |
<h2>Parse Errors</h2> |
$result->output ($out); |
120 |
|
$result->{conforming_min} = 0; |
121 |
|
$result->{conforming_max} = 1; |
122 |
|
|
123 |
<dl>]; |
$input->generate_transfer_sections ($result); |
124 |
push @nav, ['#parse-errors' => 'Parse Error']; |
$result->generate_result_section; |
125 |
|
} |
126 |
|
|
127 |
my $onerror = sub { |
$out->nav_list; |
|
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"; |
|
|
return 1; |
|
|
}; |
|
|
|
|
|
open my $fh, '<', \$s; |
|
|
$doc = Message::DOM::XMLParserTemp->parse_byte_stream |
|
|
($fh => $dom, $onerror, charset => 'utf-8'); |
|
128 |
|
|
129 |
print STDOUT qq[</dl> |
exit; |
130 |
</div> |
} |
|
]; |
|
|
} else { |
|
|
print STDOUT qq[ |
|
|
</dl> |
|
|
</div> |
|
131 |
|
|
132 |
<div id="result-summary" class="section"> |
sub check_and_print ($$$) { |
133 |
<p><em>Media type <code class="MIME" lang="en">@{[htescape $input_format]}</code> is not supported!</em></p> |
my ($input, $result, $out) = @_; |
134 |
</div> |
my $original_input = $out->input; |
135 |
]; |
$out->input ($input); |
136 |
push @nav, ['#result-summary' => 'Result']; |
|
137 |
} |
$input->generate_transfer_sections ($result); |
138 |
|
|
139 |
|
my @subdoc; |
140 |
|
|
141 |
|
my $checker_class = { |
142 |
|
'text/cache-manifest' => 'WebHACC::Language::CacheManifest', |
143 |
|
'text/css' => 'WebHACC::Language::CSS', |
144 |
|
'text/html' => 'WebHACC::Language::HTML', |
145 |
|
'text/x-webidl' => 'WebHACC::Language::WebIDL', |
146 |
|
|
147 |
|
'text/xml' => 'WebHACC::Language::XML', |
148 |
|
'application/atom+xml' => 'WebHACC::Language::XML', |
149 |
|
'application/rss+xml' => 'WebHACC::Language::XML', |
150 |
|
'image/svg+xml' => 'WebHACC::Language::XML', |
151 |
|
'application/xhtml+xml' => 'WebHACC::Language::XML', |
152 |
|
'application/xml' => 'WebHACC::Language::XML', |
153 |
|
## TODO: Should we make all XML MIME Types fall |
154 |
|
## into this category? |
155 |
|
|
156 |
|
## NOTE: This type has different model from normal XML types. |
157 |
|
'application/rdf+xml' => 'WebHACC::Language::XML', |
158 |
|
}->{$input->{media_type}} || 'WebHACC::Language::Default'; |
159 |
|
|
160 |
|
eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@"; |
161 |
|
my $checker = $checker_class->new; |
162 |
|
$checker->input ($input); |
163 |
|
$checker->output ($out); |
164 |
|
$checker->result ($result); |
165 |
|
|
166 |
|
## TODO: A cache manifest MUST be text/cache-manifest |
167 |
|
## TODO: WebIDL media type "text/x-webidl" |
168 |
|
|
169 |
|
$checker->generate_syntax_error_section; |
170 |
|
$checker->generate_source_string_section; |
171 |
|
|
172 |
|
$checker->onsubdoc (sub { |
173 |
|
push @subdoc, shift; |
174 |
|
}); |
175 |
|
|
176 |
|
$checker->generate_structure_dump_section; |
177 |
|
$checker->generate_structure_error_section; |
178 |
|
$checker->generate_additional_sections; |
179 |
|
|
180 |
|
=pod |
181 |
|
|
182 |
if (defined $doc or defined $el) { |
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); |
|
183 |
|
|
184 |
|
print_table_section ($input, $elements->{table}) if @{$elements->{table}}; |
185 |
|
print_listing_section ({ |
186 |
|
id => 'identifiers', label => 'IDs', heading => 'Identifiers', |
187 |
|
}, $input, $elements->{id}) if keys %{$elements->{id}}; |
188 |
|
print_listing_section ({ |
189 |
|
id => 'terms', label => 'Terms', heading => 'Terms', |
190 |
|
}, $input, $elements->{term}) if keys %{$elements->{term}}; |
191 |
|
print_listing_section ({ |
192 |
|
id => 'classes', label => 'Classes', heading => 'Classes', |
193 |
|
}, $input, $elements->{class}) if keys %{$elements->{class}}; |
194 |
|
|
195 |
|
print_rdf_section ($input, $elements->{rdf}) if @{$elements->{rdf}}; |
196 |
|
} |
197 |
|
|
198 |
|
=cut |
199 |
|
|
200 |
|
my $id_prefix = 0; |
201 |
|
for my $_subinput (@subdoc) { |
202 |
|
my $subinput = WebHACC::Input->new; |
203 |
|
$subinput->{$_} = $_subinput->{$_} for keys %$_subinput; |
204 |
|
$subinput->id_prefix ('subdoc-' . ++$id_prefix); |
205 |
|
$subinput->nested (1); |
206 |
|
$subinput->{base_uri} = $subinput->{container_node}->base_uri |
207 |
|
unless defined $subinput->{base_uri}; |
208 |
|
my $ebaseuri = htescape ($subinput->{base_uri}); |
209 |
|
$out->start_section (id => $subinput->id_prefix, |
210 |
|
title => qq[Subdocument #$id_prefix]); |
211 |
print STDOUT qq[ |
print STDOUT qq[ |
212 |
</div> |
<dl> |
213 |
|
<dt>Internet Media Type</dt> |
214 |
<div id="document-errors" class="section"> |
<dd><code class="MIME" lang="en">@{[htescape $subinput->{media_type}]}</code> |
215 |
<h2>Document Errors</h2> |
<dt>Container Node</dt> |
216 |
|
<dd>@{[get_node_link ($input, $subinput->{container_node})]}</dd> |
217 |
<dl>]; |
<dt>Base <abbr title="Uniform Resource Identifiers">URI</abbr></dt> |
218 |
push @nav, ['#document-errors' => 'Document Error']; |
<dd><code class=URI><<a href="$ebaseuri">$ebaseuri</a>></code></dd> |
219 |
|
</dl>]; |
220 |
require Whatpm::ContentChecker; |
|
221 |
my $onerror = sub { |
$subinput->{id_prefix} .= '-'; |
222 |
my %opt = @_; |
check_and_print ($subinput => $result => $out); |
223 |
my ($cls, $msg) = get_text ($opt{type}, $opt{level}); |
|
224 |
$opt{type} = $opt{level} . ':' . $opt{type} if defined $opt{level}; |
$out->end_section; |
225 |
$opt{type} =~ tr/ /-/; |
} |
226 |
$opt{type} =~ s/\|/%7C/g; |
|
227 |
$msg .= qq[ [<a href="../error-description#$opt{type}">Description</a>]]; |
$out->input ($original_input); |
228 |
print STDOUT qq[<dt class="$cls">] . get_node_link ($opt{node}) . |
} # check_and_print |
229 |
qq[</dt>\n<dd class="$cls">], $msg, "</dd>\n"; |
|
230 |
}; |
sub print_table_section ($$) { |
231 |
|
my ($input, $tables) = @_; |
232 |
my $elements; |
|
233 |
if ($el) { |
# push @nav, [qq[#$input->{id_prefix}tables] => 'Tables'] |
234 |
$elements = Whatpm::ContentChecker->check_element ($el, $onerror); |
# unless $input->{nested}; |
235 |
} else { |
print STDOUT qq[ |
236 |
$elements = Whatpm::ContentChecker->check_document ($doc, $onerror); |
<div id="$input->{id_prefix}tables" class="section"> |
|
} |
|
|
|
|
|
print STDOUT qq[</dl> |
|
|
</div> |
|
|
]; |
|
|
|
|
|
if (@{$elements->{table}}) { |
|
|
require JSON; |
|
|
|
|
|
print STDOUT qq[ |
|
|
<div id="tables" class="section"> |
|
237 |
<h2>Tables</h2> |
<h2>Tables</h2> |
238 |
|
|
239 |
<!--[if IE]><script type="text/javascript" src="../excanvas.js"></script><![endif]--> |
<!--[if IE]><script type="text/javascript" src="../excanvas.js"></script><![endif]--> |
242 |
<p><em>Structure of tables are visualized here if scripting is enabled.</em></p> |
<p><em>Structure of tables are visualized here if scripting is enabled.</em></p> |
243 |
</noscript> |
</noscript> |
244 |
]; |
]; |
245 |
|
|
246 |
my $i = 0; |
require JSON; |
247 |
for my $table_el (@{$elements->{table}}) { |
|
248 |
$i++; |
my $i = 0; |
249 |
print STDOUT qq[<div class="section" id="table-$i"><h3>] . |
for my $table (@$tables) { |
250 |
get_node_link ($table_el) . q[</h3>]; |
$i++; |
251 |
|
print STDOUT qq[<div class="section" id="$input->{id_prefix}table-$i"><h3>] . |
252 |
my $table = Whatpm::HTMLTable->form_table ($table_el); |
get_node_link ($input, $table->{element}) . q[</h3>]; |
253 |
|
|
254 |
for (@{$table->{column_group}}, @{$table->{column}}, $table->{caption}) { |
delete $table->{element}; |
255 |
next unless $_; |
|
256 |
delete $_->{element}; |
for (@{$table->{column_group}}, @{$table->{column}}, $table->{caption}, |
257 |
} |
@{$table->{row}}) { |
258 |
|
next unless $_; |
259 |
for (@{$table->{row_group}}) { |
delete $_->{element}; |
260 |
next unless $_; |
} |
261 |
next unless $_->{element}; |
|
262 |
$_->{type} = $_->{element}->manakai_local_name; |
for (@{$table->{row_group}}) { |
263 |
|
next unless $_; |
264 |
|
next unless $_->{element}; |
265 |
|
$_->{type} = $_->{element}->manakai_local_name; |
266 |
|
delete $_->{element}; |
267 |
|
} |
268 |
|
|
269 |
|
for (@{$table->{cell}}) { |
270 |
|
next unless $_; |
271 |
|
for (@{$_}) { |
272 |
|
next unless $_; |
273 |
|
for (@$_) { |
274 |
|
$_->{id} = refaddr $_->{element} if defined $_->{element}; |
275 |
delete $_->{element}; |
delete $_->{element}; |
276 |
|
$_->{is_header} = $_->{is_header} ? 1 : 0; |
277 |
} |
} |
|
|
|
|
for (@{$table->{cell}}) { |
|
|
next unless $_; |
|
|
for (@{$_}) { |
|
|
next unless $_; |
|
|
for (@$_) { |
|
|
$_->{id} = refaddr $_->{element} if defined $_->{element}; |
|
|
delete $_->{element}; |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
print STDOUT '</div><script type="text/javascript">tableToCanvas ('; |
|
|
print STDOUT JSON::objToJson ($table); |
|
|
print STDOUT qq[, document.getElementById ('table-$i'));</script>]; |
|
278 |
} |
} |
|
|
|
|
print STDOUT qq[</div>]; |
|
279 |
} |
} |
280 |
|
|
281 |
if (keys %{$elements->{term}}) { |
print STDOUT '</div><script type="text/javascript">tableToCanvas ('; |
282 |
print STDOUT qq[ |
print STDOUT JSON::objToJson ($table); |
283 |
<div id="terms" class="section"> |
print STDOUT qq[, document.getElementById ('$input->{id_prefix}table-$i')]; |
284 |
<h2>Terms</h2> |
print STDOUT qq[, '$input->{id_prefix}');</script>]; |
285 |
|
} |
286 |
|
|
287 |
|
print STDOUT qq[</div>]; |
288 |
|
} # print_table_section |
289 |
|
|
290 |
|
sub print_listing_section ($$$) { |
291 |
|
my ($opt, $input, $ids) = @_; |
292 |
|
|
293 |
|
# push @nav, ['#' . $input->{id_prefix} . $opt->{id} => $opt->{label}] |
294 |
|
# unless $input->{nested}; |
295 |
|
print STDOUT qq[ |
296 |
|
<div id="$input->{id_prefix}$opt->{id}" class="section"> |
297 |
|
<h2>$opt->{heading}</h2> |
298 |
|
|
299 |
<dl> |
<dl> |
300 |
]; |
]; |
301 |
for my $term (sort {$a cmp $b} keys %{$elements->{term}}) { |
for my $id (sort {$a cmp $b} keys %$ids) { |
302 |
print STDOUT qq[<dt>@{[htescape $term]}</dt>]; |
print STDOUT qq[<dt><code>@{[htescape $id]}</code></dt>]; |
303 |
for (@{$elements->{term}->{$term}}) { |
for (@{$ids->{$id}}) { |
304 |
print STDOUT qq[<dd>].get_node_link ($_).qq[</dd>]; |
print STDOUT qq[<dd>].get_node_link ($input, $_).qq[</dd>]; |
|
} |
|
|
} |
|
|
print STDOUT qq[</dl></div>]; |
|
305 |
} |
} |
306 |
} |
} |
307 |
|
print STDOUT qq[</dl></div>]; |
308 |
|
} # print_listing_section |
309 |
|
|
|
## TODO: Show result |
|
310 |
|
|
311 |
|
sub print_rdf_section ($$$) { |
312 |
|
my ($input, $rdfs) = @_; |
313 |
|
|
314 |
|
# push @nav, ['#' . $input->{id_prefix} . 'rdf' => 'RDF'] |
315 |
|
# unless $input->{nested}; |
316 |
print STDOUT qq[ |
print STDOUT qq[ |
317 |
<ul class="navigation" id="nav-items"> |
<div id="$input->{id_prefix}rdf" class="section"> |
318 |
]; |
<h2>RDF Triples</h2> |
|
for (@nav) { |
|
|
print STDOUT qq[<li><a href="$_->[0]">$_->[1]</a></li>]; |
|
|
} |
|
|
print STDOUT qq[ |
|
|
</ul> |
|
|
</body> |
|
|
</html> |
|
|
]; |
|
319 |
|
|
320 |
exit; |
<dl>]; |
321 |
|
my $i = 0; |
322 |
sub print_source_string ($) { |
for my $rdf (@$rdfs) { |
323 |
my $s = $_[0]; |
print STDOUT qq[<dt id="$input->{id_prefix}rdf-@{[$i++]}">]; |
324 |
my $i = 1; |
print STDOUT get_node_link ($input, $rdf->[0]); |
325 |
print STDOUT qq[<ol lang="">\n]; |
print STDOUT qq[<dd><dl>]; |
326 |
if (length $$s) { |
for my $triple (@{$rdf->[1]}) { |
327 |
while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { |
print STDOUT '<dt>' . get_node_link ($input, $triple->[0]) . '<dd>'; |
328 |
print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n"; |
print STDOUT get_rdf_resource_html ($triple->[1]); |
329 |
$i++; |
print STDOUT ' '; |
330 |
} |
print STDOUT get_rdf_resource_html ($triple->[2]); |
331 |
if ($$s =~ /\G([^\x0A]+)/gc) { |
print STDOUT ' '; |
332 |
print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n"; |
print STDOUT get_rdf_resource_html ($triple->[3]); |
333 |
|
} |
334 |
|
print STDOUT qq[</dl>]; |
335 |
|
} |
336 |
|
print STDOUT qq[</dl></div>]; |
337 |
|
} # print_rdf_section |
338 |
|
|
339 |
|
sub get_rdf_resource_html ($) { |
340 |
|
my $resource = shift; |
341 |
|
if (defined $resource->{uri}) { |
342 |
|
my $euri = htescape ($resource->{uri}); |
343 |
|
return '<code class=uri><<a href="' . $euri . '">' . $euri . |
344 |
|
'</a>></code>'; |
345 |
|
} elsif (defined $resource->{bnodeid}) { |
346 |
|
return htescape ('_:' . $resource->{bnodeid}); |
347 |
|
} elsif ($resource->{nodes}) { |
348 |
|
return '(rdf:XMLLiteral)'; |
349 |
|
} elsif (defined $resource->{value}) { |
350 |
|
my $elang = htescape (defined $resource->{language} |
351 |
|
? $resource->{language} : ''); |
352 |
|
my $r = qq[<q lang="$elang">] . htescape ($resource->{value}) . '</q>'; |
353 |
|
if (defined $resource->{datatype}) { |
354 |
|
my $euri = htescape ($resource->{datatype}); |
355 |
|
$r .= '^^<code class=uri><<a href="' . $euri . '">' . $euri . |
356 |
|
'</a>></code>'; |
357 |
|
} elsif (length $resource->{language}) { |
358 |
|
$r .= '@' . htescape ($resource->{language}); |
359 |
} |
} |
360 |
|
return $r; |
361 |
} else { |
} else { |
362 |
print STDOUT q[<li id="line-1"></li>]; |
return '??'; |
363 |
} |
} |
364 |
print STDOUT "</ol>"; |
} # get_rdf_resource_html |
|
} # 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) { |
|
|
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>'; |
|
|
} |
|
|
|
|
|
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><[CDATA[</code><q lang="">' . htescape ($child->data) . '</q><code>]]></code></li>'; |
|
|
} elsif ($nt == $child->COMMENT_NODE) { |
|
|
$r .= qq'<li id="$node_id" class="tree-comment"><code><!--</code><q lang="">' . htescape ($child->data) . '</q><code>--></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>]; |
|
|
$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><!DOCTYPE></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><?@{[htescape ($child->target)]}</code> <q>@{[htescape ($child->data)]}</q><code>?></code></li>'; |
|
|
} else { |
|
|
$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 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 |
|
|
|
|
|
sub get_node_link ($) { |
|
|
return qq[<a href="#node-@{[refaddr $_[0]]}">] . |
|
|
htescape (get_node_path ($_[0])) . qq[</a>]; |
|
|
} # get_node_link |
|
365 |
|
|
366 |
{ |
{ |
367 |
my $Msg = {}; |
my $Msg = {}; |
368 |
|
|
369 |
sub load_text_catalog ($) { |
sub load_text_catalog ($) { |
370 |
|
# my $self = shift; |
371 |
my $lang = shift; # MUST be a canonical lang name |
my $lang = shift; # MUST be a canonical lang name |
372 |
open my $file, '<', "cc-msg.$lang.txt" or die "$0: cc-msg.$lang.txt: $!"; |
open my $file, '<:utf8', "cc-msg.$lang.txt" |
373 |
|
or die "$0: cc-msg.$lang.txt: $!"; |
374 |
while (<$file>) { |
while (<$file>) { |
375 |
if (s/^([^;]+);([^;]*);//) { |
if (s/^([^;]+);([^;]*);//) { |
376 |
my ($type, $cls, $msg) = ($1, $2, $_); |
my ($type, $cls, $msg) = ($1, $2, $_); |
380 |
} |
} |
381 |
} # load_text_catalog |
} # load_text_catalog |
382 |
|
|
383 |
sub get_text ($) { |
sub get_text ($;$$) { |
384 |
my ($type, $level) = @_; |
# my $self = shift; |
385 |
|
my ($type, $level, $node) = @_; |
386 |
$type = $level . ':' . $type if defined $level; |
$type = $level . ':' . $type if defined $level; |
387 |
|
$level = 'm' unless defined $level; |
388 |
my @arg; |
my @arg; |
389 |
{ |
{ |
390 |
if (defined $Msg->{$type}) { |
if (defined $Msg->{$type}) { |
391 |
my $msg = $Msg->{$type}->[1]; |
my $msg = $Msg->{$type}->[1]; |
392 |
$msg =~ s/\$([0-9]+)/defined $arg[$1] ? htescape ($arg[$1]) : '(undef)'/ge; |
$msg =~ s{<var>\$([0-9]+)</var>}{ |
393 |
return ($Msg->{$type}->[0], $msg); |
defined $arg[$1] ? ($arg[$1]) : '(undef)'; |
394 |
|
}ge; ##BUG: ^ must be escaped |
395 |
|
$msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{ |
396 |
|
UNIVERSAL::can ($node, 'get_attribute_ns') |
397 |
|
? ($node->get_attribute_ns (undef, $1)) : '' |
398 |
|
}ge; ## BUG: ^ must be escaped |
399 |
|
$msg =~ s{<var>{\@}</var>}{ ## BUG: v must be escaped |
400 |
|
UNIVERSAL::can ($node, 'value') ? ($node->value) : '' |
401 |
|
}ge; |
402 |
|
$msg =~ s{<var>{local-name}</var>}{ |
403 |
|
UNIVERSAL::can ($node, 'manakai_local_name') |
404 |
|
? ($node->manakai_local_name) : '' |
405 |
|
}ge; ## BUG: ^ must be escaped |
406 |
|
$msg =~ s{<var>{element-local-name}</var>}{ |
407 |
|
(UNIVERSAL::can ($node, 'owner_element') and |
408 |
|
$node->owner_element) |
409 |
|
? ($node->owner_element->manakai_local_name) |
410 |
|
: '' ## BUG: ^ must be escaped |
411 |
|
}ge; |
412 |
|
return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg); |
413 |
} elsif ($type =~ s/:([^:]*)$//) { |
} elsif ($type =~ s/:([^:]*)$//) { |
414 |
unshift @arg, $1; |
unshift @arg, $1; |
415 |
redo; |
redo; |
416 |
} |
} |
417 |
} |
} |
418 |
return ('', htescape ($_[0])); |
return ($type, 'level-'.$level, ($_[0])); |
419 |
|
## BUG: ^ must be escaped |
420 |
} # get_text |
} # get_text |
421 |
|
|
422 |
} |
} |
423 |
|
|
424 |
|
sub get_input_document ($$) { |
425 |
|
my ($http, $dom) = @_; |
426 |
|
|
427 |
|
require Encode; |
428 |
|
my $request_uri = Encode::decode ('utf-8', $http->get_parameter ('uri')); |
429 |
|
my $r = WebHACC::Input->new; |
430 |
|
if (defined $request_uri and length $request_uri) { |
431 |
|
my $uri = $dom->create_uri_reference ($request_uri); |
432 |
|
unless ({ |
433 |
|
http => 1, |
434 |
|
}->{lc $uri->uri_scheme}) { |
435 |
|
$r = WebHACC::Input::Error->new; |
436 |
|
$r->{uri} = $request_uri; |
437 |
|
$r->{request_uri} = $request_uri; |
438 |
|
$r->{error_status_text} = 'URL scheme not allowed'; |
439 |
|
} |
440 |
|
|
441 |
|
require Message::Util::HostPermit; |
442 |
|
my $host_permit = new Message::Util::HostPermit; |
443 |
|
$host_permit->add_rule (<<EOH); |
444 |
|
Allow host=suika port=80 |
445 |
|
Deny host=suika |
446 |
|
Allow host=suika.fam.cx port=80 |
447 |
|
Deny host=suika.fam.cx |
448 |
|
Deny host=localhost |
449 |
|
Deny host=*.localdomain |
450 |
|
Deny ipv4=0.0.0.0/8 |
451 |
|
Deny ipv4=10.0.0.0/8 |
452 |
|
Deny ipv4=127.0.0.0/8 |
453 |
|
Deny ipv4=169.254.0.0/16 |
454 |
|
Deny ipv4=172.0.0.0/11 |
455 |
|
Deny ipv4=192.0.2.0/24 |
456 |
|
Deny ipv4=192.88.99.0/24 |
457 |
|
Deny ipv4=192.168.0.0/16 |
458 |
|
Deny ipv4=198.18.0.0/15 |
459 |
|
Deny ipv4=224.0.0.0/4 |
460 |
|
Deny ipv4=255.255.255.255/32 |
461 |
|
Deny ipv6=0::0/0 |
462 |
|
Allow host=* |
463 |
|
EOH |
464 |
|
unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) { |
465 |
|
my $r = WebHACC::Input::Error->new; |
466 |
|
$r->{uri} = $request_uri; |
467 |
|
$r->{request_uri} = $request_uri; |
468 |
|
$r->{error_status_text} = 'Connection to the host is forbidden'; |
469 |
|
return $r; |
470 |
|
} |
471 |
|
|
472 |
|
require LWP::UserAgent; |
473 |
|
my $ua = WDCC::LWPUA->new; |
474 |
|
$ua->{wdcc_dom} = $dom; |
475 |
|
$ua->{wdcc_host_permit} = $host_permit; |
476 |
|
$ua->agent ('Mozilla'); ## TODO: for now. |
477 |
|
$ua->parse_head (0); |
478 |
|
$ua->protocols_allowed ([qw/http/]); |
479 |
|
$ua->max_size (1000_000); |
480 |
|
my $req = HTTP::Request->new (GET => $request_uri); |
481 |
|
$req->header ('Accept-Encoding' => 'identity, *; q=0'); |
482 |
|
my $res = $ua->request ($req); |
483 |
|
## TODO: 401 sets |is_success| true. |
484 |
|
if ($res->is_success or $http->get_parameter ('error-page')) { |
485 |
|
$r->{base_uri} = $res->base; ## NOTE: It does check |Content-Base|, |Content-Location|, and <base>. ## TODO: Use our own code! |
486 |
|
$r->{uri} = $res->request->uri; |
487 |
|
$r->{request_uri} = $request_uri; |
488 |
|
|
489 |
|
## TODO: More strict parsing... |
490 |
|
my $ct = $res->header ('Content-Type'); |
491 |
|
if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) { |
492 |
|
$r->{charset} = lc $1; |
493 |
|
$r->{charset} =~ tr/\\//d; |
494 |
|
$r->{official_charset} = $r->{charset}; |
495 |
|
} |
496 |
|
|
497 |
|
my $input_charset = $http->get_parameter ('charset'); |
498 |
|
if (defined $input_charset and length $input_charset) { |
499 |
|
$r->{charset_overridden} |
500 |
|
= (not defined $r->{charset} or $r->{charset} ne $input_charset); |
501 |
|
$r->{charset} = $input_charset; |
502 |
|
} |
503 |
|
|
504 |
|
## TODO: Support for HTTP Content-Encoding |
505 |
|
|
506 |
|
$r->{s} = ''.$res->content; |
507 |
|
|
508 |
|
require Whatpm::ContentType; |
509 |
|
($r->{official_type}, $r->{media_type}) |
510 |
|
= Whatpm::ContentType->get_sniffed_type |
511 |
|
(get_file_head => sub { |
512 |
|
return substr $r->{s}, 0, shift; |
513 |
|
}, |
514 |
|
http_content_type_byte => $ct, |
515 |
|
has_http_content_encoding => |
516 |
|
defined $res->header ('Content-Encoding'), |
517 |
|
supported_image_types => {}); |
518 |
|
} else { |
519 |
|
$r->{uri} = $res->request->uri; |
520 |
|
$r->{request_uri} = $request_uri; |
521 |
|
$r->{error_status_text} = $res->status_line; |
522 |
|
} |
523 |
|
|
524 |
|
$r->{header_field} = []; |
525 |
|
$res->scan (sub { |
526 |
|
push @{$r->{header_field}}, [$_[0], $_[1]]; |
527 |
|
}); |
528 |
|
$r->{header_status_code} = $res->code; |
529 |
|
$r->{header_status_text} = $res->message; |
530 |
|
} else { |
531 |
|
$r->{s} = ''.$http->get_parameter ('s'); |
532 |
|
$r->{uri} = q<thismessage:/>; |
533 |
|
$r->{request_uri} = q<thismessage:/>; |
534 |
|
$r->{base_uri} = q<thismessage:/>; |
535 |
|
$r->{charset} = ''.$http->get_parameter ('_charset_'); |
536 |
|
$r->{charset} =~ s/\s+//g; |
537 |
|
$r->{charset} = 'utf-8' if $r->{charset} eq ''; |
538 |
|
$r->{official_charset} = $r->{charset}; |
539 |
|
$r->{header_field} = []; |
540 |
|
|
541 |
|
require Whatpm::ContentType; |
542 |
|
($r->{official_type}, $r->{media_type}) |
543 |
|
= Whatpm::ContentType->get_sniffed_type |
544 |
|
(get_file_head => sub { |
545 |
|
return substr $r->{s}, 0, shift; |
546 |
|
}, |
547 |
|
http_content_type_byte => undef, |
548 |
|
has_http_content_encoding => 0, |
549 |
|
supported_image_types => {}); |
550 |
|
} |
551 |
|
|
552 |
|
my $input_format = $http->get_parameter ('i'); |
553 |
|
if (defined $input_format and length $input_format) { |
554 |
|
$r->{media_type_overridden} |
555 |
|
= (not defined $r->{media_type} or $input_format ne $r->{media_type}); |
556 |
|
$r->{media_type} = $input_format; |
557 |
|
} |
558 |
|
if (defined $r->{s} and not defined $r->{media_type}) { |
559 |
|
$r->{media_type} = 'text/html'; |
560 |
|
$r->{media_type_overridden} = 1; |
561 |
|
} |
562 |
|
|
563 |
|
if ($r->{media_type} eq 'text/xml') { |
564 |
|
unless (defined $r->{charset}) { |
565 |
|
$r->{charset} = 'us-ascii'; |
566 |
|
$r->{official_charset} = $r->{charset}; |
567 |
|
} elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') { |
568 |
|
$r->{charset_overridden} = 0; |
569 |
|
} |
570 |
|
} |
571 |
|
|
572 |
|
if (length $r->{s} > 1000_000) { |
573 |
|
$r->{error_status_text} = 'Entity-body too large'; |
574 |
|
delete $r->{s}; |
575 |
|
return $r; |
576 |
|
} |
577 |
|
|
578 |
|
$r->{inner_html_element} = $http->get_parameter ('e'); |
579 |
|
|
580 |
|
return $r; |
581 |
|
} # get_input_document |
582 |
|
|
583 |
|
package WDCC::LWPUA; |
584 |
|
BEGIN { push our @ISA, 'LWP::UserAgent'; } |
585 |
|
|
586 |
|
sub redirect_ok { |
587 |
|
my $ua = shift; |
588 |
|
unless ($ua->SUPER::redirect_ok (@_)) { |
589 |
|
return 0; |
590 |
|
} |
591 |
|
|
592 |
|
my $uris = $_[1]->header ('Location'); |
593 |
|
return 0 unless $uris; |
594 |
|
my $uri = $ua->{wdcc_dom}->create_uri_reference ($uris); |
595 |
|
unless ({ |
596 |
|
http => 1, |
597 |
|
}->{lc $uri->uri_scheme}) { |
598 |
|
return 0; |
599 |
|
} |
600 |
|
unless ($ua->{wdcc_host_permit}->check ($uri->uri_host, $uri->uri_port || 80)) { |
601 |
|
return 0; |
602 |
|
} |
603 |
|
return 1; |
604 |
|
} # redirect_ok |
605 |
|
|
606 |
=head1 AUTHOR |
=head1 AUTHOR |
607 |
|
|
608 |
Wakaba <w@suika.fam.cx>. |
Wakaba <w@suika.fam.cx>. |
609 |
|
|
610 |
=head1 LICENSE |
=head1 LICENSE |
611 |
|
|
612 |
Copyright 2007 Wakaba <w@suika.fam.cx> |
Copyright 2007-2008 Wakaba <w@suika.fam.cx> |
613 |
|
|
614 |
This library is free software; you can redistribute it |
This library is free software; you can redistribute it |
615 |
and/or modify it under the same terms as Perl itself. |
and/or modify it under the same terms as Perl itself. |