| 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 Time::HiRes qw/time/; |
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 |
my $http = SuikaWiki::Input::HTTP->new; |
my $out; |
| 15 |
|
|
| 16 |
## TODO: _charset_ |
require Message::DOM::DOMImplementation; |
| 17 |
|
my $dom = Message::DOM::DOMImplementation->new; |
| 18 |
my @mode = split m#/#, scalar $http->meta_variable ('PATH_INFO'), -1; |
{ |
| 19 |
shift @mode if @mode and $mode[0] == ''; |
use Message::CGI::HTTP; |
| 20 |
## TODO: decode unreserved characters |
my $http = Message::CGI::HTTP->new; |
| 21 |
|
|
| 22 |
my $s = $http->parameter ('s'); |
if ($http->get_meta_variable ('PATH_INFO') ne '/') { |
| 23 |
if (length $s > 1000_000) { |
print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400"; |
|
print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long"; |
|
| 24 |
exit; |
exit; |
| 25 |
} |
} |
| 26 |
my $char_length = length $s; |
|
| 27 |
my %time; |
load_text_catalog ('en'); ## TODO: conneg |
|
my $time1; |
|
|
my $time2; |
|
| 28 |
|
|
| 29 |
require Message::DOM::DOMImplementation; |
$out = WebHACC::Output->new; |
| 30 |
my $dom = Message::DOM::DOMImplementation->____new; |
$out->handle (*STDOUT); |
| 31 |
# $| = 1; |
$out->set_utf8; |
| 32 |
my $doc; |
$out->set_flush; |
| 33 |
my $el; |
$out->html (qq[Content-Type: text/html; charset=utf-8 |
| 34 |
|
|
| 35 |
if (@mode == 3 and $mode[0] eq 'html' and |
<!DOCTYPE html> |
| 36 |
($mode[2] eq 'html' or $mode[2] eq 'test')) { |
<html lang="en"> |
| 37 |
print STDOUT "Content-Type: text/plain; charset=utf-8\n\n"; |
<head> |
| 38 |
|
<title>Web Document Conformance Checker (BETA)</title> |
| 39 |
|
<link rel="stylesheet" href="../cc-style.css" type="text/css"> |
| 40 |
|
</head> |
| 41 |
|
<body> |
| 42 |
|
<h1><a href="../cc-interface">Web Document Conformance Checker</a> |
| 43 |
|
(<em>beta</em>)</h1> |
| 44 |
|
]); |
| 45 |
|
|
| 46 |
|
my $input = get_input_document ($http, $dom); |
| 47 |
|
$out->input ($input); |
| 48 |
|
$out->unset_flush; |
| 49 |
|
|
| 50 |
|
my $char_length = 0; |
| 51 |
|
|
| 52 |
|
$out->start_section (id => 'document-info', title => 'Information'); |
| 53 |
|
$out->html (qq[<dl> |
| 54 |
|
<dt>Request URL</dt> |
| 55 |
|
<dd>]); |
| 56 |
|
$out->url ($input->{request_uri}); |
| 57 |
|
$out->html (q[<dt>Document URL<!-- HTML5 document's address? --> |
| 58 |
|
<dd>]); |
| 59 |
|
$out->url ($input->{uri}, id => 'anchor-document-url'); |
| 60 |
|
$out->html (q[ |
| 61 |
|
<script> |
| 62 |
|
document.title = '<' |
| 63 |
|
+ document.getElementById ('anchor-document-url').href + '> \\u2014 ' |
| 64 |
|
+ document.title; |
| 65 |
|
</script>]); |
| 66 |
|
## NOTE: no </dl> yet |
| 67 |
|
|
| 68 |
|
if (defined $input->{s}) { |
| 69 |
|
$char_length = length $input->{s}; |
| 70 |
|
|
| 71 |
|
$out->html (qq[<dt>Base URI<dd>]); |
| 72 |
|
$out->url ($input->{base_uri}); |
| 73 |
|
$out->html (qq[<dt>Internet Media Type</dt> |
| 74 |
|
<dd><code class="MIME" lang="en">]); |
| 75 |
|
$out->text ($input->{media_type}); |
| 76 |
|
$out->html (qq[</code> ]); |
| 77 |
|
if ($input->{media_type_overridden}) { |
| 78 |
|
$out->html ('<em>(overridden)</em>'); |
| 79 |
|
} elsif (defined $input->{official_type}) { |
| 80 |
|
if ($input->{media_type} eq $input->{official_type}) { |
| 81 |
|
# |
| 82 |
|
} else { |
| 83 |
|
$out->html ('<em>(sniffed; official type is: <code class=MIME lang=en>'); |
| 84 |
|
$out->text ($input->{official_type}); |
| 85 |
|
$out->html ('</code>)'); |
| 86 |
|
} |
| 87 |
|
} else { |
| 88 |
|
$out->html ('<em>(sniffed)</em>'); |
| 89 |
|
} |
| 90 |
|
$out->html (q[<dt>Character Encoding<dd>]); |
| 91 |
|
if (defined $input->{charset}) { |
| 92 |
|
$out->html ('<code class="charset" lang="en">'); |
| 93 |
|
$out->text ($input->{charset}); |
| 94 |
|
$out->html ('</code>'); |
| 95 |
|
} else { |
| 96 |
|
$out->text ('(none)'); |
| 97 |
|
} |
| 98 |
|
$out->html (' <em>overridden</em>') if $input->{charset_overridden}; |
| 99 |
|
$out->html (qq[ |
| 100 |
|
<dt>Length</dt> |
| 101 |
|
<dd>$char_length byte@{[$char_length == 1 ? '' : 's']}</dd> |
| 102 |
|
</dl> |
| 103 |
|
|
| 104 |
|
<script src="../cc-script.js"></script> |
| 105 |
|
]); |
| 106 |
|
$out->end_section; |
| 107 |
|
|
| 108 |
|
my $result = WebHACC::Result->new; |
| 109 |
|
$result->output ($out); |
| 110 |
|
$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 |
require Encode; |
my $result = WebHACC::Result->new; |
| 119 |
require Whatpm::HTML; |
$result->output ($out); |
| 120 |
|
$result->{conforming_min} = 0; |
| 121 |
|
$result->{conforming_max} = 1; |
| 122 |
|
|
| 123 |
$time1 = time; |
$input->generate_transfer_sections ($result); |
| 124 |
$s = Encode::decode ('utf-8', $s); |
$result->generate_result_section; |
|
$time2 = time; |
|
|
$time{decode} = $time2 - $time1; |
|
|
|
|
|
|
|
|
print STDOUT "#errors\n"; |
|
|
|
|
|
my $onerror = sub { |
|
|
my (%opt) = @_; |
|
|
print STDOUT "$opt{line},$opt{column},$opt{type}\n"; |
|
|
}; |
|
|
|
|
|
$doc = $dom->create_document; |
|
|
$time1 = time; |
|
|
if (length $mode[1]) { |
|
|
$el = $doc->create_element_ns |
|
|
('http://www.w3.org/1999/xhtml', [undef, $mode[1]]); |
|
|
Whatpm::HTML->set_inner_html ($el, $s, $onerror); |
|
|
} else { |
|
|
Whatpm::HTML->parse_string ($s => $doc, $onerror); |
|
| 125 |
} |
} |
|
$time2 = time; |
|
|
$time{parse} = $time2 - $time1; |
|
| 126 |
|
|
| 127 |
print "#document\n"; |
$out->nav_list; |
| 128 |
|
|
|
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', $$out); |
|
|
print STDOUT "\n"; |
|
|
} elsif (@mode == 3 and $mode[0] eq 'xhtml' and |
|
|
($mode[2] eq 'html' or $mode[2] eq 'test')) { |
|
|
print STDOUT "Content-Type: text/plain; charset=utf-8\n\n"; |
|
|
|
|
|
require Message::DOM::XMLParserTemp; |
|
|
print STDOUT "#errors\n"; |
|
|
|
|
|
my $onerror = sub { |
|
|
my $err = shift; |
|
|
print STDOUT $err->location->line_number, ","; |
|
|
print STDOUT $err->location->column_number, ","; |
|
|
print STDOUT $err->text, "\n"; |
|
|
return 1; |
|
|
}; |
|
|
|
|
|
open my $fh, '<', \$s; |
|
|
my $time1 = time; |
|
|
$doc = Message::DOM::XMLParserTemp->parse_byte_stream |
|
|
($fh => $dom, $onerror, charset => 'utf-8'); |
|
|
my $time2 = time; |
|
|
$time{parse_xml} = $time2 - $time1; |
|
|
|
|
|
print "#document\n"; |
|
|
|
|
|
my $out; |
|
|
if ($mode[2] eq 'html') { |
|
|
## TODO: Use XHTML serializer |
|
|
#$out = Whatpm::HTML->get_inner_html ($doc); |
|
|
} else { # test |
|
|
$time1 = time; |
|
|
$out = test_serialize ($doc); |
|
|
$time2 = time; |
|
|
$time{serialize_test} = $time2 - $time1; |
|
|
} |
|
|
print STDOUT Encode::encode ('utf-8', $$out); |
|
|
print STDOUT "\n"; |
|
|
} else { |
|
|
print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n404"; |
|
| 129 |
exit; |
exit; |
| 130 |
} |
} |
| 131 |
|
|
| 132 |
if ($http->parameter ('dom5')) { |
sub check_and_print ($$$) { |
| 133 |
require Whatpm::ContentChecker; |
my ($input, $result, $out) = @_; |
| 134 |
my $onerror = sub { |
my $original_input = $out->input; |
| 135 |
my %opt = @_; |
$out->input ($input); |
| 136 |
print STDOUT get_node_path ($opt{node}) . ';' . $opt{type} . "\n"; |
|
| 137 |
}; |
$input->generate_transfer_sections ($result); |
| 138 |
print STDOUT "#domerrors\n"; |
|
| 139 |
$time1 = time; |
my @subdoc; |
| 140 |
if ($el) { |
|
| 141 |
Whatpm::ContentChecker->check_element ($el, $onerror); |
my $checker_class = { |
| 142 |
} else { |
'text/cache-manifest' => 'WebHACC::Language::CacheManifest', |
| 143 |
Whatpm::ContentChecker->check_document ($doc, $onerror); |
'text/css' => 'WebHACC::Language::CSS', |
| 144 |
} |
'text/html' => 'WebHACC::Language::HTML', |
| 145 |
$time2 = time; |
'text/x-webidl' => 'WebHACC::Language::WebIDL', |
| 146 |
$time{check} = $time2 - $time1; |
|
| 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) { |
| 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[ |
| 212 |
|
<dl> |
| 213 |
|
<dt>Internet Media Type</dt> |
| 214 |
|
<dd><code class="MIME" lang="en">@{[htescape $subinput->{media_type}]}</code> |
| 215 |
|
<dt>Container Node</dt> |
| 216 |
|
<dd>@{[get_node_link ($input, $subinput->{container_node})]}</dd> |
| 217 |
|
<dt>Base <abbr title="Uniform Resource Identifiers">URI</abbr></dt> |
| 218 |
|
<dd><code class=URI><<a href="$ebaseuri">$ebaseuri</a>></code></dd> |
| 219 |
|
</dl>]; |
| 220 |
|
|
| 221 |
|
$subinput->{id_prefix} .= '-'; |
| 222 |
|
check_and_print ($subinput => $result => $out); |
| 223 |
|
|
| 224 |
|
$out->end_section; |
| 225 |
} |
} |
| 226 |
|
|
| 227 |
print STDOUT "#log\n"; |
$out->input ($original_input); |
| 228 |
for (qw/decode parse parse_xml serialize_html serialize_xml serialize_test |
} # check_and_print |
| 229 |
check/) { |
|
| 230 |
next unless defined $time{$_}; |
sub print_table_section ($$) { |
| 231 |
print STDOUT { |
my ($input, $tables) = @_; |
| 232 |
decode => 'bytes->chars', |
|
| 233 |
parse => 'html5(chars)->dom5', |
# push @nav, [qq[#$input->{id_prefix}tables] => 'Tables'] |
| 234 |
parse_xml => 'xml1(chars)->dom5', |
# unless $input->{nested}; |
| 235 |
serialize_html => 'dom5->html5(char)', |
print STDOUT qq[ |
| 236 |
serialize_xml => 'dom5->xml1(char)', |
<div id="$input->{id_prefix}tables" class="section"> |
| 237 |
serialize_test => 'dom5->test(char)', |
<h2>Tables</h2> |
| 238 |
check => 'dom5 check', |
|
| 239 |
}->{$_}; |
<!--[if IE]><script type="text/javascript" src="../excanvas.js"></script><![endif]--> |
| 240 |
print STDOUT "\t", $time{$_}, "s\n"; |
<script src="../table-script.js" type="text/javascript"></script> |
| 241 |
open my $file, '>>', ".manakai-$_.txt" or die ".manakai-$_.txt: $!"; |
<noscript> |
| 242 |
print $file $char_length, "\t", $time{$_}, "\n"; |
<p><em>Structure of tables are visualized here if scripting is enabled.</em></p> |
| 243 |
} |
</noscript> |
| 244 |
|
]; |
| 245 |
exit; |
|
| 246 |
|
require JSON; |
| 247 |
sub test_serialize ($) { |
|
| 248 |
my $node = shift; |
my $i = 0; |
| 249 |
my $r = ''; |
for my $table (@$tables) { |
| 250 |
|
$i++; |
| 251 |
my @node = map { [$_, ''] } @{$node->child_nodes}; |
print STDOUT qq[<div class="section" id="$input->{id_prefix}table-$i"><h3>] . |
| 252 |
while (@node) { |
get_node_link ($input, $table->{element}) . q[</h3>]; |
| 253 |
my $child = shift @node; |
|
| 254 |
my $nt = $child->[0]->node_type; |
delete $table->{element}; |
| 255 |
if ($nt == $child->[0]->ELEMENT_NODE) { |
|
| 256 |
$r .= '| ' . $child->[1] . '<' . $child->[0]->tag_name . ">\x0A"; ## ISSUE: case? |
for (@{$table->{column_group}}, @{$table->{column}}, $table->{caption}, |
| 257 |
|
@{$table->{row}}) { |
| 258 |
for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value] } |
next unless $_; |
| 259 |
@{$child->[0]->attributes}) { |
delete $_->{element}; |
| 260 |
$r .= '| ' . $child->[1] . ' ' . $attr->[0] . '="'; ## ISSUE: case? |
} |
| 261 |
$r .= $attr->[1] . '"' . "\x0A"; |
|
| 262 |
|
for (@{$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}; |
| 276 |
|
$_->{is_header} = $_->{is_header} ? 1 : 0; |
| 277 |
|
} |
| 278 |
} |
} |
|
|
|
|
unshift @node, |
|
|
map { [$_, $child->[1] . ' '] } @{$child->[0]->child_nodes}; |
|
|
} elsif ($nt == $child->[0]->TEXT_NODE) { |
|
|
$r .= '| ' . $child->[1] . '"' . $child->[0]->data . '"' . "\x0A"; |
|
|
} elsif ($nt == $child->[0]->CDATA_SECTION_NODE) { |
|
|
$r .= '| ' . $child->[1] . '<![CDATA[' . $child->[0]->data . "]]>\x0A"; |
|
|
} elsif ($nt == $child->[0]->COMMENT_NODE) { |
|
|
$r .= '| ' . $child->[1] . '<!-- ' . $child->[0]->data . " -->\x0A"; |
|
|
} elsif ($nt == $child->[0]->DOCUMENT_TYPE_NODE) { |
|
|
$r .= '| ' . $child->[1] . '<!DOCTYPE ' . $child->[0]->name . ">\x0A"; |
|
|
} elsif ($nt == $child->[0]->PROCESSING_INSTRUCTION_NODE) { |
|
|
$r .= '| ' . $child->[1] . '<?' . $child->[0]->target . ' ' . |
|
|
$child->[0]->data . "?>\x0A"; |
|
|
} else { |
|
|
$r .= '| ' . $child->[1] . $child->[0]->node_type . "\x0A"; # error |
|
| 279 |
} |
} |
| 280 |
|
|
| 281 |
|
print STDOUT '</div><script type="text/javascript">tableToCanvas ('; |
| 282 |
|
print STDOUT JSON::objToJson ($table); |
| 283 |
|
print STDOUT qq[, document.getElementById ('$input->{id_prefix}table-$i')]; |
| 284 |
|
print STDOUT qq[, '$input->{id_prefix}');</script>]; |
| 285 |
} |
} |
| 286 |
|
|
| 287 |
return \$r; |
print STDOUT qq[</div>]; |
| 288 |
} # test_serialize |
} # print_table_section |
| 289 |
|
|
| 290 |
|
sub print_listing_section ($$$) { |
| 291 |
|
my ($opt, $input, $ids) = @_; |
| 292 |
|
|
| 293 |
|
# push @nav, ['#' . $input->{id_prefix} . $opt->{id} => $opt->{label}] |
| 294 |
|
# unless $input->{nested}; |
| 295 |
|
print STDOUT qq[ |
| 296 |
|
<div id="$input->{id_prefix}$opt->{id}" class="section"> |
| 297 |
|
<h2>$opt->{heading}</h2> |
| 298 |
|
|
| 299 |
|
<dl> |
| 300 |
|
]; |
| 301 |
|
for my $id (sort {$a cmp $b} keys %$ids) { |
| 302 |
|
print STDOUT qq[<dt><code>@{[htescape $id]}</code></dt>]; |
| 303 |
|
for (@{$ids->{$id}}) { |
| 304 |
|
print STDOUT qq[<dd>].get_node_link ($input, $_).qq[</dd>]; |
| 305 |
|
} |
| 306 |
|
} |
| 307 |
|
print STDOUT qq[</dl></div>]; |
| 308 |
|
} # print_listing_section |
| 309 |
|
|
| 310 |
|
|
| 311 |
|
sub print_rdf_section ($$$) { |
| 312 |
|
my ($input, $rdfs) = @_; |
| 313 |
|
|
| 314 |
|
# push @nav, ['#' . $input->{id_prefix} . 'rdf' => 'RDF'] |
| 315 |
|
# unless $input->{nested}; |
| 316 |
|
print STDOUT qq[ |
| 317 |
|
<div id="$input->{id_prefix}rdf" class="section"> |
| 318 |
|
<h2>RDF Triples</h2> |
| 319 |
|
|
| 320 |
|
<dl>]; |
| 321 |
|
my $i = 0; |
| 322 |
|
for my $rdf (@$rdfs) { |
| 323 |
|
print STDOUT qq[<dt id="$input->{id_prefix}rdf-@{[$i++]}">]; |
| 324 |
|
print STDOUT get_node_link ($input, $rdf->[0]); |
| 325 |
|
print STDOUT qq[<dd><dl>]; |
| 326 |
|
for my $triple (@{$rdf->[1]}) { |
| 327 |
|
print STDOUT '<dt>' . get_node_link ($input, $triple->[0]) . '<dd>'; |
| 328 |
|
print STDOUT get_rdf_resource_html ($triple->[1]); |
| 329 |
|
print STDOUT ' '; |
| 330 |
|
print STDOUT get_rdf_resource_html ($triple->[2]); |
| 331 |
|
print STDOUT ' '; |
| 332 |
|
print STDOUT get_rdf_resource_html ($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 $r; |
| 361 |
|
} else { |
| 362 |
|
return '??'; |
| 363 |
|
} |
| 364 |
|
} # get_rdf_resource_html |
| 365 |
|
|
| 366 |
sub get_node_path ($) { |
{ |
| 367 |
my $node = shift; |
my $Msg = {}; |
| 368 |
my @r; |
|
| 369 |
while (defined $node) { |
sub load_text_catalog ($) { |
| 370 |
my $rs; |
# my $self = shift; |
| 371 |
if ($node->node_type == 1) { |
my $lang = shift; # MUST be a canonical lang name |
| 372 |
$rs = $node->manakai_local_name; |
open my $file, '<:utf8', "cc-msg.$lang.txt" |
| 373 |
$node = $node->parent_node; |
or die "$0: cc-msg.$lang.txt: $!"; |
| 374 |
} elsif ($node->node_type == 2) { |
while (<$file>) { |
| 375 |
$rs = '@' . $node->manakai_local_name; |
if (s/^([^;]+);([^;]*);//) { |
| 376 |
$node = $node->owner_element; |
my ($type, $cls, $msg) = ($1, $2, $_); |
| 377 |
} elsif ($node->node_type == 3) { |
$msg =~ tr/\x0D\x0A//d; |
| 378 |
$rs = '"' . $node->data . '"'; |
$Msg->{$type} = [$cls, $msg]; |
| 379 |
$node = $node->parent_node; |
} |
| 380 |
} elsif ($node->node_type == 9) { |
} |
| 381 |
$rs = ''; |
} # load_text_catalog |
| 382 |
$node = $node->parent_node; |
|
| 383 |
|
sub get_text ($;$$) { |
| 384 |
|
# my $self = shift; |
| 385 |
|
my ($type, $level, $node) = @_; |
| 386 |
|
$type = $level . ':' . $type if defined $level; |
| 387 |
|
$level = 'm' unless defined $level; |
| 388 |
|
my @arg; |
| 389 |
|
{ |
| 390 |
|
if (defined $Msg->{$type}) { |
| 391 |
|
my $msg = $Msg->{$type}->[1]; |
| 392 |
|
$msg =~ s{<var>\$([0-9]+)</var>}{ |
| 393 |
|
defined $arg[$1] ? ($arg[$1]) : '(undef)'; |
| 394 |
|
}ge; ##BUG: ^ must be escaped |
| 395 |
|
$msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{ |
| 396 |
|
UNIVERSAL::can ($node, 'get_attribute_ns') |
| 397 |
|
? ($node->get_attribute_ns (undef, $1)) : '' |
| 398 |
|
}ge; ## BUG: ^ must be escaped |
| 399 |
|
$msg =~ s{<var>{\@}</var>}{ ## BUG: v must be escaped |
| 400 |
|
UNIVERSAL::can ($node, 'value') ? ($node->value) : '' |
| 401 |
|
}ge; |
| 402 |
|
$msg =~ s{<var>{local-name}</var>}{ |
| 403 |
|
UNIVERSAL::can ($node, 'manakai_local_name') |
| 404 |
|
? ($node->manakai_local_name) : '' |
| 405 |
|
}ge; ## BUG: ^ must be escaped |
| 406 |
|
$msg =~ s{<var>{element-local-name}</var>}{ |
| 407 |
|
(UNIVERSAL::can ($node, 'owner_element') and |
| 408 |
|
$node->owner_element) |
| 409 |
|
? ($node->owner_element->manakai_local_name) |
| 410 |
|
: '' ## BUG: ^ must be escaped |
| 411 |
|
}ge; |
| 412 |
|
return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg); |
| 413 |
|
} elsif ($type =~ s/:([^:]*)$//) { |
| 414 |
|
unshift @arg, $1; |
| 415 |
|
redo; |
| 416 |
|
} |
| 417 |
|
} |
| 418 |
|
return ($type, 'level-'.$level, ($_[0])); |
| 419 |
|
## BUG: ^ must be escaped |
| 420 |
|
} # get_text |
| 421 |
|
|
| 422 |
|
} |
| 423 |
|
|
| 424 |
|
sub get_input_document ($$) { |
| 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 { |
} else { |
| 519 |
$rs = '#' . $node->node_type; |
$r->{uri} = $res->request->uri; |
| 520 |
$node = $node->parent_node; |
$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 |
} |
} |
|
unshift @r, $rs; |
|
| 570 |
} |
} |
| 571 |
return join '/', @r; |
|
| 572 |
} # get_node_path |
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 |
|
|
| 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. |
Parent Directory
| 
Revision Log
| 
Patch