| 1 |
#!/usr/bin/perl |
| 2 |
use strict; |
| 3 |
use utf8; |
| 4 |
|
| 5 |
use lib qw[/home/httpd/html/www/markup/html/whatpm |
| 6 |
/home/wakaba/work/manakai2/lib]; |
| 7 |
use CGI::Carp qw[fatalsToBrowser]; |
| 8 |
use Scalar::Util qw[refaddr]; |
| 9 |
|
| 10 |
require WebHACC::Input; |
| 11 |
require WebHACC::Result; |
| 12 |
require WebHACC::Output; |
| 13 |
|
| 14 |
my $out; |
| 15 |
|
| 16 |
require Message::DOM::DOMImplementation; |
| 17 |
my $dom = Message::DOM::DOMImplementation->new; |
| 18 |
{ |
| 19 |
use Message::CGI::HTTP; |
| 20 |
my $http = Message::CGI::HTTP->new; |
| 21 |
|
| 22 |
if ($http->get_meta_variable ('PATH_INFO') ne '/') { |
| 23 |
print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400"; |
| 24 |
exit; |
| 25 |
} |
| 26 |
|
| 27 |
load_text_catalog ('en'); ## TODO: conneg |
| 28 |
|
| 29 |
$out = WebHACC::Output->new; |
| 30 |
$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> |
| 36 |
<html lang="en"> |
| 37 |
<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 |
|
| 48 |
$out->input ($input); |
| 49 |
$out->unset_flush; |
| 50 |
|
| 51 |
my $result = WebHACC::Result->new; |
| 52 |
$result->output ($out); |
| 53 |
$result->{conforming_min} = 1; |
| 54 |
$result->{conforming_max} = 1; |
| 55 |
|
| 56 |
$out->html ('<script src="../cc-script.js"></script>'); |
| 57 |
|
| 58 |
check_and_print ($input => $result => $out); |
| 59 |
|
| 60 |
$result->generate_result_section; |
| 61 |
|
| 62 |
$out->nav_list; |
| 63 |
|
| 64 |
exit; |
| 65 |
} |
| 66 |
|
| 67 |
sub check_and_print ($$$) { |
| 68 |
my ($input, $result, $out) = @_; |
| 69 |
my $original_input = $out->input; |
| 70 |
$out->input ($input); |
| 71 |
|
| 72 |
$input->generate_info_section ($result); |
| 73 |
|
| 74 |
$input->generate_transfer_sections ($result); |
| 75 |
|
| 76 |
unless (defined $input->{s}) { |
| 77 |
$result->{conforming_min} = 0; |
| 78 |
return; |
| 79 |
} |
| 80 |
|
| 81 |
my $checker_class = { |
| 82 |
'text/cache-manifest' => 'WebHACC::Language::CacheManifest', |
| 83 |
'text/css' => 'WebHACC::Language::CSS', |
| 84 |
'text/html' => 'WebHACC::Language::HTML', |
| 85 |
'text/x-webidl' => 'WebHACC::Language::WebIDL', |
| 86 |
|
| 87 |
'text/xml' => 'WebHACC::Language::XML', |
| 88 |
'application/atom+xml' => 'WebHACC::Language::XML', |
| 89 |
'application/rss+xml' => 'WebHACC::Language::XML', |
| 90 |
'image/svg+xml' => 'WebHACC::Language::XML', |
| 91 |
'application/xhtml+xml' => 'WebHACC::Language::XML', |
| 92 |
'application/xml' => 'WebHACC::Language::XML', |
| 93 |
## TODO: Should we make all XML MIME Types fall |
| 94 |
## into this category? |
| 95 |
|
| 96 |
## NOTE: This type has different model from normal XML types. |
| 97 |
'application/rdf+xml' => 'WebHACC::Language::XML', |
| 98 |
}->{$input->{media_type}} || 'WebHACC::Language::Default'; |
| 99 |
|
| 100 |
eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@"; |
| 101 |
my $checker = $checker_class->new; |
| 102 |
$checker->input ($input); |
| 103 |
$checker->output ($out); |
| 104 |
$checker->result ($result); |
| 105 |
|
| 106 |
## TODO: A cache manifest MUST be text/cache-manifest |
| 107 |
## TODO: WebIDL media type "text/x-webidl" |
| 108 |
|
| 109 |
$checker->generate_syntax_error_section; |
| 110 |
$checker->generate_source_string_section; |
| 111 |
|
| 112 |
my @subdoc; |
| 113 |
$checker->onsubdoc (sub { |
| 114 |
push @subdoc, shift; |
| 115 |
}); |
| 116 |
|
| 117 |
$checker->generate_structure_dump_section; |
| 118 |
$checker->generate_structure_error_section; |
| 119 |
$checker->generate_additional_sections; |
| 120 |
|
| 121 |
my $id_prefix = 0; |
| 122 |
for my $_subinput (@subdoc) { |
| 123 |
my $subinput = WebHACC::Input::Subdocument->new (++$id_prefix); |
| 124 |
$subinput->{$_} = $_subinput->{$_} for keys %$_subinput; |
| 125 |
$subinput->{base_uri} = $subinput->{container_node}->base_uri |
| 126 |
unless defined $subinput->{base_uri}; |
| 127 |
$subinput->{parent_input} = $input; |
| 128 |
|
| 129 |
$subinput->start_section ($result); |
| 130 |
check_and_print ($subinput => $result => $out); |
| 131 |
$subinput->end_section ($result); |
| 132 |
} |
| 133 |
|
| 134 |
$out->input ($original_input); |
| 135 |
} # check_and_print |
| 136 |
|
| 137 |
|
| 138 |
{ |
| 139 |
my $Msg = {}; |
| 140 |
|
| 141 |
sub load_text_catalog ($) { |
| 142 |
# my $self = shift; |
| 143 |
my $lang = shift; # MUST be a canonical lang name |
| 144 |
open my $file, '<:utf8', "cc-msg.$lang.txt" |
| 145 |
or die "$0: cc-msg.$lang.txt: $!"; |
| 146 |
while (<$file>) { |
| 147 |
if (s/^([^;]+);([^;]*);//) { |
| 148 |
my ($type, $cls, $msg) = ($1, $2, $_); |
| 149 |
$msg =~ tr/\x0D\x0A//d; |
| 150 |
$Msg->{$type} = [$cls, $msg]; |
| 151 |
} |
| 152 |
} |
| 153 |
} # load_text_catalog |
| 154 |
|
| 155 |
sub get_text ($;$$) { |
| 156 |
# my $self = shift; |
| 157 |
my ($type, $level, $node) = @_; |
| 158 |
$type = $level . ':' . $type if defined $level; |
| 159 |
$level = 'm' unless defined $level; |
| 160 |
my @arg; |
| 161 |
{ |
| 162 |
if (defined $Msg->{$type}) { |
| 163 |
my $msg = $Msg->{$type}->[1]; |
| 164 |
$msg =~ s{<var>\$([0-9]+)</var>}{ |
| 165 |
defined $arg[$1] ? ($arg[$1]) : '(undef)'; |
| 166 |
}ge; ##BUG: ^ must be escaped |
| 167 |
$msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{ |
| 168 |
UNIVERSAL::can ($node, 'get_attribute_ns') |
| 169 |
? ($node->get_attribute_ns (undef, $1)) : '' |
| 170 |
}ge; ## BUG: ^ must be escaped |
| 171 |
$msg =~ s{<var>{\@}</var>}{ ## BUG: v must be escaped |
| 172 |
UNIVERSAL::can ($node, 'value') ? ($node->value) : '' |
| 173 |
}ge; |
| 174 |
$msg =~ s{<var>{local-name}</var>}{ |
| 175 |
UNIVERSAL::can ($node, 'manakai_local_name') |
| 176 |
? ($node->manakai_local_name) : '' |
| 177 |
}ge; ## BUG: ^ must be escaped |
| 178 |
$msg =~ s{<var>{element-local-name}</var>}{ |
| 179 |
(UNIVERSAL::can ($node, 'owner_element') and |
| 180 |
$node->owner_element) |
| 181 |
? ($node->owner_element->manakai_local_name) |
| 182 |
: '' ## BUG: ^ must be escaped |
| 183 |
}ge; |
| 184 |
return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg); |
| 185 |
} elsif ($type =~ s/:([^:]*)$//) { |
| 186 |
unshift @arg, $1; |
| 187 |
redo; |
| 188 |
} |
| 189 |
} |
| 190 |
return ($type, 'level-'.$level, ($_[0])); |
| 191 |
## BUG: ^ must be escaped |
| 192 |
} # get_text |
| 193 |
|
| 194 |
} |
| 195 |
|
| 196 |
sub get_input_document ($$) { |
| 197 |
my ($http, $dom) = @_; |
| 198 |
|
| 199 |
require Encode; |
| 200 |
my $request_uri = Encode::decode ('utf-8', $http->get_parameter ('uri')); |
| 201 |
my $r = WebHACC::Input->new; |
| 202 |
if (defined $request_uri and length $request_uri) { |
| 203 |
my $uri = $dom->create_uri_reference ($request_uri); |
| 204 |
unless ({ |
| 205 |
http => 1, |
| 206 |
}->{lc $uri->uri_scheme}) { |
| 207 |
$r = WebHACC::Input::Error->new; |
| 208 |
$r->{uri} = $request_uri; |
| 209 |
$r->{request_uri} = $request_uri; |
| 210 |
$r->{error_status_text} = 'URL scheme not allowed'; |
| 211 |
} |
| 212 |
|
| 213 |
require Message::Util::HostPermit; |
| 214 |
my $host_permit = new Message::Util::HostPermit; |
| 215 |
$host_permit->add_rule (<<EOH); |
| 216 |
Allow host=suika port=80 |
| 217 |
Deny host=suika |
| 218 |
Allow host=suika.fam.cx port=80 |
| 219 |
Deny host=suika.fam.cx |
| 220 |
Deny host=localhost |
| 221 |
Deny host=*.localdomain |
| 222 |
Deny ipv4=0.0.0.0/8 |
| 223 |
Deny ipv4=10.0.0.0/8 |
| 224 |
Deny ipv4=127.0.0.0/8 |
| 225 |
Deny ipv4=169.254.0.0/16 |
| 226 |
Deny ipv4=172.0.0.0/11 |
| 227 |
Deny ipv4=192.0.2.0/24 |
| 228 |
Deny ipv4=192.88.99.0/24 |
| 229 |
Deny ipv4=192.168.0.0/16 |
| 230 |
Deny ipv4=198.18.0.0/15 |
| 231 |
Deny ipv4=224.0.0.0/4 |
| 232 |
Deny ipv4=255.255.255.255/32 |
| 233 |
Deny ipv6=0::0/0 |
| 234 |
Allow host=* |
| 235 |
EOH |
| 236 |
unless ($host_permit->check ($uri->uri_host, $uri->uri_port || 80)) { |
| 237 |
my $r = WebHACC::Input::Error->new; |
| 238 |
$r->{uri} = $request_uri; |
| 239 |
$r->{request_uri} = $request_uri; |
| 240 |
$r->{error_status_text} = 'Connection to the host is forbidden'; |
| 241 |
return $r; |
| 242 |
} |
| 243 |
|
| 244 |
require LWP::UserAgent; |
| 245 |
my $ua = WDCC::LWPUA->new; |
| 246 |
$ua->{wdcc_dom} = $dom; |
| 247 |
$ua->{wdcc_host_permit} = $host_permit; |
| 248 |
$ua->agent ('Mozilla'); ## TODO: for now. |
| 249 |
$ua->parse_head (0); |
| 250 |
$ua->protocols_allowed ([qw/http/]); |
| 251 |
$ua->max_size (1000_000); |
| 252 |
my $req = HTTP::Request->new (GET => $request_uri); |
| 253 |
$req->header ('Accept-Encoding' => 'identity, *; q=0'); |
| 254 |
my $res = $ua->request ($req); |
| 255 |
## TODO: 401 sets |is_success| true. |
| 256 |
if ($res->is_success or $http->get_parameter ('error-page')) { |
| 257 |
$r->{base_uri} = $res->base; ## NOTE: It does check |Content-Base|, |Content-Location|, and <base>. ## TODO: Use our own code! |
| 258 |
$r->{uri} = $res->request->uri; |
| 259 |
$r->{request_uri} = $request_uri; |
| 260 |
|
| 261 |
## TODO: More strict parsing... |
| 262 |
my $ct = $res->header ('Content-Type'); |
| 263 |
if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) { |
| 264 |
$r->{charset} = lc $1; |
| 265 |
$r->{charset} =~ tr/\\//d; |
| 266 |
$r->{official_charset} = $r->{charset}; |
| 267 |
} |
| 268 |
|
| 269 |
my $input_charset = $http->get_parameter ('charset'); |
| 270 |
if (defined $input_charset and length $input_charset) { |
| 271 |
$r->{charset_overridden} |
| 272 |
= (not defined $r->{charset} or $r->{charset} ne $input_charset); |
| 273 |
$r->{charset} = $input_charset; |
| 274 |
} |
| 275 |
|
| 276 |
## TODO: Support for HTTP Content-Encoding |
| 277 |
|
| 278 |
$r->{s} = ''.$res->content; |
| 279 |
|
| 280 |
require Whatpm::ContentType; |
| 281 |
($r->{official_type}, $r->{media_type}) |
| 282 |
= Whatpm::ContentType->get_sniffed_type |
| 283 |
(get_file_head => sub { |
| 284 |
return substr $r->{s}, 0, shift; |
| 285 |
}, |
| 286 |
http_content_type_byte => $ct, |
| 287 |
has_http_content_encoding => |
| 288 |
defined $res->header ('Content-Encoding'), |
| 289 |
supported_image_types => {}); |
| 290 |
} else { |
| 291 |
$r->{uri} = $res->request->uri; |
| 292 |
$r->{request_uri} = $request_uri; |
| 293 |
$r->{error_status_text} = $res->status_line; |
| 294 |
} |
| 295 |
|
| 296 |
$r->{header_field} = []; |
| 297 |
$res->scan (sub { |
| 298 |
push @{$r->{header_field}}, [$_[0], $_[1]]; |
| 299 |
}); |
| 300 |
$r->{header_status_code} = $res->code; |
| 301 |
$r->{header_status_text} = $res->message; |
| 302 |
} else { |
| 303 |
$r->{s} = ''.$http->get_parameter ('s'); |
| 304 |
$r->{uri} = q<thismessage:/>; |
| 305 |
$r->{request_uri} = q<thismessage:/>; |
| 306 |
$r->{base_uri} = q<thismessage:/>; |
| 307 |
$r->{charset} = ''.$http->get_parameter ('_charset_'); |
| 308 |
$r->{charset} =~ s/\s+//g; |
| 309 |
$r->{charset} = 'utf-8' if $r->{charset} eq ''; |
| 310 |
$r->{official_charset} = $r->{charset}; |
| 311 |
$r->{header_field} = []; |
| 312 |
|
| 313 |
require Whatpm::ContentType; |
| 314 |
($r->{official_type}, $r->{media_type}) |
| 315 |
= Whatpm::ContentType->get_sniffed_type |
| 316 |
(get_file_head => sub { |
| 317 |
return substr $r->{s}, 0, shift; |
| 318 |
}, |
| 319 |
http_content_type_byte => undef, |
| 320 |
has_http_content_encoding => 0, |
| 321 |
supported_image_types => {}); |
| 322 |
} |
| 323 |
|
| 324 |
my $input_format = $http->get_parameter ('i'); |
| 325 |
if (defined $input_format and length $input_format) { |
| 326 |
$r->{media_type_overridden} |
| 327 |
= (not defined $r->{media_type} or $input_format ne $r->{media_type}); |
| 328 |
$r->{media_type} = $input_format; |
| 329 |
} |
| 330 |
if (defined $r->{s} and not defined $r->{media_type}) { |
| 331 |
$r->{media_type} = 'text/html'; |
| 332 |
$r->{media_type_overridden} = 1; |
| 333 |
} |
| 334 |
|
| 335 |
if ($r->{media_type} eq 'text/xml') { |
| 336 |
unless (defined $r->{charset}) { |
| 337 |
$r->{charset} = 'us-ascii'; |
| 338 |
$r->{official_charset} = $r->{charset}; |
| 339 |
} elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') { |
| 340 |
$r->{charset_overridden} = 0; |
| 341 |
} |
| 342 |
} |
| 343 |
|
| 344 |
if (length $r->{s} > 1000_000) { |
| 345 |
$r->{error_status_text} = 'Entity-body too large'; |
| 346 |
delete $r->{s}; |
| 347 |
return $r; |
| 348 |
} |
| 349 |
|
| 350 |
$r->{inner_html_element} = $http->get_parameter ('e'); |
| 351 |
|
| 352 |
return $r; |
| 353 |
} # get_input_document |
| 354 |
|
| 355 |
package WDCC::LWPUA; |
| 356 |
BEGIN { push our @ISA, 'LWP::UserAgent'; } |
| 357 |
|
| 358 |
sub redirect_ok { |
| 359 |
my $ua = shift; |
| 360 |
unless ($ua->SUPER::redirect_ok (@_)) { |
| 361 |
return 0; |
| 362 |
} |
| 363 |
|
| 364 |
my $uris = $_[1]->header ('Location'); |
| 365 |
return 0 unless $uris; |
| 366 |
my $uri = $ua->{wdcc_dom}->create_uri_reference ($uris); |
| 367 |
unless ({ |
| 368 |
http => 1, |
| 369 |
}->{lc $uri->uri_scheme}) { |
| 370 |
return 0; |
| 371 |
} |
| 372 |
unless ($ua->{wdcc_host_permit}->check ($uri->uri_host, $uri->uri_port || 80)) { |
| 373 |
return 0; |
| 374 |
} |
| 375 |
return 1; |
| 376 |
} # redirect_ok |
| 377 |
|
| 378 |
=head1 AUTHOR |
| 379 |
|
| 380 |
Wakaba <w@suika.fam.cx>. |
| 381 |
|
| 382 |
=head1 LICENSE |
| 383 |
|
| 384 |
Copyright 2007-2008 Wakaba <w@suika.fam.cx> |
| 385 |
|
| 386 |
This library is free software; you can redistribute it |
| 387 |
and/or modify it under the same terms as Perl itself. |
| 388 |
|
| 389 |
=cut |
| 390 |
|
| 391 |
## $Date: 2008/07/21 08:39:12 $ |
Parent Directory
| 
Revision Log