| 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; | 
| 15 | my $s = $_[0]; |  | 
| 16 | $s =~ s/&/&/g; | require Message::DOM::DOMImplementation; | 
| 17 | $s =~ s/</</g; | my $dom = Message::DOM::DOMImplementation->new; | 
| 18 | $s =~ s/>/>/g; | { | 
| 19 | $s =~ s/"/"/g; | use Message::CGI::HTTP; | 
| 20 | $s =~ s!([\x00-\x09\x0B-\x1F\x7F-\x80])!sprintf '<var>U+%04X</var>', ord $1!ge; | my $http = Message::CGI::HTTP->new; | 
| 21 | return $s; |  | 
| 22 | } # htescape | if ($http->get_meta_variable ('PATH_INFO') ne '/') { | 
| 23 |  | print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400"; | 
|  | 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"; |  | 
| 24 | exit; | exit; | 
| 25 | } | } | 
| 26 |  |  | 
| 27 |  | 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 |  | ]); | 
| 45 |  |  | 
| 46 | <div id="document-info" class="section"> | my $input = get_input_document ($http, $dom); | 
|  | <dl> |  | 
|  | <dt>Document URI</dt> |  | 
|  | <dd><code class="URI" lang=""><<a href="@{[htescape $input_uri]}">@{[htescape $input_uri]}</a>></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']; |  | 
| 47 |  |  | 
| 48 | require Message::DOM::DOMImplementation; | $out->input ($input); | 
| 49 | my $dom = Message::DOM::DOMImplementation->____new; | $out->unset_flush; | 
|  | 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> |  | 
|  |  |  | 
|  | <dl> |  | 
|  | ]; |  | 
|  | push @nav, ['#parse-errors' => 'Parse Error']; |  | 
|  |  |  | 
|  | my $onerror = sub { |  | 
|  | my (%opt) = @_; |  | 
|  | if ($opt{column} > 0) { |  | 
|  | print STDOUT qq[<dt><a href="#line-$opt{line}">Line $opt{line}</a> column $opt{column}</dt>\n]; |  | 
|  | } else { |  | 
|  | $opt{line}--; |  | 
|  | print STDOUT qq[<dt><a href="#line-$opt{line}">Line $opt{line}</a></dt>\n]; |  | 
|  | } |  | 
|  | print STDOUT qq[<dd>@{[htescape $opt{type}]}</dd>\n]; |  | 
|  | }; |  | 
| 50 |  |  | 
| 51 | $doc = $dom->create_document; | my $result = WebHACC::Result->new; | 
| 52 | if (defined $inner_html_element and length $inner_html_element) { | $result->output ($out); | 
| 53 | $el = $doc->create_element_ns | $result->{conforming_min} = 1; | 
| 54 | ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]); | $result->{conforming_max} = 1; | 
| 55 | Whatpm::HTML->set_inner_html ($el, $s, $onerror); |  | 
| 56 | } else { | $out->html ('<script src="../cc-script.js"></script>'); | 
| 57 | Whatpm::HTML->parse_string ($s => $doc, $onerror); |  | 
| 58 |  | check_and_print ($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 | print STDOUT qq[ | my $checker_class = { | 
| 82 | </dl> | 'text/cache-manifest' => 'WebHACC::Language::CacheManifest', | 
| 83 | </div> | 'text/css' => 'WebHACC::Language::CSS', | 
| 84 | ]; | 'text/html' => 'WebHACC::Language::HTML', | 
| 85 | } elsif ($input_format eq 'application/xhtml+xml') { | 'text/x-webidl' => 'WebHACC::Language::WebIDL', | 
| 86 | require Message::DOM::XMLParserTemp; |  | 
| 87 | require Encode; | 'text/xml' => 'WebHACC::Language::XML', | 
| 88 |  | 'application/atom+xml' => 'WebHACC::Language::XML', | 
| 89 | my $t = Encode::decode ('utf-8', $s); | 'application/rss+xml' => 'WebHACC::Language::XML', | 
| 90 |  | 'image/svg+xml' => 'WebHACC::Language::XML', | 
| 91 | print STDOUT qq[ | 'application/xhtml+xml' => 'WebHACC::Language::XML', | 
| 92 | <dt>Character Encoding</dt> | 'application/xml' => 'WebHACC::Language::XML', | 
| 93 | <dd>(none)</dd> | ## TODO: Should we make all XML MIME Types fall | 
| 94 | </dl> | ## into this category? | 
| 95 | </div> |  | 
| 96 |  | ## NOTE: This type has different model from normal XML types. | 
| 97 | <div id="source-string" class="section"> | 'application/rdf+xml' => 'WebHACC::Language::XML', | 
| 98 | <h2>Document Source</h2> | }->{$input->{media_type}} || 'WebHACC::Language::Default'; | 
| 99 | ]; |  | 
| 100 | push @nav, ['#source-string' => 'Source']; | eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@"; | 
| 101 | print_source_string (\$t); | my $checker = $checker_class->new; | 
| 102 | print STDOUT qq[ | $checker->input ($input); | 
| 103 | </div> | $checker->output ($out); | 
| 104 |  | $checker->result ($result); | 
| 105 | <div id="parse-errors" class="section"> |  | 
| 106 | <h2>Parse Errors</h2> | ## TODO: A cache manifest MUST be text/cache-manifest | 
| 107 |  | ## TODO: WebIDL media type "text/x-webidl" | 
| 108 | <dl> |  | 
| 109 | ]; | $checker->generate_syntax_error_section; | 
| 110 | push @nav, ['#parse-errors' => 'Parse Error']; | $checker->generate_source_string_section; | 
| 111 |  |  | 
| 112 | my $onerror = sub { | my @subdoc; | 
| 113 | my $err = shift; | $checker->onsubdoc (sub { | 
| 114 | my $line = $err->location->line_number; | push @subdoc, shift; | 
| 115 | print STDOUT qq[<dt><a href="#line-$line">Line $line</a> column ]; | }); | 
| 116 | print STDOUT $err->location->column_number, "</dt><dd>"; |  | 
| 117 | print STDOUT htescape $err->text, "</dd>\n"; | $checker->generate_structure_dump_section; | 
| 118 | return 1; | $checker->generate_structure_error_section; | 
| 119 | }; | $checker->generate_additional_sections; | 
| 120 |  |  | 
| 121 | open my $fh, '<', \$s; | my $id_prefix = 0; | 
| 122 | $doc = Message::DOM::XMLParserTemp->parse_byte_stream | for my $_subinput (@subdoc) { | 
| 123 | ($fh => $dom, $onerror, charset => 'utf-8'); | my $subinput = WebHACC::Input::Subdocument->new (++$id_prefix); | 
| 124 |  | $subinput->{$_} = $_subinput->{$_} for keys %$_subinput; | 
| 125 | print STDOUT qq[ | $subinput->{base_uri} = $subinput->{container_node}->base_uri | 
| 126 | </dl> | unless defined $subinput->{base_uri}; | 
| 127 | </div> | $subinput->{parent_input} = $input; | 
| 128 | ]; |  | 
| 129 | } else { | $subinput->start_section ($result); | 
| 130 | print STDOUT qq[ | check_and_print ($subinput => $result => $out); | 
| 131 | </dl> | $subinput->end_section ($result); | 
|  | </div> |  | 
|  |  |  | 
|  | <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']; |  | 
| 132 | } | } | 
| 133 |  |  | 
| 134 |  | $out->input ($original_input); | 
| 135 |  | } # check_and_print | 
| 136 |  |  | 
|  | 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> |  | 
|  |  |  | 
|  | <dl> |  | 
|  | ]; |  | 
|  | push @nav, ['#document-errors' => 'Document Error']; |  | 
|  |  |  | 
|  | require Whatpm::ContentChecker; |  | 
|  | my $onerror = sub { |  | 
|  | my %opt = @_; |  | 
|  | print STDOUT qq[<dt><a href="#node-@{[refaddr $opt{node}]}">], |  | 
|  | htescape get_node_path ($opt{node}), |  | 
|  | "</a></dt>\n<dd>", htescape $opt{type}, "</dd>\n"; |  | 
|  | }; |  | 
| 137 |  |  | 
| 138 | if ($el) { | { | 
| 139 | Whatpm::ContentChecker->check_element ($el, $onerror); | my $Msg = {}; | 
| 140 | } else { |  | 
| 141 | Whatpm::ContentChecker->check_document ($doc, $onerror); | 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 | print STDOUT qq[ | sub get_text ($;$$) { | 
| 156 | </dl> | #  my $self = shift; | 
| 157 | </div> | my ($type, $level, $node) = @_; | 
| 158 | ]; | $type = $level . ':' . $type if defined $level; | 
| 159 | } | $level = 'm' unless defined $level; | 
| 160 |  | my @arg; | 
| 161 | ## TODO: Show result | { | 
| 162 |  | if (defined $Msg->{$type}) { | 
| 163 | print STDOUT qq[ | my $msg = $Msg->{$type}->[1]; | 
| 164 | <ul class="navigation" id="nav-items"> | $msg =~ s{<var>\$([0-9]+)</var>}{ | 
| 165 | ]; | defined $arg[$1] ? ($arg[$1]) : '(undef)'; | 
| 166 | for (@nav) { | }ge;                 ##BUG: ^ must be escaped | 
| 167 | print STDOUT qq[<li><a href="$_->[0]">$_->[1]</a></li>]; | $msg =~ s{<var>{\@([A-Za-z0-9:_.-]+)}</var>}{ | 
| 168 | } | UNIVERSAL::can ($node, 'get_attribute_ns') | 
| 169 | print STDOUT qq[ | ?  ($node->get_attribute_ns (undef, $1)) : '' | 
| 170 | </ul> | }ge; ## BUG: ^ must be escaped | 
| 171 | </body> | $msg =~ s{<var>{\@}</var>}{        ## BUG: v must be escaped | 
| 172 | </html> | UNIVERSAL::can ($node, 'value') ? ($node->value) : '' | 
| 173 | ]; | }ge; | 
| 174 |  | $msg =~ s{<var>{local-name}</var>}{ | 
| 175 | exit; | UNIVERSAL::can ($node, 'manakai_local_name') | 
| 176 |  | ? ($node->manakai_local_name) : '' | 
| 177 | sub print_source_string ($) { | }ge;  ## BUG: ^ must be escaped | 
| 178 | my $s = $_[0]; | $msg =~ s{<var>{element-local-name}</var>}{ | 
| 179 | my $i = 1; | (UNIVERSAL::can ($node, 'owner_element') and | 
| 180 | print STDOUT qq[<ol lang="">\n]; | $node->owner_element) | 
| 181 | while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { | ?  ($node->owner_element->manakai_local_name) | 
| 182 | print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n"; | : '' ## BUG: ^ must be escaped | 
| 183 | $i++; | }ge; | 
| 184 | } | return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg); | 
| 185 | if ($$s =~ /\G([^\x0A]+)/gc) { | } elsif ($type =~ s/:([^:]*)$//) { | 
| 186 | print STDOUT qq[<li id="line-$i">], htescape $1, "</li>\n"; | unshift @arg, $1; | 
| 187 | } | redo; | 
| 188 | print STDOUT "</ol>"; | } | 
| 189 | } # print_input_string | } | 
| 190 |  | return ($type, 'level-'.$level, ($_[0])); | 
| 191 | sub print_document_tree ($) { | ## BUG: ^ must be escaped | 
| 192 | my $node = shift; | } # get_text | 
| 193 | my $r = '<ol class="xoxo">'; |  | 
| 194 |  | } | 
| 195 | my @node = ($node); |  | 
| 196 | while (@node) { | sub get_input_document ($$) { | 
| 197 | my $child = shift @node; | my ($http, $dom) = @_; | 
| 198 | unless (ref $child) { |  | 
| 199 | $r .= $child; | require Encode; | 
| 200 | next; | 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 | my $node_id = 'node-'.refaddr $child; | require Message::Util::HostPermit; | 
| 214 | my $nt = $child->node_type; | my $host_permit = new Message::Util::HostPermit; | 
| 215 | if ($nt == $child->ELEMENT_NODE) { | $host_permit->add_rule (<<EOH); | 
| 216 | my $child_nsuri = $child->namespace_uri; | Allow host=suika port=80 | 
| 217 | $r .= qq[<li id="$node_id" class="tree-element"><code title="@{[defined $child_nsuri ? $child_nsuri : '']}">] . htescape ($child->tag_name) . | Deny host=suika | 
| 218 | '</code>'; ## ISSUE: case | Allow host=suika.fam.cx port=80 | 
| 219 |  | Deny host=suika.fam.cx | 
| 220 | if ($child->has_attributes) { | Deny host=localhost | 
| 221 | $r .= '<ul class="attributes">'; | Deny host=*.localdomain | 
| 222 | for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, $_->namespace_uri, 'node-'.refaddr $_] } | Deny ipv4=0.0.0.0/8 | 
| 223 | @{$child->attributes}) { | Deny ipv4=10.0.0.0/8 | 
| 224 | $r .= qq[<li id="$attr->[3]" class="tree-attribute"><code title="@{[defined $_->[2] ? $_->[2] : '']}">] . htescape ($attr->[0]) . '</code> = '; ## ISSUE: case? | Deny ipv4=127.0.0.0/8 | 
| 225 | $r .= '<q>' . htescape ($attr->[1]) . '</q></li>'; ## TODO: children | Deny ipv4=169.254.0.0/16 | 
| 226 | } | Deny ipv4=172.0.0.0/11 | 
| 227 | $r .= '</ul>'; | Deny ipv4=192.0.2.0/24 | 
| 228 | } | Deny ipv4=192.88.99.0/24 | 
| 229 |  | Deny ipv4=192.168.0.0/16 | 
| 230 |  | Deny ipv4=198.18.0.0/15 | 
| 231 |  | Deny ipv4=224.0.0.0/4 | 
| 232 |  | Deny ipv4=255.255.255.255/32 | 
| 233 |  | Deny ipv6=0::0/0 | 
| 234 |  | Allow host=* | 
| 235 |  | EOH | 
| 236 |  | unless ($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 $r; | 
| 242 |  | } | 
| 243 |  |  | 
| 244 | if ($node->has_child_nodes) { | require LWP::UserAgent; | 
| 245 | $r .= '<ol class="children">'; | my $ua = WDCC::LWPUA->new; | 
| 246 | unshift @node, @{$child->child_nodes}, '</ol>'; | $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} = $r->{charset}; | 
| 267 | } | } | 
| 268 | } elsif ($nt == $child->TEXT_NODE) { |  | 
| 269 | $r .= qq'<li id="$node_id" class="tree-text"><q lang="">' . htescape ($child->data) . '</q></li>'; | my $input_charset = $http->get_parameter ('charset'); | 
| 270 | } elsif ($nt == $child->CDATA_SECTION_NODE) { | if (defined $input_charset and length $input_charset) { | 
| 271 | $r .= qq'<li id="$node_id" class="tree-cdata"><code><[CDATA[</code><q lang="">' . htescape ($child->data) . '</q><code>]]></code></li>'; | $r->{charset_overridden} | 
| 272 | } elsif ($nt == $child->COMMENT_NODE) { | = (not defined $r->{charset} or $r->{charset} ne $input_charset); | 
| 273 | $r .= qq'<li id="$node_id" class="tree-comment"><code><!--</code><q lang="">' . htescape ($child->data) . '</q><code>--></code></li>'; | $r->{charset} = $input_charset; | 
|  | } elsif ($nt == $child->DOCUMENT_NODE) { |  | 
|  | $r .= qq'<li id="$node_id" class="tree-document">Document</li>'; |  | 
|  | if ($child->has_child_nodes) { |  | 
|  | $r .= '<ol>'; |  | 
|  | unshift @node, @{$child->child_nodes}, '</ol>'; |  | 
| 274 | } | } | 
| 275 | } elsif ($nt == $child->DOCUMENT_TYPE_NODE) { |  | 
| 276 | $r .= qq'<li id="$node_id" class="tree-doctype"><code><!DOCTYPE></code><ul>'; | ## TODO: Support for HTTP Content-Encoding | 
| 277 | $r .= '<li class="tree-doctype-name">Name = <q>@{[htescape ($child->name)]}</q></li>'; |  | 
| 278 | $r .= '<li class="tree-doctype-publicid">Public identifier = <q>@{[htescape ($child->public_id)]}</q></li>'; | $r->{s} = ''.$res->content; | 
| 279 | $r .= '<li class="tree-doctype-systemid">System identifier = <q>@{[htescape ($child->system_id)]}</q></li>'; |  | 
| 280 | $r .= '</ul></li>'; | require Whatpm::ContentType; | 
| 281 | } elsif ($nt == $child->PROCESSING_INSTRUCTION_NODE) { | ($r->{official_type}, $r->{media_type}) | 
| 282 | $r .= qq'<li id="$node_id" class="tree-id"><code><?@{[htescape ($child->target)]}</code> <q>@{[htescape ($child->data)]}</q><code>?></code></li>'; | = 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 { | } else { | 
| 291 | $r .= qq'<li id="$node_id" class="tree-unknown">@{[$child->node_type]} @{[htescape ($child->node_name)]}</li>'; # error | $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 | $r .= '</ol>'; | my $input_format = $http->get_parameter ('i'); | 
| 325 | print STDOUT $r; | if (defined $input_format and length $input_format) { | 
| 326 | } # print_document_tree | $r->{media_type_overridden} | 
| 327 |  | = (not defined $r->{media_type} or $input_format ne $r->{media_type}); | 
| 328 | sub get_node_path ($) { | $r->{media_type} = $input_format; | 
| 329 | my $node = shift; | } | 
| 330 | my @r; | if (defined $r->{s} and not defined $r->{media_type}) { | 
| 331 | while (defined $node) { | $r->{media_type} = 'text/html'; | 
| 332 | my $rs; | $r->{media_type_overridden} = 1; | 
| 333 | if ($node->node_type == 1) { | } | 
| 334 | $rs = $node->manakai_local_name; |  | 
| 335 | $node = $node->parent_node; | if ($r->{media_type} eq 'text/xml') { | 
| 336 | } elsif ($node->node_type == 2) { | unless (defined $r->{charset}) { | 
| 337 | $rs = '@' . $node->manakai_local_name; | $r->{charset} = 'us-ascii'; | 
| 338 | $node = $node->owner_element; | $r->{official_charset} = $r->{charset}; | 
| 339 | } elsif ($node->node_type == 3) { | } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') { | 
| 340 | $rs = '"' . $node->data . '"'; | $r->{charset_overridden} = 0; | 
|  | $node = $node->parent_node; |  | 
|  | } elsif ($node->node_type == 9) { |  | 
|  | $rs = ''; |  | 
|  | $node = $node->parent_node; |  | 
|  | } else { |  | 
|  | $rs = '#' . $node->node_type; |  | 
|  | $node = $node->parent_node; |  | 
| 341 | } | } | 
|  | unshift @r, $rs; |  | 
| 342 | } | } | 
| 343 | return join '/', @r; |  | 
| 344 | } # get_node_path | 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 | =head1 AUTHOR | 
| 379 |  |  | 
| 381 |  |  | 
| 382 | =head1 LICENSE | =head1 LICENSE | 
| 383 |  |  | 
| 384 | Copyright 2007 Wakaba <w@suika.fam.cx> | Copyright 2007-2008 Wakaba <w@suika.fam.cx> | 
| 385 |  |  | 
| 386 | This library is free software; you can redistribute it | This library is free software; you can redistribute it | 
| 387 | and/or modify it under the same terms as Perl itself. | and/or modify it under the same terms as Perl itself. |