--- test/html-webhacc/cc.cgi 2007/06/27 14:36:45 1.4 +++ test/html-webhacc/cc.cgi 2008/02/10 02:42:01 1.33 @@ -1,13 +1,12 @@ #!/usr/bin/perl use strict; +use utf8; use lib qw[/home/httpd/html/www/markup/html/whatpm - /home/wakaba/work/manakai/lib - /home/wakaba/public_html/-temp/wiki/lib]; + /home/wakaba/work/manakai2/lib]; use CGI::Carp qw[fatalsToBrowser]; use Scalar::Util qw[refaddr]; - -use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module +use Time::HiRes qw/time/; sub htescape ($) { my $s = $_[0]; @@ -15,24 +14,28 @@ $s =~ s//>/g; $s =~ s/"/"/g; - $s =~ s!([\x00-\x09\x0B-\x1F\x7F-\x80])!sprintf 'U+%04X', ord $1!ge; + $s =~ s{([\x00-\x09\x0B-\x1F\x7F-\xA0\x{FEFF}\x{FFFC}-\x{FFFF}])}{ + sprintf 'U+%04X', ord $1; + }ge; return $s; } # htescape -my $http = SuikaWiki::Input::HTTP->new; - -## TODO: _charset_ - - my $input_format = $http->parameter ('i') || 'text/html'; - my $inner_html_element = $http->parameter ('e'); - my $input_uri = 'thismessage:/'; + use Message::CGI::HTTP; + my $http = Message::CGI::HTTP->new; - my $s = $http->parameter ('s'); - if (length $s > 1000_000) { - print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long"; + if ($http->get_meta_variable ('PATH_INFO') ne '/') { + print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400"; exit; } + binmode STDOUT, ':utf8'; + $| = 1; + + require Message::DOM::DOMImplementation; + my $dom = Message::DOM::DOMImplementation->new; + + load_text_catalog ('en'); ## TODO: conneg + my @nav; print STDOUT qq[Content-Type: text/html; charset=utf-8 @@ -43,99 +46,263 @@ -

Web Document Conformance Checker (beta)

Web Document Conformance Checker +(beta)

+]; + + $| = 0; + my $input = get_input_document ($http, $dom); + my $char_length = 0; + my %time; + print qq[

Request URI: <@{[htescape $input->{request_uri}]}>
Document URI: <@{[htescape $input_uri]}>
Internet Media Type: @{[htescape $input_format]}; <@{[htescape $input->{uri}]}> +

yet push @nav, ['#document-info' => 'Information']; - require Message::DOM::DOMImplementation; - my $dom = Message::DOM::DOMImplementation->____new; - my $doc; - my $el; +if (defined $input->{s}) { + $char_length = length $input->{s}; - if ($input_format eq 'text/html') { - require Encode; - require Whatpm::HTML; - - $s = Encode::decode ('utf-8', $s); - - print STDOUT qq[ + print STDOUT qq[ +

Base URI

<@{[htescape $input->{base_uri}]}>

Internet Media Type

@{[htescape $input->{media_type}]} + @{[$input->{media_type_overridden} ? '(overridden)' : defined $input->{official_type} ? $input->{media_type} eq $input->{official_type} ? '' : '(sniffed; official type is: '.htescape ($input->{official_type}).')' : '(sniffed)']}

Character Encoding
-
(none)
+
@{[defined $input->{charset} ? ''.htescape ($input->{charset}).'' : '(none)']} + @{[$input->{charset_overridden} ? '(overridden)' : '']}
+
Length
+
$char_length byte@{[$char_length == 1 ? '' : 's']}

+]; -
-
Document Source
+ my $result = {conforming_min => 1, conforming_max => 1}; + check_and_print ($input => $result); + print_result_section ($result); +} else { + print STDOUT qq[
]; + print_result_input_error_section ($input); +} + + print STDOUT qq[ +
+]; + for (@nav) { + print STDOUT qq[
$_->[1]
]; + } + print STDOUT qq[ +
+ + ]; - push @nav, ['#source-string' => 'Source']; - print_source_string (\$s); - print STDOUT qq[ - -
-
Parse Errors
+ for (qw/decode parse parse_html parse_xml parse_manifest + check check_manifest/) { + next unless defined $time{$_}; + open my $file, '>>', ".cc-$_.txt" or die ".cc-$_.txt: $!"; + print $file $char_length, "\t", $time{$_}, "\n"; + } -
+exit; + +sub add_error ($$$) { + my ($layer, $err, $result) = @_; + if (defined $err->{level}) { + if ($err->{level} eq 's') { + $result->{$layer}->{should}++; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + } elsif ($err->{level} eq 'w' or $err->{level} eq 'g') { + $result->{$layer}->{warning}++; + } elsif ($err->{level} eq 'u' or $err->{level} eq 'unsupported') { + $result->{$layer}->{unsupported}++; + $result->{unsupported} = 1; + } else { + $result->{$layer}->{must}++; + $result->{$layer}->{score_max} -= 2; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + $result->{conforming_max} = 0; + } + } else { + $result->{$layer}->{must}++; + $result->{$layer}->{score_max} -= 2; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + $result->{conforming_max} = 0; + } +} # add_error + +sub check_and_print ($$) { + my ($input, $result) = @_; + $input->{id_prefix} = ''; + #$input->{nested} = 1/0; + + print_http_header_section ($input, $result); + + my $doc; + my $el; + my $manifest; + + if ($input->{media_type} eq 'text/html') { + ($doc, $el) = print_syntax_error_html_section ($input, $result); + print_source_string_section + (\($input->{s}), $input->{charset} || $doc->input_encoding); + } elsif ({ + 'text/xml' => 1, + 'application/atom+xml' => 1, + 'application/rss+xml' => 1, + 'application/svg+xml' => 1, + 'application/xhtml+xml' => 1, + 'application/xml' => 1, + }->{$input->{media_type}}) { + ($doc, $el) = print_syntax_error_xml_section ($input, $result); + print_source_string_section (\($input->{s}), $doc->input_encoding); + } elsif ($input->{media_type} eq 'text/cache-manifest') { +## TODO: MUST be text/cache-manifest + $manifest = print_syntax_error_manifest_section ($input, $result); + print_source_string_section (\($input->{s}), 'utf-8'); + } else { + ## TODO: Change HTTP status code?? + print_result_unknown_type_section ($input, $result); + } + + if (defined $doc or defined $el) { + print_structure_dump_dom_section ($input, $doc, $el); + my $elements = print_structure_error_dom_section + ($input, $doc, $el, $result); + print_table_section ($input, $elements->{table}) if @{$elements->{table}}; + print_listing_section ({ + id => 'identifiers', label => 'IDs', heading => 'Identifiers', + }, $input, $elements->{id}) if keys %{$elements->{id}}; + print_listing_section ({ + id => 'terms', label => 'Terms', heading => 'Terms', + }, $input, $elements->{term}) if keys %{$elements->{term}}; + print_listing_section ({ + id => 'classes', label => 'Classes', heading => 'Classes', + }, $input, $elements->{class}) if keys %{$elements->{class}}; + } elsif (defined $manifest) { + print_structure_dump_manifest_section ($input, $manifest); + print_structure_error_manifest_section ($input, $manifest, $result); + } +} # check_and_print + +sub print_http_header_section ($$) { + my ($input, $result) = @_; + return unless defined $input->{header_status_code} or + defined $input->{header_status_text} or + @{$input->{header_field}}; + + push @nav, ['#source-header' => 'HTTP Header'] unless $input->{nested}; + print STDOUT qq[
+
HTTP Header
+ +
Note: Due to the limitation of the +network library in use, the content of this section might +not be the real header.
+ + ]; - push @nav, ['#parse-errors' => 'Parse Error']; + + if (defined $input->{header_status_code}) { + print STDOUT qq[]; + print STDOUT qq[]; + } + if (defined $input->{header_status_text}) { + print STDOUT qq[]; + print STDOUT qq[]; + } + + for (@{$input->{header_field}}) { + print STDOUT qq[]; + print STDOUT qq[]; + } + + print STDOUT qq[
Status code @{[htescape ($input->{header_status_code})]}
Status text @{[htescape ($input->{header_status_text})]}
@{[htescape ($_->[0])]} @{[htescape ($_->[1])]}
]; +} # print_http_header_section + +sub print_syntax_error_html_section ($$) { + my ($input, $result) = @_; + + require Encode; + require Whatpm::HTML; + + print STDOUT qq[ +
+
Parse Errors
+ +
]; + push @nav, ['#parse-errors' => 'Parse Error'] unless $input->{nested}; my $onerror = sub { my (%opt) = @_; + my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}); if ($opt{column} > 0) { - print STDOUT qq[
Line $opt{line} column $opt{column}
\n]; + print STDOUT qq[
Line $opt{line} column $opt{column}
\n]; } else { - $opt{line}--; - print STDOUT qq[
Line $opt{line}
\n]; + $opt{line} = $opt{line} - 1 || 1; + print STDOUT qq[
Line $opt{line}
\n]; } - print STDOUT qq[
@{[htescape $opt{type}]}
\n]; + $type =~ tr/ /-/; + $type =~ s/\|/%7C/g; + $msg .= qq[ [Description]]; + print STDOUT qq[
], get_error_level_label (\%opt); + print STDOUT qq[$msg
\n]; + + add_error ('syntax', \%opt => $result); }; - $doc = $dom->create_document; + my $doc = $dom->create_document; + my $el; + my $inner_html_element = $http->get_parameter ('e'); if (defined $inner_html_element and length $inner_html_element) { + $input->{charset} ||= 'windows-1252'; ## TODO: for now. + my $time1 = time; + my $t = Encode::decode ($input->{charset}, $input->{s}); + $time{decode} = time - $time1; + $el = $doc->create_element_ns ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]); - Whatpm::HTML->set_inner_html ($el, $s, $onerror); + $time1 = time; + Whatpm::HTML->set_inner_html ($el, $t, $onerror); + $time{parse} = time - $time1; } else { - Whatpm::HTML->parse_string ($s => $doc, $onerror); + my $time1 = time; + Whatpm::HTML->parse_byte_string + ($input->{charset}, $input->{s} => $doc, $onerror); + $time{parse_html} = time - $time1; } - + $doc->manakai_charset ($input->{official_charset}) + if defined $input->{official_charset}; + + print STDOUT qq[
]; + + return ($doc, $el); +} # print_syntax_error_html_section + +sub print_syntax_error_xml_section ($$) { + my ($input, $result) = @_; + + require Message::DOM::XMLParserTemp; + print STDOUT qq[ -
-
-]; - } elsif ($input_format eq 'application/xhtml+xml') { - require Message::DOM::XMLParserTemp; - require Encode; - - my $t = Encode::decode ('utf-8', $s); - - print STDOUT qq[ -
Character Encoding
-
(none)
- - - -
-
Document Source
-]; - push @nav, ['#source-string' => 'Source']; - print_source_string (\$t); - print STDOUT qq[ -
- -
+

Parse Errors
-
-]; - push @nav, ['#parse-errors' => 'Parse Error']; +
]; + push @nav, ['#parse-errors' => 'Parse Error'] unless $input->{prefix}; my $onerror = sub { my $err = shift; @@ -143,98 +310,92 @@ print STDOUT qq[
Line $line column ]; print STDOUT $err->location->column_number, "
"; print STDOUT htescape $err->text, "
\n"; - return 1; - }; - open my $fh, '<', \$s; - $doc = Message::DOM::XMLParserTemp->parse_byte_stream - ($fh => $dom, $onerror, charset => 'utf-8'); + add_error ('syntax', {type => $err->text, + level => [ + $err->SEVERITY_FATAL_ERROR => 'm', + $err->SEVERITY_ERROR => 'm', + $err->SEVERITY_WARNING => 's', + ]->[$err->severity]} => $result); - print STDOUT qq[ -
-
-]; - } else { - print STDOUT qq[ - -
+ return 1; + }; -
-
Media type @{[htescape $input_format]} is not supported!
-
-]; - push @nav, ['#result-summary' => 'Result']; - } + my $time1 = time; + open my $fh, '<', \($input->{s}); + my $doc = Message::DOM::XMLParserTemp->parse_byte_stream + ($fh => $dom, $onerror, charset => $input->{charset}); + $time{parse_xml} = time - $time1; + $doc->manakai_charset ($input->{official_charset}) + if defined $input->{official_charset}; + + print STDOUT qq[]; + return ($doc, undef); +} # print_syntax_error_xml_section - if (defined $doc or defined $el) { - print STDOUT qq[ -
-
Document Tree
-]; - push @nav, ['#document-tree' => 'Tree']; +sub print_syntax_error_manifest_section ($$) { + my ($input, $result) = @_; - print_document_tree ($el || $doc); + require Whatpm::CacheManifest; - print STDOUT qq[ -
+ print STDOUT qq[ +
+
Parse Errors
-
-
Document Errors
+
]; + push @nav, ['#parse-errors' => 'Parse Error'] unless $input->{nested}; -
-]; - push @nav, ['#document-errors' => 'Document Error']; + my $onerror = sub { + my (%opt) = @_; + my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}); + print STDOUT qq[
], get_error_label ($input, \%opt), + qq[
]; + $type =~ tr/ /-/; + $type =~ s/\|/%7C/g; + $msg .= qq[ [Description]]; + print STDOUT qq[
], get_error_level_label (\%opt); + print STDOUT qq[$msg
\n]; - require Whatpm::ContentChecker; - my $onerror = sub { - my %opt = @_; - print STDOUT qq[
], - htescape get_node_path ($opt{node}), - "
\n
", htescape $opt{type}, "
\n"; - }; + add_error ('syntax', \%opt => $result); + }; - if ($el) { - Whatpm::ContentChecker->check_element ($el, $onerror); - } else { - Whatpm::ContentChecker->check_document ($doc, $onerror); + my $time1 = time; + my $manifest = Whatpm::CacheManifest->parse_byte_string + ($input->{s}, $input->{uri}, $input->{base_uri}, $onerror); + $time{parse_manifest} = time - $time1; + + print STDOUT qq[
]; + + return $manifest; +} # print_syntax_error_manifest_section + +sub print_source_string_section ($$) { + require Encode; + my $enc = Encode::find_encoding ($_[1]); ## TODO: charset name -> Perl name + return unless $enc; + + my $s = \($enc->decode (${$_[0]})); + my $i = 1; + push @nav, ['#source-string' => 'Source'] unless $input->{nested}; + print STDOUT qq[
+
Document Source
+
\n]; + if (length $$s) { + while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { + print STDOUT qq[
], htescape $1, + "
\n"; + $i++; } - - print STDOUT qq[ - -
-]; - } - - ## TODO: Show result - - print STDOUT qq[ -
-]; - for (@nav) { - print STDOUT qq[
$_->[1]
]; - } - print STDOUT qq[ -
- - -]; - -exit; - -sub print_source_string ($) { - my $s = $_[0]; - my $i = 1; - print STDOUT qq[
\n]; - while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { - print STDOUT qq[
], htescape $1, "
\n"; - $i++; - } - if ($$s =~ /\G([^\x0A]+)/gc) { - print STDOUT qq[
], htescape $1, "
\n"; + if ($$s =~ /\G([^\x0A]+)/gc) { + print STDOUT qq[
], htescape $1, + "
\n"; + } + } else { + print STDOUT q[
]; } - print STDOUT "
"; -} # print_input_string + print STDOUT "
"; +} # print_input_string_section sub print_document_tree ($) { my $node = shift; @@ -248,7 +409,7 @@ next; } - my $node_id = 'node-'.refaddr $child; + my $node_id = $input->{id_prefix} . 'node-'.refaddr $child; my $nt = $child->node_type; if ($nt == $child->ELEMENT_NODE) { my $child_nsuri = $child->namespace_uri; @@ -259,15 +420,17 @@ $r .= '
'; for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, $_->namespace_uri, 'node-'.refaddr $_] } @{$child->attributes}) { - $r .= qq[
] . htescape ($attr->[0]) . ' = '; ## ISSUE: case? + $r .= qq[
] . htescape ($attr->[0]) . ' = '; ## ISSUE: case? $r .= '' . htescape ($attr->[1]) . '
'; ## TODO: children } $r .= '
'; } - if ($node->has_child_nodes) { + if ($child->has_child_nodes) { $r .= '
'; - unshift @node, @{$child->child_nodes}, '
'; + unshift @node, @{$child->child_nodes}, ''; + } else { + $r .= ''; } } elsif ($nt == $child->TEXT_NODE) { $r .= qq'
' . htescape ($child->data) . '
'; @@ -276,16 +439,44 @@ } elsif ($nt == $child->COMMENT_NODE) { $r .= qq'

'; } elsif ($nt == $child->DOCUMENT_NODE) { - $r .= qq'
Document
'; + $r .= qq'
Document'; + $r .= qq[
]; + my $cp = $child->manakai_charset; + if (defined $cp) { + $r .= qq[
charset parameter = ]; + $r .= htescape ($cp) . qq[
]; + } + $r .= qq[
inputEncoding = ]; + my $ie = $child->input_encoding; + if (defined $ie) { + $r .= qq[@{[htescape ($ie)]}]; + if ($child->manakai_has_bom) { + $r .= qq[ (with BOM)]; + } + } else { + $r .= qq[(null)]; + } + $r .= qq[
@{[scalar get_text ('manakaiIsHTML:'.($child->manakai_is_html?1:0))]}
]; + $r .= qq[
@{[scalar get_text ('manakaiCompatMode:'.$child->manakai_compat_mode)]}
]; + unless ($child->manakai_is_html) { + $r .= qq[
XML version = @{[htescape ($child->xml_version)]}
]; + if (defined $child->xml_encoding) { + $r .= qq[
XML encoding = @{[htescape ($child->xml_encoding)]}
]; + } else { + $r .= qq[
XML encoding = (null)
]; + } + $r .= qq[
XML standalone = @{[$child->xml_standalone ? 'true' : 'false']}
]; + } + $r .= qq[
]; if ($child->has_child_nodes) { - $r .= '
'; - unshift @node, @{$child->child_nodes}, '
'; + $r .= '
'; + unshift @node, @{$child->child_nodes}, '
'; } } elsif ($nt == $child->DOCUMENT_TYPE_NODE) { - $r .= qq'
<!DOCTYPE>
'; - $r .= '
Name = @{[htescape ($child->name)]}
'; - $r .= '
Public identifier = @{[htescape ($child->public_id)]}
'; - $r .= '
System identifier = @{[htescape ($child->system_id)]}
'; + $r .= qq'
<!DOCTYPE>
'; + $r .= qq[
Name = @{[htescape ($child->name)]}
]; + $r .= qq[
Public identifier = @{[htescape ($child->public_id)]}
]; + $r .= qq[
System identifier = @{[htescape ($child->system_id)]}
]; $r .= '
'; } elsif ($nt == $child->PROCESSING_INSTRUCTION_NODE) { $r .= qq'
<?@{[htescape ($child->target)]} @{[htescape ($child->data)]}?>
'; @@ -298,6 +489,374 @@ print STDOUT $r; } # print_document_tree +sub print_structure_dump_dom_section ($$$) { + my ($input, $doc, $el) = @_; + + print STDOUT qq[ +
+
Document Tree
+]; + push @nav, ['#document-tree' => 'Tree'] unless $input->{nested}; + + print_document_tree ($el || $doc); + + print STDOUT qq[
]; +} # print_structure_dump_dom_section + +sub print_structure_dump_manifest_section ($$) { + my ($input, $manifest) = @_; + + print STDOUT qq[ +
+
Cache Manifest
+]; + push @nav, ['#dump-manifest' => 'Caceh Manifest'] unless $input->{nested}; + + print STDOUT qq[
Explicit entries
]; + for my $uri (@{$manifest->[0]}) { + my $euri = htescape ($uri); + print STDOUT qq[
<$euri>
]; + } + + print STDOUT qq[
Fallback entries
+ + ]; + for my $uri (sort {$a cmp $b} keys %{$manifest->[1]}) { + my $euri = htescape ($uri); + my $euri2 = htescape ($manifest->[1]->{$uri}); + print STDOUT qq[ + ]; + } + + print STDOUT qq[
Oppotunistic Caching Namespace Fallback Entry
<$euri> <$euri2>
Online whitelist
]; + for my $uri (@{$manifest->[2]}) { + my $euri = htescape ($uri); + print STDOUT qq[
<$euri>
]; + } + + print STDOUT qq[
]; +} # print_structure_dump_manifest_section + +sub print_structure_error_dom_section ($$$$) { + my ($input, $doc, $el, $result) = @_; + + print STDOUT qq[
+
Document Errors
+ +
]; + push @nav, ['#document-errors' => 'Document Error'] unless $input->{nested}; + + require Whatpm::ContentChecker; + my $onerror = sub { + my %opt = @_; + my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}, $opt{node}); + $type =~ tr/ /-/; + $type =~ s/\|/%7C/g; + $msg .= qq[ [Description]]; + print STDOUT qq[
] . get_error_label ($input, \%opt) . + qq[
\n
], get_error_level_label (\%opt); + print STDOUT $msg, "
\n"; + add_error ('structure', \%opt => $result); + }; + + my $elements; + my $time1 = time; + if ($el) { + $elements = Whatpm::ContentChecker->check_element ($el, $onerror); + } else { + $elements = Whatpm::ContentChecker->check_document ($doc, $onerror); + } + $time{check} = time - $time1; + + print STDOUT qq[
]; + + return $elements; +} # print_structure_error_dom_section + +sub print_structure_error_manifest_section ($$$) { + my ($input, $manifest, $result) = @_; + + print STDOUT qq[
+
Document Errors
+ +
]; + push @nav, ['#document-errors' => 'Document Error'] unless $input->{nested}; + + require Whatpm::CacheManifest; + Whatpm::CacheManifest->check_manifest ($manifest, sub { + my %opt = @_; + my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}, $opt{node}); + $type =~ tr/ /-/; + $type =~ s/\|/%7C/g; + $msg .= qq[ [Description]]; + print STDOUT qq[
] . get_error_label ($input, \%opt) . + qq[
\n
], $msg, "
\n"; + add_error ('structure', \%opt => $result); + }); + + print STDOUT qq[
]; +} # print_structure_error_manifest_section + +sub print_table_section ($$) { + my ($input, $tables) = @_; + + push @nav, ['#tables' => 'Tables'] unless $input->{nested}; + print STDOUT qq[ +
+
Tables
+ + + + +]; + + require JSON; + + my $i = 0; + for my $table_el (@$tables) { + $i++; + print STDOUT qq[
] . + get_node_link ($input, $table_el) . q[
]; + + ## TODO: Make |ContentChecker| return |form_table| result + ## so that this script don't have to run the algorithm twice. + my $table = Whatpm::HTMLTable->form_table ($table_el); + + for (@{$table->{column_group}}, @{$table->{column}}, $table->{caption}) { + next unless $_; + delete $_->{element}; + } + + for (@{$table->{row_group}}) { + next unless $_; + next unless $_->{element}; + $_->{type} = $_->{element}->manakai_local_name; + delete $_->{element}; + } + + for (@{$table->{cell}}) { + next unless $_; + for (@{$_}) { + next unless $_; + for (@$_) { + $_->{id} = refaddr $_->{element} if defined $_->{element}; + delete $_->{element}; + $_->{is_header} = $_->{is_header} ? 1 : 0; + } + } + } + + print STDOUT '
]; + } + + print STDOUT qq[
]; +} # print_table_section + +sub print_listing_section ($$$) { + my ($opt, $input, $ids) = @_; + + push @nav, ['#' . $opt->{id} => $opt->{label}] unless $input->{nested}; + print STDOUT qq[ +
+
$opt->{heading}
+ +
+]; + for my $id (sort {$a cmp $b} keys %$ids) { + print STDOUT qq[
@{[htescape $id]}
]; + for (@{$ids->{$id}}) { + print STDOUT qq[
].get_node_link ($input, $_).qq[
]; + } + } + print STDOUT qq[
]; +} # print_listing_section + +sub print_result_section ($) { + my $result = shift; + + print STDOUT qq[ +
+
Result
]; + + if ($result->{unsupported} and $result->{conforming_max}) { + print STDOUT qq[
The conformance + checker cannot decide whether the document is conforming or + not, since the document contains one or more unsupported + features. The document might or might not be conforming.
]; + } elsif ($result->{conforming_min}) { + print STDOUT qq[
No conformance-error is + found in this document.
]; + } elsif ($result->{conforming_max}) { + print STDOUT qq[
This document + is likely non-conforming, but in rare case + it might be conforming.
]; + } else { + print STDOUT qq[
This document is + non-conforming.
]; + } + + print STDOUT qq[ ++ + + + + +]; + + my $must_error = 0; + my $should_error = 0; + my $warning = 0; + my $score_min = 0; + my $score_max = 0; + my $score_base = 20; + my $score_unit = $score_base / 100; + for ( + [Transfer => 'transfer', ''], + [Character => 'char', ''], + [Syntax => 'syntax', '#parse-errors'], + [Structure => 'structure', '#document-errors'], + ) { + $must_error += ($result->{$_->[1]}->{must} += 0); + $should_error += ($result->{$_->[1]}->{should} += 0); + $warning += ($result->{$_->[1]}->{warning} += 0); + $score_min += (($result->{$_->[1]}->{score_min} *= $score_unit) += $score_base); + $score_max += (($result->{$_->[1]}->{score_max} *= $score_unit) += $score_base); + + my $uncertain = $result->{$_->[1]}->{unsupported} ? '?' : ''; + my $label = $_->[0]; + if ($result->{$_->[1]}->{must} or + $result->{$_->[1]}->{should} or + $result->{$_->[1]}->{warning} or + $result->{$_->[1]}->{unsupported}) { + $label = qq[$label]; + } + + print STDOUT qq[]; + if ($uncertain) { + print qq[]; + } elsif ($result->{$_->[1]}->{score_min} != $result->{$_->[1]}->{score_max}) { + print qq[]; + } else { + print qq[]; + } + } + + $score_max += $score_base; + + print STDOUT qq[ + + + + + + + +

MUST‐level +Errors SHOULD‐level +Errors Warnings Score
$label $result->{$_->[1]}->{must}$uncertain $result->{$_->[1]}->{should}$uncertain $result->{$_->[1]}->{warning}$uncertain −∞..$result->{$_->[1]}->{score_max} $result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max}
$result->{$_->[1]}->{score_min}
Semantics 0? 0? 0? −∞..$score_base
Total $must_error? $should_error? $warning? −∞..$score_max
+ +
Important: This conformance checking service +is under development. The result above might be wrong.
+
]; + push @nav, ['#result-summary' => 'Result']; +} # print_result_section + +sub print_result_unknown_type_section ($$) { + my ($input, $result) = @_; + + my $euri = htescape ($input->{uri}); + print STDOUT qq[ +
+
Errors
+ +
+
<$euri>
+
Not + supported: + Media type + @{[htescape $input->{media_type}]} + is not supported.
+
+
+]; + push @nav, ['#parse-errors' => 'Errors']; + add_error (char => {level => 'u'} => $result); + add_error (syntax => {level => 'u'} => $result); + add_error (structure => {level => 'u'} => $result); +} # print_result_unknown_type_section + +sub print_result_input_error_section ($) { + my $input = shift; + print STDOUT qq[
+
Input Error: @{[htescape ($input->{error_status_text})]}
+
]; + push @nav, ['#result-summary' => 'Result']; +} # print_result_input_error_section + +sub get_error_label ($$) { + my ($input, $err) = @_; + + my $r = ''; + + if (defined $err->{line}) { + if ($err->{column} > 0) { + $r = qq[Line $err->{line} column $err->{column}]; + } else { + $err->{line} = $err->{line} - 1 || 1; + $r = qq[Line $err->{line}]; + } + } + + if (defined $err->{node}) { + $r .= ' ' if length $r; + $r = get_node_link ($input, $err->{node}); + } + + if (defined $err->{index}) { + $r .= ' ' if length $r; + $r .= 'Index ' . (0+$err->{index}); + } + + if (defined $err->{value}) { + $r .= ' ' if length $r; + $r .= '' . htescape ($err->{value}) . ''; + } + + return $r; +} # get_error_label + +sub get_error_level_label ($) { + my $err = shift; + + my $r = ''; + + if (not defined $err->{level} or $err->{level} eq 'm') { + $r = qq[MUST‐level + error: ]; + } elsif ($err->{level} eq 's') { + $r = qq[SHOULD‐level + error: ]; + } elsif ($err->{level} eq 'w') { + $r = qq[Warning: + ]; + } elsif ($err->{level} eq 'u' or $err->{level} eq 'unsupported') { + $r = qq[Not + supported: ]; + } else { + my $elevel = htescape ($err->{level}); + $r = qq[$elevel: + ]; + } + + return $r; +} # get_error_level_label + sub get_node_path ($) { my $node = shift; my @r; @@ -313,6 +872,7 @@ $rs = '"' . $node->data . '"'; $node = $node->parent_node; } elsif ($node->node_type == 9) { + @r = ('') unless @r; $rs = ''; $node = $node->parent_node; } else { @@ -324,6 +884,240 @@ return join '/', @r; } # get_node_path +sub get_node_link ($$) { + return qq[] . + htescape (get_node_path ($_[1])) . qq[]; +} # get_node_link + +{ + my $Msg = {}; + +sub load_text_catalog ($) { + my $lang = shift; # MUST be a canonical lang name + open my $file, '<:utf8', "cc-msg.$lang.txt" + or die "$0: cc-msg.$lang.txt: $!"; + while (<$file>) { + if (s/^([^;]+);([^;]*);//) { + my ($type, $cls, $msg) = ($1, $2, $_); + $msg =~ tr/\x0D\x0A//d; + $Msg->{$type} = [$cls, $msg]; + } + } +} # load_text_catalog + +sub get_text ($) { + my ($type, $level, $node) = @_; + $type = $level . ':' . $type if defined $level; + $level = 'm' unless defined $level; + my @arg; + { + if (defined $Msg->{$type}) { + my $msg = $Msg->{$type}->[1]; + $msg =~ s{\$([0-9]+)}{ + defined $arg[$1] ? htescape ($arg[$1]) : '(undef)'; + }ge; + $msg =~ s{{\@([A-Za-z0-9:_.-]+)}}{ + UNIVERSAL::can ($node, 'get_attribute_ns') + ? htescape ($node->get_attribute_ns (undef, $1)) : '' + }ge; + $msg =~ s{{\@}}{ + UNIVERSAL::can ($node, 'value') ? htescape ($node->value) : '' + }ge; + $msg =~ s{{local-name}}{ + UNIVERSAL::can ($node, 'manakai_local_name') + ? htescape ($node->manakai_local_name) : '' + }ge; + $msg =~ s{{element-local-name}}{ + (UNIVERSAL::can ($node, 'owner_element') and + $node->owner_element) + ? htescape ($node->owner_element->manakai_local_name) + : '' + }ge; + return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg); + } elsif ($type =~ s/:([^:]*)$//) { + unshift @arg, $1; + redo; + } + } + return ($type, 'level-'.$level, htescape ($_[0])); +} # get_text + +} + +sub get_input_document ($$) { + my ($http, $dom) = @_; + + my $request_uri = $http->get_parameter ('uri'); + my $r = {}; + if (defined $request_uri and length $request_uri) { + my $uri = $dom->create_uri_reference ($request_uri); + unless ({ + http => 1, + }->{lc $uri->uri_scheme}) { + return {uri => $request_uri, request_uri => $request_uri, + error_status_text => 'URI scheme not allowed'}; + } + + require Message::Util::HostPermit; + my $host_permit = new Message::Util::HostPermit; + $host_permit->add_rule (<check ($uri->uri_host, $uri->uri_port || 80)) { + return {uri => $request_uri, request_uri => $request_uri, + error_status_text => 'Connection to the host is forbidden'}; + } + + require LWP::UserAgent; + my $ua = WDCC::LWPUA->new; + $ua->{wdcc_dom} = $dom; + $ua->{wdcc_host_permit} = $host_permit; + $ua->agent ('Mozilla'); ## TODO: for now. + $ua->parse_head (0); + $ua->protocols_allowed ([qw/http/]); + $ua->max_size (1000_000); + my $req = HTTP::Request->new (GET => $request_uri); + $req->header ('Accept-Encoding' => 'identity, *; q=0'); + my $res = $ua->request ($req); + ## TODO: 401 sets |is_success| true. + if ($res->is_success or $http->get_parameter ('error-page')) { + $r->{base_uri} = $res->base; ## NOTE: It does check |Content-Base|, |Content-Location|, and . ## TODO: Use our own code! + $r->{uri} = $res->request->uri; + $r->{request_uri} = $request_uri; + + ## TODO: More strict parsing... + my $ct = $res->header ('Content-Type'); + if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?([^\s;"]+)"?/i) { + $r->{charset} = lc $1; + $r->{charset} =~ tr/\\//d; + $r->{official_charset} = $r->{charset}; + } + + my $input_charset = $http->get_parameter ('charset'); + if (defined $input_charset and length $input_charset) { + $r->{charset_overridden} + = (not defined $r->{charset} or $r->{charset} ne $input_charset); + $r->{charset} = $input_charset; + } + + ## TODO: Support for HTTP Content-Encoding + + $r->{s} = ''.$res->content; + + require Whatpm::ContentType; + ($r->{official_type}, $r->{media_type}) + = Whatpm::ContentType->get_sniffed_type + (get_file_head => sub { + return substr $r->{s}, 0, shift; + }, + http_content_type_byte => $ct, + has_http_content_encoding => + defined $res->header ('Content-Encoding'), + supported_image_types => {}); + } else { + $r->{uri} = $res->request->uri; + $r->{request_uri} = $request_uri; + $r->{error_status_text} = $res->status_line; + } + + $r->{header_field} = []; + $res->scan (sub { + push @{$r->{header_field}}, [$_[0], $_[1]]; + }); + $r->{header_status_code} = $res->code; + $r->{header_status_text} = $res->message; + } else { + $r->{s} = ''.$http->get_parameter ('s'); + $r->{uri} = q; + $r->{request_uri} = q; + $r->{base_uri} = q; + $r->{charset} = ''.$http->get_parameter ('_charset_'); + $r->{charset} =~ s/\s+//g; + $r->{charset} = 'utf-8' if $r->{charset} eq ''; + $r->{official_charset} = $r->{charset}; + $r->{header_field} = []; + + require Whatpm::ContentType; + ($r->{official_type}, $r->{media_type}) + = Whatpm::ContentType->get_sniffed_type + (get_file_head => sub { + return substr $r->{s}, 0, shift; + }, + http_content_type_byte => undef, + has_http_content_encoding => 0, + supported_image_types => {}); + } + + my $input_format = $http->get_parameter ('i'); + if (defined $input_format and length $input_format) { + $r->{media_type_overridden} + = (not defined $r->{media_type} or $input_format ne $r->{media_type}); + $r->{media_type} = $input_format; + } + if (defined $r->{s} and not defined $r->{media_type}) { + $r->{media_type} = 'text/html'; + $r->{media_type_overridden} = 1; + } + + if ($r->{media_type} eq 'text/xml') { + unless (defined $r->{charset}) { + $r->{charset} = 'us-ascii'; + $r->{official_charset} = $r->{charset}; + } elsif ($r->{charset_overridden} and $r->{charset} eq 'us-ascii') { + $r->{charset_overridden} = 0; + } + } + + if (length $r->{s} > 1000_000) { + $r->{error_status_text} = 'Entity-body too large'; + delete $r->{s}; + return $r; + } + + return $r; +} # get_input_document + +package WDCC::LWPUA; +BEGIN { push our @ISA, 'LWP::UserAgent'; } + +sub redirect_ok { + my $ua = shift; + unless ($ua->SUPER::redirect_ok (@_)) { + return 0; + } + + my $uris = $_[1]->header ('Location'); + return 0 unless $uris; + my $uri = $ua->{wdcc_dom}->create_uri_reference ($uris); + unless ({ + http => 1, + }->{lc $uri->uri_scheme}) { + return 0; + } + unless ($ua->{wdcc_host_permit}->check ($uri->uri_host, $uri->uri_port || 80)) { + return 0; + } + return 1; +} # redirect_ok + =head1 AUTHOR Wakaba . @@ -337,4 +1131,4 @@ =cut -## $Date: 2007/06/27 14:36:45 $ +## $Date: 2008/02/10 02:42:01 $

	MUST‐level +Errors	SHOULD‐level +Errors	Warnings	Score
$label	$result->{$_->[1]}->{must}$uncertain	$result->{$_->[1]}->{should}$uncertain	$result->{$_->[1]}->{warning}$uncertain	−∞..$result->{$_->[1]}->{score_max}	$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max}
$result->{$_->[1]}->{score_min}
Semantics	0?	0?	0?	−∞..$score_base
Total	$must_error?	$should_error?	$warning?	−∞..$score_max