--- test/html-webhacc/cc.cgi 2007/06/27 14:36:45 1.4 +++ test/html-webhacc/cc.cgi 2007/09/11 08:25:23 1.21 @@ -2,12 +2,10 @@ use strict; 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 +13,28 @@ $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_ + use Message::CGI::HTTP; + my $http = Message::CGI::HTTP->new; - 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 $s > 1000_000) { - print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long"; + if ($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 @@ -43,98 +45,228 @@ -

Web Document Conformance Checker (beta)

+

Web Document Conformance Checker +(beta)

+]; + $| = 0; + my $input = get_input_document ($http, $dom); + my $inner_html_element = $http->get_parameter ('e'); + 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}]}>
]; # no
yet push @nav, ['#document-info' => 'Information']; - require Message::DOM::DOMImplementation; - my $dom = Message::DOM::DOMImplementation->____new; - my $doc; - my $el; - - if ($input_format eq 'text/html') { - require Encode; - require Whatpm::HTML; - - $s = Encode::decode ('utf-8', $s); +if (defined $input->{s}) { + $char_length = length $input->{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)' : '']}
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 $result = {conforming_min => 1, conforming_max => 1}; + print_http_header_section ($input, $result); + + my $doc; + my $el; + + if ($input->{media_type} eq 'text/html') { + ($doc, $el) = print_syntax_error_html_section ($input, $result); + print_source_string_section (\($input->{s}), $input->{charset}); + } elsif ({ + 'text/xml' => 1, + 'application/atom+xml' => 1, + 'application/rss+xml' => 1, + 'application/svg+xml' => 1, + 'application/xhtml+xml' => 1, + 'application/xml' => 1, + }->{$input->{media_type}}) { + ($doc, $el) = print_syntax_error_xml_section ($input, $result); + print_source_string_section (\($input->{s}), $doc->input_encoding); + } else { + ## TODO: Change HTTP status code?? + print_result_unknown_type_section ($input); + } + + if (defined $doc or defined $el) { + print_structure_dump_section ($doc, $el); + my $elements = print_structure_error_section ($doc, $el, $result); + print_table_section ($elements->{table}) if @{$elements->{table}}; + print_id_section ($elements->{id}) if keys %{$elements->{id}}; + print_term_section ($elements->{term}) if keys %{$elements->{term}}; + print_class_section ($elements->{class}) if keys %{$elements->{class}}; + } + + print_result_section ($result); +} else { + print STDOUT qq[
]; + print_result_input_error_section ($input); +} + + print STDOUT qq[ + + + +]; + + for (qw/decode parse parse_xml check/) { + next unless defined $time{$_}; + open my $file, '>>', ".cc-$_.txt" or die ".cc-$_.txt: $!"; + print $file $char_length, "\t", $time{$_}, "\n"; + } + +exit; +sub add_error ($$$) { + my ($layer, $err, $result) = @_; + if (defined $err->{level}) { + if ($err->{level} eq 's') { + $result->{$layer}->{should}++; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + } elsif ($err->{level} eq 'w' or $err->{level} eq 'g') { + $result->{$layer}->{warning}++; + } elsif ($err->{level} eq 'unsupported') { + $result->{$layer}->{unsupported}++; + $result->{unsupported} = 1; + } else { + $result->{$layer}->{must}++; + $result->{$layer}->{score_max} -= 2; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + $result->{conforming_max} = 0; + } + } else { + $result->{$layer}->{must}++; + $result->{$layer}->{score_max} -= 2; + $result->{$layer}->{score_min} -= 2; + $result->{conforming_min} = 0; + $result->{conforming_max} = 0; + } +} # add_error + +sub print_http_header_section ($$) { + 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']; + 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.

+ + +]; + + 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; + + $input->{charset} ||= 'ISO-8859-1'; ## TODO: for now. + + my $time1 = time; + my $t = Encode::decode ($input->{charset}, $input->{s}); + $time{decode} = time - $time1; + + print STDOUT qq[

Parse Errors

-
-]; +
]; push @nav, ['#parse-errors' => 'Parse Error']; 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[
$msg
\n]; + + add_error ('syntax', \%opt => $result); }; - $doc = $dom->create_document; + my $doc = $dom->create_document; + my $el; + $time1 = time; if (defined $inner_html_element and length $inner_html_element) { $el = $doc->create_element_ns ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]); - Whatpm::HTML->set_inner_html ($el, $s, $onerror); + Whatpm::HTML->set_inner_html ($el, $t, $onerror); } else { - Whatpm::HTML->parse_string ($s => $doc, $onerror); + Whatpm::HTML->parse_string ($t => $doc, $onerror); } + $time{parse} = time - $time1; - print STDOUT qq[ -
-
-]; - } elsif ($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)
- - + print STDOUT qq[]; -
-

Document Source

-]; - push @nav, ['#source-string' => 'Source']; - print_source_string (\$t); - 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[

Parse Errors

-
-]; +
]; push @nav, ['#parse-errors' => 'Parse Error']; my $onerror = sub { @@ -143,98 +275,52 @@ 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'); - - print STDOUT qq[ -
-
-]; - } else { - print STDOUT qq[ - - - -
-

Media type @{[htescape $input_format]} is not supported!

-
-]; - push @nav, ['#result-summary' => 'Result']; - } + add_error ('syntax', {type => $err->text, + level => [ + $err->SEVERITY_FATAL_ERROR => 'm', + $err->SEVERITY_ERROR => 'm', + $err->SEVERITY_WARNING => 's', + ]->[$err->severity]} => $result); - if (defined $doc or defined $el) { - print STDOUT qq[ -
-

Document Tree

-]; - push @nav, ['#document-tree' => 'Tree']; - - print_document_tree ($el || $doc); - - print STDOUT qq[ -
- -
-

Document Errors

- -
-]; - push @nav, ['#document-errors' => 'Document Error']; - - require Whatpm::ContentChecker; - my $onerror = sub { - my %opt = @_; - print STDOUT qq[
], - htescape get_node_path ($opt{node}), - "
\n
", htescape $opt{type}, "
\n"; - }; + return 1; + }; - if ($el) { - Whatpm::ContentChecker->check_element ($el, $onerror); - } else { - Whatpm::ContentChecker->check_document ($doc, $onerror); + 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 - $time1; + + print STDOUT qq[
]; + + return ($doc, undef); +} # print_syntax_error_xml_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']; + print STDOUT qq[
+

Document Source

+
    \n]; + if (length $$s) { + while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { + print STDOUT qq[
  1. ], htescape $1, "
  2. \n"; + $i++; } - - print STDOUT qq[ - -
-]; - } - - ## TODO: Show result - - print STDOUT qq[ - - - -]; - -exit; - -sub print_source_string ($) { - my $s = $_[0]; - my $i = 1; - print STDOUT qq[
    \n]; - while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) { - print STDOUT qq[
  1. ], htescape $1, "
  2. \n"; - $i++; - } - if ($$s =~ /\G([^\x0A]+)/gc) { - print STDOUT qq[
  3. ], htescape $1, "
  4. \n"; + if ($$s =~ /\G([^\x0A]+)/gc) { + print STDOUT qq[
  5. ], htescape $1, "
  6. \n"; + } + } else { + print STDOUT q[
  7. ]; } - print STDOUT "
"; -} # print_input_string + print STDOUT ""; +} # print_input_string_section sub print_document_tree ($) { my $node = shift; @@ -265,9 +351,11 @@ $r .= ''; } - if ($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 +364,29 @@ } elsif ($nt == $child->COMMENT_NODE) { $r .= qq'
  • <!--' . htescape ($child->data) . '-->
  • '; } elsif ($nt == $child->DOCUMENT_NODE) { - $r .= qq'
  • Document
  • '; + $r .= qq'
  • Document'; + $r .= qq[]; if ($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>