--- test/html-webhacc/cc.cgi 2007/07/21 04:58:17 1.15 +++ test/html-webhacc/cc.cgi 2007/09/10 11:51:09 1.19 @@ -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]; @@ -21,11 +19,10 @@ return $s; } # htescape -my $http = SuikaWiki::Input::HTTP->new; - -## TODO: _charset_ + use Message::CGI::HTTP; + my $http = Message::CGI::HTTP->new; - if ($http->meta_variable ('PATH_INFO') ne '/') { + if ($http->get_meta_variable ('PATH_INFO') ne '/') { print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n400"; exit; } @@ -54,7 +51,9 @@ $| = 0; my $input = get_input_document ($http, $dom); - my $inner_html_element = $http->parameter ('e'); + my $inner_html_element = $http->get_parameter ('e'); + my $char_length = 0; + my %time; print qq[
@@ -67,6 +66,7 @@ push @nav, ['#document-info' => 'Information']; if (defined $input->{s}) { + $char_length = length $input->{s}; print STDOUT qq[
Base URI
@@ -77,263 +77,49 @@
Character Encoding
@{[defined $input->{charset} ? ''.htescape ($input->{charset}).'' : '(none)']} @{[$input->{charset_overridden} ? '(overridden)' : '']}
+
Length
+
$char_length byte@{[$char_length == 1 ? '' : 's']}
]; - print_http_header_section ($input); + my $result = {}; + print_http_header_section ($input, $result); my $doc; my $el; if ($input->{media_type} eq 'text/html') { - require Encode; - require Whatpm::HTML; - - $input->{charset} ||= 'ISO-8859-1'; ## TODO: for now. - - my $t = Encode::decode ($input->{charset}, $input->{s}); - - 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]; - } else { - $opt{line} = $opt{line} - 1 || 1; - print STDOUT qq[
Line $opt{line}
\n]; - } - $type =~ tr/ /-/; - $type =~ s/\|/%7C/g; - $msg .= qq[ [Description]]; - print STDOUT qq[
$msg
\n]; - }; - - $doc = $dom->create_document; - 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, $t, $onerror); - } else { - Whatpm::HTML->parse_string ($t => $doc, $onerror); - } - - print STDOUT qq[
-
-]; - + ($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}}) { - require Message::DOM::XMLParserTemp; - - print STDOUT qq[ -
-

Parse Errors

- -
]; - push @nav, ['#parse-errors' => 'Parse Error']; - - my $onerror = sub { - my $err = shift; - my $line = $err->location->line_number; - print STDOUT qq[
Line $line column ]; - print STDOUT $err->location->column_number, "
"; - print STDOUT htescape $err->text, "
\n"; - return 1; - }; - - open my $fh, '<', \($input->{s}); - $doc = Message::DOM::XMLParserTemp->parse_byte_stream - ($fh => $dom, $onerror, charset => $input->{charset}); - - print STDOUT qq[
-
- -]; + ($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 STDOUT qq[ -
-

Media type @{[htescape $input->{media_type}]} is not supported!

-
-]; - push @nav, ['#result-summary' => 'Result']; + print_result_unknown_type_section ($input); } - 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 $onerror = sub { - my %opt = @_; - my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level}, $opt{node}); - $type =~ tr/ /-/; - $type =~ s/\|/%7C/g; - $msg .= qq[ [Description]]; - print STDOUT qq[
] . get_node_link ($opt{node}) . - qq[
\n
], $msg, "
\n"; - }; - - my $elements; - if ($el) { - $elements = Whatpm::ContentChecker->check_element ($el, $onerror); - } else { - $elements = Whatpm::ContentChecker->check_document ($doc, $onerror); - } - - print STDOUT qq[
-
-]; - - if (@{$elements->{table}}) { - require JSON; - - push @nav, ['#tables' => 'Tables']; - print STDOUT qq[ -
-

Tables

- - - - -]; - - my $i = 0; - for my $table_el (@{$elements->{table}}) { - $i++; - print STDOUT qq[

] . - get_node_link ($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} = $_->{is_header} ? 1 : 0; - } - } - } - - print STDOUT '
]; - } - - print STDOUT qq[
]; - } - - if (keys %{$elements->{id}}) { - push @nav, ['#identifiers' => 'IDs']; - print STDOUT qq[ -
-

Identifiers

- -
-]; - for my $id (sort {$a cmp $b} keys %{$elements->{id}}) { - print STDOUT qq[
@{[htescape $id]}
]; - for (@{$elements->{id}->{$id}}) { - print STDOUT qq[
].get_node_link ($_).qq[
]; - } - } - print STDOUT qq[
]; - } - - if (keys %{$elements->{term}}) { - push @nav, ['#terms' => 'Terms']; - print STDOUT qq[ -
-

Terms

- -
-]; - for my $term (sort {$a cmp $b} keys %{$elements->{term}}) { - print STDOUT qq[
@{[htescape $term]}
]; - for (@{$elements->{term}->{$term}}) { - print STDOUT qq[
].get_node_link ($_).qq[
]; - } - } - print STDOUT qq[
]; - } - - if (keys %{$elements->{class}}) { - push @nav, ['#classes' => 'Classes']; - print STDOUT qq[ -
-

Classes

- -
-]; - for my $class (sort {$a cmp $b} keys %{$elements->{class}}) { - print STDOUT qq[
@{[htescape $class]}
]; - for (@{$elements->{class}->{$class}}) { - print STDOUT qq[
].get_node_link ($_).qq[
]; - } - } - print STDOUT qq[
]; - } + 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 %{$elements->{class}}; } - ## TODO: Show result + print_result_section ($result); } else { - print STDOUT qq[ - - - -
-

Input Error: @{[htescape ($input->{error_status_text})]}

-
-]; - push @nav, ['#result-summary' => 'Result']; - + print STDOUT qq[]; + print_result_input_error_section ($input); } print STDOUT qq[ @@ -348,10 +134,44 @@ ]; + 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 print_http_header_section ($) { - my $input = shift; +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}}; @@ -384,6 +204,99 @@ print STDOUT qq[]; } # 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]; + } else { + $opt{line} = $opt{line} - 1 || 1; + print STDOUT qq[
Line $opt{line}
\n]; + } + $type =~ tr/ /-/; + $type =~ s/\|/%7C/g; + $msg .= qq[ [Description]]; + print STDOUT qq[
$msg
\n]; + + add_error ('syntax', \%opt => $result); + }; + + 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, $t, $onerror); + } else { + Whatpm::HTML->parse_string ($t => $doc, $onerror); + } + $time{parse} = time - $time1; + + print STDOUT qq[
]; + + return ($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 { + my $err = shift; + my $line = $err->location->line_number; + print STDOUT qq[
Line $line column ]; + print STDOUT $err->location->column_number, "
"; + print STDOUT htescape $err->text, "
\n"; + + add_error ('syntax', {type => $err->text, + level => [ + $err->SEVERITY_FATAL_ERROR => 'm', + $err->SEVERITY_ERROR => 'm', + $err->SEVERITY_WARNING => 's', + ]->[$err->severity]} => $result); + + return 1; + }; + + 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 ($$) { require Encode; my $enc = Encode::find_encoding ($_[1]); ## TODO: charset name -> Perl name @@ -486,6 +399,273 @@ print STDOUT $r; } # print_document_tree +sub print_structure_dump_section ($$) { + my ($doc, $el) = @_; + + print STDOUT qq[ +
+

Document Tree

+]; + push @nav, ['#document-tree' => 'Tree']; + + print_document_tree ($el || $doc); + + print STDOUT qq[
]; +} # print_structure_dump_section + +sub print_structure_error_section ($$$) { + my ($doc, $el, $result) = @_; + + print STDOUT qq[
+

Document Errors

+ +
]; + push @nav, ['#document-errors' => 'Document Error']; + + 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; + $msg .= qq[ [Description]]; + print STDOUT qq[
] . get_node_link ($opt{node}) . + qq[
\n
], $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 $elements; +} # print_structure_error_section + +sub print_table_section ($) { + my $tables = shift; + + push @nav, ['#tables' => 'Tables']; + print STDOUT qq[ +
+

Tables

+ + + + +]; + + require JSON; + + my $i = 0; + for my $table_el (@$tables) { + $i++; + print STDOUT qq[

] . + get_node_link ($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} = $_->{is_header} ? 1 : 0; + } + } + } + + print STDOUT '
]; + } + + print STDOUT qq[
]; +} # print_table_section + +sub print_id_section ($) { + my $ids = shift; + + push @nav, ['#identifiers' => 'IDs']; + print STDOUT qq[ +
+

Identifiers

+ +
+]; + for my $id (sort {$a cmp $b} keys %$ids) { + print STDOUT qq[
@{[htescape $id]}
]; + for (@{$ids->{$id}}) { + print STDOUT qq[
].get_node_link ($_).qq[
]; + } + } + print STDOUT qq[
]; +} # print_id_section + +sub print_term_section ($) { + my $terms = shift; + + push @nav, ['#terms' => 'Terms']; + print STDOUT qq[ +
+

Terms

+ +
+]; + for my $term (sort {$a cmp $b} keys %$terms) { + print STDOUT qq[
@{[htescape $term]}
]; + for (@{$terms->{$term}}) { + print STDOUT qq[
].get_node_link ($_).qq[
]; + } + } + print STDOUT qq[
]; +} # print_term_section + +sub print_class_section ($) { + my $classes = shift; + + push @nav, ['#classes' => 'Classes']; + print STDOUT qq[ +
+

Classes

+ +
+]; + for my $class (sort {$a cmp $b} keys %$classes) { + print STDOUT qq[
@{[htescape $class]}
]; + for (@{$classes->{$class}}) { + print STDOUT qq[
].get_node_link ($_).qq[
]; + } + } + print STDOUT qq[
]; +} # print_class_section + +sub print_result_section ($) { + my $result = shift; + + print STDOUT qq[ +
+

Result

]; + + if ($result->{unsupported}) { + print STDOUT qq[

The conformance + checker cannot decide whether the document is conforming or + not, since the document contains one or more unsupported + features.

]; + } 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; + 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_base); + $score_max += ($result->{$_->[1]}->{score_max} += $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 +ErrorsSHOULD-level +ErrorsWarningsScore
$label$result->{$_->[1]}->{must}$uncertain$result->{$_->[1]}->{should}$uncertain$result->{$_->[1]}->{warning}$uncertain−∞..$result->{$_->[1]}->{score_max}$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max} + $score_base
$result->{$_->[1]}->{score_min}
Semantics0?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 = shift; + + print STDOUT qq[ +
+

Media type @{[htescape $input->{media_type}]} is not supported!

+
+]; + push @nav, ['#result-summary' => '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_node_path ($) { my $node = shift; my @r; @@ -550,6 +730,16 @@ $msg =~ s{{\@}}{ UNIVERSAL::can ($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, $Msg->{$type}->[0], $msg); } elsif ($type =~ s/:([^:]*)$//) { unshift @arg, $1; @@ -564,7 +754,7 @@ sub get_input_document ($$) { my ($http, $dom) = @_; - my $request_uri = $http->parameter ('uri'); + 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); @@ -613,7 +803,8 @@ $ua->max_size (1000_000); my $req = HTTP::Request->new (GET => $request_uri); my $res = $ua->request ($req); - if ($res->is_success or $http->parameter ('error-page')) { + ## TODO: 401 sets |is_success| true. + if ($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; @@ -628,7 +819,7 @@ $r->{charset} =~ tr/\\//d; } - my $input_charset = $http->parameter ('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); @@ -649,17 +840,17 @@ $r->{header_status_code} = $res->code; $r->{header_status_text} = $res->message; } else { - $r->{s} = ''.$http->parameter ('s'); + $r->{s} = ''.$http->get_parameter ('s'); $r->{uri} = q; $r->{request_uri} = q; $r->{base_uri} = q; - $r->{charset} = ''.$http->parameter ('_charset_'); + $r->{charset} = ''.$http->get_parameter ('_charset_'); $r->{charset} =~ s/\s+//g; $r->{charset} = 'utf-8' if $r->{charset} eq ''; $r->{header_field} = []; } - my $input_format = $http->parameter ('i'); + 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}); @@ -723,4 +914,4 @@ =cut -## $Date: 2007/07/21 04:58:17 $ +## $Date: 2007/09/10 11:51:09 $