--- test/html-webhacc/cc.cgi 2007/06/27 11:08:03 1.1
+++ test/html-webhacc/cc.cgi 2007/09/10 11:51:09 1.19
@@ -2,200 +2,669 @@
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 Time::HiRes qw/time/;
-use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module
+sub htescape ($) {
+ my $s = $_[0];
+ $s =~ s/&/&/g;
+ $s =~ s//>/g;
+ $s =~ s/"/"/g;
+ $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;
+ use Message::CGI::HTTP;
+ my $http = Message::CGI::HTTP->new;
-## TODO: _charset_
-
-my @mode = split m#/#, scalar $http->meta_variable ('PATH_INFO'), -1;
-shift @mode if @mode and $mode[0] == '';
-## TODO: decode unreserved characters
-
- 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;
}
- my $char_length = length $s;
- my %time;
- my $time1;
- my $time2;
+
+ binmode STDOUT, ':utf8';
+ $| = 1;
require Message::DOM::DOMImplementation;
- my $dom = Message::DOM::DOMImplementation->____new;
-# $| = 1;
+ my $dom = Message::DOM::DOMImplementation->new;
+
+ load_text_catalog ('en'); ## TODO: conneg
+
+ my @nav;
+ print STDOUT qq[Content-Type: text/html; charset=utf-8
+
+
+
+
+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}]}>
yet
+ push @nav, ['#document-info' => 'Information'];
+
+if (defined $input->{s}) {
+ $char_length = length $input->{s};
+
+ print STDOUT qq[
+
Base URI
+
<@{[htescape $input->{base_uri}]}>
+
Internet Media Type
+
@{[htescape $input->{media_type}]}
+ @{[$input->{media_type_overridden} ? '(overridden)' : '']}
+
Character Encoding
+
@{[defined $input->{charset} ? ''.htescape ($input->{charset}).'' : '(none)']}
+ @{[$input->{charset_overridden} ? '(overridden)' : '']}
+
Length
+
$char_length byte@{[$char_length == 1 ? '' : 's']}
+
+
+];
+ for (@nav) {
+ print STDOUT qq[- $_->[1]
];
+ }
+ 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[];
+} # print_http_header_section
+sub print_syntax_error_html_section ($$) {
+ my ($input, $result) = @_;
+
require Encode;
require Whatpm::HTML;
- $time1 = time;
- $s = Encode::decode ('utf-8', $s);
- $time2 = time;
- $time{decode} = $time2 - $time1;
+ $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
- print STDOUT "#errors\n";
+
];
+ push @nav, ['#parse-errors' => 'Parse Error'];
my $onerror = sub {
my (%opt) = @_;
- print STDOUT "$opt{line},$opt{column},$opt{type}\n";
+ 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);
};
- $doc = $dom->create_document;
+ my $doc = $dom->create_document;
+ my $el;
$time1 = time;
- if (length $mode[1]) {
+ if (defined $inner_html_element and length $inner_html_element) {
$el = $doc->create_element_ns
- ('http://www.w3.org/1999/xhtml', [undef, $mode[1]]);
- Whatpm::HTML->set_inner_html ($el, $s, $onerror);
+ ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
+ Whatpm::HTML->set_inner_html ($el, $t, $onerror);
} else {
- Whatpm::HTML->parse_string ($s => $doc, $onerror);
+ Whatpm::HTML->parse_string ($t => $doc, $onerror);
}
- $time2 = time;
- $time{parse} = $time2 - $time1;
+ $time{parse} = time - $time1;
+
+ print STDOUT qq[
+
Parse Errors
+
+
];
+ push @nav, ['#parse-errors' => 'Parse Error'];
my $onerror = sub {
my $err = shift;
- print STDOUT $err->location->line_number, ",";
- print STDOUT $err->location->column_number, ",";
- print STDOUT $err->text, "\n";
+ 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;
};
- open my $fh, '<', \$s;
my $time1 = time;
- $doc = Message::DOM::XMLParserTemp->parse_byte_stream
- ($fh => $dom, $onerror, charset => 'utf-8');
- my $time2 = time;
- $time{parse_xml} = $time2 - $time1;
-
- print "#document\n";
-
- my $out;
- if ($mode[2] eq 'html') {
- ## TODO: Use XHTML serializer
- #$out = Whatpm::HTML->get_inner_html ($doc);
- } else { # test
- $time1 = time;
- $out = test_serialize ($doc);
- $time2 = time;
- $time{serialize_test} = $time2 - $time1;
- }
- print STDOUT Encode::encode ('utf-8', $$out);
- print STDOUT "\n";
-} else {
- print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n404";
- exit;
-}
+ open my $fh, '<', \($input->{s});
+ my $doc = Message::DOM::XMLParserTemp->parse_byte_stream
+ ($fh => $dom, $onerror, charset => $input->{charset});
+ $time{parse_xml} = time - $time1;
- if ($http->parameter ('dom5')) {
- require Whatpm::ContentChecker;
- my $onerror = sub {
- my %opt = @_;
- print STDOUT get_node_path ($opt{node}) . ';' . $opt{type} . "\n";
- };
- print STDOUT "#domerrors\n";
- $time1 = time;
- if ($el) {
- Whatpm::ContentChecker->check_element ($el, $onerror);
- } else {
- Whatpm::ContentChecker->check_document ($doc, $onerror);
- }
- $time2 = time;
- $time{check} = $time2 - $time1;
- }
-
- print STDOUT "#log\n";
- for (qw/decode parse parse_xml serialize_html serialize_xml serialize_test
- check/) {
- next unless defined $time{$_};
- print STDOUT {
- decode => 'bytes->chars',
- parse => 'html5(chars)->dom5',
- parse_xml => 'xml1(chars)->dom5',
- serialize_html => 'dom5->html5(char)',
- serialize_xml => 'dom5->xml1(char)',
- serialize_test => 'dom5->test(char)',
- check => 'dom5 check',
- }->{$_};
- print STDOUT "\t", $time{$_}, "s\n";
- open my $file, '>>', ".manakai-$_.txt" or die ".manakai-$_.txt: $!";
- print $file $char_length, "\t", $time{$_}, "\n";
- }
+ print STDOUT qq[
+
Document Source
+
\n];
+ if (length $$s) {
+ while ($$s =~ /\G([^\x0A]*?)\x0D?\x0A/gc) {
+ print STDOUT qq[- ], htescape $1, "
\n";
+ $i++;
+ }
+ if ($$s =~ /\G([^\x0A]+)/gc) {
+ print STDOUT qq[- ], htescape $1, "
\n";
+ }
+ } else {
+ print STDOUT q[];
+ }
+ print STDOUT "
';
- my @node = map { [$_, ''] } @{$node->child_nodes};
+ my @node = ($node);
while (@node) {
my $child = shift @node;
- my $nt = $child->[0]->node_type;
- if ($nt == $child->[0]->ELEMENT_NODE) {
- $r .= '| ' . $child->[1] . '<' . $child->[0]->tag_name . ">\x0A"; ## ISSUE: case?
-
- for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value] }
- @{$child->[0]->attributes}) {
- $r .= '| ' . $child->[1] . ' ' . $attr->[0] . '="'; ## ISSUE: case?
- $r .= $attr->[1] . '"' . "\x0A";
- }
-
- unshift @node,
- map { [$_, $child->[1] . ' '] } @{$child->[0]->child_nodes};
- } elsif ($nt == $child->[0]->TEXT_NODE) {
- $r .= '| ' . $child->[1] . '"' . $child->[0]->data . '"' . "\x0A";
- } elsif ($nt == $child->[0]->CDATA_SECTION_NODE) {
- $r .= '| ' . $child->[1] . '[0]->data . "]]>\x0A";
- } elsif ($nt == $child->[0]->COMMENT_NODE) {
- $r .= '| ' . $child->[1] . '\x0A";
- } elsif ($nt == $child->[0]->DOCUMENT_TYPE_NODE) {
- $r .= '| ' . $child->[1] . '[0]->name . ">\x0A";
- } elsif ($nt == $child->[0]->PROCESSING_INSTRUCTION_NODE) {
- $r .= '| ' . $child->[1] . '[0]->target . ' ' .
- $child->[0]->data . "?>\x0A";
+ unless (ref $child) {
+ $r .= $child;
+ next;
+ }
+
+ my $node_id = 'node-'.refaddr $child;
+ my $nt = $child->node_type;
+ if ($nt == $child->ELEMENT_NODE) {
+ my $child_nsuri = $child->namespace_uri;
+ $r .= qq[] . htescape ($child->tag_name) .
+ ''; ## ISSUE: case
+
+ if ($child->has_attributes) {
+ $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 .= '' . htescape ($attr->[1]) . '
';
+ }
+
+ if ($child->has_child_nodes) {
+ $r .= '';
+ unshift @node, @{$child->child_nodes}, '
' . htescape ($child->data) . '
<[CDATA[' . htescape ($child->data) . '
]]>- Document';
+ $r .= qq[
];
+ $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}, '
';
+ }
+ } elsif ($nt == $child->DOCUMENT_TYPE_NODE) {
+ $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 .= '
<?@{[htescape ($child->target)]} @{[htescape ($child->data)]}
?>- @{[$child->node_type]} @{[htescape ($child->node_name)]}
'; # error
+ }
+ }
+
+ $r .= '
';
+ 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[
+
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[
+
Identifiers
+
+
+];
+ for my $id (sort {$a cmp $b} keys %$ids) {
+ print STDOUT qq[@{[htescape $id]}- ].get_node_link ($_).qq[
];
+ }
+ }
+ 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[
+
Classes
+
+
+];
+ for my $class (sort {$a cmp $b} keys %$classes) {
+ print STDOUT qq[@{[htescape $class]}- ].get_node_link ($_).qq[
];
+ }
+ }
+ 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[
+
+
+ | MUST-level
+Errors | SHOULD-level
+Errors | Warnings | Score |
|---|
+];
+
+ 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[| $label | $result->{$_->[1]}->{must}$uncertain | $result->{$_->[1]}->{should}$uncertain | $result->{$_->[1]}->{warning}$uncertain | ];
+ if ($uncertain) {
+ print qq[−∞..$result->{$_->[1]}->{score_max} | ];
+ } elsif ($result->{$_->[1]}->{score_min} != $result->{$_->[1]}->{score_max}) {
+ print qq[$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max} + $score_base |
|---|
];
+ } else {
+ print qq[$result->{$_->[1]}->{score_min} | ];
+ }
+ }
+
+ $score_max += $score_base;
+
+ print STDOUT qq[
+| Semantics | 0? | 0? | 0? | −∞..$score_base |
|---|
+
+| Total | $must_error? | $should_error? | $warning? | −∞..$score_max |
|---|
+
+
+
Important: This conformance checking service
+is under development. The result above might be wrong.
+
+
Media type @{[htescape $input->{media_type}]} is not supported!
+
+
Input Error: @{[htescape ($input->{error_status_text})]}
+