--- test/html-webhacc/cc.cgi 2007/07/17 14:28:20 1.14
+++ test/html-webhacc/cc.cgi 2008/07/20 16:53:10 1.54
@@ -1,45 +1,36 @@
#!/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
+ require WebHACC::Input;
+ require WebHACC::Result;
+ require WebHACC::Output;
-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 $out;
-my $http = SuikaWiki::Input::HTTP->new;
-
-## TODO: _charset_
+ require Message::DOM::DOMImplementation;
+ my $dom = Message::DOM::DOMImplementation->new;
+{
+ 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;
}
-
- 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
+ $out = WebHACC::Output->new;
+ $out->handle (*STDOUT);
+ $out->set_utf8;
+ $out->set_flush;
+ $out->html (qq[Content-Type: text/html; charset=utf-8
@@ -50,175 +41,199 @@
-];
+]);
- $| = 0;
my $input = get_input_document ($http, $dom);
- my $inner_html_element = $http->parameter ('e');
-
- print qq[
-
-
-- Request URI
- <@{[htescape $input->{request_uri}]}>- Document URI
- <@{[htescape $input->{uri}]}>
yet
- push @nav, ['#document-info' => 'Information'];
+ $out->input ($input);
+ $out->unset_flush;
-if (defined $input->{s}) {
+ my $char_length = 0;
- 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)' : '']}
+ $out->start_section (id => 'document-info', title => 'Information');
+ $out->html (qq[
+- Request URL
+ - ]);
+ $out->url ($input->{request_uri});
+ $out->html (q[
- Document URL
+
- ]);
+ $out->url ($input->{uri}, id => 'anchor-document-url');
+ $out->html (q[
+ ]);
+ ## NOTE: no
yet
+
+ if (defined $input->{s}) {
+ $char_length = length $input->{s};
+
+ $out->html (qq[
Base URI]);
+ $out->url ($input->{base_uri});
+ $out->html (qq[Internet Media Type
+
]);
+ $out->text ($input->{media_type});
+ $out->html (qq[ ]);
+ if ($input->{media_type_overridden}) {
+ $out->html ('(overridden)');
+ } elsif (defined $input->{official_type}) {
+ if ($input->{media_type} eq $input->{official_type}) {
+ #
+ } else {
+ $out->html ('(sniffed; official type is: ');
+ $out->text ($input->{official_type});
+ $out->html (')');
+ }
+ } else {
+ $out->html ('(sniffed)');
+ }
+ $out->html (q[Character Encoding]);
+ if (defined $input->{charset}) {
+ $out->html ('');
+ $out->text ($input->{charset});
+ $out->html ('');
+ } else {
+ $out->text ('(none)');
+ }
+ $out->html (' overridden') if $input->{charset_overridden};
+ $out->html (qq[
+Length
+ $char_length byte@{[$char_length == 1 ? '' : 's']}
-
-
Parse Errors
+
+]);
+ $out->end_section;
+
+ my $result = WebHACC::Result->new;
+ $result->output ($out);
+ $result->{conforming_min} = 1;
+ $result->{conforming_max} = 1;
+ check_and_print ($input => $result => $out);
+ $result->generate_result_section;
+ } else {
+ $out->html ('');
+ $out->end_section;
-
];
- push @nav, ['#parse-errors' => 'Parse Error'];
+ my $result = WebHACC::Result->new;
+ $result->output ($out);
+ $result->{conforming_min} = 0;
+ $result->{conforming_max} = 1;
- 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);
+ $input->generate_transfer_sections ($result);
+ $result->generate_result_section;
}
- print STDOUT qq[
-
-
Parse Errors
+ exit;
+}
-
];
- push @nav, ['#parse-errors' => 'Parse Error'];
+sub check_and_print ($$$) {
+ my ($input, $result, $out) = @_;
+ my $original_input = $out->input;
+ $out->input ($input);
+
+ $input->generate_transfer_sections ($result);
+
+ my @subdoc;
+
+ my $checker_class = {
+ 'text/cache-manifest' => 'WebHACC::Language::CacheManifest',
+ 'text/css' => 'WebHACC::Language::CSS',
+ 'text/html' => 'WebHACC::Language::HTML',
+ 'text/x-webidl' => 'WebHACC::Language::WebIDL',
+
+ 'text/xml' => 'WebHACC::Language::XML',
+ 'application/atom+xml' => 'WebHACC::Language::XML',
+ 'application/rss+xml' => 'WebHACC::Language::XML',
+ 'image/svg+xml' => 'WebHACC::Language::XML',
+ 'application/xhtml+xml' => 'WebHACC::Language::XML',
+ 'application/xml' => 'WebHACC::Language::XML',
+ ## TODO: Should we make all XML MIME Types fall
+ ## into this category?
+
+ ## NOTE: This type has different model from normal XML types.
+ 'application/rdf+xml' => 'WebHACC::Language::XML',
+ }->{$input->{media_type}} || 'WebHACC::Language::Default';
+
+ eval qq{ require $checker_class } or die "$0: Loading $checker_class: $@";
+ my $checker = $checker_class->new;
+ $checker->input ($input);
+ $checker->output ($out);
+ $checker->result ($result);
+
+ ## TODO: A cache manifest MUST be text/cache-manifest
+ ## TODO: WebIDL media type "text/x-webidl"
+
+ $checker->generate_syntax_error_section;
+ $checker->generate_source_string_section;
+
+ $checker->onsubdoc (sub {
+ push @subdoc, shift;
+ });
+
+ $checker->generate_structure_dump_section;
+ $checker->generate_structure_error_section;
+ $checker->generate_additional_sections;
- 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});
+=pod
- print STDOUT qq[
-
-
Media type @{[htescape $input->{media_type}]} is not supported!
-
-
Document Tree
-];
- push @nav, ['#document-tree' => 'Tree'];
-
- print_document_tree ($el || $doc);
-
+ my $id_prefix = 0;
+ for my $_subinput (@subdoc) {
+ my $subinput = WebHACC::Input->new;
+ $subinput->{$_} = $_subinput->{$_} for keys %$_subinput;
+ $subinput->id_prefix ('subdoc-' . ++$id_prefix);
+ $subinput->nested (1);
+ $subinput->{base_uri} = $subinput->{container_node}->base_uri
+ unless defined $subinput->{base_uri};
+ my $ebaseuri = htescape ($subinput->{base_uri});
+ $out->start_section (id => $subinput->id_prefix,
+ title => qq[Subdocument #$id_prefix]);
print STDOUT qq[
-
-
-
-
Document Errors
-
-
];
- push @nav, ['#document-errors' => 'Document Error'];
+
+ - Internet Media Type
+ @{[htescape $subinput->{media_type}]}
+ - Container Node
+ - @{[get_node_link ($input, $subinput->{container_node})]}
+ - Base URI
+ <$ebaseuri>
];
+
+ $subinput->{id_prefix} .= '-';
+ check_and_print ($subinput => $result => $out);
- require Whatpm::ContentChecker;
- my $onerror = sub {
- my %opt = @_;
- my ($type, $cls, $msg) = get_text ($opt{type}, $opt{level});
- $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[
-
+sub print_table_section ($$) {
+ my ($input, $tables) = @_;
+
+# push @nav, [qq[#$input->{id_prefix}tables] => 'Tables']
+# unless $input->{nested};
+ 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[
];
+
+ require JSON;
+
+ my $i = 0;
+ for my $table (@$tables) {
+ $i++;
+ print STDOUT qq[
];
}
+
+ print STDOUT qq[
-
Input Error: @{[htescape ($input->{error_status_text})]}
-
';
-
- my @node = ($node);
- while (@node) {
- my $child = shift @node;
- 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
- }
+ return '??';
}
-
- $r .= '
';
- print STDOUT $r;
-} # print_document_tree
-
-sub get_node_path ($) {
- my $node = shift;
- my @r;
- while (defined $node) {
- my $rs;
- if ($node->node_type == 1) {
- $rs = $node->manakai_local_name;
- $node = $node->parent_node;
- } elsif ($node->node_type == 2) {
- $rs = '@' . $node->manakai_local_name;
- $node = $node->owner_element;
- } elsif ($node->node_type == 3) {
- $rs = '"' . $node->data . '"';
- $node = $node->parent_node;
- } elsif ($node->node_type == 9) {
- @r = ('') unless @r;
- $rs = '';
- $node = $node->parent_node;
- } else {
- $rs = '#' . $node->node_type;
- $node = $node->parent_node;
- }
- unshift @r, $rs;
- }
- return join '/', @r;
-} # get_node_path
-
-sub get_node_link ($) {
- return qq[] .
- htescape (get_node_path ($_[0])) . qq[];
-} # get_node_link
+} # get_rdf_resource_html
{
my $Msg = {};
sub load_text_catalog ($) {
+# my $self = shift;
my $lang = shift; # MUST be a canonical lang name
- open my $file, '<', "cc-msg.$lang.txt" or die "$0: cc-msg.$lang.txt: $!";
+ 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, $_);
@@ -533,23 +380,43 @@
}
} # load_text_catalog
-sub get_text ($) {
- my ($type, $level) = @_;
+sub get_text ($;$$) {
+# my $self = shift;
+ my ($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)';
+ defined $arg[$1] ? ($arg[$1]) : '(undef)';
+ }ge; ##BUG: ^ must be escaped
+ $msg =~ s{{\@([A-Za-z0-9:_.-]+)}}{
+ UNIVERSAL::can ($node, 'get_attribute_ns')
+ ? ($node->get_attribute_ns (undef, $1)) : ''
+ }ge; ## BUG: ^ must be escaped
+ $msg =~ s{{\@}}{ ## BUG: v must be escaped
+ UNIVERSAL::can ($node, 'value') ? ($node->value) : ''
+ }ge;
+ $msg =~ s{{local-name}}{
+ UNIVERSAL::can ($node, 'manakai_local_name')
+ ? ($node->manakai_local_name) : ''
+ }ge; ## BUG: ^ must be escaped
+ $msg =~ s{{element-local-name}}{
+ (UNIVERSAL::can ($node, 'owner_element') and
+ $node->owner_element)
+ ? ($node->owner_element->manakai_local_name)
+ : '' ## BUG: ^ must be escaped
}ge;
- return ($type, $Msg->{$type}->[0], $msg);
+ return ($type, 'level-' . $level . ' ' . $Msg->{$type}->[0], $msg);
} elsif ($type =~ s/:([^:]*)$//) {
unshift @arg, $1;
redo;
}
}
- return ($type, '', htescape ($_[0]));
+ return ($type, 'level-'.$level, ($_[0]));
+ ## BUG: ^ must be escaped
} # get_text
}
@@ -557,15 +424,18 @@
sub get_input_document ($$) {
my ($http, $dom) = @_;
- my $request_uri = $http->parameter ('uri');
- my $r = {};
+ require Encode;
+ my $request_uri = Encode::decode ('utf-8', $http->get_parameter ('uri'));
+ my $r = WebHACC::Input->new;
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'};
+ $r = WebHACC::Input::Error->new;
+ $r->{uri} = $request_uri;
+ $r->{request_uri} = $request_uri;
+ $r->{error_status_text} = 'URL scheme not allowed';
}
require Message::Util::HostPermit;
@@ -592,8 +462,11 @@
Allow host=*
EOH
unless ($host_permit->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'};
+ my $r = WebHACC::Input::Error->new;
+ $r->{uri} = $request_uri;
+ $r->{request_uri} = $request_uri;
+ $r->{error_status_text} = 'Connection to the host is forbidden';
+ return $r;
}
require LWP::UserAgent;
@@ -605,30 +478,43 @@
$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);
- 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;
## TODO: More strict parsing...
my $ct = $res->header ('Content-Type');
- if (defined $ct and $ct =~ m#^([0-9A-Za-z._+-]+/[0-9A-Za-z._+-]+)#) {
- $r->{media_type} = lc $1;
- }
- if (defined $ct and $ct =~ /;\s*charset\s*=\s*"?(\S+)"?/i) {
+ 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->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);
$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;
@@ -642,17 +528,28 @@
$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->{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->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 $r->{media_type});
@@ -666,6 +563,7 @@
if ($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;
}
@@ -677,6 +575,8 @@
return $r;
}
+ $r->{inner_html_element} = $http->get_parameter ('e');
+
return $r;
} # get_input_document
@@ -709,11 +609,11 @@
=head1 LICENSE
-Copyright 2007 Wakaba
+Copyright 2007-2008 Wakaba
This library is free software; you can redistribute it
and/or modify it under the same terms as Perl itself.
=cut
-## $Date: 2007/07/17 14:28:20 $
+## $Date: 2008/07/20 16:53:10 $