--- test/html-webhacc/cc.cgi 2007/09/10 12:09:34 1.20 +++ test/html-webhacc/cc.cgi 2008/02/10 02:42:01 1.33 @@ -1,5 +1,6 @@ #!/usr/bin/perl use strict; +use utf8; use lib qw[/home/httpd/html/www/markup/html/whatpm /home/wakaba/work/manakai2/lib]; @@ -51,7 +52,6 @@ $| = 0; my $input = get_input_document ($http, $dom); - my $inner_html_element = $http->get_parameter ('e'); my $char_length = 0; my %time; @@ -61,7 +61,12 @@
Request URI
<@{[htescape $input->{request_uri}]}>
Document URI
-
<@{[htescape $input->{uri}]}>
+
<@{[htescape $input->{uri}]}> +
]; # no yet push @nav, ['#document-info' => 'Information']; @@ -73,7 +78,7 @@
<@{[htescape $input->{base_uri}]}>
Internet Media Type
@{[htescape $input->{media_type}]} - @{[$input->{media_type_overridden} ? '(overridden)' : '']}
+ @{[$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
@{[defined $input->{charset} ? ''.htescape ($input->{charset}).'' : '(none)']} @{[$input->{charset_overridden} ? '(overridden)' : '']}
@@ -84,38 +89,7 @@ ]; 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 %{$elements->{class}}; - } - + check_and_print ($input => $result); print_result_section ($result); } else { print STDOUT qq[]; @@ -134,7 +108,8 @@ ]; - for (qw/decode parse parse_xml check/) { + 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"; @@ -151,7 +126,7 @@ $result->{conforming_min} = 0; } elsif ($err->{level} eq 'w' or $err->{level} eq 'g') { $result->{$layer}->{warning}++; - } elsif ($err->{level} eq 'unsupported') { + } elsif ($err->{level} eq 'u' or $err->{level} eq 'unsupported') { $result->{$layer}->{unsupported}++; $result->{unsupported} = 1; } else { @@ -170,14 +145,68 @@ } } # 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']; - print STDOUT qq[
+ push @nav, ['#source-header' => 'HTTP Header'] unless $input->{nested}; + print STDOUT qq[

HTTP Header

Note: Due to the limitation of the @@ -209,19 +238,13 @@ 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']; + push @nav, ['#parse-errors' => 'Parse Error'] unless $input->{nested}; my $onerror = sub { my (%opt) = @_; @@ -235,23 +258,35 @@ $type =~ tr/ /-/; $type =~ s/\|/%7C/g; $msg .= qq[ [Description]]; - print STDOUT qq[
$msg
\n]; + print STDOUT qq[
], get_error_level_label (\%opt); + print STDOUT qq[$msg
\n]; add_error ('syntax', \%opt => $result); }; my $doc = $dom->create_document; my $el; - $time1 = time; + 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]); + $time1 = time; Whatpm::HTML->set_inner_html ($el, $t, $onerror); + $time{parse} = time - $time1; } else { - Whatpm::HTML->parse_string ($t => $doc, $onerror); + my $time1 = time; + Whatpm::HTML->parse_byte_string + ($input->{charset}, $input->{s} => $doc, $onerror); + $time{parse_html} = time - $time1; } - $time{parse} = time - $time1; - + $doc->manakai_charset ($input->{official_charset}) + if defined $input->{official_charset}; + print STDOUT qq[
]; return ($doc, $el); @@ -263,11 +298,11 @@ require Message::DOM::XMLParserTemp; 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; @@ -291,12 +326,50 @@ 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 +sub print_syntax_error_manifest_section ($$) { + my ($input, $result) = @_; + + require Whatpm::CacheManifest; + + 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}, $opt{level}); + print STDOUT qq[
], get_error_label ($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]; + + add_error ('syntax', \%opt => $result); + }; + + 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 @@ -304,20 +377,22 @@ my $s = \($enc->decode (${$_[0]})); my $i = 1; - push @nav, ['#source-string' => 'Source']; - print STDOUT qq[
+ 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[
  1. ], htescape $1, "
    2. \n"; + print STDOUT qq[
  2. ], htescape $1, + "
    3. \n"; $i++; } if ($$s =~ /\G([^\x0A]+)/gc) { - print STDOUT qq[
  3. ], htescape $1, "
    4. \n"; + print STDOUT qq[
  4. ], htescape $1, + "
    5. \n"; } } else { - print STDOUT q[
  5. ]; + print STDOUT q[
  6. ]; } print STDOUT "
"; } # print_input_string_section @@ -334,7 +409,7 @@ next; } - my $node_id = 'node-'.refaddr $child; + my $node_id = $input->{id_prefix} . 'node-'.refaddr $child; my $nt = $child->node_type; if ($nt == $child->ELEMENT_NODE) { my $child_nsuri = $child->namespace_uri; @@ -345,7 +420,7 @@ $r .= '
    '; for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value, $_->namespace_uri, 'node-'.refaddr $_] } @{$child->attributes}) { - $r .= qq[
  • ] . htescape ($attr->[0]) . ' = '; ## ISSUE: case? + $r .= qq[
  • ] . htescape ($attr->[0]) . ' = '; ## ISSUE: case? $r .= '' . htescape ($attr->[1]) . '
    • '; ## TODO: children } $r .= '
'; @@ -366,6 +441,21 @@ } elsif ($nt == $child->DOCUMENT_NODE) { $r .= qq'
  • Document'; $r .= qq[
      ]; + my $cp = $child->manakai_charset; + if (defined $cp) { + $r .= qq[
    • charset parameter = ]; + $r .= htescape ($cp) . qq[
      • ]; + } + $r .= qq[
    • inputEncoding = ]; + my $ie = $child->input_encoding; + if (defined $ie) { + $r .= qq[@{[htescape ($ie)]}]; + if ($child->manakai_has_bom) { + $r .= qq[ (with BOM)]; + } + } else { + $r .= qq[(null)]; + } $r .= qq[
    • @{[scalar get_text ('manakaiIsHTML:'.($child->manakai_is_html?1:0))]}
      • ]; $r .= qq[
    • @{[scalar get_text ('manakaiCompatMode:'.$child->manakai_compat_mode)]}
      • ]; unless ($child->manakai_is_html) { @@ -399,28 +489,62 @@ print STDOUT $r; } # print_document_tree -sub print_structure_dump_section ($$) { - my ($doc, $el) = @_; +sub print_structure_dump_dom_section ($$$) { + my ($input, $doc, $el) = @_; print STDOUT qq[ -
      +

      Document Tree

      ]; - push @nav, ['#document-tree' => 'Tree']; + push @nav, ['#document-tree' => 'Tree'] unless $input->{nested}; print_document_tree ($el || $doc); print STDOUT qq[
      ]; -} # print_structure_dump_section +} # print_structure_dump_dom_section -sub print_structure_error_section ($$$) { - my ($doc, $el, $result) = @_; +sub print_structure_dump_manifest_section ($$) { + my ($input, $manifest) = @_; - print STDOUT qq[
      + 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]->{$uri}); + print STDOUT qq[ + ]; + } + + print STDOUT qq[
      Oppotunistic Caching NamespaceFallback Entry
      <$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']; + push @nav, ['#document-errors' => 'Document Error'] unless $input->{nested}; require Whatpm::ContentChecker; my $onerror = sub { @@ -429,8 +553,9 @@ $type =~ tr/ /-/; $type =~ s/\|/%7C/g; $msg .= qq[ [Description]]; - print STDOUT qq[
      ] . get_node_link ($opt{node}) . - qq[
      \n
      ], $msg, "
      \n"; + print STDOUT qq[
      ] . get_error_label ($input, \%opt) . + qq[
      \n
      ], get_error_level_label (\%opt); + print STDOUT $msg, "
      \n"; add_error ('structure', \%opt => $result); }; @@ -446,14 +571,38 @@ print STDOUT qq[
      ]; return $elements; -} # print_structure_error_section +} # 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 $input->{nested}; -sub print_table_section ($) { - my $tables = shift; + require Whatpm::CacheManifest; + Whatpm::CacheManifest->check_manifest ($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']; + push @nav, ['#tables' => 'Tables'] unless $input->{nested}; print STDOUT qq[ -
      +

      Tables

      @@ -468,8 +617,8 @@ my $i = 0; for my $table_el (@$tables) { $i++; - print STDOUT qq[

      ] . - get_node_link ($table_el) . q[

      ]; + 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. @@ -501,68 +650,31 @@ print STDOUT '
      ]; + print STDOUT qq[, document.getElementById ('$input->{id_prefix}table-$i')]; + print STDOUT qq[, '$input->{id_prefix}');]; } print STDOUT qq[
      ]; } # print_table_section -sub print_id_section ($) { - my $ids = shift; +sub print_listing_section ($$$) { + my ($opt, $input, $ids) = @_; - push @nav, ['#identifiers' => 'IDs']; + push @nav, ['#' . $opt->{id} => $opt->{label}] unless $input->{nested}; print STDOUT qq[ -
      -

      Identifiers

      +
      +

      $opt->{heading}

      ]; for my $id (sort {$a cmp $b} keys %$ids) { print STDOUT qq[
      @{[htescape $id]}
      ]; for (@{$ids->{$id}}) { - print STDOUT qq[
      ].get_node_link ($_).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[
      ].get_node_link ($input, $_).qq[
      ]; } } print STDOUT qq[
      ]; -} # print_term_section - -sub print_class_section ($) { - 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 +} # print_listing_section sub print_result_section ($) { my $result = shift; @@ -571,11 +683,11 @@

      Result

      ]; - if ($result->{unsupported}) { + if ($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.

      ]; + features. The document might or might not be conforming.

      ]; } elsif ($result->{conforming_min}) { print STDOUT qq[

      No conformance-error is found in this document.

      ]; @@ -591,10 +703,13 @@ print STDOUT qq[ - -]; + + + + +]; my $must_error = 0; my $should_error = 0; @@ -602,6 +717,7 @@ my $score_min = 0; my $score_max = 0; my $score_base = 20; + my $score_unit = $score_base / 100; for ( [Transfer => 'transfer', ''], [Character => 'char', ''], @@ -611,8 +727,8 @@ $must_error += ($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); + $score_min += (($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 = $_->[0]; @@ -625,11 +741,11 @@ print STDOUT qq[]; if ($uncertain) { - print qq[]; + print qq[]; } elsif ($result->{$_->[1]}->{score_min} != $result->{$_->[1]}->{score_max}) { - print qq[]; + print qq[]; } else { - print qq[]; + print qq[]; } } @@ -638,7 +754,11 @@ print STDOUT qq[ - + + + + +
      MUST-level -ErrorsSHOULD-level -ErrorsWarningsScore
      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_max}$result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max} + $score_base
      $result->{$_->[1]}->{score_min}..$result->{$_->[1]}->{score_max}
      $result->{$_->[1]}->{score_min}
      $result->{$_->[1]}->{score_min}
      Semantics0?0?0?−∞..$score_base
      Total$must_error?$should_error?$warning?−∞..$score_max
      Total$must_error?$should_error?$warning?−∞..$score_max

      Important: This conformance checking service @@ -647,15 +767,28 @@ push @nav, ['#result-summary' => 'Result']; } # print_result_section -sub print_result_unknown_type_section ($) { - my $input = shift; +sub print_result_unknown_type_section ($$) { + my ($input, $result) = @_; + my $euri = htescape ($input->{uri}); print STDOUT qq[ -

      -

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

      +
      +

      Errors

      + +
      +
      <$euri>
      +
      Not + supported: + Media type + @{[htescape $input->{media_type}]} + is not supported.
      +
      ]; - push @nav, ['#result-summary' => 'Result']; + push @nav, ['#parse-errors' => 'Errors']; + add_error (char => {level => 'u'} => $result); + add_error (syntax => {level => 'u'} => $result); + add_error (structure => {level => 'u'} => $result); } # print_result_unknown_type_section sub print_result_input_error_section ($) { @@ -664,7 +797,65 @@

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

      ]; push @nav, ['#result-summary' => 'Result']; -} # print_Result_input_error_section +} # print_result_input_error_section + +sub get_error_label ($$) { + my ($input, $err) = @_; + + my $r = ''; + + if (defined $err->{line}) { + if ($err->{column} > 0) { + $r = qq[Line $err->{line} column $err->{column}]; + } else { + $err->{line} = $err->{line} - 1 || 1; + $r = qq[Line $err->{line}]; + } + } + + if (defined $err->{node}) { + $r .= ' ' if length $r; + $r = get_node_link ($input, $err->{node}); + } + + if (defined $err->{index}) { + $r .= ' ' if length $r; + $r .= 'Index ' . (0+$err->{index}); + } + + if (defined $err->{value}) { + $r .= ' ' if length $r; + $r .= '' . htescape ($err->{value}) . ''; + } + + return $r; +} # get_error_label + +sub get_error_level_label ($) { + my $err = shift; + + my $r = ''; + + if (not defined $err->{level} or $err->{level} eq 'm') { + $r = qq[MUST‐level + error: ]; + } elsif ($err->{level} eq 's') { + $r = qq[SHOULD‐level + error: ]; + } elsif ($err->{level} eq 'w') { + $r = qq[Warning: + ]; + } elsif ($err->{level} eq 'u' or $err->{level} eq 'unsupported') { + $r = qq[Not + supported: ]; + } else { + my $elevel = htescape ($err->{level}); + $r = qq[$elevel: + ]; + } + + return $r; +} # get_error_level_label sub get_node_path ($) { my $node = shift; @@ -693,9 +884,9 @@ return join '/', @r; } # get_node_path -sub get_node_link ($) { - return qq[] . - htescape (get_node_path ($_[0])) . qq[]; +sub get_node_link ($$) { + return qq[] . + htescape (get_node_path ($_[1])) . qq[]; } # get_node_link { @@ -703,7 +894,8 @@ sub load_text_catalog ($) { 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, $_); @@ -716,6 +908,7 @@ sub get_text ($) { my ($type, $level, $node) = @_; $type = $level . ':' . $type if defined $level; + $level = 'm' unless defined $level; my @arg; { if (defined $Msg->{$type}) { @@ -740,13 +933,13 @@ ? htescape ($node->owner_element->manakai_local_name) : '' }ge; - return ($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-'.$level, htescape ($_[0])); } # get_text } @@ -802,6 +995,7 @@ $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')) { @@ -811,12 +1005,10 @@ ## 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->get_parameter ('charset'); @@ -824,9 +1016,22 @@ $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; @@ -847,7 +1052,18 @@ $r->{charset} = ''.$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 = $http->get_parameter ('i'); @@ -864,6 +1080,7 @@ if ($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; } @@ -914,4 +1131,4 @@ =cut -## $Date: 2007/09/10 12:09:34 $ +## $Date: 2008/02/10 02:42:01 $