WebHACC/Language/HTML.pm

package WebHACC::Language::HTML;
use strict;
require WebHACC::Language::DOM;
push our @ISA, 'WebHACC::Language::DOM';

require Message::DOM::DOMImplementation;

sub new ($) {
  return bless {}, shift;
} # new

sub generate_syntax_error_section ($) {
  my $self = shift;
  
  require Encode;
  require Whatpm::HTML;
  
  my $out = $self->output;
  $out->start_section (role => 'parse-errors');
  $out->start_error_list (role => 'parse-errors');

  my $input = $self->input;
  my $result = $self->result;

  my $onerror = sub {
    $result->add_error (@_, layer => 'syntax');
  };

  my $dom = Message::DOM::DOMImplementation->new;
  my $doc = $dom->create_document;
  my $el;
  my $inner_html_element = $input->{inner_html_element};
  if (defined $inner_html_element and length $inner_html_element) {
    $input->{charset} ||= 'windows-1252'; ## TODO: for now.
    my $t = \($input->{s});
    unless ($input->{is_char_string}) {
      $t = \(Encode::decode ($input->{charset}, $$t));
    }
    
    $el = $doc->create_element_ns
        ('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
    Whatpm::HTML->set_inner_html ($el, $$t, $onerror);

    $self->{structure} = $el;
  } else {
    if ($input->{is_char_string}) {
      Whatpm::HTML->parse_char_string ($input->{s} => $doc, $onerror);
    } else {
      Whatpm::HTML->parse_byte_string
          ($input->{charset}, $input->{s} => $doc, $onerror);
    }

    $self->{structure} = $doc;
  }
  $doc->manakai_charset ($input->{official_charset})
      if defined $input->{official_charset};

  $doc->document_uri ($input->{uri});
  $doc->manakai_entity_base_uri ($input->{base_uri});

  $out->end_error_list (role => 'parse-errors');
  $out->end_section;
} # generate_syntax_error_section

sub source_charset ($) {
  my $self = shift;
  return $self->input->{charset} || ($self->{structure}->owner_document || $self->{structure})->input_encoding;
  ## TODO: Can we always use input_encoding?
} # source_charset

1;
1	wakaba	1.1	package WebHACC::Language::HTML;
2			use strict;
3			require WebHACC::Language::DOM;
4			push our @ISA, 'WebHACC::Language::DOM';
5
6			require Message::DOM::DOMImplementation;
7
8			sub new ($) {
9			return bless {}, shift;
10			} # new
11
12			sub generate_syntax_error_section ($) {
13			my $self = shift;
14
15			require Encode;
16			require Whatpm::HTML;
17
18			my $out = $self->output;
19	wakaba	1.3	$out->start_section (role => 'parse-errors');
20			$out->start_error_list (role => 'parse-errors');
21	wakaba	1.1
22			my $input = $self->input;
23			my $result = $self->result;
24
25			my $onerror = sub {
26	wakaba	1.2	$result->add_error (@_, layer => 'syntax');
27	wakaba	1.1	};
28
29			my $dom = Message::DOM::DOMImplementation->new;
30			my $doc = $dom->create_document;
31			my $el;
32			my $inner_html_element = $input->{inner_html_element};
33			if (defined $inner_html_element and length $inner_html_element) {
34			$input->{charset} \|\|= 'windows-1252'; ## TODO: for now.
35			my $t = \($input->{s});
36			unless ($input->{is_char_string}) {
37			$t = \(Encode::decode ($input->{charset}, $$t));
38			}
39
40			$el = $doc->create_element_ns
41			('http://www.w3.org/1999/xhtml', [undef, $inner_html_element]);
42			Whatpm::HTML->set_inner_html ($el, $$t, $onerror);
43
44			$self->{structure} = $el;
45			} else {
46			if ($input->{is_char_string}) {
47			Whatpm::HTML->parse_char_string ($input->{s} => $doc, $onerror);
48			} else {
49			Whatpm::HTML->parse_byte_string
50			($input->{charset}, $input->{s} => $doc, $onerror);
51			}
52
53			$self->{structure} = $doc;
54			}
55			$doc->manakai_charset ($input->{official_charset})
56			if defined $input->{official_charset};
57
58			$doc->document_uri ($input->{uri});
59			$doc->manakai_entity_base_uri ($input->{base_uri});
60
61	wakaba	1.3	$out->end_error_list (role => 'parse-errors');
62	wakaba	1.1	$out->end_section;
63			} # generate_syntax_error_section
64
65			sub source_charset ($) {
66			my $self = shift;
67			return $self->input->{charset} \|\| ($self->{structure}->owner_document \|\| $self->{structure})->input_encoding;
68			## TODO: Can we always use input_encoding?
69			} # source_charset
70
71			1;