whatpm/t/HTML-tokenizer.t

#!/usr/bin/perl
use strict;

my $dir_name;
BEGIN {
  $dir_name = 't/tokenizer/';
  my $skip = "You don't have JSON module";
  eval q{
         use JSON 1.00;
         $skip = "You don't have make command";
         system ('cd $dir_name; cd ..; make tokenizer-files') == 0 or die
           unless -f $dir_name.'test1.test';
         $skip = '';
        };
  if ($skip) {
    print "1..1\n";
    print "ok 1 # $skip\n";
    exit;
  }
  $JSON::UnMapping = 1;
  $JSON::UTF8 = 1;
}

use Test;
BEGIN { plan tests => 38 }

use Data::Dumper;
$Data::Dumper::Useqq = 1;
sub Data::Dumper::qquote {
  my $s = shift;
  $s =~ s/([^\x20\x21-\x26\x28-\x5B\x5D-\x7E])/sprintf '\x{%02X}', ord $1/ge;
  return q<qq'> . $s . q<'>;
} # Data::Dumper::qquote

use What::HTML;
use Encode;

for my $file_name (qw[
                      test1.test
                      test2.test
                      contentModelFlags.test
                     ]) {
  open my $file, '<', $dir_name.$file_name
    or die "$0: $dir_name$file_name: $!";
  local $/ = undef;
  my $js = <$file>;
  close $file;
  
  my $tests = jsonToObj ($js)->{tests};
  TEST: for my $test (@$tests) {
    my $s = $test->{input};
    
    my $j = 1;
    while ($j < @{$test->{output}}) {
      if (ref $test->{output}->[$j - 1] and
          $test->{output}->[$j - 1]->[0] eq 'Character' and
          ref $test->{output}->[$j] and 
          $test->{output}->[$j]->[0] eq 'Character') {
        $test->{output}->[$j - 1]->[1]
          .= $test->{output}->[$j]->[1];
        splice @{$test->{output}}, $j, 1;
      }
      $j++;
    }

    my @cm = @{$test->{contentModelFlags} || ['PCDATA']};
    my $last_start_tag = $test->{lastStartTag};
    for my $cm (@cm) {
      my $p = What::HTML->new;
      my $i = 0;
      $p->{set_next_input_character} = sub {
        my $self = shift;
        $self->{next_input_character} = -1 and return if $i >= length $s;
        $self->{next_input_character} = ord substr $s, $i++, 1;

        if ($self->{next_input_character} == 0x000D) { # CR
          if ($i >= length $s) {
            #
          } else {
            my $next_char = ord substr $s, $i++, 1;
            if ($next_char == 0x000A) { # LF
              #
            } else {
              push @{$self->{char}}, $next_char;
            }
          }
          $self->{next_input_character} = 0x000A; # LF # MUST
        } elsif ($self->{next_input_character} > 0x10FFFF) {
          $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
        } elsif ($self->{next_input_character} == 0x0000) { # NULL
          $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
        }
      };
      
      my @token;
      $p->{parse_error} = sub {
        push @token, 'ParseError';
      };
      
      $p->_initialize_tokenizer;
      $p->{content_model_flag} = $cm;
      $p->{last_emitted_start_tag_name} = $last_start_tag;

      while (1) {
        my $token = $p->_get_next_token;
        last if $token->{type} eq 'end-of-file';
        
        my $test_token = [
         {
          DOCTYPE => 'DOCTYPE',
          'start tag' => 'StartTag',
          'end tag' => 'EndTag',
          comment => 'Comment',
          character => 'Character',
         }->{$token->{type}} || $token->{type},
        ];
        $test_token->[1] = $token->{name} if defined $token->{name};
        $test_token->[1] = $token->{tag_name} if defined $token->{tag_name};
        $test_token->[1] = $token->{data} if defined $token->{data};
        $test_token->[2] = $token->{error} ? 1 : 0 if $token->{type} eq 'DOCTYPE';
        $test_token->[2] = {map {$_->{name} => $_->{value}} values %{$token->{attributes}}}
          if $token->{type} eq 'start tag';
        
        if (@token and ref $token[-1] and $token[-1]->[0] eq 'Character' and
            $test_token->[0] eq 'Character') {
          $token[-1]->[1] .= $test_token->[1];
        } else {
          push @token, $test_token;
        }
      }
      
      my $expected_dump = Dumper ($test->{output});
      my $parser_dump = Dumper (\@token);
      ok $parser_dump, $expected_dump,
        $test->{description} . ': ' . $test->{input};
    }
  }
}

## $Date:$
1	wakaba	1.1	#!/usr/bin/perl
2			use strict;
3
4	wakaba	1.2	my $dir_name;
5	wakaba	1.1	BEGIN {
6	wakaba	1.2	$dir_name = 't/tokenizer/';
7	wakaba	1.1	my $skip = "You don't have JSON module";
8			eval q{
9			use JSON 1.00;
10			$skip = "You don't have make command";
11	wakaba	1.2	system ('cd $dir_name; cd ..; make tokenizer-files') == 0 or die
12			unless -f $dir_name.'test1.test';
13	wakaba	1.1	$skip = '';
14			};
15			if ($skip) {
16			print "1..1\n";
17			print "ok 1 # $skip\n";
18			exit;
19			}
20			$JSON::UnMapping = 1;
21	wakaba	1.2	$JSON::UTF8 = 1;
22	wakaba	1.1	}
23
24			use Test;
25	wakaba	1.2	BEGIN { plan tests => 38 }
26
27	wakaba	1.1	use Data::Dumper;
28	wakaba	1.2	$Data::Dumper::Useqq = 1;
29			sub Data::Dumper::qquote {
30			my $s = shift;
31			$s =~ s/([^\x20\x21-\x26\x28-\x5B\x5D-\x7E])/sprintf '\x{%02X}', ord $1/ge;
32			return q<qq'> . $s . q<'>;
33			} # Data::Dumper::qquote
34	wakaba	1.1
35			use What::HTML;
36	wakaba	1.2	use Encode;
37	wakaba	1.1
38			for my $file_name (qw[
39	wakaba	1.2	test1.test
40			test2.test
41			contentModelFlags.test
42	wakaba	1.1	]) {
43	wakaba	1.2	open my $file, '<', $dir_name.$file_name
44			or die "$0: $dir_name$file_name: $!";
45	wakaba	1.1	local $/ = undef;
46			my $js = <$file>;
47			close $file;
48
49			my $tests = jsonToObj ($js)->{tests};
50			TEST: for my $test (@$tests) {
51			my $s = $test->{input};
52
53			my $j = 1;
54			while ($j < @{$test->{output}}) {
55			if (ref $test->{output}->[$j - 1] and
56			$test->{output}->[$j - 1]->[0] eq 'Character' and
57			ref $test->{output}->[$j] and
58			$test->{output}->[$j]->[0] eq 'Character') {
59			$test->{output}->[$j - 1]->[1]
60			.= $test->{output}->[$j]->[1];
61			splice @{$test->{output}}, $j, 1;
62			}
63			$j++;
64			}
65
66	wakaba	1.2	my @cm = @{$test->{contentModelFlags} \|\| ['PCDATA']};
67			my $last_start_tag = $test->{lastStartTag};
68	wakaba	1.1	for my $cm (@cm) {
69			my $p = What::HTML->new;
70			my $i = 0;
71			$p->{set_next_input_character} = sub {
72			my $self = shift;
73			$self->{next_input_character} = -1 and return if $i >= length $s;
74			$self->{next_input_character} = ord substr $s, $i++, 1;
75	wakaba	1.2
76			if ($self->{next_input_character} == 0x000D) { # CR
77			if ($i >= length $s) {
78			#
79			} else {
80			my $next_char = ord substr $s, $i++, 1;
81			if ($next_char == 0x000A) { # LF
82			#
83			} else {
84			push @{$self->{char}}, $next_char;
85			}
86			}
87			$self->{next_input_character} = 0x000A; # LF # MUST
88			} elsif ($self->{next_input_character} > 0x10FFFF) {
89			$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
90			} elsif ($self->{next_input_character} == 0x0000) { # NULL
91			$self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
92			}
93	wakaba	1.1	};
94
95			my @token;
96			$p->{parse_error} = sub {
97			push @token, 'ParseError';
98			};
99
100			$p->_initialize_tokenizer;
101			$p->{content_model_flag} = $cm;
102	wakaba	1.2	$p->{last_emitted_start_tag_name} = $last_start_tag;
103	wakaba	1.1
104			while (1) {
105			my $token = $p->_get_next_token;
106			last if $token->{type} eq 'end-of-file';
107
108			my $test_token = [
109			{
110			DOCTYPE => 'DOCTYPE',
111			'start tag' => 'StartTag',
112			'end tag' => 'EndTag',
113			comment => 'Comment',
114			character => 'Character',
115			}->{$token->{type}} \|\| $token->{type},
116			];
117			$test_token->[1] = $token->{name} if defined $token->{name};
118			$test_token->[1] = $token->{tag_name} if defined $token->{tag_name};
119			$test_token->[1] = $token->{data} if defined $token->{data};
120			$test_token->[2] = $token->{error} ? 1 : 0 if $token->{type} eq 'DOCTYPE';
121			$test_token->[2] = {map {$_->{name} => $_->{value}} values %{$token->{attributes}}}
122			if $token->{type} eq 'start tag';
123
124			if (@token and ref $token[-1] and $token[-1]->[0] eq 'Character' and
125			$test_token->[0] eq 'Character') {
126			$token[-1]->[1] .= $test_token->[1];
127			} else {
128			push @token, $test_token;
129			}
130			}
131
132			my $expected_dump = Dumper ($test->{output});
133			my $parser_dump = Dumper (\@token);
134	wakaba	1.2	ok $parser_dump, $expected_dump,
135			$test->{description} . ': ' . $test->{input};
136	wakaba	1.1	}
137			}
138			}
139
140	wakaba	1.2	## $Date:$