/[suikacvs]/markup/html/whatpm/Whatpm/HTML.pm.src
Suika

Diff of /markup/html/whatpm/Whatpm/HTML.pm.src

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.112 by wakaba, Sun Mar 16 06:39:57 2008 UTC revision 1.196 by wakaba, Sat Oct 4 07:58:58 2008 UTC
# Line 3  use strict; Line 3  use strict;
3  our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};  our $VERSION=do{my @r=(q$Revision$=~/\d+/g);sprintf "%d."."%02d" x $#r,@r};
4  use Error qw(:try);  use Error qw(:try);
5    
6    ## NOTE: This module don't check all HTML5 parse errors; character
7    ## encoding related parse errors are expected to be handled by relevant
8    ## modules.
9    ## Parse errors for control characters that are not allowed in HTML5
10    ## documents, for surrogate code points, and for noncharacter code
11    ## points, as well as U+FFFD substitions for characters whose code points
12    ## is higher than U+10FFFF may be detected by combining the parser with
13    ## the checker implemented by Whatpm::Charset::UnicodeChecker (for its
14    ## usage example, see |t/HTML-tree.t| in the Whatpm package or the
15    ## WebHACC::Language::HTML module in the WebHACC package).
16    
17  ## ISSUE:  ## ISSUE:
18  ## var doc = implementation.createDocument (null, null, null);  ## var doc = implementation.createDocument (null, null, null);
19  ## doc.write ('');  ## doc.write ('');
20  ## alert (doc.compatMode);  ## alert (doc.compatMode);
21    
22  ## TODO: Control charcters and noncharacters are not allowed (HTML5 revision 1263)  require IO::Handle;
23  ## TODO: 1252 parse error (revision 1264)  
24  ## TODO: 8859-11 = 874 (revision 1271)  my $HTML_NS = q<http://www.w3.org/1999/xhtml>;
25    my $MML_NS = q<http://www.w3.org/1998/Math/MathML>;
26  my $permitted_slash_tag_name = {  my $SVG_NS = q<http://www.w3.org/2000/svg>;
27    base => 1,  my $XLINK_NS = q<http://www.w3.org/1999/xlink>;
28    link => 1,  my $XML_NS = q<http://www.w3.org/XML/1998/namespace>;
29    meta => 1,  my $XMLNS_NS = q<http://www.w3.org/2000/xmlns/>;
30    hr => 1,  
31    br => 1,  sub A_EL () { 0b1 }
32    img => 1,  sub ADDRESS_EL () { 0b10 }
33    embed => 1,  sub BODY_EL () { 0b100 }
34    param => 1,  sub BUTTON_EL () { 0b1000 }
35    area => 1,  sub CAPTION_EL () { 0b10000 }
36    col => 1,  sub DD_EL () { 0b100000 }
37    input => 1,  sub DIV_EL () { 0b1000000 }
38    sub DT_EL () { 0b10000000 }
39    sub FORM_EL () { 0b100000000 }
40    sub FORMATTING_EL () { 0b1000000000 }
41    sub FRAMESET_EL () { 0b10000000000 }
42    sub HEADING_EL () { 0b100000000000 }
43    sub HTML_EL () { 0b1000000000000 }
44    sub LI_EL () { 0b10000000000000 }
45    sub NOBR_EL () { 0b100000000000000 }
46    sub OPTION_EL () { 0b1000000000000000 }
47    sub OPTGROUP_EL () { 0b10000000000000000 }
48    sub P_EL () { 0b100000000000000000 }
49    sub SELECT_EL () { 0b1000000000000000000 }
50    sub TABLE_EL () { 0b10000000000000000000 }
51    sub TABLE_CELL_EL () { 0b100000000000000000000 }
52    sub TABLE_ROW_EL () { 0b1000000000000000000000 }
53    sub TABLE_ROW_GROUP_EL () { 0b10000000000000000000000 }
54    sub MISC_SCOPING_EL () { 0b100000000000000000000000 }
55    sub MISC_SPECIAL_EL () { 0b1000000000000000000000000 }
56    sub FOREIGN_EL () { 0b10000000000000000000000000 }
57    sub FOREIGN_FLOW_CONTENT_EL () { 0b100000000000000000000000000 }
58    sub MML_AXML_EL () { 0b1000000000000000000000000000 }
59    sub RUBY_EL () { 0b10000000000000000000000000000 }
60    sub RUBY_COMPONENT_EL () { 0b100000000000000000000000000000 }
61    
62    sub TABLE_ROWS_EL () {
63      TABLE_EL |
64      TABLE_ROW_EL |
65      TABLE_ROW_GROUP_EL
66    }
67    
68    ## NOTE: Used in "generate implied end tags" algorithm.
69    ## NOTE: There is a code where a modified version of
70    ## END_TAG_OPTIONAL_EL is used in "generate implied end tags"
71    ## implementation (search for the algorithm name).
72    sub END_TAG_OPTIONAL_EL () {
73      DD_EL |
74      DT_EL |
75      LI_EL |
76      OPTION_EL |
77      OPTGROUP_EL |
78      P_EL |
79      RUBY_COMPONENT_EL
80    }
81    
82    ## NOTE: Used in </body> and EOF algorithms.
83    sub ALL_END_TAG_OPTIONAL_EL () {
84      DD_EL |
85      DT_EL |
86      LI_EL |
87      P_EL |
88    
89      BODY_EL |
90      HTML_EL |
91      TABLE_CELL_EL |
92      TABLE_ROW_EL |
93      TABLE_ROW_GROUP_EL
94    }
95    
96    sub SCOPING_EL () {
97      BUTTON_EL |
98      CAPTION_EL |
99      HTML_EL |
100      TABLE_EL |
101      TABLE_CELL_EL |
102      MISC_SCOPING_EL
103    }
104    
105    sub TABLE_SCOPING_EL () {
106      HTML_EL |
107      TABLE_EL
108    }
109    
110    sub TABLE_ROWS_SCOPING_EL () {
111      HTML_EL |
112      TABLE_ROW_GROUP_EL
113    }
114    
115    sub TABLE_ROW_SCOPING_EL () {
116      HTML_EL |
117      TABLE_ROW_EL
118    }
119    
120    sub SPECIAL_EL () {
121      ADDRESS_EL |
122      BODY_EL |
123      DIV_EL |
124    
125      DD_EL |
126      DT_EL |
127      LI_EL |
128      P_EL |
129    
130      FORM_EL |
131      FRAMESET_EL |
132      HEADING_EL |
133      OPTION_EL |
134      OPTGROUP_EL |
135      SELECT_EL |
136      TABLE_ROW_EL |
137      TABLE_ROW_GROUP_EL |
138      MISC_SPECIAL_EL
139    }
140    
141    my $el_category = {
142      a => A_EL | FORMATTING_EL,
143      address => ADDRESS_EL,
144      applet => MISC_SCOPING_EL,
145      area => MISC_SPECIAL_EL,
146      article => MISC_SPECIAL_EL,
147      aside => MISC_SPECIAL_EL,
148      b => FORMATTING_EL,
149      base => MISC_SPECIAL_EL,
150      basefont => MISC_SPECIAL_EL,
151      bgsound => MISC_SPECIAL_EL,
152      big => FORMATTING_EL,
153      blockquote => MISC_SPECIAL_EL,
154      body => BODY_EL,
155      br => MISC_SPECIAL_EL,
156      button => BUTTON_EL,
157      caption => CAPTION_EL,
158      center => MISC_SPECIAL_EL,
159      col => MISC_SPECIAL_EL,
160      colgroup => MISC_SPECIAL_EL,
161      command => MISC_SPECIAL_EL,
162      datagrid => MISC_SPECIAL_EL,
163      dd => DD_EL,
164      details => MISC_SPECIAL_EL,
165      dialog => MISC_SPECIAL_EL,
166      dir => MISC_SPECIAL_EL,
167      div => DIV_EL,
168      dl => MISC_SPECIAL_EL,
169      dt => DT_EL,
170      em => FORMATTING_EL,
171      embed => MISC_SPECIAL_EL,
172      eventsource => MISC_SPECIAL_EL,
173      fieldset => MISC_SPECIAL_EL,
174      figure => MISC_SPECIAL_EL,
175      font => FORMATTING_EL,
176      footer => MISC_SPECIAL_EL,
177      form => FORM_EL,
178      frame => MISC_SPECIAL_EL,
179      frameset => FRAMESET_EL,
180      h1 => HEADING_EL,
181      h2 => HEADING_EL,
182      h3 => HEADING_EL,
183      h4 => HEADING_EL,
184      h5 => HEADING_EL,
185      h6 => HEADING_EL,
186      head => MISC_SPECIAL_EL,
187      header => MISC_SPECIAL_EL,
188      hr => MISC_SPECIAL_EL,
189      html => HTML_EL,
190      i => FORMATTING_EL,
191      iframe => MISC_SPECIAL_EL,
192      img => MISC_SPECIAL_EL,
193      #image => MISC_SPECIAL_EL, ## NOTE: Commented out in the spec.
194      input => MISC_SPECIAL_EL,
195      isindex => MISC_SPECIAL_EL,
196      li => LI_EL,
197      link => MISC_SPECIAL_EL,
198      listing => MISC_SPECIAL_EL,
199      marquee => MISC_SCOPING_EL,
200      menu => MISC_SPECIAL_EL,
201      meta => MISC_SPECIAL_EL,
202      nav => MISC_SPECIAL_EL,
203      nobr => NOBR_EL | FORMATTING_EL,
204      noembed => MISC_SPECIAL_EL,
205      noframes => MISC_SPECIAL_EL,
206      noscript => MISC_SPECIAL_EL,
207      object => MISC_SCOPING_EL,
208      ol => MISC_SPECIAL_EL,
209      optgroup => OPTGROUP_EL,
210      option => OPTION_EL,
211      p => P_EL,
212      param => MISC_SPECIAL_EL,
213      plaintext => MISC_SPECIAL_EL,
214      pre => MISC_SPECIAL_EL,
215      rp => RUBY_COMPONENT_EL,
216      rt => RUBY_COMPONENT_EL,
217      ruby => RUBY_EL,
218      s => FORMATTING_EL,
219      script => MISC_SPECIAL_EL,
220      select => SELECT_EL,
221      section => MISC_SPECIAL_EL,
222      small => FORMATTING_EL,
223      spacer => MISC_SPECIAL_EL,
224      strike => FORMATTING_EL,
225      strong => FORMATTING_EL,
226      style => MISC_SPECIAL_EL,
227      table => TABLE_EL,
228      tbody => TABLE_ROW_GROUP_EL,
229      td => TABLE_CELL_EL,
230      textarea => MISC_SPECIAL_EL,
231      tfoot => TABLE_ROW_GROUP_EL,
232      th => TABLE_CELL_EL,
233      thead => TABLE_ROW_GROUP_EL,
234      title => MISC_SPECIAL_EL,
235      tr => TABLE_ROW_EL,
236      tt => FORMATTING_EL,
237      u => FORMATTING_EL,
238      ul => MISC_SPECIAL_EL,
239      wbr => MISC_SPECIAL_EL,
240  };  };
241    
242  my $c1_entity_char = {  my $el_category_f = {
243      $MML_NS => {
244        'annotation-xml' => MML_AXML_EL,
245        mi => FOREIGN_FLOW_CONTENT_EL,
246        mo => FOREIGN_FLOW_CONTENT_EL,
247        mn => FOREIGN_FLOW_CONTENT_EL,
248        ms => FOREIGN_FLOW_CONTENT_EL,
249        mtext => FOREIGN_FLOW_CONTENT_EL,
250      },
251      $SVG_NS => {
252        foreignObject => FOREIGN_FLOW_CONTENT_EL,
253        desc => FOREIGN_FLOW_CONTENT_EL,
254        title => FOREIGN_FLOW_CONTENT_EL,
255      },
256      ## NOTE: In addition, FOREIGN_EL is set to non-HTML elements.
257    };
258    
259    my $svg_attr_name = {
260      attributename => 'attributeName',
261      attributetype => 'attributeType',
262      basefrequency => 'baseFrequency',
263      baseprofile => 'baseProfile',
264      calcmode => 'calcMode',
265      clippathunits => 'clipPathUnits',
266      contentscripttype => 'contentScriptType',
267      contentstyletype => 'contentStyleType',
268      diffuseconstant => 'diffuseConstant',
269      edgemode => 'edgeMode',
270      externalresourcesrequired => 'externalResourcesRequired',
271      filterres => 'filterRes',
272      filterunits => 'filterUnits',
273      glyphref => 'glyphRef',
274      gradienttransform => 'gradientTransform',
275      gradientunits => 'gradientUnits',
276      kernelmatrix => 'kernelMatrix',
277      kernelunitlength => 'kernelUnitLength',
278      keypoints => 'keyPoints',
279      keysplines => 'keySplines',
280      keytimes => 'keyTimes',
281      lengthadjust => 'lengthAdjust',
282      limitingconeangle => 'limitingConeAngle',
283      markerheight => 'markerHeight',
284      markerunits => 'markerUnits',
285      markerwidth => 'markerWidth',
286      maskcontentunits => 'maskContentUnits',
287      maskunits => 'maskUnits',
288      numoctaves => 'numOctaves',
289      pathlength => 'pathLength',
290      patterncontentunits => 'patternContentUnits',
291      patterntransform => 'patternTransform',
292      patternunits => 'patternUnits',
293      pointsatx => 'pointsAtX',
294      pointsaty => 'pointsAtY',
295      pointsatz => 'pointsAtZ',
296      preservealpha => 'preserveAlpha',
297      preserveaspectratio => 'preserveAspectRatio',
298      primitiveunits => 'primitiveUnits',
299      refx => 'refX',
300      refy => 'refY',
301      repeatcount => 'repeatCount',
302      repeatdur => 'repeatDur',
303      requiredextensions => 'requiredExtensions',
304      requiredfeatures => 'requiredFeatures',
305      specularconstant => 'specularConstant',
306      specularexponent => 'specularExponent',
307      spreadmethod => 'spreadMethod',
308      startoffset => 'startOffset',
309      stddeviation => 'stdDeviation',
310      stitchtiles => 'stitchTiles',
311      surfacescale => 'surfaceScale',
312      systemlanguage => 'systemLanguage',
313      tablevalues => 'tableValues',
314      targetx => 'targetX',
315      targety => 'targetY',
316      textlength => 'textLength',
317      viewbox => 'viewBox',
318      viewtarget => 'viewTarget',
319      xchannelselector => 'xChannelSelector',
320      ychannelselector => 'yChannelSelector',
321      zoomandpan => 'zoomAndPan',
322    };
323    
324    my $foreign_attr_xname = {
325      'xlink:actuate' => [$XLINK_NS, ['xlink', 'actuate']],
326      'xlink:arcrole' => [$XLINK_NS, ['xlink', 'arcrole']],
327      'xlink:href' => [$XLINK_NS, ['xlink', 'href']],
328      'xlink:role' => [$XLINK_NS, ['xlink', 'role']],
329      'xlink:show' => [$XLINK_NS, ['xlink', 'show']],
330      'xlink:title' => [$XLINK_NS, ['xlink', 'title']],
331      'xlink:type' => [$XLINK_NS, ['xlink', 'type']],
332      'xml:base' => [$XML_NS, ['xml', 'base']],
333      'xml:lang' => [$XML_NS, ['xml', 'lang']],
334      'xml:space' => [$XML_NS, ['xml', 'space']],
335      'xmlns' => [$XMLNS_NS, [undef, 'xmlns']],
336      'xmlns:xlink' => [$XMLNS_NS, ['xmlns', 'xlink']],
337    };
338    
339    ## ISSUE: xmlns:xlink="non-xlink-ns" is not an error.
340    
341    my $charref_map = {
342      0x0D => 0x000A,
343    0x80 => 0x20AC,    0x80 => 0x20AC,
344    0x81 => 0xFFFD,    0x81 => 0xFFFD,
345    0x82 => 0x201A,    0x82 => 0x201A,
# Line 59  my $c1_entity_char = { Line 372  my $c1_entity_char = {
372    0x9D => 0xFFFD,    0x9D => 0xFFFD,
373    0x9E => 0x017E,    0x9E => 0x017E,
374    0x9F => 0x0178,    0x9F => 0x0178,
375  }; # $c1_entity_char  }; # $charref_map
376    $charref_map->{$_} = 0xFFFD
377        for 0x0000..0x0008, 0x000B, 0x000E..0x001F, 0x007F,
378            0xD800..0xDFFF, 0xFDD0..0xFDDF, ## ISSUE: 0xFDEF
379            0xFFFE, 0xFFFF, 0x1FFFE, 0x1FFFF, 0x2FFFE, 0x2FFFF, 0x3FFFE, 0x3FFFF,
380            0x4FFFE, 0x4FFFF, 0x5FFFE, 0x5FFFF, 0x6FFFE, 0x6FFFF, 0x7FFFE,
381            0x7FFFF, 0x8FFFE, 0x8FFFF, 0x9FFFE, 0x9FFFF, 0xAFFFE, 0xAFFFF,
382            0xBFFFE, 0xBFFFF, 0xCFFFE, 0xCFFFF, 0xDFFFE, 0xDFFFF, 0xEFFFE,
383            0xEFFFF, 0xFFFFE, 0xFFFFF, 0x10FFFE, 0x10FFFF;
384    
385  my $special_category = {  ## TODO: Invoke the reset algorithm when a resettable element is
386    address => 1, area => 1, base => 1, basefont => 1, bgsound => 1,  ## created (cf. HTML5 revision 2259).
   blockquote => 1, body => 1, br => 1, center => 1, col => 1, colgroup => 1,  
   dd => 1, dir => 1, div => 1, dl => 1, dt => 1, embed => 1, fieldset => 1,  
   form => 1, frame => 1, frameset => 1, h1 => 1, h2 => 1, h3 => 1,  
   h4 => 1, h5 => 1, h6 => 1, head => 1, hr => 1, iframe => 1, image => 1,  
   img => 1, input => 1, isindex => 1, li => 1, link => 1, listing => 1,  
   menu => 1, meta => 1, noembed => 1, noframes => 1, noscript => 1,  
   ol => 1, optgroup => 1, option => 1, p => 1, param => 1, plaintext => 1,  
   pre => 1, script => 1, select => 1, spacer => 1, style => 1, tbody => 1,  
   textarea => 1, tfoot => 1, thead => 1, title => 1, tr => 1, ul => 1, wbr => 1,  
 };  
 my $scoping_category = {  
   applet => 1, button => 1, caption => 1, html => 1, marquee => 1, object => 1,  
   table => 1, td => 1, th => 1,  
 };  
 my $formatting_category = {  
   a => 1, b => 1, big => 1, em => 1, font => 1, i => 1, nobr => 1,  
   s => 1, small => 1, strile => 1, strong => 1, tt => 1, u => 1,  
 };  
 # $phrasing_category: all other elements  
387    
388  sub parse_byte_string ($$$$;$) {  sub parse_byte_string ($$$$;$) {
389      my $self = shift;
390      my $charset_name = shift;
391      open my $input, '<', ref $_[0] ? $_[0] : \($_[0]);
392      return $self->parse_byte_stream ($charset_name, $input, @_[1..$#_]);
393    } # parse_byte_string
394    
395    sub parse_byte_stream ($$$$;$$) {
396      # my ($self, $charset_name, $byte_stream, $doc, $onerror, $get_wrapper) = @_;
397    my $self = ref $_[0] ? shift : shift->new;    my $self = ref $_[0] ? shift : shift->new;
398    my $charset = shift;    my $charset_name = shift;
399    my $bytes_s = ref $_[0] ? $_[0] : \($_[0]);    my $byte_stream = $_[0];
   my $s;  
     
   if (defined $charset) {  
     require Encode; ## TODO: decode(utf8) don't delete BOM  
     $s = \ (Encode::decode ($charset, $$bytes_s));  
     $self->{input_encoding} = lc $charset; ## TODO: normalize name  
     $self->{confident} = 1;  
   } else {  
     ## TODO: Implement HTML5 detection algorithm  
     require Whatpm::Charset::UniversalCharDet;  
     $charset = Whatpm::Charset::UniversalCharDet->detect_byte_string  
         (substr ($$bytes_s, 0, 1024));  
     $charset ||= 'windows-1252';  
     $s = \ (Encode::decode ($charset, $$bytes_s));  
     $self->{input_encoding} = $charset;  
     $self->{confident} = 0;  
   }  
400    
401    $self->{change_encoding} = sub {    my $onerror = $_[2] || sub {
402      my $self = shift;      my (%opt) = @_;
403      my $charset = lc shift;      warn "Parse error ($opt{type})\n";
404      ## TODO: if $charset is supported    };
405      ## TODO: normalize charset name    $self->{parse_error} = $onerror; # updated later by parse_char_string
406    
407      ## "Change the encoding" algorithm:    my $get_wrapper = $_[3] || sub ($) {
408        return $_[0]; # $_[0] = byte stream handle, returned = arg to char handle
409      ## Step 1        };
410      if ($charset eq 'utf-16') { ## ISSUE: UTF-16BE -> UTF-8? UTF-16LE -> UTF-8?  
411        $charset = 'utf-8';    ## HTML5 encoding sniffing algorithm
412      require Message::Charset::Info;
413      my $charset;
414      my $buffer;
415      my ($char_stream, $e_status);
416    
417      SNIFFING: {
418        ## NOTE: By setting |allow_fallback| option true when the
419        ## |get_decode_handle| method is invoked, we ignore what the HTML5
420        ## spec requires, i.e. unsupported encoding should be ignored.
421          ## TODO: We should not do this unless the parser is invoked
422          ## in the conformance checking mode, in which this behavior
423          ## would be useful.
424    
425        ## Step 1
426        if (defined $charset_name) {
427          $charset = Message::Charset::Info->get_by_html_name ($charset_name);
428              ## TODO: Is this ok?  Transfer protocol's parameter should be
429              ## interpreted in its semantics?
430    
431          ($char_stream, $e_status) = $charset->get_decode_handle
432              ($byte_stream, allow_error_reporting => 1,
433               allow_fallback => 1);
434          if ($char_stream) {
435            $self->{confident} = 1;
436            last SNIFFING;
437          } else {
438            !!!parse-error (type => 'charset:not supported',
439                            layer => 'encode',
440                            line => 1, column => 1,
441                            value => $charset_name,
442                            level => $self->{level}->{uncertain});
443          }
444      }      }
445    
446      ## Step 2      ## Step 2
447      if (defined $self->{input_encoding} and      my $byte_buffer = '';
448          $self->{input_encoding} eq $charset) {      for (1..1024) {
449          my $char = $byte_stream->getc;
450          last unless defined $char;
451          $byte_buffer .= $char;
452        } ## TODO: timeout
453    
454        ## Step 3
455        if ($byte_buffer =~ /^\xFE\xFF/) {
456          $charset = Message::Charset::Info->get_by_html_name ('utf-16be');
457          ($char_stream, $e_status) = $charset->get_decode_handle
458              ($byte_stream, allow_error_reporting => 1,
459               allow_fallback => 1, byte_buffer => \$byte_buffer);
460        $self->{confident} = 1;        $self->{confident} = 1;
461        return;        last SNIFFING;
462        } elsif ($byte_buffer =~ /^\xFF\xFE/) {
463          $charset = Message::Charset::Info->get_by_html_name ('utf-16le');
464          ($char_stream, $e_status) = $charset->get_decode_handle
465              ($byte_stream, allow_error_reporting => 1,
466               allow_fallback => 1, byte_buffer => \$byte_buffer);
467          $self->{confident} = 1;
468          last SNIFFING;
469        } elsif ($byte_buffer =~ /^\xEF\xBB\xBF/) {
470          $charset = Message::Charset::Info->get_by_html_name ('utf-8');
471          ($char_stream, $e_status) = $charset->get_decode_handle
472              ($byte_stream, allow_error_reporting => 1,
473               allow_fallback => 1, byte_buffer => \$byte_buffer);
474          $self->{confident} = 1;
475          last SNIFFING;
476      }      }
477    
478      !!!parse-error (type => 'charset label detected:'.$self->{input_encoding}.      ## Step 4
479          ':'.$charset, level => 'w');      ## TODO: <meta charset>
480    
481      ## Step 3      ## Step 5
482      # if (can) {      ## TODO: from history
       ## change the encoding on the fly.  
       #$self->{confident} = 1;  
       #return;  
     # }  
483    
484      ## Step 4      ## Step 6
485      throw Whatpm::HTML::RestartParser (charset => $charset);      require Whatpm::Charset::UniversalCharDet;
486        $charset_name = Whatpm::Charset::UniversalCharDet->detect_byte_string
487            ($byte_buffer);
488        if (defined $charset_name) {
489          $charset = Message::Charset::Info->get_by_html_name ($charset_name);
490    
491          ## ISSUE: Unsupported encoding is not ignored according to the spec.
492          require Whatpm::Charset::DecodeHandle;
493          $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new
494              ($byte_stream);
495          ($char_stream, $e_status) = $charset->get_decode_handle
496              ($buffer, allow_error_reporting => 1,
497               allow_fallback => 1, byte_buffer => \$byte_buffer);
498          if ($char_stream) {
499            $buffer->{buffer} = $byte_buffer;
500            !!!parse-error (type => 'sniffing:chardet',
501                            text => $charset_name,
502                            level => $self->{level}->{info},
503                            layer => 'encode',
504                            line => 1, column => 1);
505            $self->{confident} = 0;
506            last SNIFFING;
507          }
508        }
509    
510        ## Step 7: default
511        ## TODO: Make this configurable.
512        $charset = Message::Charset::Info->get_by_html_name ('windows-1252');
513            ## NOTE: We choose |windows-1252| here, since |utf-8| should be
514            ## detectable in the step 6.
515        require Whatpm::Charset::DecodeHandle;
516        $buffer = Whatpm::Charset::DecodeHandle::ByteBuffer->new
517            ($byte_stream);
518        ($char_stream, $e_status)
519            = $charset->get_decode_handle ($buffer,
520                                           allow_error_reporting => 1,
521                                           allow_fallback => 1,
522                                           byte_buffer => \$byte_buffer);
523        $buffer->{buffer} = $byte_buffer;
524        !!!parse-error (type => 'sniffing:default',
525                        text => 'windows-1252',
526                        level => $self->{level}->{info},
527                        line => 1, column => 1,
528                        layer => 'encode');
529        $self->{confident} = 0;
530      } # SNIFFING
531    
532      if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) {
533        $self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name?
534        !!!parse-error (type => 'chardecode:fallback',
535                        #text => $self->{input_encoding},
536                        level => $self->{level}->{uncertain},
537                        line => 1, column => 1,
538                        layer => 'encode');
539      } elsif (not ($e_status &
540                    Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) {
541        $self->{input_encoding} = $charset->get_iana_name;
542        !!!parse-error (type => 'chardecode:no error',
543                        text => $self->{input_encoding},
544                        level => $self->{level}->{uncertain},
545                        line => 1, column => 1,
546                        layer => 'encode');
547      } else {
548        $self->{input_encoding} = $charset->get_iana_name;
549      }
550    
551      $self->{change_encoding} = sub {
552        my $self = shift;
553        $charset_name = shift;
554        my $token = shift;
555    
556        $charset = Message::Charset::Info->get_by_html_name ($charset_name);
557        ($char_stream, $e_status) = $charset->get_decode_handle
558            ($byte_stream, allow_error_reporting => 1, allow_fallback => 1,
559             byte_buffer => \ $buffer->{buffer});
560        
561        if ($char_stream) { # if supported
562          ## "Change the encoding" algorithm:
563    
564          ## Step 1    
565          if ($charset->{category} &
566              Message::Charset::Info::CHARSET_CATEGORY_UTF16 ()) {
567            $charset = Message::Charset::Info->get_by_html_name ('utf-8');
568            ($char_stream, $e_status) = $charset->get_decode_handle
569                ($byte_stream,
570                 byte_buffer => \ $buffer->{buffer});
571          }
572          $charset_name = $charset->get_iana_name;
573          
574          ## Step 2
575          if (defined $self->{input_encoding} and
576              $self->{input_encoding} eq $charset_name) {
577            !!!parse-error (type => 'charset label:matching',
578                            text => $charset_name,
579                            level => $self->{level}->{info});
580            $self->{confident} = 1;
581            return;
582          }
583    
584          !!!parse-error (type => 'charset label detected',
585                          text => $self->{input_encoding},
586                          value => $charset_name,
587                          level => $self->{level}->{warn},
588                          token => $token);
589          
590          ## Step 3
591          # if (can) {
592            ## change the encoding on the fly.
593            #$self->{confident} = 1;
594            #return;
595          # }
596          
597          ## Step 4
598          throw Whatpm::HTML::RestartParser ();
599        }
600    }; # $self->{change_encoding}    }; # $self->{change_encoding}
601    
602    my @args = @_; shift @args; # $s    my $char_onerror = sub {
603        my (undef, $type, %opt) = @_;
604        !!!parse-error (layer => 'encode',
605                        line => $self->{line}, column => $self->{column} + 1,
606                        %opt, type => $type);
607        if ($opt{octets}) {
608          ${$opt{octets}} = "\x{FFFD}"; # relacement character
609        }
610      };
611    
612      my $wrapped_char_stream = $get_wrapper->($char_stream);
613      $wrapped_char_stream->onerror ($char_onerror);
614    
615      my @args = ($_[1], $_[2]); # $doc, $onerror - $get_wrapper = undef;
616    my $return;    my $return;
617    try {    try {
618      $return = $self->parse_char_string ($s, @args);        $return = $self->parse_char_stream ($wrapped_char_stream, @args);  
619    } catch Whatpm::HTML::RestartParser with {    } catch Whatpm::HTML::RestartParser with {
620      my $charset = shift->{charset};      ## NOTE: Invoked after {change_encoding}.
621      $s = \ (Encode::decode ($charset, $$bytes_s));      
622      $self->{input_encoding} = $charset; ## TODO: normalize      if ($e_status & Message::Charset::Info::FALLBACK_ENCODING_IMPL ()) {
623          $self->{input_encoding} = $charset->get_iana_name; ## TODO: Should we set actual charset decoder's encoding name?
624          !!!parse-error (type => 'chardecode:fallback',
625                          level => $self->{level}->{uncertain},
626                          #text => $self->{input_encoding},
627                          line => 1, column => 1,
628                          layer => 'encode');
629        } elsif (not ($e_status &
630                      Message::Charset::Info::ERROR_REPORTING_ENCODING_IMPL ())) {
631          $self->{input_encoding} = $charset->get_iana_name;
632          !!!parse-error (type => 'chardecode:no error',
633                          text => $self->{input_encoding},
634                          level => $self->{level}->{uncertain},
635                          line => 1, column => 1,
636                          layer => 'encode');
637        } else {
638          $self->{input_encoding} = $charset->get_iana_name;
639        }
640      $self->{confident} = 1;      $self->{confident} = 1;
641      $return = $self->parse_char_string ($s, @args);  
642        $wrapped_char_stream = $get_wrapper->($char_stream);
643        $wrapped_char_stream->onerror ($char_onerror);
644    
645        $return = $self->parse_char_stream ($wrapped_char_stream, @args);
646    };    };
647    return $return;    return $return;
648  } # parse_byte_string  } # parse_byte_stream
649    
650  ## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM  ## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM
651  ## and the HTML layer MUST ignore it.  However, we does strip BOM in  ## and the HTML layer MUST ignore it.  However, we does strip BOM in
# Line 162  sub parse_byte_string ($$$$;$) { Line 656  sub parse_byte_string ($$$$;$) {
656  ## such as |parse_byte_string| in this module, must ensure that it does  ## such as |parse_byte_string| in this module, must ensure that it does
657  ## strip the BOM and never strip any ZWNBSP.  ## strip the BOM and never strip any ZWNBSP.
658    
659  *parse_char_string = \&parse_string;  sub parse_char_string ($$$;$$) {
660      #my ($self, $s, $doc, $onerror, $get_wrapper) = @_;
661      my $self = shift;
662      my $s = ref $_[0] ? $_[0] : \($_[0]);
663      require Whatpm::Charset::DecodeHandle;
664      my $input = Whatpm::Charset::DecodeHandle::CharString->new ($s);
665      return $self->parse_char_stream ($input, @_[1..$#_]);
666    } # parse_char_string
667    *parse_string = \&parse_char_string; ## NOTE: Alias for backward compatibility.
668    
669  sub parse_string ($$$;$) {  sub parse_char_stream ($$$;$$) {
670    my $self = ref $_[0] ? shift : shift->new;    my $self = ref $_[0] ? shift : shift->new;
671    my $s = ref $_[0] ? $_[0] : \($_[0]);    my $input = $_[0];
672    $self->{document} = $_[1];    $self->{document} = $_[1];
673    @{$self->{document}->child_nodes} = ();    @{$self->{document}->child_nodes} = ();
674    
# Line 175  sub parse_string ($$$;$) { Line 677  sub parse_string ($$$;$) {
677    $self->{confident} = 1 unless exists $self->{confident};    $self->{confident} = 1 unless exists $self->{confident};
678    $self->{document}->input_encoding ($self->{input_encoding})    $self->{document}->input_encoding ($self->{input_encoding})
679        if defined $self->{input_encoding};        if defined $self->{input_encoding};
680    ## TODO: |{input_encoding}| is needless?
681    
   my $i = 0;  
682    $self->{line_prev} = $self->{line} = 1;    $self->{line_prev} = $self->{line} = 1;
683    $self->{column_prev} = $self->{column} = 0;    $self->{column_prev} = -1;
684    $self->{set_next_char} = sub {    $self->{column} = 0;
685      $self->{set_nc} = sub {
686      my $self = shift;      my $self = shift;
687    
688      pop @{$self->{prev_char}};      my $char = '';
689      unshift @{$self->{prev_char}}, $self->{next_char};      if (defined $self->{next_nc}) {
690          $char = $self->{next_nc};
691          delete $self->{next_nc};
692          $self->{nc} = ord $char;
693        } else {
694          $self->{char_buffer} = '';
695          $self->{char_buffer_pos} = 0;
696    
697      $self->{next_char} = -1 and return if $i >= length $$s;        my $count = $input->manakai_read_until
698      $self->{next_char} = ord substr $$s, $i++, 1;           ($self->{char_buffer}, qr/[^\x00\x0A\x0D]/, $self->{char_buffer_pos});
699          if ($count) {
700            $self->{line_prev} = $self->{line};
701            $self->{column_prev} = $self->{column};
702            $self->{column}++;
703            $self->{nc}
704                = ord substr ($self->{char_buffer}, $self->{char_buffer_pos}++, 1);
705            return;
706          }
707    
708          if ($input->read ($char, 1)) {
709            $self->{nc} = ord $char;
710          } else {
711            $self->{nc} = -1;
712            return;
713          }
714        }
715    
716      ($self->{line_prev}, $self->{column_prev})      ($self->{line_prev}, $self->{column_prev})
717          = ($self->{line}, $self->{column});          = ($self->{line}, $self->{column});
718      $self->{column}++;      $self->{column}++;
719            
720      if ($self->{next_char} == 0x000A) { # LF      if ($self->{nc} == 0x000A) { # LF
721          !!!cp ('j1');
722        $self->{line}++;        $self->{line}++;
723        $self->{column} = 0;        $self->{column} = 0;
724      } elsif ($self->{next_char} == 0x000D) { # CR      } elsif ($self->{nc} == 0x000D) { # CR
725        $i++ if substr ($$s, $i, 1) eq "\x0A";        !!!cp ('j2');
726        $self->{next_char} = 0x000A; # LF # MUST  ## TODO: support for abort/streaming
727          my $next = '';
728          if ($input->read ($next, 1) and $next ne "\x0A") {
729            $self->{next_nc} = $next;
730          }
731          $self->{nc} = 0x000A; # LF # MUST
732        $self->{line}++;        $self->{line}++;
733        $self->{column} = 0;        $self->{column} = 0;
734      } elsif ($self->{next_char} > 0x10FFFF) {      } elsif ($self->{nc} == 0x0000) { # NULL
735        $self->{next_char} = 0xFFFD; # REPLACEMENT CHARACTER # MUST        !!!cp ('j4');
     } elsif ($self->{next_char} == 0x0000) { # NULL  
736        !!!parse-error (type => 'NULL');        !!!parse-error (type => 'NULL');
737        $self->{next_char} = 0xFFFD; # REPLACEMENT CHARACTER # MUST        $self->{nc} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
738      }      }
739    };    };
740    $self->{prev_char} = [-1, -1, -1];  
741    $self->{next_char} = -1;    $self->{read_until} = sub {
742        #my ($scalar, $specials_range, $offset) = @_;
743        return 0 if defined $self->{next_nc};
744    
745        my $pattern = qr/[^$_[1]\x00\x0A\x0D]/;
746        my $offset = $_[2] || 0;
747    
748        if ($self->{char_buffer_pos} < length $self->{char_buffer}) {
749          pos ($self->{char_buffer}) = $self->{char_buffer_pos};
750          if ($self->{char_buffer} =~ /\G(?>$pattern)+/) {
751            substr ($_[0], $offset)
752                = substr ($self->{char_buffer}, $-[0], $+[0] - $-[0]);
753            my $count = $+[0] - $-[0];
754            if ($count) {
755              $self->{column} += $count;
756              $self->{char_buffer_pos} += $count;
757              $self->{line_prev} = $self->{line};
758              $self->{column_prev} = $self->{column} - 1;
759              $self->{nc} = -1;
760            }
761            return $count;
762          } else {
763            return 0;
764          }
765        } else {
766          my $count = $input->manakai_read_until ($_[0], $pattern, $_[2]);
767          if ($count) {
768            $self->{column} += $count;
769            $self->{line_prev} = $self->{line};
770            $self->{column_prev} = $self->{column} - 1;
771            $self->{nc} = -1;
772          }
773          return $count;
774        }
775      }; # $self->{read_until}
776    
777    my $onerror = $_[2] || sub {    my $onerror = $_[2] || sub {
778      my (%opt) = @_;      my (%opt) = @_;
# Line 220  sub parse_string ($$$;$) { Line 784  sub parse_string ($$$;$) {
784      $onerror->(line => $self->{line}, column => $self->{column}, @_);      $onerror->(line => $self->{line}, column => $self->{column}, @_);
785    };    };
786    
787      my $char_onerror = sub {
788        my (undef, $type, %opt) = @_;
789        !!!parse-error (layer => 'encode',
790                        line => $self->{line}, column => $self->{column} + 1,
791                        %opt, type => $type);
792      }; # $char_onerror
793    
794      if ($_[3]) {
795        $input = $_[3]->($input);
796        $input->onerror ($char_onerror);
797      } else {
798        $input->onerror ($char_onerror) unless defined $input->onerror;
799      }
800    
801    $self->_initialize_tokenizer;    $self->_initialize_tokenizer;
802    $self->_initialize_tree_constructor;    $self->_initialize_tree_constructor;
803    $self->_construct_tree;    $self->_construct_tree;
# Line 228  sub parse_string ($$$;$) { Line 806  sub parse_string ($$$;$) {
806    delete $self->{parse_error}; # remove loop    delete $self->{parse_error}; # remove loop
807    
808    return $self->{document};    return $self->{document};
809  } # parse_string  } # parse_char_stream
810    
811  sub new ($) {  sub new ($) {
812    my $class = shift;    my $class = shift;
813    my $self = bless {}, $class;    my $self = bless {
814    $self->{set_next_char} = sub {      level => {must => 'm',
815      $self->{next_char} = -1;                should => 's',
816                  warn => 'w',
817                  info => 'i',
818                  uncertain => 'u'},
819      }, $class;
820      $self->{set_nc} = sub {
821        $self->{nc} = -1;
822    };    };
823    $self->{parse_error} = sub {    $self->{parse_error} = sub {
824      #      #
# Line 261  sub RCDATA_CONTENT_MODEL () { CM_ENTITY Line 845  sub RCDATA_CONTENT_MODEL () { CM_ENTITY
845  sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP }  sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP }
846    
847  sub DATA_STATE () { 0 }  sub DATA_STATE () { 0 }
848  sub ENTITY_DATA_STATE () { 1 }  #sub ENTITY_DATA_STATE () { 1 }
849  sub TAG_OPEN_STATE () { 2 }  sub TAG_OPEN_STATE () { 2 }
850  sub CLOSE_TAG_OPEN_STATE () { 3 }  sub CLOSE_TAG_OPEN_STATE () { 3 }
851  sub TAG_NAME_STATE () { 4 }  sub TAG_NAME_STATE () { 4 }
# Line 272  sub BEFORE_ATTRIBUTE_VALUE_STATE () { 8 Line 856  sub BEFORE_ATTRIBUTE_VALUE_STATE () { 8
856  sub ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE () { 9 }  sub ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE () { 9 }
857  sub ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE () { 10 }  sub ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE () { 10 }
858  sub ATTRIBUTE_VALUE_UNQUOTED_STATE () { 11 }  sub ATTRIBUTE_VALUE_UNQUOTED_STATE () { 11 }
859  sub ENTITY_IN_ATTRIBUTE_VALUE_STATE () { 12 }  #sub ENTITY_IN_ATTRIBUTE_VALUE_STATE () { 12 }
860  sub MARKUP_DECLARATION_OPEN_STATE () { 13 }  sub MARKUP_DECLARATION_OPEN_STATE () { 13 }
861  sub COMMENT_START_STATE () { 14 }  sub COMMENT_START_STATE () { 14 }
862  sub COMMENT_START_DASH_STATE () { 15 }  sub COMMENT_START_DASH_STATE () { 15 }
# Line 294  sub DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUO Line 878  sub DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUO
878  sub AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 31 }  sub AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 31 }
879  sub BOGUS_DOCTYPE_STATE () { 32 }  sub BOGUS_DOCTYPE_STATE () { 32 }
880  sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 }  sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 }
881    sub SELF_CLOSING_START_TAG_STATE () { 34 }
882    sub CDATA_SECTION_STATE () { 35 }
883    sub MD_HYPHEN_STATE () { 36 } # "markup declaration open state" in the spec
884    sub MD_DOCTYPE_STATE () { 37 } # "markup declaration open state" in the spec
885    sub MD_CDATA_STATE () { 38 } # "markup declaration open state" in the spec
886    sub CDATA_RCDATA_CLOSE_TAG_STATE () { 39 } # "close tag open state" in the spec
887    sub CDATA_SECTION_MSE1_STATE () { 40 } # "CDATA section state" in the spec
888    sub CDATA_SECTION_MSE2_STATE () { 41 } # "CDATA section state" in the spec
889    sub PUBLIC_STATE () { 42 } # "after DOCTYPE name state" in the spec
890    sub SYSTEM_STATE () { 43 } # "after DOCTYPE name state" in the spec
891    ## NOTE: "Entity data state", "entity in attribute value state", and
892    ## "consume a character reference" algorithm are jointly implemented
893    ## using the following six states:
894    sub ENTITY_STATE () { 44 }
895    sub ENTITY_HASH_STATE () { 45 }
896    sub NCR_NUM_STATE () { 46 }
897    sub HEXREF_X_STATE () { 47 }
898    sub HEXREF_HEX_STATE () { 48 }
899    sub ENTITY_NAME_STATE () { 49 }
900    sub PCDATA_STATE () { 50 } # "data state" in the spec
901    
902  sub DOCTYPE_TOKEN () { 1 }  sub DOCTYPE_TOKEN () { 1 }
903  sub COMMENT_TOKEN () { 2 }  sub COMMENT_TOKEN () { 2 }
# Line 311  sub ROW_IMS ()        { 0b10000000 } Line 915  sub ROW_IMS ()        { 0b10000000 }
915  sub BODY_AFTER_IMS () { 0b100000000 }  sub BODY_AFTER_IMS () { 0b100000000 }
916  sub FRAME_IMS ()      { 0b1000000000 }  sub FRAME_IMS ()      { 0b1000000000 }
917  sub SELECT_IMS ()     { 0b10000000000 }  sub SELECT_IMS ()     { 0b10000000000 }
918    sub IN_FOREIGN_CONTENT_IM () { 0b100000000000 }
919        ## NOTE: "in foreign content" insertion mode is special; it is combined
920        ## with the secondary insertion mode.  In this parser, they are stored
921        ## together in the bit-or'ed form.
922    
923  ## NOTE: "initial" and "before html" insertion modes have no constants.  ## NOTE: "initial" and "before html" insertion modes have no constants.
924    
# Line 342  sub IN_COLUMN_GROUP_IM () { 0b10 } Line 950  sub IN_COLUMN_GROUP_IM () { 0b10 }
950  sub _initialize_tokenizer ($) {  sub _initialize_tokenizer ($) {
951    my $self = shift;    my $self = shift;
952    $self->{state} = DATA_STATE; # MUST    $self->{state} = DATA_STATE; # MUST
953      #$self->{s_kwd}; # state keyword - initialized when used
954      #$self->{entity__value}; # initialized when used
955      #$self->{entity__match}; # initialized when used
956    $self->{content_model} = PCDATA_CONTENT_MODEL; # be    $self->{content_model} = PCDATA_CONTENT_MODEL; # be
957    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE    undef $self->{ct}; # current token
958    undef $self->{current_attribute};    undef $self->{ca}; # current attribute
959    undef $self->{last_emitted_start_tag_name};    undef $self->{last_stag_name}; # last emitted start tag name
960    undef $self->{last_attribute_value_state};    #$self->{prev_state}; # initialized when used
961    $self->{char} = [];    delete $self->{self_closing};
962    # $self->{next_char}    $self->{char_buffer} = '';
963      $self->{char_buffer_pos} = 0;
964      $self->{nc} = -1; # next input character
965      #$self->{next_nc}
966    !!!next-input-character;    !!!next-input-character;
967    $self->{token} = [];    $self->{token} = [];
968    # $self->{escape}    # $self->{escape}
# Line 359  sub _initialize_tokenizer ($) { Line 973  sub _initialize_tokenizer ($) {
973  ##       CHARACTER_TOKEN, or END_OF_FILE_TOKEN  ##       CHARACTER_TOKEN, or END_OF_FILE_TOKEN
974  ##   ->{name} (DOCTYPE_TOKEN)  ##   ->{name} (DOCTYPE_TOKEN)
975  ##   ->{tag_name} (START_TAG_TOKEN, END_TAG_TOKEN)  ##   ->{tag_name} (START_TAG_TOKEN, END_TAG_TOKEN)
976  ##   ->{public_identifier} (DOCTYPE_TOKEN)  ##   ->{pubid} (DOCTYPE_TOKEN)
977  ##   ->{system_identifier} (DOCTYPE_TOKEN)  ##   ->{sysid} (DOCTYPE_TOKEN)
978  ##   ->{quirks} == 1 or 0 (DOCTYPE_TOKEN): "force-quirks" flag  ##   ->{quirks} == 1 or 0 (DOCTYPE_TOKEN): "force-quirks" flag
979  ##   ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN)  ##   ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN)
980  ##        ->{name}  ##        ->{name}
981  ##        ->{value}  ##        ->{value}
982  ##        ->{has_reference} == 1 or 0  ##        ->{has_reference} == 1 or 0
983  ##   ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN)  ##   ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN)
984    ## NOTE: The "self-closing flag" is hold as |$self->{self_closing}|.
985    ##     |->{self_closing}| is used to save the value of |$self->{self_closing}|
986    ##     while the token is pushed back to the stack.
987    
988  ## Emitted token MUST immediately be handled by the tree construction state.  ## Emitted token MUST immediately be handled by the tree construction state.
989    
# Line 376  sub _initialize_tokenizer ($) { Line 993  sub _initialize_tokenizer ($) {
993  ## has completed loading.  If one has, then it MUST be executed  ## has completed loading.  If one has, then it MUST be executed
994  ## and removed from the list.  ## and removed from the list.
995    
996  ## NOTE: HTML5 "Writing HTML documents" section, applied to  ## TODO: Polytheistic slash SHOULD NOT be used. (Applied only to atheists.)
997  ## documents and not to user agents and conformance checkers,  ## (This requirement was dropped from HTML5 spec, unfortunately.)
998  ## contains some requirements that are not detected by the  
999  ## parsing algorithm:  my $is_space = {
1000  ## - Some requirements on character encoding declarations. ## TODO    0x0009 => 1, # CHARACTER TABULATION (HT)
1001  ## - "Elements MUST NOT contain content that their content model disallows."    0x000A => 1, # LINE FEED (LF)
1002  ##   ... Some are parse error, some are not (will be reported by c.c.).    #0x000B => 0, # LINE TABULATION (VT)
1003  ## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO    0x000C => 1, # FORM FEED (FF)
1004  ## - Text (in elements, attributes, and comments) SHOULD NOT contain    #0x000D => 1, # CARRIAGE RETURN (CR)
1005  ##   control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL?  Unicode control character?)    0x0020 => 1, # SPACE (SP)
1006    };
 ## TODO: HTML5 poses authors two SHOULD-level requirements that cannot  
 ## be detected by the HTML5 parsing algorithm:  
 ## - Text,  
1007    
1008  sub _get_next_token ($) {  sub _get_next_token ($) {
1009    my $self = shift;    my $self = shift;
1010    
1011      if ($self->{self_closing}) {
1012        !!!parse-error (type => 'nestc', token => $self->{ct});
1013        ## NOTE: The |self_closing| flag is only set by start tag token.
1014        ## In addition, when a start tag token is emitted, it is always set to
1015        ## |ct|.
1016        delete $self->{self_closing};
1017      }
1018    
1019    if (@{$self->{token}}) {    if (@{$self->{token}}) {
1020        $self->{self_closing} = $self->{token}->[0]->{self_closing};
1021      return shift @{$self->{token}};      return shift @{$self->{token}};
1022    }    }
1023    
1024    A: {    A: {
1025      if ($self->{state} == DATA_STATE) {      if ($self->{state} == PCDATA_STATE) {
1026        if ($self->{next_char} == 0x0026) { # &        ## NOTE: Same as |DATA_STATE|, but only for |PCDATA| content model.
1027    
1028          if ($self->{nc} == 0x0026) { # &
1029            !!!cp (0.1);
1030            ## NOTE: In the spec, the tokenizer is switched to the
1031            ## "entity data state".  In this implementation, the tokenizer
1032            ## is switched to the |ENTITY_STATE|, which is an implementation
1033            ## of the "consume a character reference" algorithm.
1034            $self->{entity_add} = -1;
1035            $self->{prev_state} = DATA_STATE;
1036            $self->{state} = ENTITY_STATE;
1037            !!!next-input-character;
1038            redo A;
1039          } elsif ($self->{nc} == 0x003C) { # <
1040            !!!cp (0.2);
1041            $self->{state} = TAG_OPEN_STATE;
1042            !!!next-input-character;
1043            redo A;
1044          } elsif ($self->{nc} == -1) {
1045            !!!cp (0.3);
1046            !!!emit ({type => END_OF_FILE_TOKEN,
1047                      line => $self->{line}, column => $self->{column}});
1048            last A; ## TODO: ok?
1049          } else {
1050            !!!cp (0.4);
1051            #
1052          }
1053    
1054          # Anything else
1055          my $token = {type => CHARACTER_TOKEN,
1056                       data => chr $self->{nc},
1057                       line => $self->{line}, column => $self->{column},
1058                      };
1059          $self->{read_until}->($token->{data}, q[<&], length $token->{data});
1060    
1061          ## Stay in the state.
1062          !!!next-input-character;
1063          !!!emit ($token);
1064          redo A;
1065        } elsif ($self->{state} == DATA_STATE) {
1066          $self->{s_kwd} = '' unless defined $self->{s_kwd};
1067          if ($self->{nc} == 0x0026) { # &
1068            $self->{s_kwd} = '';
1069          if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA          if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA
1070              not $self->{escape}) {              not $self->{escape}) {
1071            !!!cp (1);            !!!cp (1);
1072            $self->{state} = ENTITY_DATA_STATE;            ## NOTE: In the spec, the tokenizer is switched to the
1073              ## "entity data state".  In this implementation, the tokenizer
1074              ## is switched to the |ENTITY_STATE|, which is an implementation
1075              ## of the "consume a character reference" algorithm.
1076              $self->{entity_add} = -1;
1077              $self->{prev_state} = DATA_STATE;
1078              $self->{state} = ENTITY_STATE;
1079            !!!next-input-character;            !!!next-input-character;
1080            redo A;            redo A;
1081          } else {          } else {
1082            !!!cp (2);            !!!cp (2);
1083            #            #
1084          }          }
1085        } elsif ($self->{next_char} == 0x002D) { # -        } elsif ($self->{nc} == 0x002D) { # -
1086          if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA          if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
1087            unless ($self->{escape}) {            $self->{s_kwd} .= '-';
1088              if ($self->{prev_char}->[0] == 0x002D and # -            
1089                  $self->{prev_char}->[1] == 0x0021 and # !            if ($self->{s_kwd} eq '<!--') {
1090                  $self->{prev_char}->[2] == 0x003C) { # <              !!!cp (3);
1091                !!!cp (3);              $self->{escape} = 1; # unless $self->{escape};
1092                $self->{escape} = 1;              $self->{s_kwd} = '--';
1093              } else {              #
1094                !!!cp (4);            } elsif ($self->{s_kwd} eq '---') {
1095              }              !!!cp (4);
1096                $self->{s_kwd} = '--';
1097                #
1098            } else {            } else {
1099              !!!cp (5);              !!!cp (5);
1100                #
1101            }            }
1102          }          }
1103                    
1104          #          #
1105        } elsif ($self->{next_char} == 0x003C) { # <        } elsif ($self->{nc} == 0x0021) { # !
1106            if (length $self->{s_kwd}) {
1107              !!!cp (5.1);
1108              $self->{s_kwd} .= '!';
1109              #
1110            } else {
1111              !!!cp (5.2);
1112              #$self->{s_kwd} = '';
1113              #
1114            }
1115            #
1116          } elsif ($self->{nc} == 0x003C) { # <
1117          if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA          if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA
1118              (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA              (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA
1119               not $self->{escape})) {               not $self->{escape})) {
# Line 437  sub _get_next_token ($) { Line 1123  sub _get_next_token ($) {
1123            redo A;            redo A;
1124          } else {          } else {
1125            !!!cp (7);            !!!cp (7);
1126              $self->{s_kwd} = '';
1127            #            #
1128          }          }
1129        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1130          if ($self->{escape} and          if ($self->{escape} and
1131              ($self->{content_model} & CM_LIMITED_MARKUP)) { # RCDATA | CDATA              ($self->{content_model} & CM_LIMITED_MARKUP)) { # RCDATA | CDATA
1132            if ($self->{prev_char}->[0] == 0x002D and # -            if ($self->{s_kwd} eq '--') {
               $self->{prev_char}->[1] == 0x002D) { # -  
1133              !!!cp (8);              !!!cp (8);
1134              delete $self->{escape};              delete $self->{escape};
1135            } else {            } else {
# Line 453  sub _get_next_token ($) { Line 1139  sub _get_next_token ($) {
1139            !!!cp (10);            !!!cp (10);
1140          }          }
1141                    
1142            $self->{s_kwd} = '';
1143          #          #
1144        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1145          !!!cp (11);          !!!cp (11);
1146            $self->{s_kwd} = '';
1147          !!!emit ({type => END_OF_FILE_TOKEN,          !!!emit ({type => END_OF_FILE_TOKEN,
1148                    line => $self->{line}, column => $self->{column}});                    line => $self->{line}, column => $self->{column}});
1149          last A; ## TODO: ok?          last A; ## TODO: ok?
1150        } else {        } else {
1151          !!!cp (12);          !!!cp (12);
1152            $self->{s_kwd} = '';
1153            #
1154        }        }
1155    
1156        # Anything else        # Anything else
1157        my $token = {type => CHARACTER_TOKEN,        my $token = {type => CHARACTER_TOKEN,
1158                     data => chr $self->{next_char},                     data => chr $self->{nc},
1159                     line => $self->{line}, column => $self->{column}};                     line => $self->{line}, column => $self->{column},
1160        ## Stay in the data state                    };
1161        !!!next-input-character;        if ($self->{read_until}->($token->{data}, q[-!<>&],
1162                                    length $token->{data})) {
1163        !!!emit ($token);          $self->{s_kwd} = '';
1164          }
       redo A;  
     } elsif ($self->{state} == ENTITY_DATA_STATE) {  
       ## (cannot happen in CDATA state)  
   
       my ($l, $c) = ($self->{line_prev}, $self->{column_prev});  
         
       my $token = $self->_tokenize_attempt_to_consume_an_entity (0, -1);  
   
       $self->{state} = DATA_STATE;  
       # next-input-character is already done  
1165    
1166        unless (defined $token) {        ## Stay in the data state.
1167          if ($self->{content_model} == PCDATA_CONTENT_MODEL) {
1168          !!!cp (13);          !!!cp (13);
1169          !!!emit ({type => CHARACTER_TOKEN, data => '&',          $self->{state} = PCDATA_STATE;
                   line => $l, column => $c});  
1170        } else {        } else {
1171          !!!cp (14);          !!!cp (14);
1172          !!!emit ($token);          ## Stay in the state.
1173        }        }
1174          !!!next-input-character;
1175          !!!emit ($token);
1176        redo A;        redo A;
1177      } elsif ($self->{state} == TAG_OPEN_STATE) {      } elsif ($self->{state} == TAG_OPEN_STATE) {
1178        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
1179          if ($self->{next_char} == 0x002F) { # /          if ($self->{nc} == 0x002F) { # /
1180            !!!cp (15);            !!!cp (15);
1181            !!!next-input-character;            !!!next-input-character;
1182            $self->{state} = CLOSE_TAG_OPEN_STATE;            $self->{state} = CLOSE_TAG_OPEN_STATE;
1183            redo A;            redo A;
1184            } elsif ($self->{nc} == 0x0021) { # !
1185              !!!cp (15.1);
1186              $self->{s_kwd} = '<' unless $self->{escape};
1187              #
1188          } else {          } else {
1189            !!!cp (16);            !!!cp (16);
1190            ## reconsume            #
           $self->{state} = DATA_STATE;  
   
           !!!emit ({type => CHARACTER_TOKEN, data => '<',  
                     line => $self->{line_prev},  
                     column => $self->{column_prev}});  
   
           redo A;  
1191          }          }
1192    
1193            ## reconsume
1194            $self->{state} = DATA_STATE;
1195            !!!emit ({type => CHARACTER_TOKEN, data => '<',
1196                      line => $self->{line_prev},
1197                      column => $self->{column_prev},
1198                     });
1199            redo A;
1200        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA
1201          if ($self->{next_char} == 0x0021) { # !          if ($self->{nc} == 0x0021) { # !
1202            !!!cp (17);            !!!cp (17);
1203            $self->{state} = MARKUP_DECLARATION_OPEN_STATE;            $self->{state} = MARKUP_DECLARATION_OPEN_STATE;
1204            !!!next-input-character;            !!!next-input-character;
1205            redo A;            redo A;
1206          } elsif ($self->{next_char} == 0x002F) { # /          } elsif ($self->{nc} == 0x002F) { # /
1207            !!!cp (18);            !!!cp (18);
1208            $self->{state} = CLOSE_TAG_OPEN_STATE;            $self->{state} = CLOSE_TAG_OPEN_STATE;
1209            !!!next-input-character;            !!!next-input-character;
1210            redo A;            redo A;
1211          } elsif (0x0041 <= $self->{next_char} and          } elsif (0x0041 <= $self->{nc} and
1212                   $self->{next_char} <= 0x005A) { # A..Z                   $self->{nc} <= 0x005A) { # A..Z
1213            !!!cp (19);            !!!cp (19);
1214            $self->{current_token}            $self->{ct}
1215              = {type => START_TAG_TOKEN,              = {type => START_TAG_TOKEN,
1216                 tag_name => chr ($self->{next_char} + 0x0020),                 tag_name => chr ($self->{nc} + 0x0020),
1217                 line => $self->{line_prev},                 line => $self->{line_prev},
1218                 column => $self->{column_prev}};                 column => $self->{column_prev}};
1219            $self->{state} = TAG_NAME_STATE;            $self->{state} = TAG_NAME_STATE;
1220            !!!next-input-character;            !!!next-input-character;
1221            redo A;            redo A;
1222          } elsif (0x0061 <= $self->{next_char} and          } elsif (0x0061 <= $self->{nc} and
1223                   $self->{next_char} <= 0x007A) { # a..z                   $self->{nc} <= 0x007A) { # a..z
1224            !!!cp (20);            !!!cp (20);
1225            $self->{current_token} = {type => START_TAG_TOKEN,            $self->{ct} = {type => START_TAG_TOKEN,
1226                                      tag_name => chr ($self->{next_char}),                                      tag_name => chr ($self->{nc}),
1227                                      line => $self->{line_prev},                                      line => $self->{line_prev},
1228                                      column => $self->{column_prev}};                                      column => $self->{column_prev}};
1229            $self->{state} = TAG_NAME_STATE;            $self->{state} = TAG_NAME_STATE;
1230            !!!next-input-character;            !!!next-input-character;
1231            redo A;            redo A;
1232          } elsif ($self->{next_char} == 0x003E) { # >          } elsif ($self->{nc} == 0x003E) { # >
1233            !!!cp (21);            !!!cp (21);
1234            !!!parse-error (type => 'empty start tag');            !!!parse-error (type => 'empty start tag',
1235                              line => $self->{line_prev},
1236                              column => $self->{column_prev});
1237            $self->{state} = DATA_STATE;            $self->{state} = DATA_STATE;
1238            !!!next-input-character;            !!!next-input-character;
1239    
1240            !!!emit ({type => CHARACTER_TOKEN, data => '<>',            !!!emit ({type => CHARACTER_TOKEN, data => '<>',
1241                      line => $self->{line_prev},                      line => $self->{line_prev},
1242                      column => $self->{column_prev}});                      column => $self->{column_prev},
1243                       });
1244    
1245            redo A;            redo A;
1246          } elsif ($self->{next_char} == 0x003F) { # ?          } elsif ($self->{nc} == 0x003F) { # ?
1247            !!!cp (22);            !!!cp (22);
1248            !!!parse-error (type => 'pio');            !!!parse-error (type => 'pio',
1249                              line => $self->{line_prev},
1250                              column => $self->{column_prev});
1251            $self->{state} = BOGUS_COMMENT_STATE;            $self->{state} = BOGUS_COMMENT_STATE;
1252            $self->{current_token} = {type => COMMENT_TOKEN, data => '',            $self->{ct} = {type => COMMENT_TOKEN, data => '',
1253                                      line => $self->{line_prev},                                      line => $self->{line_prev},
1254                                      column => $self->{column_prev}};                                      column => $self->{column_prev},
1255            ## $self->{next_char} is intentionally left as is                                     };
1256              ## $self->{nc} is intentionally left as is
1257            redo A;            redo A;
1258          } else {          } else {
1259            !!!cp (23);            !!!cp (23);
1260            !!!parse-error (type => 'bare stago');            !!!parse-error (type => 'bare stago',
1261                              line => $self->{line_prev},
1262                              column => $self->{column_prev});
1263            $self->{state} = DATA_STATE;            $self->{state} = DATA_STATE;
1264            ## reconsume            ## reconsume
1265    
1266            !!!emit ({type => CHARACTER_TOKEN, data => '<',            !!!emit ({type => CHARACTER_TOKEN, data => '<',
1267                      line => $self->{line_prev},                      line => $self->{line_prev},
1268                      column => $self->{column_prev}});                      column => $self->{column_prev},
1269                       });
1270    
1271            redo A;            redo A;
1272          }          }
# Line 578  sub _get_next_token ($) { Line 1274  sub _get_next_token ($) {
1274          die "$0: $self->{content_model} in tag open";          die "$0: $self->{content_model} in tag open";
1275        }        }
1276      } elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) {      } elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) {
1277        my ($l, $c) = ($self->{line_prev}, $self->{column_prev});        ## NOTE: The "close tag open state" in the spec is implemented as
1278        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA        ## |CLOSE_TAG_OPEN_STATE| and |CDATA_RCDATA_CLOSE_TAG_STATE|.
         if (defined $self->{last_emitted_start_tag_name}) {  
   
           ## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>  
           my @next_char;  
           TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) {  
             push @next_char, $self->{next_char};  
             my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1);  
             my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c;  
             if ($self->{next_char} == $c or $self->{next_char} == $C) {  
               !!!cp (24);  
               !!!next-input-character;  
               next TAGNAME;  
             } else {  
               !!!cp (25);  
               $self->{next_char} = shift @next_char; # reconsume  
               !!!back-next-input-character (@next_char);  
               $self->{state} = DATA_STATE;  
1279    
1280                !!!emit ({type => CHARACTER_TOKEN, data => '</',        my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1); # "<"of"</"
1281                          line => $l, column => $c});        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
1282              if (defined $self->{last_stag_name}) {
1283                redo A;            $self->{state} = CDATA_RCDATA_CLOSE_TAG_STATE;
1284              }            $self->{s_kwd} = '';
1285            }            ## Reconsume.
1286            push @next_char, $self->{next_char};            redo A;
         
           unless ($self->{next_char} == 0x0009 or # HT  
                   $self->{next_char} == 0x000A or # LF  
                   $self->{next_char} == 0x000B or # VT  
                   $self->{next_char} == 0x000C or # FF  
                   $self->{next_char} == 0x0020 or # SP  
                   $self->{next_char} == 0x003E or # >  
                   $self->{next_char} == 0x002F or # /  
                   $self->{next_char} == -1) {  
             !!!cp (26);  
             $self->{next_char} = shift @next_char; # reconsume  
             !!!back-next-input-character (@next_char);  
             $self->{state} = DATA_STATE;  
             !!!emit ({type => CHARACTER_TOKEN, data => '</',  
                       line => $l, column => $c});  
             redo A;  
           } else {  
             !!!cp (27);  
             $self->{next_char} = shift @next_char;  
             !!!back-next-input-character (@next_char);  
             # and consume...  
           }  
1287          } else {          } else {
1288            ## No start tag token has ever been emitted            ## No start tag token has ever been emitted
1289              ## NOTE: See <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>.
1290            !!!cp (28);            !!!cp (28);
           # next-input-character is already done  
1291            $self->{state} = DATA_STATE;            $self->{state} = DATA_STATE;
1292              ## Reconsume.
1293            !!!emit ({type => CHARACTER_TOKEN, data => '</',            !!!emit ({type => CHARACTER_TOKEN, data => '</',
1294                      line => $l, column => $c});                      line => $l, column => $c,
1295                       });
1296            redo A;            redo A;
1297          }          }
1298        }        }
1299          
1300        if (0x0041 <= $self->{next_char} and        if (0x0041 <= $self->{nc} and
1301            $self->{next_char} <= 0x005A) { # A..Z            $self->{nc} <= 0x005A) { # A..Z
1302          !!!cp (29);          !!!cp (29);
1303          $self->{current_token}          $self->{ct}
1304              = {type => END_TAG_TOKEN,              = {type => END_TAG_TOKEN,
1305                 tag_name => chr ($self->{next_char} + 0x0020),                 tag_name => chr ($self->{nc} + 0x0020),
1306                 line => $l, column => $c};                 line => $l, column => $c};
1307          $self->{state} = TAG_NAME_STATE;          $self->{state} = TAG_NAME_STATE;
1308          !!!next-input-character;          !!!next-input-character;
1309          redo A;          redo A;
1310        } elsif (0x0061 <= $self->{next_char} and        } elsif (0x0061 <= $self->{nc} and
1311                 $self->{next_char} <= 0x007A) { # a..z                 $self->{nc} <= 0x007A) { # a..z
1312          !!!cp (30);          !!!cp (30);
1313          $self->{current_token} = {type => END_TAG_TOKEN,          $self->{ct} = {type => END_TAG_TOKEN,
1314                                    tag_name => chr ($self->{next_char}),                                    tag_name => chr ($self->{nc}),
1315                                    line => $l, column => $c};                                    line => $l, column => $c};
1316          $self->{state} = TAG_NAME_STATE;          $self->{state} = TAG_NAME_STATE;
1317          !!!next-input-character;          !!!next-input-character;
1318          redo A;          redo A;
1319        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1320          !!!cp (31);          !!!cp (31);
1321          !!!parse-error (type => 'empty end tag');          !!!parse-error (type => 'empty end tag',
1322                            line => $self->{line_prev}, ## "<" in "</>"
1323                            column => $self->{column_prev} - 1);
1324          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1325          !!!next-input-character;          !!!next-input-character;
1326          redo A;          redo A;
1327        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1328          !!!cp (32);          !!!cp (32);
1329          !!!parse-error (type => 'bare etago');          !!!parse-error (type => 'bare etago');
1330          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1331          # reconsume          # reconsume
1332    
1333          !!!emit ({type => CHARACTER_TOKEN, data => '</',          !!!emit ({type => CHARACTER_TOKEN, data => '</',
1334                    line => $l, column => $c});                    line => $l, column => $c,
1335                     });
1336    
1337          redo A;          redo A;
1338        } else {        } else {
1339          !!!cp (33);          !!!cp (33);
1340          !!!parse-error (type => 'bogus end tag');          !!!parse-error (type => 'bogus end tag');
1341          $self->{state} = BOGUS_COMMENT_STATE;          $self->{state} = BOGUS_COMMENT_STATE;
1342          $self->{current_token} = {type => COMMENT_TOKEN, data => '',          $self->{ct} = {type => COMMENT_TOKEN, data => '',
1343                                    line => $self->{line_prev}, # "<" of "</"                                    line => $self->{line_prev}, # "<" of "</"
1344                                    column => $self->{column_prev} - 1};                                    column => $self->{column_prev} - 1,
1345          ## $self->{next_char} is intentionally left as is                                   };
1346          redo A;          ## NOTE: $self->{nc} is intentionally left as is.
1347            ## Although the "anything else" case of the spec not explicitly
1348            ## states that the next input character is to be reconsumed,
1349            ## it will be included to the |data| of the comment token
1350            ## generated from the bogus end tag, as defined in the
1351            ## "bogus comment state" entry.
1352            redo A;
1353          }
1354        } elsif ($self->{state} == CDATA_RCDATA_CLOSE_TAG_STATE) {
1355          my $ch = substr $self->{last_stag_name}, length $self->{s_kwd}, 1;
1356          if (length $ch) {
1357            my $CH = $ch;
1358            $ch =~ tr/a-z/A-Z/;
1359            my $nch = chr $self->{nc};
1360            if ($nch eq $ch or $nch eq $CH) {
1361              !!!cp (24);
1362              ## Stay in the state.
1363              $self->{s_kwd} .= $nch;
1364              !!!next-input-character;
1365              redo A;
1366            } else {
1367              !!!cp (25);
1368              $self->{state} = DATA_STATE;
1369              ## Reconsume.
1370              !!!emit ({type => CHARACTER_TOKEN,
1371                        data => '</' . $self->{s_kwd},
1372                        line => $self->{line_prev},
1373                        column => $self->{column_prev} - 1 - length $self->{s_kwd},
1374                       });
1375              redo A;
1376            }
1377          } else { # after "<{tag-name}"
1378            unless ($is_space->{$self->{nc}} or
1379                    {
1380                     0x003E => 1, # >
1381                     0x002F => 1, # /
1382                     -1 => 1, # EOF
1383                    }->{$self->{nc}}) {
1384              !!!cp (26);
1385              ## Reconsume.
1386              $self->{state} = DATA_STATE;
1387              !!!emit ({type => CHARACTER_TOKEN,
1388                        data => '</' . $self->{s_kwd},
1389                        line => $self->{line_prev},
1390                        column => $self->{column_prev} - 1 - length $self->{s_kwd},
1391                       });
1392              redo A;
1393            } else {
1394              !!!cp (27);
1395              $self->{ct}
1396                  = {type => END_TAG_TOKEN,
1397                     tag_name => $self->{last_stag_name},
1398                     line => $self->{line_prev},
1399                     column => $self->{column_prev} - 1 - length $self->{s_kwd}};
1400              $self->{state} = TAG_NAME_STATE;
1401              ## Reconsume.
1402              redo A;
1403            }
1404        }        }
1405      } elsif ($self->{state} == TAG_NAME_STATE) {      } elsif ($self->{state} == TAG_NAME_STATE) {
1406        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
1407          !!!cp (34);          !!!cp (34);
1408          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1409          !!!next-input-character;          !!!next-input-character;
1410          redo A;          redo A;
1411        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1412          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1413            !!!cp (35);            !!!cp (35);
1414            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1415                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1416            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1417            #if ($self->{current_token}->{attributes}) {            #if ($self->{ct}->{attributes}) {
1418            #  ## NOTE: This should never be reached.            #  ## NOTE: This should never be reached.
1419            #  !!! cp (36);            #  !!! cp (36);
1420            #  !!! parse-error (type => 'end tag attribute');            #  !!! parse-error (type => 'end tag attribute');
# Line 709  sub _get_next_token ($) { Line 1422  sub _get_next_token ($) {
1422              !!!cp (37);              !!!cp (37);
1423            #}            #}
1424          } else {          } else {
1425            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1426          }          }
1427          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1428          !!!next-input-character;          !!!next-input-character;
1429    
1430          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1431    
1432          redo A;          redo A;
1433        } elsif (0x0041 <= $self->{next_char} and        } elsif (0x0041 <= $self->{nc} and
1434                 $self->{next_char} <= 0x005A) { # A..Z                 $self->{nc} <= 0x005A) { # A..Z
1435          !!!cp (38);          !!!cp (38);
1436          $self->{current_token}->{tag_name} .= chr ($self->{next_char} + 0x0020);          $self->{ct}->{tag_name} .= chr ($self->{nc} + 0x0020);
1437            # start tag or end tag            # start tag or end tag
1438          ## Stay in this state          ## Stay in this state
1439          !!!next-input-character;          !!!next-input-character;
1440          redo A;          redo A;
1441        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1442          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1443          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1444            !!!cp (39);            !!!cp (39);
1445            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1446                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1447            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1448            #if ($self->{current_token}->{attributes}) {            #if ($self->{ct}->{attributes}) {
1449            #  ## NOTE: This state should never be reached.            #  ## NOTE: This state should never be reached.
1450            #  !!! cp (40);            #  !!! cp (40);
1451            #  !!! parse-error (type => 'end tag attribute');            #  !!! parse-error (type => 'end tag attribute');
# Line 742  sub _get_next_token ($) { Line 1453  sub _get_next_token ($) {
1453              !!!cp (41);              !!!cp (41);
1454            #}            #}
1455          } else {          } else {
1456            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1457          }          }
1458          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1459          # reconsume          # reconsume
1460    
1461          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1462    
1463          redo A;          redo A;
1464        } elsif ($self->{next_char} == 0x002F) { # /        } elsif ($self->{nc} == 0x002F) { # /
1465            !!!cp (42);
1466            $self->{state} = SELF_CLOSING_START_TAG_STATE;
1467          !!!next-input-character;          !!!next-input-character;
         if ($self->{next_char} == 0x003E and # >  
             $self->{current_token}->{type} == START_TAG_TOKEN and  
             $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {  
           # permitted slash  
           !!!cp (42);  
           #  
         } else {  
           !!!cp (43);  
           !!!parse-error (type => 'nestc');  
         }  
         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;  
         # next-input-character is already done  
1468          redo A;          redo A;
1469        } else {        } else {
1470          !!!cp (44);          !!!cp (44);
1471          $self->{current_token}->{tag_name} .= chr $self->{next_char};          $self->{ct}->{tag_name} .= chr $self->{nc};
1472            # start tag or end tag            # start tag or end tag
1473          ## Stay in the state          ## Stay in the state
1474          !!!next-input-character;          !!!next-input-character;
1475          redo A;          redo A;
1476        }        }
1477      } elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) {
1478        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
1479          !!!cp (45);          !!!cp (45);
1480          ## Stay in the state          ## Stay in the state
1481          !!!next-input-character;          !!!next-input-character;
1482          redo A;          redo A;
1483        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1484          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1485            !!!cp (46);            !!!cp (46);
1486            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1487                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1488            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1489            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1490              !!!cp (47);              !!!cp (47);
1491              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1492            } else {            } else {
1493              !!!cp (48);              !!!cp (48);
1494            }            }
1495          } else {          } else {
1496            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1497          }          }
1498          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1499          !!!next-input-character;          !!!next-input-character;
1500    
1501          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1502    
1503          redo A;          redo A;
1504        } elsif (0x0041 <= $self->{next_char} and        } elsif (0x0041 <= $self->{nc} and
1505                 $self->{next_char} <= 0x005A) { # A..Z                 $self->{nc} <= 0x005A) { # A..Z
1506          !!!cp (49);          !!!cp (49);
1507          $self->{current_attribute} = {name => chr ($self->{next_char} + 0x0020),          $self->{ca}
1508                                value => ''};              = {name => chr ($self->{nc} + 0x0020),
1509                   value => '',
1510                   line => $self->{line}, column => $self->{column}};
1511          $self->{state} = ATTRIBUTE_NAME_STATE;          $self->{state} = ATTRIBUTE_NAME_STATE;
1512          !!!next-input-character;          !!!next-input-character;
1513          redo A;          redo A;
1514        } elsif ($self->{next_char} == 0x002F) { # /        } elsif ($self->{nc} == 0x002F) { # /
1515            !!!cp (50);
1516            $self->{state} = SELF_CLOSING_START_TAG_STATE;
1517          !!!next-input-character;          !!!next-input-character;
         if ($self->{next_char} == 0x003E and # >  
             $self->{current_token}->{type} == START_TAG_TOKEN and  
             $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {  
           # permitted slash  
           !!!cp (50);  
           #  
         } else {  
           !!!cp (51);  
           !!!parse-error (type => 'nestc');  
         }  
         ## Stay in the state  
         # next-input-character is already done  
1518          redo A;          redo A;
1519        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1520          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1521          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1522            !!!cp (52);            !!!cp (52);
1523            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1524                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1525            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1526            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1527              !!!cp (53);              !!!cp (53);
1528              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1529            } else {            } else {
1530              !!!cp (54);              !!!cp (54);
1531            }            }
1532          } else {          } else {
1533            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1534          }          }
1535          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1536          # reconsume          # reconsume
1537    
1538          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1539    
1540          redo A;          redo A;
1541        } else {        } else {
# Line 858  sub _get_next_token ($) { Line 1543  sub _get_next_token ($) {
1543               0x0022 => 1, # "               0x0022 => 1, # "
1544               0x0027 => 1, # '               0x0027 => 1, # '
1545               0x003D => 1, # =               0x003D => 1, # =
1546              }->{$self->{next_char}}) {              }->{$self->{nc}}) {
1547            !!!cp (55);            !!!cp (55);
1548            !!!parse-error (type => 'bad attribute name');            !!!parse-error (type => 'bad attribute name');
1549          } else {          } else {
1550            !!!cp (56);            !!!cp (56);
1551          }          }
1552          $self->{current_attribute} = {name => chr ($self->{next_char}),          $self->{ca}
1553                                value => ''};              = {name => chr ($self->{nc}),
1554                   value => '',
1555                   line => $self->{line}, column => $self->{column}};
1556          $self->{state} = ATTRIBUTE_NAME_STATE;          $self->{state} = ATTRIBUTE_NAME_STATE;
1557          !!!next-input-character;          !!!next-input-character;
1558          redo A;          redo A;
1559        }        }
1560      } elsif ($self->{state} == ATTRIBUTE_NAME_STATE) {      } elsif ($self->{state} == ATTRIBUTE_NAME_STATE) {
1561        my $before_leave = sub {        my $before_leave = sub {
1562          if (exists $self->{current_token}->{attributes} # start tag or end tag          if (exists $self->{ct}->{attributes} # start tag or end tag
1563              ->{$self->{current_attribute}->{name}}) { # MUST              ->{$self->{ca}->{name}}) { # MUST
1564            !!!cp (57);            !!!cp (57);
1565            !!!parse-error (type => 'duplicate attribute:'.$self->{current_attribute}->{name});            !!!parse-error (type => 'duplicate attribute', text => $self->{ca}->{name}, line => $self->{ca}->{line}, column => $self->{ca}->{column});
1566            ## Discard $self->{current_attribute} # MUST            ## Discard $self->{ca} # MUST
1567          } else {          } else {
1568            !!!cp (58);            !!!cp (58);
1569            $self->{current_token}->{attributes}->{$self->{current_attribute}->{name}}            $self->{ct}->{attributes}->{$self->{ca}->{name}}
1570              = $self->{current_attribute};              = $self->{ca};
1571          }          }
1572        }; # $before_leave        }; # $before_leave
1573    
1574        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
1575          !!!cp (59);          !!!cp (59);
1576          $before_leave->();          $before_leave->();
1577          $self->{state} = AFTER_ATTRIBUTE_NAME_STATE;          $self->{state} = AFTER_ATTRIBUTE_NAME_STATE;
1578          !!!next-input-character;          !!!next-input-character;
1579          redo A;          redo A;
1580        } elsif ($self->{next_char} == 0x003D) { # =        } elsif ($self->{nc} == 0x003D) { # =
1581          !!!cp (60);          !!!cp (60);
1582          $before_leave->();          $before_leave->();
1583          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
1584          !!!next-input-character;          !!!next-input-character;
1585          redo A;          redo A;
1586        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1587          $before_leave->();          $before_leave->();
1588          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1589            !!!cp (61);            !!!cp (61);
1590            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1591                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1592            !!!cp (62);            !!!cp (62);
1593            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1594            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1595              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1596            }            }
1597          } else {          } else {
1598            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1599          }          }
1600          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1601          !!!next-input-character;          !!!next-input-character;
1602    
1603          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1604    
1605          redo A;          redo A;
1606        } elsif (0x0041 <= $self->{next_char} and        } elsif (0x0041 <= $self->{nc} and
1607                 $self->{next_char} <= 0x005A) { # A..Z                 $self->{nc} <= 0x005A) { # A..Z
1608          !!!cp (63);          !!!cp (63);
1609          $self->{current_attribute}->{name} .= chr ($self->{next_char} + 0x0020);          $self->{ca}->{name} .= chr ($self->{nc} + 0x0020);
1610          ## Stay in the state          ## Stay in the state
1611          !!!next-input-character;          !!!next-input-character;
1612          redo A;          redo A;
1613        } elsif ($self->{next_char} == 0x002F) { # /        } elsif ($self->{nc} == 0x002F) { # /
1614            !!!cp (64);
1615          $before_leave->();          $before_leave->();
1616            $self->{state} = SELF_CLOSING_START_TAG_STATE;
1617          !!!next-input-character;          !!!next-input-character;
         if ($self->{next_char} == 0x003E and # >  
             $self->{current_token}->{type} == START_TAG_TOKEN and  
             $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {  
           # permitted slash  
           !!!cp (64);  
           #  
         } else {  
           !!!cp (65);  
           !!!parse-error (type => 'nestc');  
         }  
         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;  
         # next-input-character is already done  
1618          redo A;          redo A;
1619        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1620          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1621          $before_leave->();          $before_leave->();
1622          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1623            !!!cp (66);            !!!cp (66);
1624            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1625                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1626            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1627            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1628              !!!cp (67);              !!!cp (67);
1629              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1630            } else {            } else {
# Line 963  sub _get_next_token ($) { Line 1632  sub _get_next_token ($) {
1632              !!!cp (68);              !!!cp (68);
1633            }            }
1634          } else {          } else {
1635            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1636          }          }
1637          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1638          # reconsume          # reconsume
1639    
1640          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1641    
1642          redo A;          redo A;
1643        } else {        } else {
1644          if ($self->{next_char} == 0x0022 or # "          if ($self->{nc} == 0x0022 or # "
1645              $self->{next_char} == 0x0027) { # '              $self->{nc} == 0x0027) { # '
1646            !!!cp (69);            !!!cp (69);
1647            !!!parse-error (type => 'bad attribute name');            !!!parse-error (type => 'bad attribute name');
1648          } else {          } else {
1649            !!!cp (70);            !!!cp (70);
1650          }          }
1651          $self->{current_attribute}->{name} .= chr ($self->{next_char});          $self->{ca}->{name} .= chr ($self->{nc});
1652          ## Stay in the state          ## Stay in the state
1653          !!!next-input-character;          !!!next-input-character;
1654          redo A;          redo A;
1655        }        }
1656      } elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) {      } elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) {
1657        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
1658          !!!cp (71);          !!!cp (71);
1659          ## Stay in the state          ## Stay in the state
1660          !!!next-input-character;          !!!next-input-character;
1661          redo A;          redo A;
1662        } elsif ($self->{next_char} == 0x003D) { # =        } elsif ($self->{nc} == 0x003D) { # =
1663          !!!cp (72);          !!!cp (72);
1664          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
1665          !!!next-input-character;          !!!next-input-character;
1666          redo A;          redo A;
1667        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1668          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1669            !!!cp (73);            !!!cp (73);
1670            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1671                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1672            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1673            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1674              !!!cp (74);              !!!cp (74);
1675              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1676            } else {            } else {
# Line 1015  sub _get_next_token ($) { Line 1678  sub _get_next_token ($) {
1678              !!!cp (75);              !!!cp (75);
1679            }            }
1680          } else {          } else {
1681            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1682          }          }
1683          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1684          !!!next-input-character;          !!!next-input-character;
1685    
1686          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1687    
1688          redo A;          redo A;
1689        } elsif (0x0041 <= $self->{next_char} and        } elsif (0x0041 <= $self->{nc} and
1690                 $self->{next_char} <= 0x005A) { # A..Z                 $self->{nc} <= 0x005A) { # A..Z
1691          !!!cp (76);          !!!cp (76);
1692          $self->{current_attribute} = {name => chr ($self->{next_char} + 0x0020),          $self->{ca}
1693                                value => ''};              = {name => chr ($self->{nc} + 0x0020),
1694                   value => '',
1695                   line => $self->{line}, column => $self->{column}};
1696          $self->{state} = ATTRIBUTE_NAME_STATE;          $self->{state} = ATTRIBUTE_NAME_STATE;
1697          !!!next-input-character;          !!!next-input-character;
1698          redo A;          redo A;
1699        } elsif ($self->{next_char} == 0x002F) { # /        } elsif ($self->{nc} == 0x002F) { # /
1700            !!!cp (77);
1701            $self->{state} = SELF_CLOSING_START_TAG_STATE;
1702          !!!next-input-character;          !!!next-input-character;
         if ($self->{next_char} == 0x003E and # >  
             $self->{current_token}->{type} == START_TAG_TOKEN and  
             $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {  
           # permitted slash  
           !!!cp (77);  
           #  
         } else {  
           !!!cp (78);  
           !!!parse-error (type => 'nestc');  
           ## TODO: Different error type for <aa / bb> than <aa/>  
         }  
         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;  
         # next-input-character is already done  
1703          redo A;          redo A;
1704        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1705          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1706          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1707            !!!cp (79);            !!!cp (79);
1708            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1709                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1710            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1711            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1712              !!!cp (80);              !!!cp (80);
1713              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1714            } else {            } else {
# Line 1064  sub _get_next_token ($) { Line 1716  sub _get_next_token ($) {
1716              !!!cp (81);              !!!cp (81);
1717            }            }
1718          } else {          } else {
1719            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1720          }          }
1721          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1722          # reconsume          # reconsume
1723    
1724          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1725    
1726          redo A;          redo A;
1727        } else {        } else {
1728          !!!cp (82);          if ($self->{nc} == 0x0022 or # "
1729          $self->{current_attribute} = {name => chr ($self->{next_char}),              $self->{nc} == 0x0027) { # '
1730                                value => ''};            !!!cp (78);
1731              !!!parse-error (type => 'bad attribute name');
1732            } else {
1733              !!!cp (82);
1734            }
1735            $self->{ca}
1736                = {name => chr ($self->{nc}),
1737                   value => '',
1738                   line => $self->{line}, column => $self->{column}};
1739          $self->{state} = ATTRIBUTE_NAME_STATE;          $self->{state} = ATTRIBUTE_NAME_STATE;
1740          !!!next-input-character;          !!!next-input-character;
1741          redo A;                  redo A;        
1742        }        }
1743      } elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) {
1744        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP        
1745          !!!cp (83);          !!!cp (83);
1746          ## Stay in the state          ## Stay in the state
1747          !!!next-input-character;          !!!next-input-character;
1748          redo A;          redo A;
1749        } elsif ($self->{next_char} == 0x0022) { # "        } elsif ($self->{nc} == 0x0022) { # "
1750          !!!cp (84);          !!!cp (84);
1751          $self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE;          $self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE;
1752          !!!next-input-character;          !!!next-input-character;
1753          redo A;          redo A;
1754        } elsif ($self->{next_char} == 0x0026) { # &        } elsif ($self->{nc} == 0x0026) { # &
1755          !!!cp (85);          !!!cp (85);
1756          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
1757          ## reconsume          ## reconsume
1758          redo A;          redo A;
1759        } elsif ($self->{next_char} == 0x0027) { # '        } elsif ($self->{nc} == 0x0027) { # '
1760          !!!cp (86);          !!!cp (86);
1761          $self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE;          $self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE;
1762          !!!next-input-character;          !!!next-input-character;
1763          redo A;          redo A;
1764        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1765          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          !!!parse-error (type => 'empty unquoted attribute value');
1766            if ($self->{ct}->{type} == START_TAG_TOKEN) {
1767            !!!cp (87);            !!!cp (87);
1768            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1769                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1770            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1771            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1772              !!!cp (88);              !!!cp (88);
1773              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1774            } else {            } else {
# Line 1121  sub _get_next_token ($) { Line 1776  sub _get_next_token ($) {
1776              !!!cp (89);              !!!cp (89);
1777            }            }
1778          } else {          } else {
1779            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1780          }          }
1781          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1782          !!!next-input-character;          !!!next-input-character;
1783    
1784          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1785    
1786          redo A;          redo A;
1787        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1788          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1789          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1790            !!!cp (90);            !!!cp (90);
1791            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1792                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1793            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1794            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1795              !!!cp (91);              !!!cp (91);
1796              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1797            } else {            } else {
# Line 1146  sub _get_next_token ($) { Line 1799  sub _get_next_token ($) {
1799              !!!cp (92);              !!!cp (92);
1800            }            }
1801          } else {          } else {
1802            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1803          }          }
1804          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1805          ## reconsume          ## reconsume
1806    
1807          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1808    
1809          redo A;          redo A;
1810        } else {        } else {
1811          if ($self->{next_char} == 0x003D) { # =          if ($self->{nc} == 0x003D) { # =
1812            !!!cp (93);            !!!cp (93);
1813            !!!parse-error (type => 'bad attribute value');            !!!parse-error (type => 'bad attribute value');
1814          } else {          } else {
1815            !!!cp (94);            !!!cp (94);
1816          }          }
1817          $self->{current_attribute}->{value} .= chr ($self->{next_char});          $self->{ca}->{value} .= chr ($self->{nc});
1818          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
1819          !!!next-input-character;          !!!next-input-character;
1820          redo A;          redo A;
1821        }        }
1822      } elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) {      } elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) {
1823        if ($self->{next_char} == 0x0022) { # "        if ($self->{nc} == 0x0022) { # "
1824          !!!cp (95);          !!!cp (95);
1825          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1826          !!!next-input-character;          !!!next-input-character;
1827          redo A;          redo A;
1828        } elsif ($self->{next_char} == 0x0026) { # &        } elsif ($self->{nc} == 0x0026) { # &
1829          !!!cp (96);          !!!cp (96);
1830          $self->{last_attribute_value_state} = $self->{state};          ## NOTE: In the spec, the tokenizer is switched to the
1831          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;          ## "entity in attribute value state".  In this implementation, the
1832            ## tokenizer is switched to the |ENTITY_STATE|, which is an
1833            ## implementation of the "consume a character reference" algorithm.
1834            $self->{prev_state} = $self->{state};
1835            $self->{entity_add} = 0x0022; # "
1836            $self->{state} = ENTITY_STATE;
1837          !!!next-input-character;          !!!next-input-character;
1838          redo A;          redo A;
1839        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1840          !!!parse-error (type => 'unclosed attribute value');          !!!parse-error (type => 'unclosed attribute value');
1841          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1842            !!!cp (97);            !!!cp (97);
1843            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1844                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1845            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1846            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1847              !!!cp (98);              !!!cp (98);
1848              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1849            } else {            } else {
# Line 1195  sub _get_next_token ($) { Line 1851  sub _get_next_token ($) {
1851              !!!cp (99);              !!!cp (99);
1852            }            }
1853          } else {          } else {
1854            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1855          }          }
1856          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1857          ## reconsume          ## reconsume
1858    
1859          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1860    
1861          redo A;          redo A;
1862        } else {        } else {
1863          !!!cp (100);          !!!cp (100);
1864          $self->{current_attribute}->{value} .= chr ($self->{next_char});          $self->{ca}->{value} .= chr ($self->{nc});
1865            $self->{read_until}->($self->{ca}->{value},
1866                                  q["&],
1867                                  length $self->{ca}->{value});
1868    
1869          ## Stay in the state          ## Stay in the state
1870          !!!next-input-character;          !!!next-input-character;
1871          redo A;          redo A;
1872        }        }
1873      } elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) {      } elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) {
1874        if ($self->{next_char} == 0x0027) { # '        if ($self->{nc} == 0x0027) { # '
1875          !!!cp (101);          !!!cp (101);
1876          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1877          !!!next-input-character;          !!!next-input-character;
1878          redo A;          redo A;
1879        } elsif ($self->{next_char} == 0x0026) { # &        } elsif ($self->{nc} == 0x0026) { # &
1880          !!!cp (102);          !!!cp (102);
1881          $self->{last_attribute_value_state} = $self->{state};          ## NOTE: In the spec, the tokenizer is switched to the
1882          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;          ## "entity in attribute value state".  In this implementation, the
1883            ## tokenizer is switched to the |ENTITY_STATE|, which is an
1884            ## implementation of the "consume a character reference" algorithm.
1885            $self->{entity_add} = 0x0027; # '
1886            $self->{prev_state} = $self->{state};
1887            $self->{state} = ENTITY_STATE;
1888          !!!next-input-character;          !!!next-input-character;
1889          redo A;          redo A;
1890        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1891          !!!parse-error (type => 'unclosed attribute value');          !!!parse-error (type => 'unclosed attribute value');
1892          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1893            !!!cp (103);            !!!cp (103);
1894            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1895                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1896            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1897            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1898              !!!cp (104);              !!!cp (104);
1899              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1900            } else {            } else {
# Line 1239  sub _get_next_token ($) { Line 1902  sub _get_next_token ($) {
1902              !!!cp (105);              !!!cp (105);
1903            }            }
1904          } else {          } else {
1905            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1906          }          }
1907          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1908          ## reconsume          ## reconsume
1909    
1910          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1911    
1912          redo A;          redo A;
1913        } else {        } else {
1914          !!!cp (106);          !!!cp (106);
1915          $self->{current_attribute}->{value} .= chr ($self->{next_char});          $self->{ca}->{value} .= chr ($self->{nc});
1916            $self->{read_until}->($self->{ca}->{value},
1917                                  q['&],
1918                                  length $self->{ca}->{value});
1919    
1920          ## Stay in the state          ## Stay in the state
1921          !!!next-input-character;          !!!next-input-character;
1922          redo A;          redo A;
1923        }        }
1924      } elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) {      } elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) {
1925        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # HT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
1926          !!!cp (107);          !!!cp (107);
1927          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1928          !!!next-input-character;          !!!next-input-character;
1929          redo A;          redo A;
1930        } elsif ($self->{next_char} == 0x0026) { # &        } elsif ($self->{nc} == 0x0026) { # &
1931          !!!cp (108);          !!!cp (108);
1932          $self->{last_attribute_value_state} = $self->{state};          ## NOTE: In the spec, the tokenizer is switched to the
1933          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;          ## "entity in attribute value state".  In this implementation, the
1934            ## tokenizer is switched to the |ENTITY_STATE|, which is an
1935            ## implementation of the "consume a character reference" algorithm.
1936            $self->{entity_add} = -1;
1937            $self->{prev_state} = $self->{state};
1938            $self->{state} = ENTITY_STATE;
1939          !!!next-input-character;          !!!next-input-character;
1940          redo A;          redo A;
1941        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
1942          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1943            !!!cp (109);            !!!cp (109);
1944            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1945                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1946            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1947            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1948              !!!cp (110);              !!!cp (110);
1949              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1950            } else {            } else {
# Line 1286  sub _get_next_token ($) { Line 1952  sub _get_next_token ($) {
1952              !!!cp (111);              !!!cp (111);
1953            }            }
1954          } else {          } else {
1955            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1956          }          }
1957          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1958          !!!next-input-character;          !!!next-input-character;
1959    
1960          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1961    
1962          redo A;          redo A;
1963        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
1964          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1965          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
1966            !!!cp (112);            !!!cp (112);
1967            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
1968                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
1969            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1970            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
1971              !!!cp (113);              !!!cp (113);
1972              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1973            } else {            } else {
# Line 1311  sub _get_next_token ($) { Line 1975  sub _get_next_token ($) {
1975              !!!cp (114);              !!!cp (114);
1976            }            }
1977          } else {          } else {
1978            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
1979          }          }
1980          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
1981          ## reconsume          ## reconsume
1982    
1983          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
1984    
1985          redo A;          redo A;
1986        } else {        } else {
# Line 1324  sub _get_next_token ($) { Line 1988  sub _get_next_token ($) {
1988               0x0022 => 1, # "               0x0022 => 1, # "
1989               0x0027 => 1, # '               0x0027 => 1, # '
1990               0x003D => 1, # =               0x003D => 1, # =
1991              }->{$self->{next_char}}) {              }->{$self->{nc}}) {
1992            !!!cp (115);            !!!cp (115);
1993            !!!parse-error (type => 'bad attribute value');            !!!parse-error (type => 'bad attribute value');
1994          } else {          } else {
1995            !!!cp (116);            !!!cp (116);
1996          }          }
1997          $self->{current_attribute}->{value} .= chr ($self->{next_char});          $self->{ca}->{value} .= chr ($self->{nc});
1998            $self->{read_until}->($self->{ca}->{value},
1999                                  q["'=& >],
2000                                  length $self->{ca}->{value});
2001    
2002          ## Stay in the state          ## Stay in the state
2003          !!!next-input-character;          !!!next-input-character;
2004          redo A;          redo A;
2005        }        }
     } elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) {  
       my $token = $self->_tokenize_attempt_to_consume_an_entity  
           (1,  
            $self->{last_attribute_value_state}  
              == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE ? 0x0022 : # "  
            $self->{last_attribute_value_state}  
              == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE ? 0x0027 : # '  
            -1);  
   
       unless (defined $token) {  
         !!!cp (117);  
         $self->{current_attribute}->{value} .= '&';  
       } else {  
         !!!cp (118);  
         $self->{current_attribute}->{value} .= $token->{data};  
         $self->{current_attribute}->{has_reference} = $token->{has_reference};  
         ## ISSUE: spec says "append the returned character token to the current attribute's value"  
       }  
   
       $self->{state} = $self->{last_attribute_value_state};  
       # next-input-character is already done  
       redo A;  
2006      } elsif ($self->{state} == AFTER_ATTRIBUTE_VALUE_QUOTED_STATE) {      } elsif ($self->{state} == AFTER_ATTRIBUTE_VALUE_QUOTED_STATE) {
2007        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
2008          !!!cp (118);          !!!cp (118);
2009          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
2010          !!!next-input-character;          !!!next-input-character;
2011          redo A;          redo A;
2012        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2013          if ($self->{current_token}->{type} == START_TAG_TOKEN) {          if ($self->{ct}->{type} == START_TAG_TOKEN) {
2014            !!!cp (119);            !!!cp (119);
2015            $self->{current_token}->{first_start_tag}            $self->{last_stag_name} = $self->{ct}->{tag_name};
2016                = not defined $self->{last_emitted_start_tag_name};          } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
           $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};  
         } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {  
2017            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
2018            if ($self->{current_token}->{attributes}) {            if ($self->{ct}->{attributes}) {
2019              !!!cp (120);              !!!cp (120);
2020              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
2021            } else {            } else {
# Line 1383  sub _get_next_token ($) { Line 2023  sub _get_next_token ($) {
2023              !!!cp (121);              !!!cp (121);
2024            }            }
2025          } else {          } else {
2026            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{ct}->{type}: Unknown token type";
2027          }          }
2028          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2029          !!!next-input-character;          !!!next-input-character;
2030    
2031          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{ct}); # start tag or end tag
2032    
2033          redo A;          redo A;
2034        } elsif ($self->{next_char} == 0x002F) { # /        } elsif ($self->{nc} == 0x002F) { # /
2035            !!!cp (122);
2036            $self->{state} = SELF_CLOSING_START_TAG_STATE;
2037          !!!next-input-character;          !!!next-input-character;
2038          if ($self->{next_char} == 0x003E and # >          redo A;
2039              $self->{current_token}->{type} == START_TAG_TOKEN and        } elsif ($self->{nc} == -1) {
2040              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {          !!!parse-error (type => 'unclosed tag');
2041            # permitted slash          if ($self->{ct}->{type} == START_TAG_TOKEN) {
2042            !!!cp (122);            !!!cp (122.3);
2043            #            $self->{last_stag_name} = $self->{ct}->{tag_name};
2044            } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
2045              if ($self->{ct}->{attributes}) {
2046                !!!cp (122.1);
2047                !!!parse-error (type => 'end tag attribute');
2048              } else {
2049                ## NOTE: This state should never be reached.
2050                !!!cp (122.2);
2051              }
2052          } else {          } else {
2053            !!!cp (123);            die "$0: $self->{ct}->{type}: Unknown token type";
           !!!parse-error (type => 'nestc');  
2054          }          }
2055          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;          $self->{state} = DATA_STATE;
2056          # next-input-character is already done          ## Reconsume.
2057            !!!emit ($self->{ct}); # start tag or end tag
2058          redo A;          redo A;
2059        } else {        } else {
2060          !!!cp (124);          !!!cp ('124.1');
2061          !!!parse-error (type => 'no space between attributes');          !!!parse-error (type => 'no space between attributes');
2062          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
2063          ## reconsume          ## reconsume
2064          redo A;          redo A;
2065        }        }
2066      } elsif ($self->{state} == BOGUS_COMMENT_STATE) {      } elsif ($self->{state} == SELF_CLOSING_START_TAG_STATE) {
2067        ## (only happen if PCDATA state)        if ($self->{nc} == 0x003E) { # >
2068                  if ($self->{ct}->{type} == END_TAG_TOKEN) {
2069        ## NOTE: Set by the previous state            !!!cp ('124.2');
2070        #my $token = {type => COMMENT_TOKEN, data => ''};            !!!parse-error (type => 'nestc', token => $self->{ct});
2071              ## TODO: Different type than slash in start tag
2072        BC: {            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
2073          if ($self->{next_char} == 0x003E) { # >            if ($self->{ct}->{attributes}) {
2074            !!!cp (124);              !!!cp ('124.4');
2075            $self->{state} = DATA_STATE;              !!!parse-error (type => 'end tag attribute');
2076            !!!next-input-character;            } else {
2077                !!!cp ('124.5');
2078            !!!emit ($self->{current_token}); # comment            }
2079              ## TODO: Test |<title></title/>|
2080            } else {
2081              !!!cp ('124.3');
2082              $self->{self_closing} = 1;
2083            }
2084    
2085            redo A;          $self->{state} = DATA_STATE;
2086          } elsif ($self->{next_char} == -1) {          !!!next-input-character;
           !!!cp (125);  
           $self->{state} = DATA_STATE;  
           ## reconsume  
2087    
2088            !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # start tag or end tag
2089    
2090            redo A;          redo A;
2091          } elsif ($self->{nc} == -1) {
2092            !!!parse-error (type => 'unclosed tag');
2093            if ($self->{ct}->{type} == START_TAG_TOKEN) {
2094              !!!cp (124.7);
2095              $self->{last_stag_name} = $self->{ct}->{tag_name};
2096            } elsif ($self->{ct}->{type} == END_TAG_TOKEN) {
2097              if ($self->{ct}->{attributes}) {
2098                !!!cp (124.5);
2099                !!!parse-error (type => 'end tag attribute');
2100              } else {
2101                ## NOTE: This state should never be reached.
2102                !!!cp (124.6);
2103              }
2104          } else {          } else {
2105            !!!cp (126);            die "$0: $self->{ct}->{type}: Unknown token type";
           $self->{current_token}->{data} .= chr ($self->{next_char}); # comment  
           !!!next-input-character;  
           redo BC;  
2106          }          }
2107        } # BC          $self->{state} = DATA_STATE;
2108            ## Reconsume.
2109        die "$0: _get_next_token: unexpected case [BC]";          !!!emit ($self->{ct}); # start tag or end tag
2110      } elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) {          redo A;
2111          } else {
2112            !!!cp ('124.4');
2113            !!!parse-error (type => 'nestc');
2114            ## TODO: This error type is wrong.
2115            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
2116            ## Reconsume.
2117            redo A;
2118          }
2119        } elsif ($self->{state} == BOGUS_COMMENT_STATE) {
2120        ## (only happen if PCDATA state)        ## (only happen if PCDATA state)
2121    
2122        my ($l, $c) = ($self->{line_prev}, $self->{column_prev} - 1);        ## NOTE: Unlike spec's "bogus comment state", this implementation
2123          ## consumes characters one-by-one basis.
       my @next_char;  
       push @next_char, $self->{next_char};  
2124                
2125        if ($self->{next_char} == 0x002D) { # -        if ($self->{nc} == 0x003E) { # >
2126            !!!cp (124);
2127            $self->{state} = DATA_STATE;
2128          !!!next-input-character;          !!!next-input-character;
2129          push @next_char, $self->{next_char};  
2130          if ($self->{next_char} == 0x002D) { # -          !!!emit ($self->{ct}); # comment
2131            !!!cp (127);          redo A;
2132            $self->{current_token} = {type => COMMENT_TOKEN, data => '',        } elsif ($self->{nc} == -1) {
2133                                      line => $l, column => $c};          !!!cp (125);
2134            $self->{state} = COMMENT_START_STATE;          $self->{state} = DATA_STATE;
2135            !!!next-input-character;          ## reconsume
2136            redo A;  
2137          } else {          !!!emit ($self->{ct}); # comment
2138            !!!cp (128);          redo A;
2139          }        } else {
2140        } elsif ($self->{next_char} == 0x0044 or # D          !!!cp (126);
2141                 $self->{next_char} == 0x0064) { # d          $self->{ct}->{data} .= chr ($self->{nc}); # comment
2142            $self->{read_until}->($self->{ct}->{data},
2143                                  q[>],
2144                                  length $self->{ct}->{data});
2145    
2146            ## Stay in the state.
2147          !!!next-input-character;          !!!next-input-character;
2148          push @next_char, $self->{next_char};          redo A;
2149          if ($self->{next_char} == 0x004F or # O        }
2150              $self->{next_char} == 0x006F) { # o      } elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) {
2151            !!!next-input-character;        ## (only happen if PCDATA state)
2152            push @next_char, $self->{next_char};        
2153            if ($self->{next_char} == 0x0043 or # C        if ($self->{nc} == 0x002D) { # -
2154                $self->{next_char} == 0x0063) { # c          !!!cp (133);
2155              !!!next-input-character;          $self->{state} = MD_HYPHEN_STATE;
2156              push @next_char, $self->{next_char};          !!!next-input-character;
2157              if ($self->{next_char} == 0x0054 or # T          redo A;
2158                  $self->{next_char} == 0x0074) { # t        } elsif ($self->{nc} == 0x0044 or # D
2159                !!!next-input-character;                 $self->{nc} == 0x0064) { # d
2160                push @next_char, $self->{next_char};          ## ASCII case-insensitive.
2161                if ($self->{next_char} == 0x0059 or # Y          !!!cp (130);
2162                    $self->{next_char} == 0x0079) { # y          $self->{state} = MD_DOCTYPE_STATE;
2163                  !!!next-input-character;          $self->{s_kwd} = chr $self->{nc};
2164                  push @next_char, $self->{next_char};          !!!next-input-character;
2165                  if ($self->{next_char} == 0x0050 or # P          redo A;
2166                      $self->{next_char} == 0x0070) { # p        } elsif ($self->{insertion_mode} & IN_FOREIGN_CONTENT_IM and
2167                    !!!next-input-character;                 $self->{open_elements}->[-1]->[1] & FOREIGN_EL and
2168                    push @next_char, $self->{next_char};                 $self->{nc} == 0x005B) { # [
2169                    if ($self->{next_char} == 0x0045 or # E          !!!cp (135.4);                
2170                        $self->{next_char} == 0x0065) { # e          $self->{state} = MD_CDATA_STATE;
2171                      !!!cp (129);          $self->{s_kwd} = '[';
2172                      ## TODO: What a stupid code this is!          !!!next-input-character;
2173                      $self->{state} = DOCTYPE_STATE;          redo A;
                     $self->{current_token} = {type => DOCTYPE_TOKEN,  
                                               quirks => 1,  
                                               line => $l, column => $c};  
                     !!!next-input-character;  
                     redo A;  
                   } else {  
                     !!!cp (130);  
                   }  
                 } else {  
                   !!!cp (131);  
                 }  
               } else {  
                 !!!cp (132);  
               }  
             } else {  
               !!!cp (133);  
             }  
           } else {  
             !!!cp (134);  
           }  
         } else {  
           !!!cp (135);  
         }  
2174        } else {        } else {
2175          !!!cp (136);          !!!cp (136);
2176        }        }
2177    
2178        !!!parse-error (type => 'bogus comment');        !!!parse-error (type => 'bogus comment',
2179        $self->{next_char} = shift @next_char;                        line => $self->{line_prev},
2180        !!!back-next-input-character (@next_char);                        column => $self->{column_prev} - 1);
2181          ## Reconsume.
2182        $self->{state} = BOGUS_COMMENT_STATE;        $self->{state} = BOGUS_COMMENT_STATE;
2183        $self->{current_token} = {type => COMMENT_TOKEN, data => '',        $self->{ct} = {type => COMMENT_TOKEN, data => '',
2184                                  line => $l, column => $c};                                  line => $self->{line_prev},
2185                                    column => $self->{column_prev} - 1,
2186                                   };
2187        redo A;        redo A;
2188              } elsif ($self->{state} == MD_HYPHEN_STATE) {
2189        ## ISSUE: typos in spec: chacacters, is is a parse error        if ($self->{nc} == 0x002D) { # -
2190        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?          !!!cp (127);
2191            $self->{ct} = {type => COMMENT_TOKEN, data => '',
2192                                      line => $self->{line_prev},
2193                                      column => $self->{column_prev} - 2,
2194                                     };
2195            $self->{state} = COMMENT_START_STATE;
2196            !!!next-input-character;
2197            redo A;
2198          } else {
2199            !!!cp (128);
2200            !!!parse-error (type => 'bogus comment',
2201                            line => $self->{line_prev},
2202                            column => $self->{column_prev} - 2);
2203            $self->{state} = BOGUS_COMMENT_STATE;
2204            ## Reconsume.
2205            $self->{ct} = {type => COMMENT_TOKEN,
2206                                      data => '-',
2207                                      line => $self->{line_prev},
2208                                      column => $self->{column_prev} - 2,
2209                                     };
2210            redo A;
2211          }
2212        } elsif ($self->{state} == MD_DOCTYPE_STATE) {
2213          ## ASCII case-insensitive.
2214          if ($self->{nc} == [
2215                undef,
2216                0x004F, # O
2217                0x0043, # C
2218                0x0054, # T
2219                0x0059, # Y
2220                0x0050, # P
2221              ]->[length $self->{s_kwd}] or
2222              $self->{nc} == [
2223                undef,
2224                0x006F, # o
2225                0x0063, # c
2226                0x0074, # t
2227                0x0079, # y
2228                0x0070, # p
2229              ]->[length $self->{s_kwd}]) {
2230            !!!cp (131);
2231            ## Stay in the state.
2232            $self->{s_kwd} .= chr $self->{nc};
2233            !!!next-input-character;
2234            redo A;
2235          } elsif ((length $self->{s_kwd}) == 6 and
2236                   ($self->{nc} == 0x0045 or # E
2237                    $self->{nc} == 0x0065)) { # e
2238            !!!cp (129);
2239            $self->{state} = DOCTYPE_STATE;
2240            $self->{ct} = {type => DOCTYPE_TOKEN,
2241                                      quirks => 1,
2242                                      line => $self->{line_prev},
2243                                      column => $self->{column_prev} - 7,
2244                                     };
2245            !!!next-input-character;
2246            redo A;
2247          } else {
2248            !!!cp (132);        
2249            !!!parse-error (type => 'bogus comment',
2250                            line => $self->{line_prev},
2251                            column => $self->{column_prev} - 1 - length $self->{s_kwd});
2252            $self->{state} = BOGUS_COMMENT_STATE;
2253            ## Reconsume.
2254            $self->{ct} = {type => COMMENT_TOKEN,
2255                                      data => $self->{s_kwd},
2256                                      line => $self->{line_prev},
2257                                      column => $self->{column_prev} - 1 - length $self->{s_kwd},
2258                                     };
2259            redo A;
2260          }
2261        } elsif ($self->{state} == MD_CDATA_STATE) {
2262          if ($self->{nc} == {
2263                '[' => 0x0043, # C
2264                '[C' => 0x0044, # D
2265                '[CD' => 0x0041, # A
2266                '[CDA' => 0x0054, # T
2267                '[CDAT' => 0x0041, # A
2268              }->{$self->{s_kwd}}) {
2269            !!!cp (135.1);
2270            ## Stay in the state.
2271            $self->{s_kwd} .= chr $self->{nc};
2272            !!!next-input-character;
2273            redo A;
2274          } elsif ($self->{s_kwd} eq '[CDATA' and
2275                   $self->{nc} == 0x005B) { # [
2276            !!!cp (135.2);
2277            $self->{ct} = {type => CHARACTER_TOKEN,
2278                                      data => '',
2279                                      line => $self->{line_prev},
2280                                      column => $self->{column_prev} - 7};
2281            $self->{state} = CDATA_SECTION_STATE;
2282            !!!next-input-character;
2283            redo A;
2284          } else {
2285            !!!cp (135.3);
2286            !!!parse-error (type => 'bogus comment',
2287                            line => $self->{line_prev},
2288                            column => $self->{column_prev} - 1 - length $self->{s_kwd});
2289            $self->{state} = BOGUS_COMMENT_STATE;
2290            ## Reconsume.
2291            $self->{ct} = {type => COMMENT_TOKEN,
2292                                      data => $self->{s_kwd},
2293                                      line => $self->{line_prev},
2294                                      column => $self->{column_prev} - 1 - length $self->{s_kwd},
2295                                     };
2296            redo A;
2297          }
2298      } elsif ($self->{state} == COMMENT_START_STATE) {      } elsif ($self->{state} == COMMENT_START_STATE) {
2299        if ($self->{next_char} == 0x002D) { # -        if ($self->{nc} == 0x002D) { # -
2300          !!!cp (137);          !!!cp (137);
2301          $self->{state} = COMMENT_START_DASH_STATE;          $self->{state} = COMMENT_START_DASH_STATE;
2302          !!!next-input-character;          !!!next-input-character;
2303          redo A;          redo A;
2304        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2305          !!!cp (138);          !!!cp (138);
2306          !!!parse-error (type => 'bogus comment');          !!!parse-error (type => 'bogus comment');
2307          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2308          !!!next-input-character;          !!!next-input-character;
2309    
2310          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2311    
2312          redo A;          redo A;
2313        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2314          !!!cp (139);          !!!cp (139);
2315          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
2316          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2317          ## reconsume          ## reconsume
2318    
2319          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2320    
2321          redo A;          redo A;
2322        } else {        } else {
2323          !!!cp (140);          !!!cp (140);
2324          $self->{current_token}->{data} # comment          $self->{ct}->{data} # comment
2325              .= chr ($self->{next_char});              .= chr ($self->{nc});
2326          $self->{state} = COMMENT_STATE;          $self->{state} = COMMENT_STATE;
2327          !!!next-input-character;          !!!next-input-character;
2328          redo A;          redo A;
2329        }        }
2330      } elsif ($self->{state} == COMMENT_START_DASH_STATE) {      } elsif ($self->{state} == COMMENT_START_DASH_STATE) {
2331        if ($self->{next_char} == 0x002D) { # -        if ($self->{nc} == 0x002D) { # -
2332          !!!cp (141);          !!!cp (141);
2333          $self->{state} = COMMENT_END_STATE;          $self->{state} = COMMENT_END_STATE;
2334          !!!next-input-character;          !!!next-input-character;
2335          redo A;          redo A;
2336        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2337          !!!cp (142);          !!!cp (142);
2338          !!!parse-error (type => 'bogus comment');          !!!parse-error (type => 'bogus comment');
2339          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2340          !!!next-input-character;          !!!next-input-character;
2341    
2342          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2343    
2344          redo A;          redo A;
2345        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2346          !!!cp (143);          !!!cp (143);
2347          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
2348          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2349          ## reconsume          ## reconsume
2350    
2351          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2352    
2353          redo A;          redo A;
2354        } else {        } else {
2355          !!!cp (144);          !!!cp (144);
2356          $self->{current_token}->{data} # comment          $self->{ct}->{data} # comment
2357              .= '-' . chr ($self->{next_char});              .= '-' . chr ($self->{nc});
2358          $self->{state} = COMMENT_STATE;          $self->{state} = COMMENT_STATE;
2359          !!!next-input-character;          !!!next-input-character;
2360          redo A;          redo A;
2361        }        }
2362      } elsif ($self->{state} == COMMENT_STATE) {      } elsif ($self->{state} == COMMENT_STATE) {
2363        if ($self->{next_char} == 0x002D) { # -        if ($self->{nc} == 0x002D) { # -
2364          !!!cp (145);          !!!cp (145);
2365          $self->{state} = COMMENT_END_DASH_STATE;          $self->{state} = COMMENT_END_DASH_STATE;
2366          !!!next-input-character;          !!!next-input-character;
2367          redo A;          redo A;
2368        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2369          !!!cp (146);          !!!cp (146);
2370          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
2371          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2372          ## reconsume          ## reconsume
2373    
2374          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2375    
2376          redo A;          redo A;
2377        } else {        } else {
2378          !!!cp (147);          !!!cp (147);
2379          $self->{current_token}->{data} .= chr ($self->{next_char}); # comment          $self->{ct}->{data} .= chr ($self->{nc}); # comment
2380            $self->{read_until}->($self->{ct}->{data},
2381                                  q[-],
2382                                  length $self->{ct}->{data});
2383    
2384          ## Stay in the state          ## Stay in the state
2385          !!!next-input-character;          !!!next-input-character;
2386          redo A;          redo A;
2387        }        }
2388      } elsif ($self->{state} == COMMENT_END_DASH_STATE) {      } elsif ($self->{state} == COMMENT_END_DASH_STATE) {
2389        if ($self->{next_char} == 0x002D) { # -        if ($self->{nc} == 0x002D) { # -
2390          !!!cp (148);          !!!cp (148);
2391          $self->{state} = COMMENT_END_STATE;          $self->{state} = COMMENT_END_STATE;
2392          !!!next-input-character;          !!!next-input-character;
2393          redo A;          redo A;
2394        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2395          !!!cp (149);          !!!cp (149);
2396          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
2397          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2398          ## reconsume          ## reconsume
2399    
2400          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2401    
2402          redo A;          redo A;
2403        } else {        } else {
2404          !!!cp (150);          !!!cp (150);
2405          $self->{current_token}->{data} .= '-' . chr ($self->{next_char}); # comment          $self->{ct}->{data} .= '-' . chr ($self->{nc}); # comment
2406          $self->{state} = COMMENT_STATE;          $self->{state} = COMMENT_STATE;
2407          !!!next-input-character;          !!!next-input-character;
2408          redo A;          redo A;
2409        }        }
2410      } elsif ($self->{state} == COMMENT_END_STATE) {      } elsif ($self->{state} == COMMENT_END_STATE) {
2411        if ($self->{next_char} == 0x003E) { # >        if ($self->{nc} == 0x003E) { # >
2412          !!!cp (151);          !!!cp (151);
2413          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2414          !!!next-input-character;          !!!next-input-character;
2415    
2416          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2417    
2418          redo A;          redo A;
2419        } elsif ($self->{next_char} == 0x002D) { # -        } elsif ($self->{nc} == 0x002D) { # -
2420          !!!cp (152);          !!!cp (152);
2421          !!!parse-error (type => 'dash in comment');          !!!parse-error (type => 'dash in comment',
2422          $self->{current_token}->{data} .= '-'; # comment                          line => $self->{line_prev},
2423                            column => $self->{column_prev});
2424            $self->{ct}->{data} .= '-'; # comment
2425          ## Stay in the state          ## Stay in the state
2426          !!!next-input-character;          !!!next-input-character;
2427          redo A;          redo A;
2428        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2429          !!!cp (153);          !!!cp (153);
2430          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
2431          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2432          ## reconsume          ## reconsume
2433    
2434          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{ct}); # comment
2435    
2436          redo A;          redo A;
2437        } else {        } else {
2438          !!!cp (154);          !!!cp (154);
2439          !!!parse-error (type => 'dash in comment');          !!!parse-error (type => 'dash in comment',
2440          $self->{current_token}->{data} .= '--' . chr ($self->{next_char}); # comment                          line => $self->{line_prev},
2441                            column => $self->{column_prev});
2442            $self->{ct}->{data} .= '--' . chr ($self->{nc}); # comment
2443          $self->{state} = COMMENT_STATE;          $self->{state} = COMMENT_STATE;
2444          !!!next-input-character;          !!!next-input-character;
2445          redo A;          redo A;
2446        }        }
2447      } elsif ($self->{state} == DOCTYPE_STATE) {      } elsif ($self->{state} == DOCTYPE_STATE) {
2448        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
2449          !!!cp (155);          !!!cp (155);
2450          $self->{state} = BEFORE_DOCTYPE_NAME_STATE;          $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
2451          !!!next-input-character;          !!!next-input-character;
# Line 1691  sub _get_next_token ($) { Line 2458  sub _get_next_token ($) {
2458          redo A;          redo A;
2459        }        }
2460      } elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) {      } elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) {
2461        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
2462          !!!cp (157);          !!!cp (157);
2463          ## Stay in the state          ## Stay in the state
2464          !!!next-input-character;          !!!next-input-character;
2465          redo A;          redo A;
2466        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2467          !!!cp (158);          !!!cp (158);
2468          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
2469          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2470          !!!next-input-character;          !!!next-input-character;
2471    
2472          !!!emit ($self->{current_token}); # DOCTYPE (quirks)          !!!emit ($self->{ct}); # DOCTYPE (quirks)
2473    
2474          redo A;          redo A;
2475        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2476          !!!cp (159);          !!!cp (159);
2477          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
2478          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2479          ## reconsume          ## reconsume
2480    
2481          !!!emit ($self->{current_token}); # DOCTYPE (quirks)          !!!emit ($self->{ct}); # DOCTYPE (quirks)
2482    
2483          redo A;          redo A;
2484        } else {        } else {
2485          !!!cp (160);          !!!cp (160);
2486          $self->{current_token}->{name} = chr $self->{next_char};          $self->{ct}->{name} = chr $self->{nc};
2487          delete $self->{current_token}->{quirks};          delete $self->{ct}->{quirks};
2488  ## ISSUE: "Set the token's name name to the" in the spec  ## ISSUE: "Set the token's name name to the" in the spec
2489          $self->{state} = DOCTYPE_NAME_STATE;          $self->{state} = DOCTYPE_NAME_STATE;
2490          !!!next-input-character;          !!!next-input-character;
# Line 1729  sub _get_next_token ($) { Line 2492  sub _get_next_token ($) {
2492        }        }
2493      } elsif ($self->{state} == DOCTYPE_NAME_STATE) {      } elsif ($self->{state} == DOCTYPE_NAME_STATE) {
2494  ## ISSUE: Redundant "First," in the spec.  ## ISSUE: Redundant "First," in the spec.
2495        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
2496          !!!cp (161);          !!!cp (161);
2497          $self->{state} = AFTER_DOCTYPE_NAME_STATE;          $self->{state} = AFTER_DOCTYPE_NAME_STATE;
2498          !!!next-input-character;          !!!next-input-character;
2499          redo A;          redo A;
2500        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2501          !!!cp (162);          !!!cp (162);
2502          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2503          !!!next-input-character;          !!!next-input-character;
2504    
2505          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2506    
2507          redo A;          redo A;
2508        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2509          !!!cp (163);          !!!cp (163);
2510          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2511          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2512          ## reconsume          ## reconsume
2513    
2514          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2515          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2516    
2517          redo A;          redo A;
2518        } else {        } else {
2519          !!!cp (164);          !!!cp (164);
2520          $self->{current_token}->{name}          $self->{ct}->{name}
2521            .= chr ($self->{next_char}); # DOCTYPE            .= chr ($self->{nc}); # DOCTYPE
2522          ## Stay in the state          ## Stay in the state
2523          !!!next-input-character;          !!!next-input-character;
2524          redo A;          redo A;
2525        }        }
2526      } elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) {      } elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) {
2527        if ($self->{next_char} == 0x0009 or # HT        if ($is_space->{$self->{nc}}) {
           $self->{next_char} == 0x000A or # LF  
           $self->{next_char} == 0x000B or # VT  
           $self->{next_char} == 0x000C or # FF  
           $self->{next_char} == 0x0020) { # SP  
2528          !!!cp (165);          !!!cp (165);
2529          ## Stay in the state          ## Stay in the state
2530          !!!next-input-character;          !!!next-input-character;
2531          redo A;          redo A;
2532        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2533          !!!cp (166);          !!!cp (166);
2534          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2535          !!!next-input-character;          !!!next-input-character;
2536    
2537          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2538    
2539          redo A;          redo A;
2540        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2541          !!!cp (167);          !!!cp (167);
2542          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2543          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2544          ## reconsume          ## reconsume
2545    
2546          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2547          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2548    
2549          redo A;          redo A;
2550        } elsif ($self->{next_char} == 0x0050 or # P        } elsif ($self->{nc} == 0x0050 or # P
2551                 $self->{next_char} == 0x0070) { # p                 $self->{nc} == 0x0070) { # p
2552            $self->{state} = PUBLIC_STATE;
2553            $self->{s_kwd} = chr $self->{nc};
2554          !!!next-input-character;          !!!next-input-character;
2555          if ($self->{next_char} == 0x0055 or # U          redo A;
2556              $self->{next_char} == 0x0075) { # u        } elsif ($self->{nc} == 0x0053 or # S
2557            !!!next-input-character;                 $self->{nc} == 0x0073) { # s
2558            if ($self->{next_char} == 0x0042 or # B          $self->{state} = SYSTEM_STATE;
2559                $self->{next_char} == 0x0062) { # b          $self->{s_kwd} = chr $self->{nc};
             !!!next-input-character;  
             if ($self->{next_char} == 0x004C or # L  
                 $self->{next_char} == 0x006C) { # l  
               !!!next-input-character;  
               if ($self->{next_char} == 0x0049 or # I  
                   $self->{next_char} == 0x0069) { # i  
                 !!!next-input-character;  
                 if ($self->{next_char} == 0x0043 or # C  
                     $self->{next_char} == 0x0063) { # c  
                   !!!cp (168);  
                   $self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE;  
                   !!!next-input-character;  
                   redo A;  
                 } else {  
                   !!!cp (169);  
                 }  
               } else {  
                 !!!cp (170);  
               }  
             } else {  
               !!!cp (171);  
             }  
           } else {  
             !!!cp (172);  
           }  
         } else {  
           !!!cp (173);  
         }  
   
         #  
       } elsif ($self->{next_char} == 0x0053 or # S  
                $self->{next_char} == 0x0073) { # s  
2560          !!!next-input-character;          !!!next-input-character;
2561          if ($self->{next_char} == 0x0059 or # Y          redo A;
             $self->{next_char} == 0x0079) { # y  
           !!!next-input-character;  
           if ($self->{next_char} == 0x0053 or # S  
               $self->{next_char} == 0x0073) { # s  
             !!!next-input-character;  
             if ($self->{next_char} == 0x0054 or # T  
                 $self->{next_char} == 0x0074) { # t  
               !!!next-input-character;  
               if ($self->{next_char} == 0x0045 or # E  
                   $self->{next_char} == 0x0065) { # e  
                 !!!next-input-character;  
                 if ($self->{next_char} == 0x004D or # M  
                     $self->{next_char} == 0x006D) { # m  
                   !!!cp (174);  
                   $self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE;  
                   !!!next-input-character;  
                   redo A;  
                 } else {  
                   !!!cp (175);  
                 }  
               } else {  
                 !!!cp (176);  
               }  
             } else {  
               !!!cp (177);  
             }  
           } else {  
             !!!cp (178);  
           }  
         } else {  
           !!!cp (179);  
         }  
   
         #  
2562        } else {        } else {
2563          !!!cp (180);          !!!cp (180);
2564            !!!parse-error (type => 'string after DOCTYPE name');
2565            $self->{ct}->{quirks} = 1;
2566    
2567            $self->{state} = BOGUS_DOCTYPE_STATE;
2568          !!!next-input-character;          !!!next-input-character;
2569          #          redo A;
2570        }        }
2571        } elsif ($self->{state} == PUBLIC_STATE) {
2572          ## ASCII case-insensitive
2573          if ($self->{nc} == [
2574                undef,
2575                0x0055, # U
2576                0x0042, # B
2577                0x004C, # L
2578                0x0049, # I
2579              ]->[length $self->{s_kwd}] or
2580              $self->{nc} == [
2581                undef,
2582                0x0075, # u
2583                0x0062, # b
2584                0x006C, # l
2585                0x0069, # i
2586              ]->[length $self->{s_kwd}]) {
2587            !!!cp (175);
2588            ## Stay in the state.
2589            $self->{s_kwd} .= chr $self->{nc};
2590            !!!next-input-character;
2591            redo A;
2592          } elsif ((length $self->{s_kwd}) == 5 and
2593                   ($self->{nc} == 0x0043 or # C
2594                    $self->{nc} == 0x0063)) { # c
2595            !!!cp (168);
2596            $self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
2597            !!!next-input-character;
2598            redo A;
2599          } else {
2600            !!!cp (169);
2601            !!!parse-error (type => 'string after DOCTYPE name',
2602                            line => $self->{line_prev},
2603                            column => $self->{column_prev} + 1 - length $self->{s_kwd});
2604            $self->{ct}->{quirks} = 1;
2605    
2606        !!!parse-error (type => 'string after DOCTYPE name');          $self->{state} = BOGUS_DOCTYPE_STATE;
2607        $self->{current_token}->{quirks} = 1;          ## Reconsume.
2608            redo A;
2609          }
2610        } elsif ($self->{state} == SYSTEM_STATE) {
2611          ## ASCII case-insensitive
2612          if ($self->{nc} == [
2613                undef,
2614                0x0059, # Y
2615                0x0053, # S
2616                0x0054, # T
2617                0x0045, # E
2618              ]->[length $self->{s_kwd}] or
2619              $self->{nc} == [
2620                undef,
2621                0x0079, # y
2622                0x0073, # s
2623                0x0074, # t
2624                0x0065, # e
2625              ]->[length $self->{s_kwd}]) {
2626            !!!cp (170);
2627            ## Stay in the state.
2628            $self->{s_kwd} .= chr $self->{nc};
2629            !!!next-input-character;
2630            redo A;
2631          } elsif ((length $self->{s_kwd}) == 5 and
2632                   ($self->{nc} == 0x004D or # M
2633                    $self->{nc} == 0x006D)) { # m
2634            !!!cp (171);
2635            $self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
2636            !!!next-input-character;
2637            redo A;
2638          } else {
2639            !!!cp (172);
2640            !!!parse-error (type => 'string after DOCTYPE name',
2641                            line => $self->{line_prev},
2642                            column => $self->{column_prev} + 1 - length $self->{s_kwd});
2643            $self->{ct}->{quirks} = 1;
2644    
2645        $self->{state} = BOGUS_DOCTYPE_STATE;          $self->{state} = BOGUS_DOCTYPE_STATE;
2646        # next-input-character is already done          ## Reconsume.
2647        redo A;          redo A;
2648          }
2649      } elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {      } elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
2650        if ({        if ($is_space->{$self->{nc}}) {
             0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,  
             #0x000D => 1, # HT, LF, VT, FF, SP, CR  
           }->{$self->{next_char}}) {  
2651          !!!cp (181);          !!!cp (181);
2652          ## Stay in the state          ## Stay in the state
2653          !!!next-input-character;          !!!next-input-character;
2654          redo A;          redo A;
2655        } elsif ($self->{next_char} eq 0x0022) { # "        } elsif ($self->{nc} eq 0x0022) { # "
2656          !!!cp (182);          !!!cp (182);
2657          $self->{current_token}->{public_identifier} = ''; # DOCTYPE          $self->{ct}->{pubid} = ''; # DOCTYPE
2658          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE;          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE;
2659          !!!next-input-character;          !!!next-input-character;
2660          redo A;          redo A;
2661        } elsif ($self->{next_char} eq 0x0027) { # '        } elsif ($self->{nc} eq 0x0027) { # '
2662          !!!cp (183);          !!!cp (183);
2663          $self->{current_token}->{public_identifier} = ''; # DOCTYPE          $self->{ct}->{pubid} = ''; # DOCTYPE
2664          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE;          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE;
2665          !!!next-input-character;          !!!next-input-character;
2666          redo A;          redo A;
2667        } elsif ($self->{next_char} eq 0x003E) { # >        } elsif ($self->{nc} eq 0x003E) { # >
2668          !!!cp (184);          !!!cp (184);
2669          !!!parse-error (type => 'no PUBLIC literal');          !!!parse-error (type => 'no PUBLIC literal');
2670    
2671          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2672          !!!next-input-character;          !!!next-input-character;
2673    
2674          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2675          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2676    
2677          redo A;          redo A;
2678        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2679          !!!cp (185);          !!!cp (185);
2680          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2681    
2682          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2683          ## reconsume          ## reconsume
2684    
2685          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2686          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2687    
2688          redo A;          redo A;
2689        } else {        } else {
2690          !!!cp (186);          !!!cp (186);
2691          !!!parse-error (type => 'string after PUBLIC');          !!!parse-error (type => 'string after PUBLIC');
2692          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2693    
2694          $self->{state} = BOGUS_DOCTYPE_STATE;          $self->{state} = BOGUS_DOCTYPE_STATE;
2695          !!!next-input-character;          !!!next-input-character;
2696          redo A;          redo A;
2697        }        }
2698      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) {      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) {
2699        if ($self->{next_char} == 0x0022) { # "        if ($self->{nc} == 0x0022) { # "
2700          !!!cp (187);          !!!cp (187);
2701          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
2702          !!!next-input-character;          !!!next-input-character;
2703          redo A;          redo A;
2704        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2705          !!!cp (188);          !!!cp (188);
2706          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
2707    
2708          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2709          !!!next-input-character;          !!!next-input-character;
2710    
2711          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2712          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2713    
2714          redo A;          redo A;
2715        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2716          !!!cp (189);          !!!cp (189);
2717          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
2718    
2719          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2720          ## reconsume          ## reconsume
2721    
2722          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2723          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2724    
2725          redo A;          redo A;
2726        } else {        } else {
2727          !!!cp (190);          !!!cp (190);
2728          $self->{current_token}->{public_identifier} # DOCTYPE          $self->{ct}->{pubid} # DOCTYPE
2729              .= chr $self->{next_char};              .= chr $self->{nc};
2730            $self->{read_until}->($self->{ct}->{pubid}, q[">],
2731                                  length $self->{ct}->{pubid});
2732    
2733          ## Stay in the state          ## Stay in the state
2734          !!!next-input-character;          !!!next-input-character;
2735          redo A;          redo A;
2736        }        }
2737      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) {      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) {
2738        if ($self->{next_char} == 0x0027) { # '        if ($self->{nc} == 0x0027) { # '
2739          !!!cp (191);          !!!cp (191);
2740          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
2741          !!!next-input-character;          !!!next-input-character;
2742          redo A;          redo A;
2743        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2744          !!!cp (192);          !!!cp (192);
2745          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
2746    
2747          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2748          !!!next-input-character;          !!!next-input-character;
2749    
2750          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2751          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2752    
2753          redo A;          redo A;
2754        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2755          !!!cp (193);          !!!cp (193);
2756          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
2757    
2758          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2759          ## reconsume          ## reconsume
2760    
2761          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2762          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2763    
2764          redo A;          redo A;
2765        } else {        } else {
2766          !!!cp (194);          !!!cp (194);
2767          $self->{current_token}->{public_identifier} # DOCTYPE          $self->{ct}->{pubid} # DOCTYPE
2768              .= chr $self->{next_char};              .= chr $self->{nc};
2769            $self->{read_until}->($self->{ct}->{pubid}, q['>],
2770                                  length $self->{ct}->{pubid});
2771    
2772          ## Stay in the state          ## Stay in the state
2773          !!!next-input-character;          !!!next-input-character;
2774          redo A;          redo A;
2775        }        }
2776      } elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {      } elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
2777        if ({        if ($is_space->{$self->{nc}}) {
             0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,  
             #0x000D => 1, # HT, LF, VT, FF, SP, CR  
           }->{$self->{next_char}}) {  
2778          !!!cp (195);          !!!cp (195);
2779          ## Stay in the state          ## Stay in the state
2780          !!!next-input-character;          !!!next-input-character;
2781          redo A;          redo A;
2782        } elsif ($self->{next_char} == 0x0022) { # "        } elsif ($self->{nc} == 0x0022) { # "
2783          !!!cp (196);          !!!cp (196);
2784          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{ct}->{sysid} = ''; # DOCTYPE
2785          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
2786          !!!next-input-character;          !!!next-input-character;
2787          redo A;          redo A;
2788        } elsif ($self->{next_char} == 0x0027) { # '        } elsif ($self->{nc} == 0x0027) { # '
2789          !!!cp (197);          !!!cp (197);
2790          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{ct}->{sysid} = ''; # DOCTYPE
2791          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
2792          !!!next-input-character;          !!!next-input-character;
2793          redo A;          redo A;
2794        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2795          !!!cp (198);          !!!cp (198);
2796          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2797          !!!next-input-character;          !!!next-input-character;
2798    
2799          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2800    
2801          redo A;          redo A;
2802        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2803          !!!cp (199);          !!!cp (199);
2804          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2805    
2806          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2807          ## reconsume          ## reconsume
2808    
2809          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2810          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2811    
2812          redo A;          redo A;
2813        } else {        } else {
2814          !!!cp (200);          !!!cp (200);
2815          !!!parse-error (type => 'string after PUBLIC literal');          !!!parse-error (type => 'string after PUBLIC literal');
2816          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2817    
2818          $self->{state} = BOGUS_DOCTYPE_STATE;          $self->{state} = BOGUS_DOCTYPE_STATE;
2819          !!!next-input-character;          !!!next-input-character;
2820          redo A;          redo A;
2821        }        }
2822      } elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {      } elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
2823        if ({        if ($is_space->{$self->{nc}}) {
             0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,  
             #0x000D => 1, # HT, LF, VT, FF, SP, CR  
           }->{$self->{next_char}}) {  
2824          !!!cp (201);          !!!cp (201);
2825          ## Stay in the state          ## Stay in the state
2826          !!!next-input-character;          !!!next-input-character;
2827          redo A;          redo A;
2828        } elsif ($self->{next_char} == 0x0022) { # "        } elsif ($self->{nc} == 0x0022) { # "
2829          !!!cp (202);          !!!cp (202);
2830          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{ct}->{sysid} = ''; # DOCTYPE
2831          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
2832          !!!next-input-character;          !!!next-input-character;
2833          redo A;          redo A;
2834        } elsif ($self->{next_char} == 0x0027) { # '        } elsif ($self->{nc} == 0x0027) { # '
2835          !!!cp (203);          !!!cp (203);
2836          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{ct}->{sysid} = ''; # DOCTYPE
2837          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
2838          !!!next-input-character;          !!!next-input-character;
2839          redo A;          redo A;
2840        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2841          !!!cp (204);          !!!cp (204);
2842          !!!parse-error (type => 'no SYSTEM literal');          !!!parse-error (type => 'no SYSTEM literal');
2843          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2844          !!!next-input-character;          !!!next-input-character;
2845    
2846          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2847          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2848    
2849          redo A;          redo A;
2850        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2851          !!!cp (205);          !!!cp (205);
2852          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2853    
2854          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2855          ## reconsume          ## reconsume
2856    
2857          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2858          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2859    
2860          redo A;          redo A;
2861        } else {        } else {
2862          !!!cp (206);          !!!cp (206);
2863          !!!parse-error (type => 'string after SYSTEM');          !!!parse-error (type => 'string after SYSTEM');
2864          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2865    
2866          $self->{state} = BOGUS_DOCTYPE_STATE;          $self->{state} = BOGUS_DOCTYPE_STATE;
2867          !!!next-input-character;          !!!next-input-character;
2868          redo A;          redo A;
2869        }        }
2870      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) {      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) {
2871        if ($self->{next_char} == 0x0022) { # "        if ($self->{nc} == 0x0022) { # "
2872          !!!cp (207);          !!!cp (207);
2873          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
2874          !!!next-input-character;          !!!next-input-character;
2875          redo A;          redo A;
2876        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2877          !!!cp (208);          !!!cp (208);
2878          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
2879    
2880          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2881          !!!next-input-character;          !!!next-input-character;
2882    
2883          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2884          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2885    
2886          redo A;          redo A;
2887        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2888          !!!cp (209);          !!!cp (209);
2889          !!!parse-error (type => 'unclosed SYSTEM literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
2890    
2891          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2892          ## reconsume          ## reconsume
2893    
2894          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2895          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2896    
2897          redo A;          redo A;
2898        } else {        } else {
2899          !!!cp (210);          !!!cp (210);
2900          $self->{current_token}->{system_identifier} # DOCTYPE          $self->{ct}->{sysid} # DOCTYPE
2901              .= chr $self->{next_char};              .= chr $self->{nc};
2902            $self->{read_until}->($self->{ct}->{sysid}, q[">],
2903                                  length $self->{ct}->{sysid});
2904    
2905          ## Stay in the state          ## Stay in the state
2906          !!!next-input-character;          !!!next-input-character;
2907          redo A;          redo A;
2908        }        }
2909      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) {      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) {
2910        if ($self->{next_char} == 0x0027) { # '        if ($self->{nc} == 0x0027) { # '
2911          !!!cp (211);          !!!cp (211);
2912          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
2913          !!!next-input-character;          !!!next-input-character;
2914          redo A;          redo A;
2915        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2916          !!!cp (212);          !!!cp (212);
2917          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
2918    
2919          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2920          !!!next-input-character;          !!!next-input-character;
2921    
2922          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2923          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2924    
2925          redo A;          redo A;
2926        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2927          !!!cp (213);          !!!cp (213);
2928          !!!parse-error (type => 'unclosed SYSTEM literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
2929    
2930          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2931          ## reconsume          ## reconsume
2932    
2933          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2934          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2935    
2936          redo A;          redo A;
2937        } else {        } else {
2938          !!!cp (214);          !!!cp (214);
2939          $self->{current_token}->{system_identifier} # DOCTYPE          $self->{ct}->{sysid} # DOCTYPE
2940              .= chr $self->{next_char};              .= chr $self->{nc};
2941            $self->{read_until}->($self->{ct}->{sysid}, q['>],
2942                                  length $self->{ct}->{sysid});
2943    
2944          ## Stay in the state          ## Stay in the state
2945          !!!next-input-character;          !!!next-input-character;
2946          redo A;          redo A;
2947        }        }
2948      } elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {      } elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
2949        if ({        if ($is_space->{$self->{nc}}) {
             0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,  
             #0x000D => 1, # HT, LF, VT, FF, SP, CR  
           }->{$self->{next_char}}) {  
2950          !!!cp (215);          !!!cp (215);
2951          ## Stay in the state          ## Stay in the state
2952          !!!next-input-character;          !!!next-input-character;
2953          redo A;          redo A;
2954        } elsif ($self->{next_char} == 0x003E) { # >        } elsif ($self->{nc} == 0x003E) { # >
2955          !!!cp (216);          !!!cp (216);
2956          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2957          !!!next-input-character;          !!!next-input-character;
2958    
2959          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2960    
2961          redo A;          redo A;
2962        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2963          !!!cp (217);          !!!cp (217);
2964          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
   
2965          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2966          ## reconsume          ## reconsume
2967    
2968          $self->{current_token}->{quirks} = 1;          $self->{ct}->{quirks} = 1;
2969          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2970    
2971          redo A;          redo A;
2972        } else {        } else {
2973          !!!cp (218);          !!!cp (218);
2974          !!!parse-error (type => 'string after SYSTEM literal');          !!!parse-error (type => 'string after SYSTEM literal');
2975          #$self->{current_token}->{quirks} = 1;          #$self->{ct}->{quirks} = 1;
2976    
2977          $self->{state} = BOGUS_DOCTYPE_STATE;          $self->{state} = BOGUS_DOCTYPE_STATE;
2978          !!!next-input-character;          !!!next-input-character;
2979          redo A;          redo A;
2980        }        }
2981      } elsif ($self->{state} == BOGUS_DOCTYPE_STATE) {      } elsif ($self->{state} == BOGUS_DOCTYPE_STATE) {
2982        if ($self->{next_char} == 0x003E) { # >        if ($self->{nc} == 0x003E) { # >
2983          !!!cp (219);          !!!cp (219);
2984          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2985          !!!next-input-character;          !!!next-input-character;
2986    
2987          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2988    
2989          redo A;          redo A;
2990        } elsif ($self->{next_char} == -1) {        } elsif ($self->{nc} == -1) {
2991          !!!cp (220);          !!!cp (220);
2992          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2993          $self->{state} = DATA_STATE;          $self->{state} = DATA_STATE;
2994          ## reconsume          ## reconsume
2995    
2996          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{ct}); # DOCTYPE
2997    
2998          redo A;          redo A;
2999        } else {        } else {
3000          !!!cp (221);          !!!cp (221);
3001            my $s = '';
3002            $self->{read_until}->($s, q[>], 0);
3003    
3004          ## Stay in the state          ## Stay in the state
3005          !!!next-input-character;          !!!next-input-character;
3006          redo A;          redo A;
3007        }        }
3008      } else {      } elsif ($self->{state} == CDATA_SECTION_STATE) {
3009        die "$0: $self->{state}: Unknown state";        ## NOTE: "CDATA section state" in the state is jointly implemented
3010      }        ## by three states, |CDATA_SECTION_STATE|, |CDATA_SECTION_MSE1_STATE|,
3011    } # A          ## and |CDATA_SECTION_MSE2_STATE|.
3012          
3013          if ($self->{nc} == 0x005D) { # ]
3014            !!!cp (221.1);
3015            $self->{state} = CDATA_SECTION_MSE1_STATE;
3016            !!!next-input-character;
3017            redo A;
3018          } elsif ($self->{nc} == -1) {
3019            $self->{state} = DATA_STATE;
3020            !!!next-input-character;
3021            if (length $self->{ct}->{data}) { # character
3022              !!!cp (221.2);
3023              !!!emit ($self->{ct}); # character
3024            } else {
3025              !!!cp (221.3);
3026              ## No token to emit. $self->{ct} is discarded.
3027            }        
3028            redo A;
3029          } else {
3030            !!!cp (221.4);
3031            $self->{ct}->{data} .= chr $self->{nc};
3032            $self->{read_until}->($self->{ct}->{data},
3033                                  q<]>,
3034                                  length $self->{ct}->{data});
3035    
3036    die "$0: _get_next_token: unexpected case";          ## Stay in the state.
3037  } # _get_next_token          !!!next-input-character;
3038            redo A;
3039          }
3040    
3041  sub _tokenize_attempt_to_consume_an_entity ($$$) {        ## ISSUE: "text tokens" in spec.
3042    my ($self, $in_attr, $additional) = @_;      } elsif ($self->{state} == CDATA_SECTION_MSE1_STATE) {
3043          if ($self->{nc} == 0x005D) { # ]
3044            !!!cp (221.5);
3045            $self->{state} = CDATA_SECTION_MSE2_STATE;
3046            !!!next-input-character;
3047            redo A;
3048          } else {
3049            !!!cp (221.6);
3050            $self->{ct}->{data} .= ']';
3051            $self->{state} = CDATA_SECTION_STATE;
3052            ## Reconsume.
3053            redo A;
3054          }
3055        } elsif ($self->{state} == CDATA_SECTION_MSE2_STATE) {
3056          if ($self->{nc} == 0x003E) { # >
3057            $self->{state} = DATA_STATE;
3058            !!!next-input-character;
3059            if (length $self->{ct}->{data}) { # character
3060              !!!cp (221.7);
3061              !!!emit ($self->{ct}); # character
3062            } else {
3063              !!!cp (221.8);
3064              ## No token to emit. $self->{ct} is discarded.
3065            }
3066            redo A;
3067          } elsif ($self->{nc} == 0x005D) { # ]
3068            !!!cp (221.9); # character
3069            $self->{ct}->{data} .= ']'; ## Add first "]" of "]]]".
3070            ## Stay in the state.
3071            !!!next-input-character;
3072            redo A;
3073          } else {
3074            !!!cp (221.11);
3075            $self->{ct}->{data} .= ']]'; # character
3076            $self->{state} = CDATA_SECTION_STATE;
3077            ## Reconsume.
3078            redo A;
3079          }
3080        } elsif ($self->{state} == ENTITY_STATE) {
3081          if ($is_space->{$self->{nc}} or
3082              {
3083                0x003C => 1, 0x0026 => 1, -1 => 1, # <, &
3084                $self->{entity_add} => 1,
3085              }->{$self->{nc}}) {
3086            !!!cp (1001);
3087            ## Don't consume
3088            ## No error
3089            ## Return nothing.
3090            #
3091          } elsif ($self->{nc} == 0x0023) { # #
3092            !!!cp (999);
3093            $self->{state} = ENTITY_HASH_STATE;
3094            $self->{s_kwd} = '#';
3095            !!!next-input-character;
3096            redo A;
3097          } elsif ((0x0041 <= $self->{nc} and
3098                    $self->{nc} <= 0x005A) or # A..Z
3099                   (0x0061 <= $self->{nc} and
3100                    $self->{nc} <= 0x007A)) { # a..z
3101            !!!cp (998);
3102            require Whatpm::_NamedEntityList;
3103            $self->{state} = ENTITY_NAME_STATE;
3104            $self->{s_kwd} = chr $self->{nc};
3105            $self->{entity__value} = $self->{s_kwd};
3106            $self->{entity__match} = 0;
3107            !!!next-input-character;
3108            redo A;
3109          } else {
3110            !!!cp (1027);
3111            !!!parse-error (type => 'bare ero');
3112            ## Return nothing.
3113            #
3114          }
3115    
3116    my ($l, $c) = ($self->{line_prev}, $self->{column_prev});        ## NOTE: No character is consumed by the "consume a character
3117          ## reference" algorithm.  In other word, there is an "&" character
3118          ## that does not introduce a character reference, which would be
3119          ## appended to the parent element or the attribute value in later
3120          ## process of the tokenizer.
3121    
3122          if ($self->{prev_state} == DATA_STATE) {
3123            !!!cp (997);
3124            $self->{state} = $self->{prev_state};
3125            ## Reconsume.
3126            !!!emit ({type => CHARACTER_TOKEN, data => '&',
3127                      line => $self->{line_prev},
3128                      column => $self->{column_prev},
3129                     });
3130            redo A;
3131          } else {
3132            !!!cp (996);
3133            $self->{ca}->{value} .= '&';
3134            $self->{state} = $self->{prev_state};
3135            ## Reconsume.
3136            redo A;
3137          }
3138        } elsif ($self->{state} == ENTITY_HASH_STATE) {
3139          if ($self->{nc} == 0x0078 or # x
3140              $self->{nc} == 0x0058) { # X
3141            !!!cp (995);
3142            $self->{state} = HEXREF_X_STATE;
3143            $self->{s_kwd} .= chr $self->{nc};
3144            !!!next-input-character;
3145            redo A;
3146          } elsif (0x0030 <= $self->{nc} and
3147                   $self->{nc} <= 0x0039) { # 0..9
3148            !!!cp (994);
3149            $self->{state} = NCR_NUM_STATE;
3150            $self->{s_kwd} = $self->{nc} - 0x0030;
3151            !!!next-input-character;
3152            redo A;
3153          } else {
3154            !!!parse-error (type => 'bare nero',
3155                            line => $self->{line_prev},
3156                            column => $self->{column_prev} - 1);
3157    
3158    if ({          ## NOTE: According to the spec algorithm, nothing is returned,
3159         0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF,          ## and then "&#" is appended to the parent element or the attribute
3160         0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR          ## value in the later processing.
3161         $additional => 1,  
3162        }->{$self->{next_char}}) {          if ($self->{prev_state} == DATA_STATE) {
3163      !!!cp (1001);            !!!cp (1019);
3164      ## Don't consume            $self->{state} = $self->{prev_state};
3165      ## No error            ## Reconsume.
3166      return undef;            !!!emit ({type => CHARACTER_TOKEN,
3167    } elsif ($self->{next_char} == 0x0023) { # #                      data => '&#',
3168      !!!next-input-character;                      line => $self->{line_prev},
3169      if ($self->{next_char} == 0x0078 or # x                      column => $self->{column_prev} - 1,
3170          $self->{next_char} == 0x0058) { # X                     });
3171        my $code;            redo A;
       X: {  
         my $x_char = $self->{next_char};  
         !!!next-input-character;  
         if (0x0030 <= $self->{next_char} and  
             $self->{next_char} <= 0x0039) { # 0..9  
           !!!cp (1002);  
           $code ||= 0;  
           $code *= 0x10;  
           $code += $self->{next_char} - 0x0030;  
           redo X;  
         } elsif (0x0061 <= $self->{next_char} and  
                  $self->{next_char} <= 0x0066) { # a..f  
           !!!cp (1003);  
           $code ||= 0;  
           $code *= 0x10;  
           $code += $self->{next_char} - 0x0060 + 9;  
           redo X;  
         } elsif (0x0041 <= $self->{next_char} and  
                  $self->{next_char} <= 0x0046) { # A..F  
           !!!cp (1004);  
           $code ||= 0;  
           $code *= 0x10;  
           $code += $self->{next_char} - 0x0040 + 9;  
           redo X;  
         } elsif (not defined $code) { # no hexadecimal digit  
           !!!cp (1005);  
           !!!parse-error (type => 'bare hcro', line => $l, column => $c);  
           !!!back-next-input-character ($x_char, $self->{next_char});  
           $self->{next_char} = 0x0023; # #  
           return undef;  
         } elsif ($self->{next_char} == 0x003B) { # ;  
           !!!cp (1006);  
           !!!next-input-character;  
3172          } else {          } else {
3173            !!!cp (1007);            !!!cp (993);
3174            !!!parse-error (type => 'no refc', line => $l, column => $c);            $self->{ca}->{value} .= '&#';
3175              $self->{state} = $self->{prev_state};
3176              ## Reconsume.
3177              redo A;
3178          }          }
3179          }
3180          if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) {      } elsif ($self->{state} == NCR_NUM_STATE) {
3181            !!!cp (1008);        if (0x0030 <= $self->{nc} and
3182            !!!parse-error (type => (sprintf 'invalid character reference:U+%04X', $code), line => $l, column => $c);            $self->{nc} <= 0x0039) { # 0..9
           $code = 0xFFFD;  
         } elsif ($code > 0x10FFFF) {  
           !!!cp (1009);  
           !!!parse-error (type => (sprintf 'invalid character reference:U-%08X', $code), line => $l, column => $c);  
           $code = 0xFFFD;  
         } elsif ($code == 0x000D) {  
           !!!cp (1010);  
           !!!parse-error (type => 'CR character reference', line => $l, column => $c);  
           $code = 0x000A;  
         } elsif (0x80 <= $code and $code <= 0x9F) {  
           !!!cp (1011);  
           !!!parse-error (type => (sprintf 'C1 character reference:U+%04X', $code), line => $l, column => $c);  
           $code = $c1_entity_char->{$code};  
         }  
   
         return {type => CHARACTER_TOKEN, data => chr $code,  
                 has_reference => 1, line => $l, column => $c};  
       } # X  
     } elsif (0x0030 <= $self->{next_char} and  
              $self->{next_char} <= 0x0039) { # 0..9  
       my $code = $self->{next_char} - 0x0030;  
       !!!next-input-character;  
         
       while (0x0030 <= $self->{next_char} and  
                 $self->{next_char} <= 0x0039) { # 0..9  
3183          !!!cp (1012);          !!!cp (1012);
3184          $code *= 10;          $self->{s_kwd} *= 10;
3185          $code += $self->{next_char} - 0x0030;          $self->{s_kwd} += $self->{nc} - 0x0030;
3186                    
3187            ## Stay in the state.
3188          !!!next-input-character;          !!!next-input-character;
3189        }          redo A;
3190          } elsif ($self->{nc} == 0x003B) { # ;
       if ($self->{next_char} == 0x003B) { # ;  
3191          !!!cp (1013);          !!!cp (1013);
3192          !!!next-input-character;          !!!next-input-character;
3193            #
3194        } else {        } else {
3195          !!!cp (1014);          !!!cp (1014);
3196          !!!parse-error (type => 'no refc', line => $l, column => $c);          !!!parse-error (type => 'no refc');
3197            ## Reconsume.
3198            #
3199        }        }
3200    
3201        if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) {        my $code = $self->{s_kwd};
3202          my $l = $self->{line_prev};
3203          my $c = $self->{column_prev};
3204          if ($charref_map->{$code}) {
3205          !!!cp (1015);          !!!cp (1015);
3206          !!!parse-error (type => (sprintf 'invalid character reference:U+%04X', $code), line => $l, column => $c);          !!!parse-error (type => 'invalid character reference',
3207          $code = 0xFFFD;                          text => (sprintf 'U+%04X', $code),
3208                            line => $l, column => $c);
3209            $code = $charref_map->{$code};
3210        } elsif ($code > 0x10FFFF) {        } elsif ($code > 0x10FFFF) {
3211          !!!cp (1016);          !!!cp (1016);
3212          !!!parse-error (type => (sprintf 'invalid character reference:U-%08X', $code), line => $l, column => $c);          !!!parse-error (type => 'invalid character reference',
3213                            text => (sprintf 'U-%08X', $code),
3214                            line => $l, column => $c);
3215          $code = 0xFFFD;          $code = 0xFFFD;
       } elsif ($code == 0x000D) {  
         !!!cp (1017);  
         !!!parse-error (type => 'CR character reference', line => $l, column => $c);  
         $code = 0x000A;  
       } elsif (0x80 <= $code and $code <= 0x9F) {  
         !!!cp (1018);  
         !!!parse-error (type => (sprintf 'C1 character reference:U+%04X', $code), line => $l, column => $c);  
         $code = $c1_entity_char->{$code};  
3216        }        }
3217          
3218        return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1,        if ($self->{prev_state} == DATA_STATE) {
3219                line => $l, column => $c};          !!!cp (992);
3220      } else {          $self->{state} = $self->{prev_state};
3221        !!!cp (1019);          ## Reconsume.
3222        !!!parse-error (type => 'bare nero', line => $l, column => $c);          !!!emit ({type => CHARACTER_TOKEN, data => chr $code,
3223        !!!back-next-input-character ($self->{next_char});                    line => $l, column => $c,
3224        $self->{next_char} = 0x0023; # #                   });
3225        return undef;          redo A;
3226      }        } else {
3227    } elsif ((0x0041 <= $self->{next_char} and          !!!cp (991);
3228              $self->{next_char} <= 0x005A) or          $self->{ca}->{value} .= chr $code;
3229             (0x0061 <= $self->{next_char} and          $self->{ca}->{has_reference} = 1;
3230              $self->{next_char} <= 0x007A)) {          $self->{state} = $self->{prev_state};
3231      my $entity_name = chr $self->{next_char};          ## Reconsume.
3232      !!!next-input-character;          redo A;
3233          }
3234      my $value = $entity_name;      } elsif ($self->{state} == HEXREF_X_STATE) {
3235      my $match = 0;        if ((0x0030 <= $self->{nc} and $self->{nc} <= 0x0039) or
3236      require Whatpm::_NamedEntityList;            (0x0041 <= $self->{nc} and $self->{nc} <= 0x0046) or
3237      our $EntityChar;            (0x0061 <= $self->{nc} and $self->{nc} <= 0x0066)) {
3238            # 0..9, A..F, a..f
3239      while (length $entity_name < 10 and          !!!cp (990);
3240             ## NOTE: Some number greater than the maximum length of entity name          $self->{state} = HEXREF_HEX_STATE;
3241             ((0x0041 <= $self->{next_char} and # a          $self->{s_kwd} = 0;
3242               $self->{next_char} <= 0x005A) or # x          ## Reconsume.
3243              (0x0061 <= $self->{next_char} and # a          redo A;
3244               $self->{next_char} <= 0x007A) or # z        } else {
3245              (0x0030 <= $self->{next_char} and # 0          !!!parse-error (type => 'bare hcro',
3246               $self->{next_char} <= 0x0039) or # 9                          line => $self->{line_prev},
3247              $self->{next_char} == 0x003B)) { # ;                          column => $self->{column_prev} - 2);
3248        $entity_name .= chr $self->{next_char};  
3249        if (defined $EntityChar->{$entity_name}) {          ## NOTE: According to the spec algorithm, nothing is returned,
3250          if ($self->{next_char} == 0x003B) { # ;          ## and then "&#" followed by "X" or "x" is appended to the parent
3251            !!!cp (1020);          ## element or the attribute value in the later processing.
3252            $value = $EntityChar->{$entity_name};  
3253            $match = 1;          if ($self->{prev_state} == DATA_STATE) {
3254            !!!next-input-character;            !!!cp (1005);
3255            last;            $self->{state} = $self->{prev_state};
3256              ## Reconsume.
3257              !!!emit ({type => CHARACTER_TOKEN,
3258                        data => '&' . $self->{s_kwd},
3259                        line => $self->{line_prev},
3260                        column => $self->{column_prev} - length $self->{s_kwd},
3261                       });
3262              redo A;
3263          } else {          } else {
3264            !!!cp (1021);            !!!cp (989);
3265            $value = $EntityChar->{$entity_name};            $self->{ca}->{value} .= '&' . $self->{s_kwd};
3266            $match = -1;            $self->{state} = $self->{prev_state};
3267              ## Reconsume.
3268              redo A;
3269            }
3270          }
3271        } elsif ($self->{state} == HEXREF_HEX_STATE) {
3272          if (0x0030 <= $self->{nc} and $self->{nc} <= 0x0039) {
3273            # 0..9
3274            !!!cp (1002);
3275            $self->{s_kwd} *= 0x10;
3276            $self->{s_kwd} += $self->{nc} - 0x0030;
3277            ## Stay in the state.
3278            !!!next-input-character;
3279            redo A;
3280          } elsif (0x0061 <= $self->{nc} and
3281                   $self->{nc} <= 0x0066) { # a..f
3282            !!!cp (1003);
3283            $self->{s_kwd} *= 0x10;
3284            $self->{s_kwd} += $self->{nc} - 0x0060 + 9;
3285            ## Stay in the state.
3286            !!!next-input-character;
3287            redo A;
3288          } elsif (0x0041 <= $self->{nc} and
3289                   $self->{nc} <= 0x0046) { # A..F
3290            !!!cp (1004);
3291            $self->{s_kwd} *= 0x10;
3292            $self->{s_kwd} += $self->{nc} - 0x0040 + 9;
3293            ## Stay in the state.
3294            !!!next-input-character;
3295            redo A;
3296          } elsif ($self->{nc} == 0x003B) { # ;
3297            !!!cp (1006);
3298            !!!next-input-character;
3299            #
3300          } else {
3301            !!!cp (1007);
3302            !!!parse-error (type => 'no refc',
3303                            line => $self->{line},
3304                            column => $self->{column});
3305            ## Reconsume.
3306            #
3307          }
3308    
3309          my $code = $self->{s_kwd};
3310          my $l = $self->{line_prev};
3311          my $c = $self->{column_prev};
3312          if ($charref_map->{$code}) {
3313            !!!cp (1008);
3314            !!!parse-error (type => 'invalid character reference',
3315                            text => (sprintf 'U+%04X', $code),
3316                            line => $l, column => $c);
3317            $code = $charref_map->{$code};
3318          } elsif ($code > 0x10FFFF) {
3319            !!!cp (1009);
3320            !!!parse-error (type => 'invalid character reference',
3321                            text => (sprintf 'U-%08X', $code),
3322                            line => $l, column => $c);
3323            $code = 0xFFFD;
3324          }
3325    
3326          if ($self->{prev_state} == DATA_STATE) {
3327            !!!cp (988);
3328            $self->{state} = $self->{prev_state};
3329            ## Reconsume.
3330            !!!emit ({type => CHARACTER_TOKEN, data => chr $code,
3331                      line => $l, column => $c,
3332                     });
3333            redo A;
3334          } else {
3335            !!!cp (987);
3336            $self->{ca}->{value} .= chr $code;
3337            $self->{ca}->{has_reference} = 1;
3338            $self->{state} = $self->{prev_state};
3339            ## Reconsume.
3340            redo A;
3341          }
3342        } elsif ($self->{state} == ENTITY_NAME_STATE) {
3343          if (length $self->{s_kwd} < 30 and
3344              ## NOTE: Some number greater than the maximum length of entity name
3345              ((0x0041 <= $self->{nc} and # a
3346                $self->{nc} <= 0x005A) or # x
3347               (0x0061 <= $self->{nc} and # a
3348                $self->{nc} <= 0x007A) or # z
3349               (0x0030 <= $self->{nc} and # 0
3350                $self->{nc} <= 0x0039) or # 9
3351               $self->{nc} == 0x003B)) { # ;
3352            our $EntityChar;
3353            $self->{s_kwd} .= chr $self->{nc};
3354            if (defined $EntityChar->{$self->{s_kwd}}) {
3355              if ($self->{nc} == 0x003B) { # ;
3356                !!!cp (1020);
3357                $self->{entity__value} = $EntityChar->{$self->{s_kwd}};
3358                $self->{entity__match} = 1;
3359                !!!next-input-character;
3360                #
3361              } else {
3362                !!!cp (1021);
3363                $self->{entity__value} = $EntityChar->{$self->{s_kwd}};
3364                $self->{entity__match} = -1;
3365                ## Stay in the state.
3366                !!!next-input-character;
3367                redo A;
3368              }
3369            } else {
3370              !!!cp (1022);
3371              $self->{entity__value} .= chr $self->{nc};
3372              $self->{entity__match} *= 2;
3373              ## Stay in the state.
3374            !!!next-input-character;            !!!next-input-character;
3375              redo A;
3376            }
3377          }
3378    
3379          my $data;
3380          my $has_ref;
3381          if ($self->{entity__match} > 0) {
3382            !!!cp (1023);
3383            $data = $self->{entity__value};
3384            $has_ref = 1;
3385            #
3386          } elsif ($self->{entity__match} < 0) {
3387            !!!parse-error (type => 'no refc');
3388            if ($self->{prev_state} != DATA_STATE and # in attribute
3389                $self->{entity__match} < -1) {
3390              !!!cp (1024);
3391              $data = '&' . $self->{s_kwd};
3392              #
3393            } else {
3394              !!!cp (1025);
3395              $data = $self->{entity__value};
3396              $has_ref = 1;
3397              #
3398          }          }
3399        } else {        } else {
3400          !!!cp (1022);          !!!cp (1026);
3401          $value .= chr $self->{next_char};          !!!parse-error (type => 'bare ero',
3402          $match *= 2;                          line => $self->{line_prev},
3403          !!!next-input-character;                          column => $self->{column_prev} - length $self->{s_kwd});
3404            $data = '&' . $self->{s_kwd};
3405            #
3406        }        }
3407      }    
3408              ## NOTE: In these cases, when a character reference is found,
3409      if ($match > 0) {        ## it is consumed and a character token is returned, or, otherwise,
3410        !!!cp (1023);        ## nothing is consumed and returned, according to the spec algorithm.
3411        return {type => CHARACTER_TOKEN, data => $value, has_reference => 1,        ## In this implementation, anything that has been examined by the
3412                line => $l, column => $c};        ## tokenizer is appended to the parent element or the attribute value
3413      } elsif ($match < 0) {        ## as string, either literal string when no character reference or
3414        !!!parse-error (type => 'no refc', line => $l, column => $c);        ## entity-replaced string otherwise, in this stage, since any characters
3415        if ($in_attr and $match < -1) {        ## that would not be consumed are appended in the data state or in an
3416          !!!cp (1024);        ## appropriate attribute value state anyway.
3417          return {type => CHARACTER_TOKEN, data => '&'.$entity_name,  
3418                  line => $l, column => $c};        if ($self->{prev_state} == DATA_STATE) {
3419        } else {          !!!cp (986);
3420          !!!cp (1025);          $self->{state} = $self->{prev_state};
3421          return {type => CHARACTER_TOKEN, data => $value, has_reference => 1,          ## Reconsume.
3422                  line => $l, column => $c};          !!!emit ({type => CHARACTER_TOKEN,
3423                      data => $data,
3424                      line => $self->{line_prev},
3425                      column => $self->{column_prev} + 1 - length $self->{s_kwd},
3426                     });
3427            redo A;
3428          } else {
3429            !!!cp (985);
3430            $self->{ca}->{value} .= $data;
3431            $self->{ca}->{has_reference} = 1 if $has_ref;
3432            $self->{state} = $self->{prev_state};
3433            ## Reconsume.
3434            redo A;
3435        }        }
3436      } else {      } else {
3437        !!!cp (1026);        die "$0: $self->{state}: Unknown state";
       !!!parse-error (type => 'bare ero', line => $l, column => $c);  
       ## NOTE: "No characters are consumed" in the spec.  
       return {type => CHARACTER_TOKEN, data => '&'.$value,  
               line => $l, column => $c};  
3438      }      }
3439    } else {    } # A  
3440      !!!cp (1027);  
3441      ## no characters are consumed    die "$0: _get_next_token: unexpected case";
3442      !!!parse-error (type => 'bare ero', line => $l, column => $c);  } # _get_next_token
     return undef;  
   }  
 } # _tokenize_attempt_to_consume_an_entity  
3443    
3444  sub _initialize_tree_constructor ($) {  sub _initialize_tree_constructor ($) {
3445    my $self = shift;    my $self = shift;
# Line 2452  sub _initialize_tree_constructor ($) { Line 3448  sub _initialize_tree_constructor ($) {
3448    ## TODO: Turn mutation events off # MUST    ## TODO: Turn mutation events off # MUST
3449    ## TODO: Turn loose Document option (manakai extension) on    ## TODO: Turn loose Document option (manakai extension) on
3450    $self->{document}->manakai_is_html (1); # MUST    $self->{document}->manakai_is_html (1); # MUST
3451      $self->{document}->set_user_data (manakai_source_line => 1);
3452      $self->{document}->set_user_data (manakai_source_column => 1);
3453  } # _initialize_tree_constructor  } # _initialize_tree_constructor
3454    
3455  sub _terminate_tree_constructor ($) {  sub _terminate_tree_constructor ($) {
# Line 2506  sub _tree_construction_initial ($) { Line 3504  sub _tree_construction_initial ($) {
3504        ## language.        ## language.
3505        my $doctype_name = $token->{name};        my $doctype_name = $token->{name};
3506        $doctype_name = '' unless defined $doctype_name;        $doctype_name = '' unless defined $doctype_name;
3507        $doctype_name =~ tr/a-z/A-Z/;        $doctype_name =~ tr/a-z/A-Z/; # ASCII case-insensitive
3508        if (not defined $token->{name} or # <!DOCTYPE>        if (not defined $token->{name} or # <!DOCTYPE>
3509            defined $token->{public_identifier} or            defined $token->{sysid}) {
           defined $token->{system_identifier}) {  
3510          !!!cp ('t1');          !!!cp ('t1');
3511          !!!parse-error (type => 'not HTML5');          !!!parse-error (type => 'not HTML5', token => $token);
3512        } elsif ($doctype_name ne 'HTML') {        } elsif ($doctype_name ne 'HTML') {
3513          !!!cp ('t2');          !!!cp ('t2');
3514          ## ISSUE: ASCII case-insensitive? (in fact it does not matter)          !!!parse-error (type => 'not HTML5', token => $token);
3515          !!!parse-error (type => 'not HTML5');        } elsif (defined $token->{pubid}) {
3516            if ($token->{pubid} eq 'XSLT-compat') {
3517              !!!cp ('t1.2');
3518              !!!parse-error (type => 'XSLT-compat', token => $token,
3519                              level => $self->{level}->{should});
3520            } else {
3521              !!!parse-error (type => 'not HTML5', token => $token);
3522            }
3523        } else {        } else {
3524          !!!cp ('t3');          !!!cp ('t3');
3525            #
3526        }        }
3527                
3528        my $doctype = $self->{document}->create_document_type_definition        my $doctype = $self->{document}->create_document_type_definition
3529          ($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)?          ($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)?
3530        $doctype->public_id ($token->{public_identifier})        ## NOTE: Default value for both |public_id| and |system_id| attributes
3531            if defined $token->{public_identifier};        ## are empty strings, so that we don't set any value in missing cases.
3532        $doctype->system_id ($token->{system_identifier})        $doctype->public_id ($token->{pubid}) if defined $token->{pubid};
3533            if defined $token->{system_identifier};        $doctype->system_id ($token->{sysid}) if defined $token->{sysid};
3534        ## NOTE: Other DocumentType attributes are null or empty lists.        ## NOTE: Other DocumentType attributes are null or empty lists.
3535        ## ISSUE: internalSubset = null??        ## ISSUE: internalSubset = null??
3536        $self->{document}->append_child ($doctype);        $self->{document}->append_child ($doctype);
# Line 2533  sub _tree_construction_initial ($) { Line 3538  sub _tree_construction_initial ($) {
3538        if ($token->{quirks} or $doctype_name ne 'HTML') {        if ($token->{quirks} or $doctype_name ne 'HTML') {
3539          !!!cp ('t4');          !!!cp ('t4');
3540          $self->{document}->manakai_compat_mode ('quirks');          $self->{document}->manakai_compat_mode ('quirks');
3541        } elsif (defined $token->{public_identifier}) {        } elsif (defined $token->{pubid}) {
3542          my $pubid = $token->{public_identifier};          my $pubid = $token->{pubid};
3543          $pubid =~ tr/a-z/A-z/;          $pubid =~ tr/a-z/A-z/;
3544          if ({          my $prefix = [
3545            "+//SILMARIL//DTD HTML PRO V0R11 19970101//EN" => 1,            "+//SILMARIL//DTD HTML PRO V0R11 19970101//",
3546            "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1,            "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//",
3547            "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1,            "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//",
3548            "-//IETF//DTD HTML 2.0 LEVEL 1//EN" => 1,            "-//IETF//DTD HTML 2.0 LEVEL 1//",
3549            "-//IETF//DTD HTML 2.0 LEVEL 2//EN" => 1,            "-//IETF//DTD HTML 2.0 LEVEL 2//",
3550            "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//EN" => 1,            "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//",
3551            "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//EN" => 1,            "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//",
3552            "-//IETF//DTD HTML 2.0 STRICT//EN" => 1,            "-//IETF//DTD HTML 2.0 STRICT//",
3553            "-//IETF//DTD HTML 2.0//EN" => 1,            "-//IETF//DTD HTML 2.0//",
3554            "-//IETF//DTD HTML 2.1E//EN" => 1,            "-//IETF//DTD HTML 2.1E//",
3555            "-//IETF//DTD HTML 3.0//EN" => 1,            "-//IETF//DTD HTML 3.0//",
3556            "-//IETF//DTD HTML 3.0//EN//" => 1,            "-//IETF//DTD HTML 3.2 FINAL//",
3557            "-//IETF//DTD HTML 3.2 FINAL//EN" => 1,            "-//IETF//DTD HTML 3.2//",
3558            "-//IETF//DTD HTML 3.2//EN" => 1,            "-//IETF//DTD HTML 3//",
3559            "-//IETF//DTD HTML 3//EN" => 1,            "-//IETF//DTD HTML LEVEL 0//",
3560            "-//IETF//DTD HTML LEVEL 0//EN" => 1,            "-//IETF//DTD HTML LEVEL 1//",
3561            "-//IETF//DTD HTML LEVEL 0//EN//2.0" => 1,            "-//IETF//DTD HTML LEVEL 2//",
3562            "-//IETF//DTD HTML LEVEL 1//EN" => 1,            "-//IETF//DTD HTML LEVEL 3//",
3563            "-//IETF//DTD HTML LEVEL 1//EN//2.0" => 1,            "-//IETF//DTD HTML STRICT LEVEL 0//",
3564            "-//IETF//DTD HTML LEVEL 2//EN" => 1,            "-//IETF//DTD HTML STRICT LEVEL 1//",
3565            "-//IETF//DTD HTML LEVEL 2//EN//2.0" => 1,            "-//IETF//DTD HTML STRICT LEVEL 2//",
3566            "-//IETF//DTD HTML LEVEL 3//EN" => 1,            "-//IETF//DTD HTML STRICT LEVEL 3//",
3567            "-//IETF//DTD HTML LEVEL 3//EN//3.0" => 1,            "-//IETF//DTD HTML STRICT//",
3568            "-//IETF//DTD HTML STRICT LEVEL 0//EN" => 1,            "-//IETF//DTD HTML//",
3569            "-//IETF//DTD HTML STRICT LEVEL 0//EN//2.0" => 1,            "-//METRIUS//DTD METRIUS PRESENTATIONAL//",
3570            "-//IETF//DTD HTML STRICT LEVEL 1//EN" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//",
3571            "-//IETF//DTD HTML STRICT LEVEL 1//EN//2.0" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//",
3572            "-//IETF//DTD HTML STRICT LEVEL 2//EN" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//",
3573            "-//IETF//DTD HTML STRICT LEVEL 2//EN//2.0" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//",
3574            "-//IETF//DTD HTML STRICT LEVEL 3//EN" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//",
3575            "-//IETF//DTD HTML STRICT LEVEL 3//EN//3.0" => 1,            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//",
3576            "-//IETF//DTD HTML STRICT//EN" => 1,            "-//NETSCAPE COMM. CORP.//DTD HTML//",
3577            "-//IETF//DTD HTML STRICT//EN//2.0" => 1,            "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//",
3578            "-//IETF//DTD HTML STRICT//EN//3.0" => 1,            "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//",
3579            "-//IETF//DTD HTML//EN" => 1,            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//",
3580            "-//IETF//DTD HTML//EN//2.0" => 1,            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//",
3581            "-//IETF//DTD HTML//EN//3.0" => 1,            "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//",
3582            "-//METRIUS//DTD METRIUS PRESENTATIONAL//EN" => 1,            "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//",
3583            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//EN" => 1,            "-//SPYGLASS//DTD HTML 2.0 EXTENDED//",
3584            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//EN" => 1,            "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//",
3585            "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//EN" => 1,            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//",
3586            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//EN" => 1,            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//",
3587            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//EN" => 1,            "-//W3C//DTD HTML 3 1995-03-24//",
3588            "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//EN" => 1,            "-//W3C//DTD HTML 3.2 DRAFT//",
3589            "-//NETSCAPE COMM. CORP.//DTD HTML//EN" => 1,            "-//W3C//DTD HTML 3.2 FINAL//",
3590            "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1,            "-//W3C//DTD HTML 3.2//",
3591            "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1,            "-//W3C//DTD HTML 3.2S DRAFT//",
3592            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1,            "-//W3C//DTD HTML 4.0 FRAMESET//",
3593            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//EN" => 1,            "-//W3C//DTD HTML 4.0 TRANSITIONAL//",
3594            "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//EN" => 1,            "-//W3C//DTD HTML EXPERIMETNAL 19960712//",
3595            "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//EN" => 1,            "-//W3C//DTD HTML EXPERIMENTAL 970421//",
3596            "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1,            "-//W3C//DTD W3 HTML//",
3597            "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1,            "-//W3O//DTD W3 HTML 3.0//",
3598            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1,            "-//WEBTECHS//DTD MOZILLA HTML 2.0//",
3599            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//EN" => 1,            "-//WEBTECHS//DTD MOZILLA HTML//",
3600            "-//W3C//DTD HTML 3 1995-03-24//EN" => 1,          ]; # $prefix
3601            "-//W3C//DTD HTML 3.2 DRAFT//EN" => 1,          my $match;
3602            "-//W3C//DTD HTML 3.2 FINAL//EN" => 1,          for (@$prefix) {
3603            "-//W3C//DTD HTML 3.2//EN" => 1,            if (substr ($prefix, 0, length $_) eq $_) {
3604            "-//W3C//DTD HTML 3.2S DRAFT//EN" => 1,              $match = 1;
3605            "-//W3C//DTD HTML 4.0 FRAMESET//EN" => 1,              last;
3606            "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN" => 1,            }
3607            "-//W3C//DTD HTML EXPERIMETNAL 19960712//EN" => 1,          }
3608            "-//W3C//DTD HTML EXPERIMENTAL 970421//EN" => 1,          if ($match or
3609            "-//W3C//DTD W3 HTML//EN" => 1,              $pubid eq "-//W3O//DTD W3 HTML STRICT 3.0//EN//" or
3610            "-//W3O//DTD W3 HTML 3.0//EN" => 1,              $pubid eq "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" or
3611            "-//W3O//DTD W3 HTML 3.0//EN//" => 1,              $pubid eq "HTML") {
           "-//W3O//DTD W3 HTML STRICT 3.0//EN//" => 1,  
           "-//WEBTECHS//DTD MOZILLA HTML 2.0//EN" => 1,  
           "-//WEBTECHS//DTD MOZILLA HTML//EN" => 1,  
           "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" => 1,  
           "HTML" => 1,