/[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 1.55 by wakaba, Sat Aug 11 06:53:38 2007 UTC revision 1.72 by wakaba, Sun Mar 2 14:32:26 2008 UTC
# Line 1  Line 1 
1  package Whatpm::HTML;  package Whatpm::HTML;
2  use strict;  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);
5    
6  ## ISSUE:  ## ISSUE:
7  ## var doc = implementation.createDocument (null, null, null);  ## var doc = implementation.createDocument (null, null, null);
8  ## doc.write ('');  ## doc.write ('');
9  ## alert (doc.compatMode);  ## alert (doc.compatMode);
10    
11  ## ISSUE: HTML5 revision 967 says that the encoding layer MUST NOT  ## TODO: Control charcters and noncharacters are not allowed (HTML5 revision 1263)
12  ## strip BOM and the HTML layer MUST ignore it.  Whether we can do it  ## TODO: 1252 parse error (revision 1264)
13  ## is not yet clear.  ## TODO: 8859-11 = 874 (revision 1271)
 ## "{U+FEFF}..." in UTF-16BE/UTF-16LE is three or four characters?  
 ## "{U+FEFF}..." in GB18030?  
14    
15  my $permitted_slash_tag_name = {  my $permitted_slash_tag_name = {
16    base => 1,    base => 1,
# Line 19  my $permitted_slash_tag_name = { Line 18  my $permitted_slash_tag_name = {
18    meta => 1,    meta => 1,
19    hr => 1,    hr => 1,
20    br => 1,    br => 1,
21    img=> 1,    img => 1,
22    embed => 1,    embed => 1,
23    param => 1,    param => 1,
24    area => 1,    area => 1,
# Line 84  my $formatting_category = { Line 83  my $formatting_category = {
83  };  };
84  # $phrasing_category: all other elements  # $phrasing_category: all other elements
85    
86    sub parse_byte_string ($$$$;$) {
87      my $self = ref $_[0] ? shift : shift->new;
88      my $charset = shift;
89      my $bytes_s = ref $_[0] ? $_[0] : \($_[0]);
90      my $s;
91      
92      if (defined $charset) {
93        require Encode; ## TODO: decode(utf8) don't delete BOM
94        $s = \ (Encode::decode ($charset, $$bytes_s));
95        $self->{input_encoding} = lc $charset; ## TODO: normalize name
96        $self->{confident} = 1;
97      } else {
98        ## TODO: Implement HTML5 detection algorithm
99        require Whatpm::Charset::UniversalCharDet;
100        $charset = Whatpm::Charset::UniversalCharDet->detect_byte_string
101            (substr ($$bytes_s, 0, 1024));
102        $charset ||= 'windows-1252';
103        $s = \ (Encode::decode ($charset, $$bytes_s));
104        $self->{input_encoding} = $charset;
105        $self->{confident} = 0;
106      }
107    
108      $self->{change_encoding} = sub {
109        my $self = shift;
110        my $charset = lc shift;
111        ## TODO: if $charset is supported
112        ## TODO: normalize charset name
113    
114        ## "Change the encoding" algorithm:
115    
116        ## Step 1    
117        if ($charset eq 'utf-16') { ## ISSUE: UTF-16BE -> UTF-8? UTF-16LE -> UTF-8?
118          $charset = 'utf-8';
119        }
120    
121        ## Step 2
122        if (defined $self->{input_encoding} and
123            $self->{input_encoding} eq $charset) {
124          $self->{confident} = 1;
125          return;
126        }
127    
128        !!!parse-error (type => 'charset label detected:'.$self->{input_encoding}.
129            ':'.$charset, level => 'w');
130    
131        ## Step 3
132        # if (can) {
133          ## change the encoding on the fly.
134          #$self->{confident} = 1;
135          #return;
136        # }
137    
138        ## Step 4
139        throw Whatpm::HTML::RestartParser (charset => $charset);
140      }; # $self->{change_encoding}
141    
142      my @args = @_; shift @args; # $s
143      my $return;
144      try {
145        $return = $self->parse_char_string ($s, @args);  
146      } catch Whatpm::HTML::RestartParser with {
147        my $charset = shift->{charset};
148        $s = \ (Encode::decode ($charset, $$bytes_s));    
149        $self->{input_encoding} = $charset; ## TODO: normalize
150        $self->{confident} = 1;
151        $return = $self->parse_char_string ($s, @args);
152      };
153      return $return;
154    } # parse_byte_string
155    
156    ## NOTE: HTML5 spec says that the encoding layer MUST NOT strip BOM
157    ## and the HTML layer MUST ignore it.  However, we does strip BOM in
158    ## the encoding layer and the HTML layer does not ignore any U+FEFF,
159    ## because the core part of our HTML parser expects a string of character,
160    ## not a string of bytes or code units or anything which might contain a BOM.
161    ## Therefore, any parser interface that accepts a string of bytes,
162    ## such as |parse_byte_string| in this module, must ensure that it does
163    ## strip the BOM and never strip any ZWNBSP.
164    
165    *parse_char_string = \&parse_string;
166    
167  sub parse_string ($$$;$) {  sub parse_string ($$$;$) {
168    my $self = shift->new;    my $self = ref $_[0] ? shift : shift->new;
169    my $s = \$_[0];    my $s = ref $_[0] ? $_[0] : \($_[0]);
170    $self->{document} = $_[1];    $self->{document} = $_[1];
171      @{$self->{document}->child_nodes} = ();
172    
173    ## NOTE: |set_inner_html| copies most of this method's code    ## NOTE: |set_inner_html| copies most of this method's code
174    
175      $self->{confident} = 1 unless exists $self->{confident};
176      $self->{document}->input_encoding ($self->{input_encoding})
177          if defined $self->{input_encoding};
178    
179    my $i = 0;    my $i = 0;
180    my $line = 1;    my $line = 1;
181    my $column = 0;    my $column = 0;
# Line 147  sub new ($) { Line 232  sub new ($) {
232    $self->{parse_error} = sub {    $self->{parse_error} = sub {
233      #      #
234    };    };
235      $self->{change_encoding} = sub {
236        # if ($_[0] is a supported encoding) {
237        #   run "change the encoding" algorithm;
238        #   throw Whatpm::HTML::RestartParser (charset => $new_encoding);
239        # }
240      };
241      $self->{application_cache_selection} = sub {
242        #
243      };
244    return $self;    return $self;
245  } # new  } # new
246    
# Line 159  sub CDATA_CONTENT_MODEL () { CM_LIMITED_ Line 253  sub CDATA_CONTENT_MODEL () { CM_LIMITED_
253  sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP }  sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP }
254  sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP }  sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP }
255    
256    sub DATA_STATE () { 0 }
257    sub ENTITY_DATA_STATE () { 1 }
258    sub TAG_OPEN_STATE () { 2 }
259    sub CLOSE_TAG_OPEN_STATE () { 3 }
260    sub TAG_NAME_STATE () { 4 }
261    sub BEFORE_ATTRIBUTE_NAME_STATE () { 5 }
262    sub ATTRIBUTE_NAME_STATE () { 6 }
263    sub AFTER_ATTRIBUTE_NAME_STATE () { 7 }
264    sub BEFORE_ATTRIBUTE_VALUE_STATE () { 8 }
265    sub ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE () { 9 }
266    sub ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE () { 10 }
267    sub ATTRIBUTE_VALUE_UNQUOTED_STATE () { 11 }
268    sub ENTITY_IN_ATTRIBUTE_VALUE_STATE () { 12 }
269    sub MARKUP_DECLARATION_OPEN_STATE () { 13 }
270    sub COMMENT_START_STATE () { 14 }
271    sub COMMENT_START_DASH_STATE () { 15 }
272    sub COMMENT_STATE () { 16 }
273    sub COMMENT_END_STATE () { 17 }
274    sub COMMENT_END_DASH_STATE () { 18 }
275    sub BOGUS_COMMENT_STATE () { 19 }
276    sub DOCTYPE_STATE () { 20 }
277    sub BEFORE_DOCTYPE_NAME_STATE () { 21 }
278    sub DOCTYPE_NAME_STATE () { 22 }
279    sub AFTER_DOCTYPE_NAME_STATE () { 23 }
280    sub BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 24 }
281    sub DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE () { 25 }
282    sub DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE () { 26 }
283    sub AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 27 }
284    sub BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 28 }
285    sub DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE () { 29 }
286    sub DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE () { 30 }
287    sub AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 31 }
288    sub BOGUS_DOCTYPE_STATE () { 32 }
289    sub AFTER_ATTRIBUTE_VALUE_QUOTED_STATE () { 33 }
290    
291  sub DOCTYPE_TOKEN () { 1 }  sub DOCTYPE_TOKEN () { 1 }
292  sub COMMENT_TOKEN () { 2 }  sub COMMENT_TOKEN () { 2 }
293  sub START_TAG_TOKEN () { 3 }  sub START_TAG_TOKEN () { 3 }
# Line 169  sub CHARACTER_TOKEN () { 6 } Line 298  sub CHARACTER_TOKEN () { 6 }
298  sub AFTER_HTML_IMS () { 0b100 }  sub AFTER_HTML_IMS () { 0b100 }
299  sub HEAD_IMS ()       { 0b1000 }  sub HEAD_IMS ()       { 0b1000 }
300  sub BODY_IMS ()       { 0b10000 }  sub BODY_IMS ()       { 0b10000 }
301  sub BODY_TABLE_IMS () { 0b100000 | BODY_IMS }  sub BODY_TABLE_IMS () { 0b100000 }
302  sub TABLE_IMS ()      { 0b1000000 }  sub TABLE_IMS ()      { 0b1000000 }
303  sub ROW_IMS ()        { 0b10000000 | TABLE_IMS }  sub ROW_IMS ()        { 0b10000000 }
304  sub BODY_AFTER_IMS () { 0b100000000 }  sub BODY_AFTER_IMS () { 0b100000000 }
305  sub FRAME_IMS ()      { 0b1000000000 }  sub FRAME_IMS ()      { 0b1000000000 }
306    
# Line 182  sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS | Line 311  sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS |
311  sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 }  sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 }
312  sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 }  sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 }
313  sub IN_BODY_IM () { BODY_IMS }  sub IN_BODY_IM () { BODY_IMS }
314  sub IN_CELL_IM () { BODY_TABLE_IMS | 0b01 }  sub IN_CELL_IM () { BODY_IMS | BODY_TABLE_IMS | 0b01 }
315  sub IN_CAPTION_IM () { BODY_TABLE_IMS | 0b10 }  sub IN_CAPTION_IM () { BODY_IMS | BODY_TABLE_IMS | 0b10 }
316  sub IN_ROW_IM () { ROW_IMS | 0b01 }  sub IN_ROW_IM () { TABLE_IMS | ROW_IMS | 0b01 }
317  sub IN_TABLE_BODY_IM () { ROW_IMS | 0b10 }  sub IN_TABLE_BODY_IM () { TABLE_IMS | ROW_IMS | 0b10 }
318  sub IN_TABLE_IM () { TABLE_IMS }  sub IN_TABLE_IM () { TABLE_IMS }
319  sub AFTER_BODY_IM () { BODY_AFTER_IMS }  sub AFTER_BODY_IM () { BODY_AFTER_IMS }
320  sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 }  sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 }
# Line 197  sub IN_COLUMN_GROUP_IM () { 0b10 } Line 326  sub IN_COLUMN_GROUP_IM () { 0b10 }
326    
327  sub _initialize_tokenizer ($) {  sub _initialize_tokenizer ($) {
328    my $self = shift;    my $self = shift;
329    $self->{state} = 'data'; # MUST    $self->{state} = DATA_STATE; # MUST
330    $self->{content_model} = PCDATA_CONTENT_MODEL; # be    $self->{content_model} = PCDATA_CONTENT_MODEL; # be
331    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE
332    undef $self->{current_attribute};    undef $self->{current_attribute};
# Line 219  sub _initialize_tokenizer ($) { Line 348  sub _initialize_tokenizer ($) {
348  ##   ->{system_identifier} (DOCTYPE_TOKEN)  ##   ->{system_identifier} (DOCTYPE_TOKEN)
349  ##   ->{correct} == 1 or 0 (DOCTYPE_TOKEN)  ##   ->{correct} == 1 or 0 (DOCTYPE_TOKEN)
350  ##   ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN)  ##   ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN)
351    ##        ->{name}
352    ##        ->{value}
353    ##        ->{has_reference} == 1 or 0
354  ##   ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN)  ##   ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN)
355    
356  ## Emitted token MUST immediately be handled by the tree construction state.  ## Emitted token MUST immediately be handled by the tree construction state.
# Line 229  sub _initialize_tokenizer ($) { Line 361  sub _initialize_tokenizer ($) {
361  ## has completed loading.  If one has, then it MUST be executed  ## has completed loading.  If one has, then it MUST be executed
362  ## and removed from the list.  ## and removed from the list.
363    
364    ## NOTE: HTML5 "Writing HTML documents" section, applied to
365    ## documents and not to user agents and conformance checkers,
366    ## contains some requirements that are not detected by the
367    ## parsing algorithm:
368    ## - Some requirements on character encoding declarations. ## TODO
369    ## - "Elements MUST NOT contain content that their content model disallows."
370    ##   ... Some are parse error, some are not (will be reported by c.c.).
371    ## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO
372    ## - Text (in elements, attributes, and comments) SHOULD NOT contain
373    ##   control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL?  Unicode control character?)
374    
375    ## TODO: HTML5 poses authors two SHOULD-level requirements that cannot
376    ## be detected by the HTML5 parsing algorithm:
377    ## - Text,
378    
379  sub _get_next_token ($) {  sub _get_next_token ($) {
380    my $self = shift;    my $self = shift;
381    if (@{$self->{token}}) {    if (@{$self->{token}}) {
# Line 236  sub _get_next_token ($) { Line 383  sub _get_next_token ($) {
383    }    }
384    
385    A: {    A: {
386      if ($self->{state} eq 'data') {      if ($self->{state} == DATA_STATE) {
387        if ($self->{next_input_character} == 0x0026) { # &        if ($self->{next_input_character} == 0x0026) { # &
388          if ($self->{content_model} & CM_ENTITY) { # PCDATA | RCDATA          if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA
389            $self->{state} = 'entity data';              not $self->{escape}) {
390              $self->{state} = ENTITY_DATA_STATE;
391            !!!next-input-character;            !!!next-input-character;
392            redo A;            redo A;
393          } else {          } else {
# Line 261  sub _get_next_token ($) { Line 409  sub _get_next_token ($) {
409          if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA          if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA
410              (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA              (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA
411               not $self->{escape})) {               not $self->{escape})) {
412            $self->{state} = 'tag open';            $self->{state} = TAG_OPEN_STATE;
413            !!!next-input-character;            !!!next-input-character;
414            redo A;            redo A;
415          } else {          } else {
# Line 290  sub _get_next_token ($) { Line 438  sub _get_next_token ($) {
438        !!!emit ($token);        !!!emit ($token);
439    
440        redo A;        redo A;
441      } elsif ($self->{state} eq 'entity data') {      } elsif ($self->{state} == ENTITY_DATA_STATE) {
442        ## (cannot happen in CDATA state)        ## (cannot happen in CDATA state)
443                
444        my $token = $self->_tokenize_attempt_to_consume_an_entity (0);        my $token = $self->_tokenize_attempt_to_consume_an_entity (0, -1);
445    
446        $self->{state} = 'data';        $self->{state} = DATA_STATE;
447        # next-input-character is already done        # next-input-character is already done
448    
449        unless (defined $token) {        unless (defined $token) {
# Line 305  sub _get_next_token ($) { Line 453  sub _get_next_token ($) {
453        }        }
454    
455        redo A;        redo A;
456      } elsif ($self->{state} eq 'tag open') {      } elsif ($self->{state} == TAG_OPEN_STATE) {
457        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
458          if ($self->{next_input_character} == 0x002F) { # /          if ($self->{next_input_character} == 0x002F) { # /
459            !!!next-input-character;            !!!next-input-character;
460            $self->{state} = 'close tag open';            $self->{state} = CLOSE_TAG_OPEN_STATE;
461            redo A;            redo A;
462          } else {          } else {
463            ## reconsume            ## reconsume
464            $self->{state} = 'data';            $self->{state} = DATA_STATE;
465    
466            !!!emit ({type => CHARACTER_TOKEN, data => '<'});            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
467    
# Line 321  sub _get_next_token ($) { Line 469  sub _get_next_token ($) {
469          }          }
470        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA
471          if ($self->{next_input_character} == 0x0021) { # !          if ($self->{next_input_character} == 0x0021) { # !
472            $self->{state} = 'markup declaration open';            $self->{state} = MARKUP_DECLARATION_OPEN_STATE;
473            !!!next-input-character;            !!!next-input-character;
474            redo A;            redo A;
475          } elsif ($self->{next_input_character} == 0x002F) { # /          } elsif ($self->{next_input_character} == 0x002F) { # /
476            $self->{state} = 'close tag open';            $self->{state} = CLOSE_TAG_OPEN_STATE;
477            !!!next-input-character;            !!!next-input-character;
478            redo A;            redo A;
479          } elsif (0x0041 <= $self->{next_input_character} and          } elsif (0x0041 <= $self->{next_input_character} and
# Line 333  sub _get_next_token ($) { Line 481  sub _get_next_token ($) {
481            $self->{current_token}            $self->{current_token}
482              = {type => START_TAG_TOKEN,              = {type => START_TAG_TOKEN,
483                 tag_name => chr ($self->{next_input_character} + 0x0020)};                 tag_name => chr ($self->{next_input_character} + 0x0020)};
484            $self->{state} = 'tag name';            $self->{state} = TAG_NAME_STATE;
485            !!!next-input-character;            !!!next-input-character;
486            redo A;            redo A;
487          } elsif (0x0061 <= $self->{next_input_character} and          } elsif (0x0061 <= $self->{next_input_character} and
488                   $self->{next_input_character} <= 0x007A) { # a..z                   $self->{next_input_character} <= 0x007A) { # a..z
489            $self->{current_token} = {type => START_TAG_TOKEN,            $self->{current_token} = {type => START_TAG_TOKEN,
490                              tag_name => chr ($self->{next_input_character})};                              tag_name => chr ($self->{next_input_character})};
491            $self->{state} = 'tag name';            $self->{state} = TAG_NAME_STATE;
492            !!!next-input-character;            !!!next-input-character;
493            redo A;            redo A;
494          } elsif ($self->{next_input_character} == 0x003E) { # >          } elsif ($self->{next_input_character} == 0x003E) { # >
495            !!!parse-error (type => 'empty start tag');            !!!parse-error (type => 'empty start tag');
496            $self->{state} = 'data';            $self->{state} = DATA_STATE;
497            !!!next-input-character;            !!!next-input-character;
498    
499            !!!emit ({type => CHARACTER_TOKEN, data => '<>'});            !!!emit ({type => CHARACTER_TOKEN, data => '<>'});
# Line 353  sub _get_next_token ($) { Line 501  sub _get_next_token ($) {
501            redo A;            redo A;
502          } elsif ($self->{next_input_character} == 0x003F) { # ?          } elsif ($self->{next_input_character} == 0x003F) { # ?
503            !!!parse-error (type => 'pio');            !!!parse-error (type => 'pio');
504            $self->{state} = 'bogus comment';            $self->{state} = BOGUS_COMMENT_STATE;
505            ## $self->{next_input_character} is intentionally left as is            ## $self->{next_input_character} is intentionally left as is
506            redo A;            redo A;
507          } else {          } else {
508            !!!parse-error (type => 'bare stago');            !!!parse-error (type => 'bare stago');
509            $self->{state} = 'data';            $self->{state} = DATA_STATE;
510            ## reconsume            ## reconsume
511    
512            !!!emit ({type => CHARACTER_TOKEN, data => '<'});            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
# Line 368  sub _get_next_token ($) { Line 516  sub _get_next_token ($) {
516        } else {        } else {
517          die "$0: $self->{content_model} in tag open";          die "$0: $self->{content_model} in tag open";
518        }        }
519      } elsif ($self->{state} eq 'close tag open') {      } elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) {
520        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
521          if (defined $self->{last_emitted_start_tag_name}) {          if (defined $self->{last_emitted_start_tag_name}) {
522            ## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>            ## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>
# Line 383  sub _get_next_token ($) { Line 531  sub _get_next_token ($) {
531              } else {              } else {
532                $self->{next_input_character} = shift @next_char; # reconsume                $self->{next_input_character} = shift @next_char; # reconsume
533                !!!back-next-input-character (@next_char);                !!!back-next-input-character (@next_char);
534                $self->{state} = 'data';                $self->{state} = DATA_STATE;
535    
536                !!!emit ({type => CHARACTER_TOKEN, data => '</'});                !!!emit ({type => CHARACTER_TOKEN, data => '</'});
537        
# Line 402  sub _get_next_token ($) { Line 550  sub _get_next_token ($) {
550                    $self->{next_input_character} == -1) {                    $self->{next_input_character} == -1) {
551              $self->{next_input_character} = shift @next_char; # reconsume              $self->{next_input_character} = shift @next_char; # reconsume
552              !!!back-next-input-character (@next_char);              !!!back-next-input-character (@next_char);
553              $self->{state} = 'data';              $self->{state} = DATA_STATE;
554              !!!emit ({type => CHARACTER_TOKEN, data => '</'});              !!!emit ({type => CHARACTER_TOKEN, data => '</'});
555              redo A;              redo A;
556            } else {            } else {
# Line 413  sub _get_next_token ($) { Line 561  sub _get_next_token ($) {
561          } else {          } else {
562            ## No start tag token has ever been emitted            ## No start tag token has ever been emitted
563            # next-input-character is already done            # next-input-character is already done
564            $self->{state} = 'data';            $self->{state} = DATA_STATE;
565            !!!emit ({type => CHARACTER_TOKEN, data => '</'});            !!!emit ({type => CHARACTER_TOKEN, data => '</'});
566            redo A;            redo A;
567          }          }
# Line 423  sub _get_next_token ($) { Line 571  sub _get_next_token ($) {
571            $self->{next_input_character} <= 0x005A) { # A..Z            $self->{next_input_character} <= 0x005A) { # A..Z
572          $self->{current_token} = {type => END_TAG_TOKEN,          $self->{current_token} = {type => END_TAG_TOKEN,
573                            tag_name => chr ($self->{next_input_character} + 0x0020)};                            tag_name => chr ($self->{next_input_character} + 0x0020)};
574          $self->{state} = 'tag name';          $self->{state} = TAG_NAME_STATE;
575          !!!next-input-character;          !!!next-input-character;
576          redo A;          redo A;
577        } elsif (0x0061 <= $self->{next_input_character} and        } elsif (0x0061 <= $self->{next_input_character} and
578                 $self->{next_input_character} <= 0x007A) { # a..z                 $self->{next_input_character} <= 0x007A) { # a..z
579          $self->{current_token} = {type => END_TAG_TOKEN,          $self->{current_token} = {type => END_TAG_TOKEN,
580                            tag_name => chr ($self->{next_input_character})};                            tag_name => chr ($self->{next_input_character})};
581          $self->{state} = 'tag name';          $self->{state} = TAG_NAME_STATE;
582          !!!next-input-character;          !!!next-input-character;
583          redo A;          redo A;
584        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
585          !!!parse-error (type => 'empty end tag');          !!!parse-error (type => 'empty end tag');
586          $self->{state} = 'data';          $self->{state} = DATA_STATE;
587          !!!next-input-character;          !!!next-input-character;
588          redo A;          redo A;
589        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
590          !!!parse-error (type => 'bare etago');          !!!parse-error (type => 'bare etago');
591          $self->{state} = 'data';          $self->{state} = DATA_STATE;
592          # reconsume          # reconsume
593    
594          !!!emit ({type => CHARACTER_TOKEN, data => '</'});          !!!emit ({type => CHARACTER_TOKEN, data => '</'});
# Line 448  sub _get_next_token ($) { Line 596  sub _get_next_token ($) {
596          redo A;          redo A;
597        } else {        } else {
598          !!!parse-error (type => 'bogus end tag');          !!!parse-error (type => 'bogus end tag');
599          $self->{state} = 'bogus comment';          $self->{state} = BOGUS_COMMENT_STATE;
600          ## $self->{next_input_character} is intentionally left as is          ## $self->{next_input_character} is intentionally left as is
601          redo A;          redo A;
602        }        }
603      } elsif ($self->{state} eq 'tag name') {      } elsif ($self->{state} == TAG_NAME_STATE) {
604        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
605            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
606            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
607            $self->{next_input_character} == 0x000C or # FF            $self->{next_input_character} == 0x000C or # FF
608            $self->{next_input_character} == 0x0020) { # SP            $self->{next_input_character} == 0x0020) { # SP
609          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
610          !!!next-input-character;          !!!next-input-character;
611          redo A;          redo A;
612        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
# Line 474  sub _get_next_token ($) { Line 622  sub _get_next_token ($) {
622          } else {          } else {
623            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
624          }          }
625          $self->{state} = 'data';          $self->{state} = DATA_STATE;
626          !!!next-input-character;          !!!next-input-character;
627    
628          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 501  sub _get_next_token ($) { Line 649  sub _get_next_token ($) {
649          } else {          } else {
650            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
651          }          }
652          $self->{state} = 'data';          $self->{state} = DATA_STATE;
653          # reconsume          # reconsume
654    
655          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 517  sub _get_next_token ($) { Line 665  sub _get_next_token ($) {
665          } else {          } else {
666            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
667          }          }
668          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
669          # next-input-character is already done          # next-input-character is already done
670          redo A;          redo A;
671        } else {        } else {
# Line 527  sub _get_next_token ($) { Line 675  sub _get_next_token ($) {
675          !!!next-input-character;          !!!next-input-character;
676          redo A;          redo A;
677        }        }
678      } elsif ($self->{state} eq 'before attribute name') {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) {
679        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
680            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
681            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
# Line 549  sub _get_next_token ($) { Line 697  sub _get_next_token ($) {
697          } else {          } else {
698            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
699          }          }
700          $self->{state} = 'data';          $self->{state} = DATA_STATE;
701          !!!next-input-character;          !!!next-input-character;
702    
703          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 559  sub _get_next_token ($) { Line 707  sub _get_next_token ($) {
707                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_input_character} <= 0x005A) { # A..Z
708          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),
709                                value => ''};                                value => ''};
710          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
711          !!!next-input-character;          !!!next-input-character;
712          redo A;          redo A;
713        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_input_character} == 0x002F) { # /
# Line 589  sub _get_next_token ($) { Line 737  sub _get_next_token ($) {
737          } else {          } else {
738            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
739          }          }
740          $self->{state} = 'data';          $self->{state} = DATA_STATE;
741          # reconsume          # reconsume
742    
743          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
744    
745          redo A;          redo A;
746        } else {        } else {
747            if ({
748                 0x0022 => 1, # "
749                 0x0027 => 1, # '
750                 0x003D => 1, # =
751                }->{$self->{next_input_character}}) {
752              !!!parse-error (type => 'bad attribute name');
753            }
754          $self->{current_attribute} = {name => chr ($self->{next_input_character}),          $self->{current_attribute} = {name => chr ($self->{next_input_character}),
755                                value => ''};                                value => ''};
756          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
757          !!!next-input-character;          !!!next-input-character;
758          redo A;          redo A;
759        }        }
760      } elsif ($self->{state} eq 'attribute name') {      } elsif ($self->{state} == ATTRIBUTE_NAME_STATE) {
761        my $before_leave = sub {        my $before_leave = sub {
762          if (exists $self->{current_token}->{attributes} # start tag or end tag          if (exists $self->{current_token}->{attributes} # start tag or end tag
763              ->{$self->{current_attribute}->{name}}) { # MUST              ->{$self->{current_attribute}->{name}}) { # MUST
# Line 620  sub _get_next_token ($) { Line 775  sub _get_next_token ($) {
775            $self->{next_input_character} == 0x000C or # FF            $self->{next_input_character} == 0x000C or # FF
776            $self->{next_input_character} == 0x0020) { # SP            $self->{next_input_character} == 0x0020) { # SP
777          $before_leave->();          $before_leave->();
778          $self->{state} = 'after attribute name';          $self->{state} = AFTER_ATTRIBUTE_NAME_STATE;
779          !!!next-input-character;          !!!next-input-character;
780          redo A;          redo A;
781        } elsif ($self->{next_input_character} == 0x003D) { # =        } elsif ($self->{next_input_character} == 0x003D) { # =
782          $before_leave->();          $before_leave->();
783          $self->{state} = 'before attribute value';          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
784          !!!next-input-character;          !!!next-input-character;
785          redo A;          redo A;
786        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
# Line 642  sub _get_next_token ($) { Line 797  sub _get_next_token ($) {
797          } else {          } else {
798            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
799          }          }
800          $self->{state} = 'data';          $self->{state} = DATA_STATE;
801          !!!next-input-character;          !!!next-input-character;
802    
803          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 665  sub _get_next_token ($) { Line 820  sub _get_next_token ($) {
820          } else {          } else {
821            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
822          }          }
823          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
824          # next-input-character is already done          # next-input-character is already done
825          redo A;          redo A;
826        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
# Line 683  sub _get_next_token ($) { Line 838  sub _get_next_token ($) {
838          } else {          } else {
839            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
840          }          }
841          $self->{state} = 'data';          $self->{state} = DATA_STATE;
842          # reconsume          # reconsume
843    
844          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
845    
846          redo A;          redo A;
847        } else {        } else {
848            if ($self->{next_input_character} == 0x0022 or # "
849                $self->{next_input_character} == 0x0027) { # '
850              !!!parse-error (type => 'bad attribute name');
851            }
852          $self->{current_attribute}->{name} .= chr ($self->{next_input_character});          $self->{current_attribute}->{name} .= chr ($self->{next_input_character});
853          ## Stay in the state          ## Stay in the state
854          !!!next-input-character;          !!!next-input-character;
855          redo A;          redo A;
856        }        }
857      } elsif ($self->{state} eq 'after attribute name') {      } elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) {
858        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
859            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
860            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
# Line 705  sub _get_next_token ($) { Line 864  sub _get_next_token ($) {
864          !!!next-input-character;          !!!next-input-character;
865          redo A;          redo A;
866        } elsif ($self->{next_input_character} == 0x003D) { # =        } elsif ($self->{next_input_character} == 0x003D) { # =
867          $self->{state} = 'before attribute value';          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
868          !!!next-input-character;          !!!next-input-character;
869          redo A;          redo A;
870        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
# Line 721  sub _get_next_token ($) { Line 880  sub _get_next_token ($) {
880          } else {          } else {
881            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
882          }          }
883          $self->{state} = 'data';          $self->{state} = DATA_STATE;
884          !!!next-input-character;          !!!next-input-character;
885    
886          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 731  sub _get_next_token ($) { Line 890  sub _get_next_token ($) {
890                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_input_character} <= 0x005A) { # A..Z
891          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),
892                                value => ''};                                value => ''};
893          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
894          !!!next-input-character;          !!!next-input-character;
895          redo A;          redo A;
896        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_input_character} == 0x002F) { # /
# Line 745  sub _get_next_token ($) { Line 904  sub _get_next_token ($) {
904            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
905            ## TODO: Different error type for <aa / bb> than <aa/>            ## TODO: Different error type for <aa / bb> than <aa/>
906          }          }
907          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
908          # next-input-character is already done          # next-input-character is already done
909          redo A;          redo A;
910        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
# Line 762  sub _get_next_token ($) { Line 921  sub _get_next_token ($) {
921          } else {          } else {
922            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
923          }          }
924          $self->{state} = 'data';          $self->{state} = DATA_STATE;
925          # reconsume          # reconsume
926    
927          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 771  sub _get_next_token ($) { Line 930  sub _get_next_token ($) {
930        } else {        } else {
931          $self->{current_attribute} = {name => chr ($self->{next_input_character}),          $self->{current_attribute} = {name => chr ($self->{next_input_character}),
932                                value => ''};                                value => ''};
933          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
934          !!!next-input-character;          !!!next-input-character;
935          redo A;                  redo A;        
936        }        }
937      } elsif ($self->{state} eq 'before attribute value') {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) {
938        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
939            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
940            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
# Line 785  sub _get_next_token ($) { Line 944  sub _get_next_token ($) {
944          !!!next-input-character;          !!!next-input-character;
945          redo A;          redo A;
946        } elsif ($self->{next_input_character} == 0x0022) { # "        } elsif ($self->{next_input_character} == 0x0022) { # "
947          $self->{state} = 'attribute value (double-quoted)';          $self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE;
948          !!!next-input-character;          !!!next-input-character;
949          redo A;          redo A;
950        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_input_character} == 0x0026) { # &
951          $self->{state} = 'attribute value (unquoted)';          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
952          ## reconsume          ## reconsume
953          redo A;          redo A;
954        } elsif ($self->{next_input_character} == 0x0027) { # '        } elsif ($self->{next_input_character} == 0x0027) { # '
955          $self->{state} = 'attribute value (single-quoted)';          $self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE;
956          !!!next-input-character;          !!!next-input-character;
957          redo A;          redo A;
958        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
# Line 809  sub _get_next_token ($) { Line 968  sub _get_next_token ($) {
968          } else {          } else {
969            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
970          }          }
971          $self->{state} = 'data';          $self->{state} = DATA_STATE;
972          !!!next-input-character;          !!!next-input-character;
973    
974          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 829  sub _get_next_token ($) { Line 988  sub _get_next_token ($) {
988          } else {          } else {
989            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
990          }          }
991          $self->{state} = 'data';          $self->{state} = DATA_STATE;
992          ## reconsume          ## reconsume
993    
994          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
995    
996          redo A;          redo A;
997        } else {        } else {
998            if ($self->{next_input_character} == 0x003D) { # =
999              !!!parse-error (type => 'bad attribute value');
1000            }
1001          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});
1002          $self->{state} = 'attribute value (unquoted)';          $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
1003          !!!next-input-character;          !!!next-input-character;
1004          redo A;          redo A;
1005        }        }
1006      } elsif ($self->{state} eq 'attribute value (double-quoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) {
1007        if ($self->{next_input_character} == 0x0022) { # "        if ($self->{next_input_character} == 0x0022) { # "
1008          $self->{state} = 'before attribute name';          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1009          !!!next-input-character;          !!!next-input-character;
1010          redo A;          redo A;
1011        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_input_character} == 0x0026) { # &
1012          $self->{last_attribute_value_state} = 'attribute value (double-quoted)';          $self->{last_attribute_value_state} = $self->{state};
1013          $self->{state} = 'entity in attribute value';          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1014          !!!next-input-character;          !!!next-input-character;
1015          redo A;          redo A;
1016        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
# Line 865  sub _get_next_token ($) { Line 1027  sub _get_next_token ($) {
1027          } else {          } else {
1028            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1029          }          }
1030          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1031          ## reconsume          ## reconsume
1032    
1033          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 877  sub _get_next_token ($) { Line 1039  sub _get_next_token ($) {
1039          !!!next-input-character;          !!!next-input-character;
1040          redo A;          redo A;
1041        }        }
1042      } elsif ($self->{state} eq 'attribute value (single-quoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) {
1043        if ($self->{next_input_character} == 0x0027) { # '        if ($self->{next_input_character} == 0x0027) { # '
1044          $self->{state} = 'before attribute name';          $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1045          !!!next-input-character;          !!!next-input-character;
1046          redo A;          redo A;
1047        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_input_character} == 0x0026) { # &
1048          $self->{last_attribute_value_state} = 'attribute value (single-quoted)';          $self->{last_attribute_value_state} = $self->{state};
1049          $self->{state} = 'entity in attribute value';          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1050          !!!next-input-character;          !!!next-input-character;
1051          redo A;          redo A;
1052        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
# Line 901  sub _get_next_token ($) { Line 1063  sub _get_next_token ($) {
1063          } else {          } else {
1064            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1065          }          }
1066          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1067          ## reconsume          ## reconsume
1068    
1069          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 913  sub _get_next_token ($) { Line 1075  sub _get_next_token ($) {
1075          !!!next-input-character;          !!!next-input-character;
1076          redo A;          redo A;
1077        }        }
1078      } elsif ($self->{state} eq 'attribute value (unquoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) {
1079        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
1080            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
1081            $self->{next_input_character} == 0x000B or # HT            $self->{next_input_character} == 0x000B or # HT
1082            $self->{next_input_character} == 0x000C or # FF            $self->{next_input_character} == 0x000C or # FF
1083            $self->{next_input_character} == 0x0020) { # SP            $self->{next_input_character} == 0x0020) { # SP
1084          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1085          !!!next-input-character;          !!!next-input-character;
1086          redo A;          redo A;
1087        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_input_character} == 0x0026) { # &
1088          $self->{last_attribute_value_state} = 'attribute value (unquoted)';          $self->{last_attribute_value_state} = $self->{state};
1089          $self->{state} = 'entity in attribute value';          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1090          !!!next-input-character;          !!!next-input-character;
1091          redo A;          redo A;
1092        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
# Line 940  sub _get_next_token ($) { Line 1102  sub _get_next_token ($) {
1102          } else {          } else {
1103            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1104          }          }
1105          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1106          !!!next-input-character;          !!!next-input-character;
1107    
1108          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
# Line 960  sub _get_next_token ($) { Line 1122  sub _get_next_token ($) {
1122          } else {          } else {
1123            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1124          }          }
1125          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1126          ## reconsume          ## reconsume
1127    
1128          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
1129    
1130          redo A;          redo A;
1131        } else {        } else {
1132            if ({
1133                 0x0022 => 1, # "
1134                 0x0027 => 1, # '
1135                 0x003D => 1, # =
1136                }->{$self->{next_input_character}}) {
1137              !!!parse-error (type => 'bad attribute value');
1138            }
1139          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});
1140          ## Stay in the state          ## Stay in the state
1141          !!!next-input-character;          !!!next-input-character;
1142          redo A;          redo A;
1143        }        }
1144      } elsif ($self->{state} eq 'entity in attribute value') {      } elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) {
1145        my $token = $self->_tokenize_attempt_to_consume_an_entity (1);        my $token = $self->_tokenize_attempt_to_consume_an_entity
1146              (1,
1147               $self->{last_attribute_value_state}
1148                 == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE ? 0x0022 : # "
1149               $self->{last_attribute_value_state}
1150                 == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE ? 0x0027 : # '
1151               -1);
1152    
1153        unless (defined $token) {        unless (defined $token) {
1154          $self->{current_attribute}->{value} .= '&';          $self->{current_attribute}->{value} .= '&';
1155        } else {        } else {
1156          $self->{current_attribute}->{value} .= $token->{data};          $self->{current_attribute}->{value} .= $token->{data};
1157            $self->{current_attribute}->{has_reference} = $token->{has_reference};
1158          ## ISSUE: spec says "append the returned character token to the current attribute's value"          ## ISSUE: spec says "append the returned character token to the current attribute's value"
1159        }        }
1160    
1161        $self->{state} = $self->{last_attribute_value_state};        $self->{state} = $self->{last_attribute_value_state};
1162        # next-input-character is already done        # next-input-character is already done
1163        redo A;        redo A;
1164      } elsif ($self->{state} eq 'bogus comment') {      } elsif ($self->{state} == AFTER_ATTRIBUTE_VALUE_QUOTED_STATE) {
1165          if ($self->{next_input_character} == 0x0009 or # HT
1166              $self->{next_input_character} == 0x000A or # LF
1167              $self->{next_input_character} == 0x000B or # VT
1168              $self->{next_input_character} == 0x000C or # FF
1169              $self->{next_input_character} == 0x0020) { # SP
1170            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1171            !!!next-input-character;
1172            redo A;
1173          } elsif ($self->{next_input_character} == 0x003E) { # >
1174            if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1175              $self->{current_token}->{first_start_tag}
1176                  = not defined $self->{last_emitted_start_tag_name};
1177              $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1178            } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1179              $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1180              if ($self->{current_token}->{attributes}) {
1181                !!!parse-error (type => 'end tag attribute');
1182              }
1183            } else {
1184              die "$0: $self->{current_token}->{type}: Unknown token type";
1185            }
1186            $self->{state} = DATA_STATE;
1187            !!!next-input-character;
1188    
1189            !!!emit ($self->{current_token}); # start tag or end tag
1190    
1191            redo A;
1192          } elsif ($self->{next_input_character} == 0x002F) { # /
1193            !!!next-input-character;
1194            if ($self->{next_input_character} == 0x003E and # >
1195                $self->{current_token}->{type} == START_TAG_TOKEN and
1196                $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
1197              # permitted slash
1198              #
1199            } else {
1200              !!!parse-error (type => 'nestc');
1201            }
1202            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1203            # next-input-character is already done
1204            redo A;
1205          } else {
1206            !!!parse-error (type => 'no space between attributes');
1207            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1208            ## reconsume
1209            redo A;
1210          }
1211        } elsif ($self->{state} == BOGUS_COMMENT_STATE) {
1212        ## (only happen if PCDATA state)        ## (only happen if PCDATA state)
1213                
1214        my $token = {type => COMMENT_TOKEN, data => ''};        my $token = {type => COMMENT_TOKEN, data => ''};
1215    
1216        BC: {        BC: {
1217          if ($self->{next_input_character} == 0x003E) { # >          if ($self->{next_input_character} == 0x003E) { # >
1218            $self->{state} = 'data';            $self->{state} = DATA_STATE;
1219            !!!next-input-character;            !!!next-input-character;
1220    
1221            !!!emit ($token);            !!!emit ($token);
1222    
1223            redo A;            redo A;
1224          } elsif ($self->{next_input_character} == -1) {          } elsif ($self->{next_input_character} == -1) {
1225            $self->{state} = 'data';            $self->{state} = DATA_STATE;
1226            ## reconsume            ## reconsume
1227    
1228            !!!emit ($token);            !!!emit ($token);
# Line 1011  sub _get_next_token ($) { Line 1234  sub _get_next_token ($) {
1234            redo BC;            redo BC;
1235          }          }
1236        } # BC        } # BC
1237      } elsif ($self->{state} eq 'markup declaration open') {      } elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) {
1238        ## (only happen if PCDATA state)        ## (only happen if PCDATA state)
1239    
1240        my @next_char;        my @next_char;
# Line 1022  sub _get_next_token ($) { Line 1245  sub _get_next_token ($) {
1245          push @next_char, $self->{next_input_character};          push @next_char, $self->{next_input_character};
1246          if ($self->{next_input_character} == 0x002D) { # -          if ($self->{next_input_character} == 0x002D) { # -
1247            $self->{current_token} = {type => COMMENT_TOKEN, data => ''};            $self->{current_token} = {type => COMMENT_TOKEN, data => ''};
1248            $self->{state} = 'comment start';            $self->{state} = COMMENT_START_STATE;
1249            !!!next-input-character;            !!!next-input-character;
1250            redo A;            redo A;
1251          }          }
# Line 1053  sub _get_next_token ($) { Line 1276  sub _get_next_token ($) {
1276                    if ($self->{next_input_character} == 0x0045 or # E                    if ($self->{next_input_character} == 0x0045 or # E
1277                        $self->{next_input_character} == 0x0065) { # e                        $self->{next_input_character} == 0x0065) { # e
1278                      ## ISSUE: What a stupid code this is!                      ## ISSUE: What a stupid code this is!
1279                      $self->{state} = 'DOCTYPE';                      $self->{state} = DOCTYPE_STATE;
1280                      !!!next-input-character;                      !!!next-input-character;
1281                      redo A;                      redo A;
1282                    }                    }
# Line 1067  sub _get_next_token ($) { Line 1290  sub _get_next_token ($) {
1290        !!!parse-error (type => 'bogus comment');        !!!parse-error (type => 'bogus comment');
1291        $self->{next_input_character} = shift @next_char;        $self->{next_input_character} = shift @next_char;
1292        !!!back-next-input-character (@next_char);        !!!back-next-input-character (@next_char);
1293        $self->{state} = 'bogus comment';        $self->{state} = BOGUS_COMMENT_STATE;
1294        redo A;        redo A;
1295                
1296        ## ISSUE: typos in spec: chacacters, is is a parse error        ## ISSUE: typos in spec: chacacters, is is a parse error
1297        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?
1298      } elsif ($self->{state} eq 'comment start') {      } elsif ($self->{state} == COMMENT_START_STATE) {
1299        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_input_character} == 0x002D) { # -
1300          $self->{state} = 'comment start dash';          $self->{state} = COMMENT_START_DASH_STATE;
1301          !!!next-input-character;          !!!next-input-character;
1302          redo A;          redo A;
1303        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1304          !!!parse-error (type => 'bogus comment');          !!!parse-error (type => 'bogus comment');
1305          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1306          !!!next-input-character;          !!!next-input-character;
1307    
1308          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1087  sub _get_next_token ($) { Line 1310  sub _get_next_token ($) {
1310          redo A;          redo A;
1311        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1312          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1313          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1314          ## reconsume          ## reconsume
1315    
1316          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1096  sub _get_next_token ($) { Line 1319  sub _get_next_token ($) {
1319        } else {        } else {
1320          $self->{current_token}->{data} # comment          $self->{current_token}->{data} # comment
1321              .= chr ($self->{next_input_character});              .= chr ($self->{next_input_character});
1322          $self->{state} = 'comment';          $self->{state} = COMMENT_STATE;
1323          !!!next-input-character;          !!!next-input-character;
1324          redo A;          redo A;
1325        }        }
1326      } elsif ($self->{state} eq 'comment start dash') {      } elsif ($self->{state} == COMMENT_START_DASH_STATE) {
1327        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_input_character} == 0x002D) { # -
1328          $self->{state} = 'comment end';          $self->{state} = COMMENT_END_STATE;
1329          !!!next-input-character;          !!!next-input-character;
1330          redo A;          redo A;
1331        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1332          !!!parse-error (type => 'bogus comment');          !!!parse-error (type => 'bogus comment');
1333          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1334          !!!next-input-character;          !!!next-input-character;
1335    
1336          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1115  sub _get_next_token ($) { Line 1338  sub _get_next_token ($) {
1338          redo A;          redo A;
1339        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1340          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1341          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1342          ## reconsume          ## reconsume
1343    
1344          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1124  sub _get_next_token ($) { Line 1347  sub _get_next_token ($) {
1347        } else {        } else {
1348          $self->{current_token}->{data} # comment          $self->{current_token}->{data} # comment
1349              .= '-' . chr ($self->{next_input_character});              .= '-' . chr ($self->{next_input_character});
1350          $self->{state} = 'comment';          $self->{state} = COMMENT_STATE;
1351          !!!next-input-character;          !!!next-input-character;
1352          redo A;          redo A;
1353        }        }
1354      } elsif ($self->{state} eq 'comment') {      } elsif ($self->{state} == COMMENT_STATE) {
1355        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_input_character} == 0x002D) { # -
1356          $self->{state} = 'comment end dash';          $self->{state} = COMMENT_END_DASH_STATE;
1357          !!!next-input-character;          !!!next-input-character;
1358          redo A;          redo A;
1359        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1360          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1361          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1362          ## reconsume          ## reconsume
1363    
1364          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1147  sub _get_next_token ($) { Line 1370  sub _get_next_token ($) {
1370          !!!next-input-character;          !!!next-input-character;
1371          redo A;          redo A;
1372        }        }
1373      } elsif ($self->{state} eq 'comment end dash') {      } elsif ($self->{state} == COMMENT_END_DASH_STATE) {
1374        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_input_character} == 0x002D) { # -
1375          $self->{state} = 'comment end';          $self->{state} = COMMENT_END_STATE;
1376          !!!next-input-character;          !!!next-input-character;
1377          redo A;          redo A;
1378        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1379          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1380          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1381          ## reconsume          ## reconsume
1382    
1383          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1162  sub _get_next_token ($) { Line 1385  sub _get_next_token ($) {
1385          redo A;          redo A;
1386        } else {        } else {
1387          $self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment          $self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment
1388          $self->{state} = 'comment';          $self->{state} = COMMENT_STATE;
1389          !!!next-input-character;          !!!next-input-character;
1390          redo A;          redo A;
1391        }        }
1392      } elsif ($self->{state} eq 'comment end') {      } elsif ($self->{state} == COMMENT_END_STATE) {
1393        if ($self->{next_input_character} == 0x003E) { # >        if ($self->{next_input_character} == 0x003E) { # >
1394          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1395          !!!next-input-character;          !!!next-input-character;
1396    
1397          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1182  sub _get_next_token ($) { Line 1405  sub _get_next_token ($) {
1405          redo A;          redo A;
1406        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1407          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1408          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1409          ## reconsume          ## reconsume
1410    
1411          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
# Line 1191  sub _get_next_token ($) { Line 1414  sub _get_next_token ($) {
1414        } else {        } else {
1415          !!!parse-error (type => 'dash in comment');          !!!parse-error (type => 'dash in comment');
1416          $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment          $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment
1417          $self->{state} = 'comment';          $self->{state} = COMMENT_STATE;
1418          !!!next-input-character;          !!!next-input-character;
1419          redo A;          redo A;
1420        }        }
1421      } elsif ($self->{state} eq 'DOCTYPE') {      } elsif ($self->{state} == DOCTYPE_STATE) {
1422        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
1423            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
1424            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
1425            $self->{next_input_character} == 0x000C or # FF            $self->{next_input_character} == 0x000C or # FF
1426            $self->{next_input_character} == 0x0020) { # SP            $self->{next_input_character} == 0x0020) { # SP
1427          $self->{state} = 'before DOCTYPE name';          $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
1428          !!!next-input-character;          !!!next-input-character;
1429          redo A;          redo A;
1430        } else {        } else {
1431          !!!parse-error (type => 'no space before DOCTYPE name');          !!!parse-error (type => 'no space before DOCTYPE name');
1432          $self->{state} = 'before DOCTYPE name';          $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
1433          ## reconsume          ## reconsume
1434          redo A;          redo A;
1435        }        }
1436      } elsif ($self->{state} eq 'before DOCTYPE name') {      } elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) {
1437        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
1438            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
1439            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
# Line 1221  sub _get_next_token ($) { Line 1444  sub _get_next_token ($) {
1444          redo A;          redo A;
1445        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1446          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
1447          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1448          !!!next-input-character;          !!!next-input-character;
1449    
1450          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect
# Line 1229  sub _get_next_token ($) { Line 1452  sub _get_next_token ($) {
1452          redo A;          redo A;
1453        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1454          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
1455          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1456          ## reconsume          ## reconsume
1457    
1458          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect
# Line 1241  sub _get_next_token ($) { Line 1464  sub _get_next_token ($) {
1464                 name => chr ($self->{next_input_character}),                 name => chr ($self->{next_input_character}),
1465                 correct => 1};                 correct => 1};
1466  ## ISSUE: "Set the token's name name to the" in the spec  ## ISSUE: "Set the token's name name to the" in the spec
1467          $self->{state} = 'DOCTYPE name';          $self->{state} = DOCTYPE_NAME_STATE;
1468          !!!next-input-character;          !!!next-input-character;
1469          redo A;          redo A;
1470        }        }
1471      } elsif ($self->{state} eq 'DOCTYPE name') {      } elsif ($self->{state} == DOCTYPE_NAME_STATE) {
1472  ## ISSUE: Redundant "First," in the spec.  ## ISSUE: Redundant "First," in the spec.
1473        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
1474            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
1475            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
1476            $self->{next_input_character} == 0x000C or # FF            $self->{next_input_character} == 0x000C or # FF
1477            $self->{next_input_character} == 0x0020) { # SP            $self->{next_input_character} == 0x0020) { # SP
1478          $self->{state} = 'after DOCTYPE name';          $self->{state} = AFTER_DOCTYPE_NAME_STATE;
1479          !!!next-input-character;          !!!next-input-character;
1480          redo A;          redo A;
1481        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1482          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1483          !!!next-input-character;          !!!next-input-character;
1484    
1485          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
# Line 1264  sub _get_next_token ($) { Line 1487  sub _get_next_token ($) {
1487          redo A;          redo A;
1488        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1489          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1490          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1491          ## reconsume          ## reconsume
1492    
1493          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1278  sub _get_next_token ($) { Line 1501  sub _get_next_token ($) {
1501          !!!next-input-character;          !!!next-input-character;
1502          redo A;          redo A;
1503        }        }
1504      } elsif ($self->{state} eq 'after DOCTYPE name') {      } elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) {
1505        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_input_character} == 0x0009 or # HT
1506            $self->{next_input_character} == 0x000A or # LF            $self->{next_input_character} == 0x000A or # LF
1507            $self->{next_input_character} == 0x000B or # VT            $self->{next_input_character} == 0x000B or # VT
# Line 1288  sub _get_next_token ($) { Line 1511  sub _get_next_token ($) {
1511          !!!next-input-character;          !!!next-input-character;
1512          redo A;          redo A;
1513        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1514          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1515          !!!next-input-character;          !!!next-input-character;
1516    
1517          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
# Line 1296  sub _get_next_token ($) { Line 1519  sub _get_next_token ($) {
1519          redo A;          redo A;
1520        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1521          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1522          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1523          ## reconsume          ## reconsume
1524    
1525          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1320  sub _get_next_token ($) { Line 1543  sub _get_next_token ($) {
1543                  !!!next-input-character;                  !!!next-input-character;
1544                  if ($self->{next_input_character} == 0x0043 or # C                  if ($self->{next_input_character} == 0x0043 or # C
1545                      $self->{next_input_character} == 0x0063) { # c                      $self->{next_input_character} == 0x0063) { # c
1546                    $self->{state} = 'before DOCTYPE public identifier';                    $self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1547                    !!!next-input-character;                    !!!next-input-character;
1548                    redo A;                    redo A;
1549                  }                  }
# Line 1347  sub _get_next_token ($) { Line 1570  sub _get_next_token ($) {
1570                  !!!next-input-character;                  !!!next-input-character;
1571                  if ($self->{next_input_character} == 0x004D or # M                  if ($self->{next_input_character} == 0x004D or # M
1572                      $self->{next_input_character} == 0x006D) { # m                      $self->{next_input_character} == 0x006D) { # m
1573                    $self->{state} = 'before DOCTYPE system identifier';                    $self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
1574                    !!!next-input-character;                    !!!next-input-character;
1575                    redo A;                    redo A;
1576                  }                  }
# Line 1363  sub _get_next_token ($) { Line 1586  sub _get_next_token ($) {
1586        }        }
1587    
1588        !!!parse-error (type => 'string after DOCTYPE name');        !!!parse-error (type => 'string after DOCTYPE name');
1589        $self->{state} = 'bogus DOCTYPE';        $self->{state} = BOGUS_DOCTYPE_STATE;
1590        # next-input-character is already done        # next-input-character is already done
1591        redo A;        redo A;
1592      } elsif ($self->{state} eq 'before DOCTYPE public identifier') {      } elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
1593        if ({        if ({
1594              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1595              #0x000D => 1, # HT, LF, VT, FF, SP, CR              #0x000D => 1, # HT, LF, VT, FF, SP, CR
# Line 1376  sub _get_next_token ($) { Line 1599  sub _get_next_token ($) {
1599          redo A;          redo A;
1600        } elsif ($self->{next_input_character} eq 0x0022) { # "        } elsif ($self->{next_input_character} eq 0x0022) { # "
1601          $self->{current_token}->{public_identifier} = ''; # DOCTYPE          $self->{current_token}->{public_identifier} = ''; # DOCTYPE
1602          $self->{state} = 'DOCTYPE public identifier (double-quoted)';          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE;
1603          !!!next-input-character;          !!!next-input-character;
1604          redo A;          redo A;
1605        } elsif ($self->{next_input_character} eq 0x0027) { # '        } elsif ($self->{next_input_character} eq 0x0027) { # '
1606          $self->{current_token}->{public_identifier} = ''; # DOCTYPE          $self->{current_token}->{public_identifier} = ''; # DOCTYPE
1607          $self->{state} = 'DOCTYPE public identifier (single-quoted)';          $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE;
1608          !!!next-input-character;          !!!next-input-character;
1609          redo A;          redo A;
1610        } elsif ($self->{next_input_character} eq 0x003E) { # >        } elsif ($self->{next_input_character} eq 0x003E) { # >
1611          !!!parse-error (type => 'no PUBLIC literal');          !!!parse-error (type => 'no PUBLIC literal');
1612    
1613          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1614          !!!next-input-character;          !!!next-input-character;
1615    
1616          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1397  sub _get_next_token ($) { Line 1620  sub _get_next_token ($) {
1620        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1621          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1622    
1623          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1624          ## reconsume          ## reconsume
1625    
1626          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1406  sub _get_next_token ($) { Line 1629  sub _get_next_token ($) {
1629          redo A;          redo A;
1630        } else {        } else {
1631          !!!parse-error (type => 'string after PUBLIC');          !!!parse-error (type => 'string after PUBLIC');
1632          $self->{state} = 'bogus DOCTYPE';          $self->{state} = BOGUS_DOCTYPE_STATE;
1633          !!!next-input-character;          !!!next-input-character;
1634          redo A;          redo A;
1635        }        }
1636      } elsif ($self->{state} eq 'DOCTYPE public identifier (double-quoted)') {      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) {
1637        if ($self->{next_input_character} == 0x0022) { # "        if ($self->{next_input_character} == 0x0022) { # "
1638          $self->{state} = 'after DOCTYPE public identifier';          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1639          !!!next-input-character;          !!!next-input-character;
1640          redo A;          redo A;
1641          } elsif ($self->{next_input_character} == 0x003E) { # >
1642            !!!parse-error (type => 'unclosed PUBLIC literal');
1643    
1644            $self->{state} = DATA_STATE;
1645            !!!next-input-character;
1646    
1647            delete $self->{current_token}->{correct};
1648            !!!emit ($self->{current_token}); # DOCTYPE
1649    
1650            redo A;
1651        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1652          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
1653    
1654          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1655          ## reconsume          ## reconsume
1656    
1657          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1432  sub _get_next_token ($) { Line 1665  sub _get_next_token ($) {
1665          !!!next-input-character;          !!!next-input-character;
1666          redo A;          redo A;
1667        }        }
1668      } elsif ($self->{state} eq 'DOCTYPE public identifier (single-quoted)') {      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) {
1669        if ($self->{next_input_character} == 0x0027) { # '        if ($self->{next_input_character} == 0x0027) { # '
1670          $self->{state} = 'after DOCTYPE public identifier';          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1671          !!!next-input-character;          !!!next-input-character;
1672          redo A;          redo A;
1673          } elsif ($self->{next_input_character} == 0x003E) { # >
1674            !!!parse-error (type => 'unclosed PUBLIC literal');
1675    
1676            $self->{state} = DATA_STATE;
1677            !!!next-input-character;
1678    
1679            delete $self->{current_token}->{correct};
1680            !!!emit ($self->{current_token}); # DOCTYPE
1681    
1682            redo A;
1683        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1684          !!!parse-error (type => 'unclosed PUBLIC literal');          !!!parse-error (type => 'unclosed PUBLIC literal');
1685    
1686          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1687          ## reconsume          ## reconsume
1688    
1689          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1454  sub _get_next_token ($) { Line 1697  sub _get_next_token ($) {
1697          !!!next-input-character;          !!!next-input-character;
1698          redo A;          redo A;
1699        }        }
1700      } elsif ($self->{state} eq 'after DOCTYPE public identifier') {      } elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
1701        if ({        if ({
1702              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1703              #0x000D => 1, # HT, LF, VT, FF, SP, CR              #0x000D => 1, # HT, LF, VT, FF, SP, CR
# Line 1464  sub _get_next_token ($) { Line 1707  sub _get_next_token ($) {
1707          redo A;          redo A;
1708        } elsif ($self->{next_input_character} == 0x0022) { # "        } elsif ($self->{next_input_character} == 0x0022) { # "
1709          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1710          $self->{state} = 'DOCTYPE system identifier (double-quoted)';          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
1711          !!!next-input-character;          !!!next-input-character;
1712          redo A;          redo A;
1713        } elsif ($self->{next_input_character} == 0x0027) { # '        } elsif ($self->{next_input_character} == 0x0027) { # '
1714          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1715          $self->{state} = 'DOCTYPE system identifier (single-quoted)';          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
1716          !!!next-input-character;          !!!next-input-character;
1717          redo A;          redo A;
1718        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1719          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1720          !!!next-input-character;          !!!next-input-character;
1721    
1722          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
# Line 1482  sub _get_next_token ($) { Line 1725  sub _get_next_token ($) {
1725        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1726          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1727    
1728          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1729          ## reconsume          ## reconsume
1730    
1731          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1491  sub _get_next_token ($) { Line 1734  sub _get_next_token ($) {
1734          redo A;          redo A;
1735        } else {        } else {
1736          !!!parse-error (type => 'string after PUBLIC literal');          !!!parse-error (type => 'string after PUBLIC literal');
1737          $self->{state} = 'bogus DOCTYPE';          $self->{state} = BOGUS_DOCTYPE_STATE;
1738          !!!next-input-character;          !!!next-input-character;
1739          redo A;          redo A;
1740        }        }
1741      } elsif ($self->{state} eq 'before DOCTYPE system identifier') {      } elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
1742        if ({        if ({
1743              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1744              #0x000D => 1, # HT, LF, VT, FF, SP, CR              #0x000D => 1, # HT, LF, VT, FF, SP, CR
# Line 1505  sub _get_next_token ($) { Line 1748  sub _get_next_token ($) {
1748          redo A;          redo A;
1749        } elsif ($self->{next_input_character} == 0x0022) { # "        } elsif ($self->{next_input_character} == 0x0022) { # "
1750          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1751          $self->{state} = 'DOCTYPE system identifier (double-quoted)';          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
1752          !!!next-input-character;          !!!next-input-character;
1753          redo A;          redo A;
1754        } elsif ($self->{next_input_character} == 0x0027) { # '        } elsif ($self->{next_input_character} == 0x0027) { # '
1755          $self->{current_token}->{system_identifier} = ''; # DOCTYPE          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1756          $self->{state} = 'DOCTYPE system identifier (single-quoted)';          $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
1757          !!!next-input-character;          !!!next-input-character;
1758          redo A;          redo A;
1759        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1760          !!!parse-error (type => 'no SYSTEM literal');          !!!parse-error (type => 'no SYSTEM literal');
1761          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1762          !!!next-input-character;          !!!next-input-character;
1763    
1764          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1525  sub _get_next_token ($) { Line 1768  sub _get_next_token ($) {
1768        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1769          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1770    
1771          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1772          ## reconsume          ## reconsume
1773    
1774          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1534  sub _get_next_token ($) { Line 1777  sub _get_next_token ($) {
1777          redo A;          redo A;
1778        } else {        } else {
1779          !!!parse-error (type => 'string after SYSTEM');          !!!parse-error (type => 'string after SYSTEM');
1780          $self->{state} = 'bogus DOCTYPE';          $self->{state} = BOGUS_DOCTYPE_STATE;
1781          !!!next-input-character;          !!!next-input-character;
1782          redo A;          redo A;
1783        }        }
1784      } elsif ($self->{state} eq 'DOCTYPE system identifier (double-quoted)') {      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) {
1785        if ($self->{next_input_character} == 0x0022) { # "        if ($self->{next_input_character} == 0x0022) { # "
1786          $self->{state} = 'after DOCTYPE system identifier';          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
1787            !!!next-input-character;
1788            redo A;
1789          } elsif ($self->{next_input_character} == 0x003E) { # >
1790            !!!parse-error (type => 'unclosed PUBLIC literal');
1791    
1792            $self->{state} = DATA_STATE;
1793          !!!next-input-character;          !!!next-input-character;
1794    
1795            delete $self->{current_token}->{correct};
1796            !!!emit ($self->{current_token}); # DOCTYPE
1797    
1798          redo A;          redo A;
1799        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1800          !!!parse-error (type => 'unclosed SYSTEM literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
1801    
1802          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1803          ## reconsume          ## reconsume
1804    
1805          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1560  sub _get_next_token ($) { Line 1813  sub _get_next_token ($) {
1813          !!!next-input-character;          !!!next-input-character;
1814          redo A;          redo A;
1815        }        }
1816      } elsif ($self->{state} eq 'DOCTYPE system identifier (single-quoted)') {      } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) {
1817        if ($self->{next_input_character} == 0x0027) { # '        if ($self->{next_input_character} == 0x0027) { # '
1818          $self->{state} = 'after DOCTYPE system identifier';          $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
1819          !!!next-input-character;          !!!next-input-character;
1820          redo A;          redo A;
1821          } elsif ($self->{next_input_character} == 0x003E) { # >
1822            !!!parse-error (type => 'unclosed PUBLIC literal');
1823    
1824            $self->{state} = DATA_STATE;
1825            !!!next-input-character;
1826    
1827            delete $self->{current_token}->{correct};
1828            !!!emit ($self->{current_token}); # DOCTYPE
1829    
1830            redo A;
1831        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1832          !!!parse-error (type => 'unclosed SYSTEM literal');          !!!parse-error (type => 'unclosed SYSTEM literal');
1833    
1834          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1835          ## reconsume          ## reconsume
1836    
1837          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1582  sub _get_next_token ($) { Line 1845  sub _get_next_token ($) {
1845          !!!next-input-character;          !!!next-input-character;
1846          redo A;          redo A;
1847        }        }
1848      } elsif ($self->{state} eq 'after DOCTYPE system identifier') {      } elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
1849        if ({        if ({
1850              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1851              #0x000D => 1, # HT, LF, VT, FF, SP, CR              #0x000D => 1, # HT, LF, VT, FF, SP, CR
# Line 1591  sub _get_next_token ($) { Line 1854  sub _get_next_token ($) {
1854          !!!next-input-character;          !!!next-input-character;
1855          redo A;          redo A;
1856        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_input_character} == 0x003E) { # >
1857          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1858          !!!next-input-character;          !!!next-input-character;
1859    
1860          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
# Line 1600  sub _get_next_token ($) { Line 1863  sub _get_next_token ($) {
1863        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1864          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1865    
1866          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1867          ## reconsume          ## reconsume
1868    
1869          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1609  sub _get_next_token ($) { Line 1872  sub _get_next_token ($) {
1872          redo A;          redo A;
1873        } else {        } else {
1874          !!!parse-error (type => 'string after SYSTEM literal');          !!!parse-error (type => 'string after SYSTEM literal');
1875          $self->{state} = 'bogus DOCTYPE';          $self->{state} = BOGUS_DOCTYPE_STATE;
1876          !!!next-input-character;          !!!next-input-character;
1877          redo A;          redo A;
1878        }        }
1879      } elsif ($self->{state} eq 'bogus DOCTYPE') {      } elsif ($self->{state} == BOGUS_DOCTYPE_STATE) {
1880        if ($self->{next_input_character} == 0x003E) { # >        if ($self->{next_input_character} == 0x003E) { # >
1881          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1882          !!!next-input-character;          !!!next-input-character;
1883    
1884          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1624  sub _get_next_token ($) { Line 1887  sub _get_next_token ($) {
1887          redo A;          redo A;
1888        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_input_character} == -1) {
1889          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1890          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1891          ## reconsume          ## reconsume
1892    
1893          delete $self->{current_token}->{correct};          delete $self->{current_token}->{correct};
# Line 1644  sub _get_next_token ($) { Line 1907  sub _get_next_token ($) {
1907    die "$0: _get_next_token: unexpected case";    die "$0: _get_next_token: unexpected case";
1908  } # _get_next_token  } # _get_next_token
1909    
1910  sub _tokenize_attempt_to_consume_an_entity ($$) {  sub _tokenize_attempt_to_consume_an_entity ($$$) {
1911    my ($self, $in_attr) = @_;    my ($self, $in_attr, $additional) = @_;
1912    
1913    if ({    if ({
1914         0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF,         0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF,
1915         0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR         0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR
1916           $additional => 1,
1917        }->{$self->{next_input_character}}) {        }->{$self->{next_input_character}}) {
1918      ## Don't consume      ## Don't consume
1919      ## No error      ## No error
# Line 1705  sub _tokenize_attempt_to_consume_an_enti Line 1969  sub _tokenize_attempt_to_consume_an_enti
1969            $code = $c1_entity_char->{$code};            $code = $c1_entity_char->{$code};
1970          }          }
1971    
1972          return {type => CHARACTER_TOKEN, data => chr $code};          return {type => CHARACTER_TOKEN, data => chr $code,
1973                    has_reference => 1};
1974        } # X        } # X
1975      } elsif (0x0030 <= $self->{next_input_character} and      } elsif (0x0030 <= $self->{next_input_character} and
1976               $self->{next_input_character} <= 0x0039) { # 0..9               $self->{next_input_character} <= 0x0039) { # 0..9
# Line 1740  sub _tokenize_attempt_to_consume_an_enti Line 2005  sub _tokenize_attempt_to_consume_an_enti
2005          $code = $c1_entity_char->{$code};          $code = $c1_entity_char->{$code};
2006        }        }
2007                
2008        return {type => CHARACTER_TOKEN, data => chr $code};        return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1};
2009      } else {      } else {
2010        !!!parse-error (type => 'bare nero');        !!!parse-error (type => 'bare nero');
2011        !!!back-next-input-character ($self->{next_input_character});        !!!back-next-input-character ($self->{next_input_character});
# Line 1788  sub _tokenize_attempt_to_consume_an_enti Line 2053  sub _tokenize_attempt_to_consume_an_enti
2053      }      }
2054            
2055      if ($match > 0) {      if ($match > 0) {
2056        return {type => CHARACTER_TOKEN, data => $value};        return {type => CHARACTER_TOKEN, data => $value, has_reference => 1};
2057      } elsif ($match < 0) {      } elsif ($match < 0) {
2058        !!!parse-error (type => 'no refc');        !!!parse-error (type => 'no refc');
2059        if ($in_attr and $match < -1) {        if ($in_attr and $match < -1) {
2060          return {type => CHARACTER_TOKEN, data => '&'.$entity_name};          return {type => CHARACTER_TOKEN, data => '&'.$entity_name};
2061        } else {        } else {
2062          return {type => CHARACTER_TOKEN, data => $value};          return {type => CHARACTER_TOKEN, data => $value, has_reference => 1};
2063        }        }
2064      } else {      } else {
2065        !!!parse-error (type => 'bare ero');        !!!parse-error (type => 'bare ero');
2066        ## NOTE: No characters are consumed in the spec.        ## NOTE: "No characters are consumed" in the spec.
2067        return {type => CHARACTER_TOKEN, data => '&'.$value};        return {type => CHARACTER_TOKEN, data => '&'.$value};
2068      }      }
2069    } else {    } else {
# Line 1935  sub _tree_construction_initial ($) { Line 2200  sub _tree_construction_initial ($) {
2200            "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1,            "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1,
2201            "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1,            "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1,
2202            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1,            "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1,
2203              "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//EN" => 1,
2204              "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//EN" => 1,
2205              "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//EN" => 1,
2206            "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1,            "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1,
2207            "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1,            "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1,
2208            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1,            "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1,
# Line 2046  sub _tree_construction_root_element ($) Line 2314  sub _tree_construction_root_element ($)
2314              redo B;              redo B;
2315            }            }
2316          }          }
2317    
2318            $self->{application_cache_selection}->(undef);
2319    
2320            #
2321          } elsif ($token->{type} == START_TAG_TOKEN) {
2322            if ($token->{tag_name} eq 'html' and
2323                $token->{attributes}->{manifest}) {
2324              $self->{application_cache_selection}
2325                   ->($token->{attributes}->{manifest}->{value});
2326              ## ISSUE: No relative reference resolution?
2327            } else {
2328              $self->{application_cache_selection}->(undef);
2329            }
2330    
2331            ## ISSUE: There is an issue in the spec
2332          #          #
2333        } elsif ({        } elsif ({
                 START_TAG_TOKEN, 1,  
2334                  END_TAG_TOKEN, 1,                  END_TAG_TOKEN, 1,
2335                  END_OF_FILE_TOKEN, 1,                  END_OF_FILE_TOKEN, 1,
2336                 }->{$token->{type}}) {                 }->{$token->{type}}) {
2337            $self->{application_cache_selection}->(undef);
2338    
2339          ## ISSUE: There is an issue in the spec          ## ISSUE: There is an issue in the spec
2340          #          #
2341        } else {        } else {
2342          die "$0: $token->{type}: Unknown token type";          die "$0: $token->{type}: Unknown token type";
2343        }        }
2344    
2345        my $root_element; !!!create-element ($root_element, 'html');        my $root_element; !!!create-element ($root_element, 'html');
2346        $self->{document}->append_child ($root_element);        $self->{document}->append_child ($root_element);
2347        push @{$self->{open_elements}}, [$root_element, 'html'];        push @{$self->{open_elements}}, [$root_element, 'html'];
# Line 2539  sub _tree_construction_main ($) { Line 2824  sub _tree_construction_main ($) {
2824        !!!next-token;        !!!next-token;
2825        redo B;        redo B;
2826      } elsif ($token->{type} == END_OF_FILE_TOKEN) {      } elsif ($token->{type} == END_OF_FILE_TOKEN) {
2827        if ($self->{insertion_mode} == AFTER_HTML_BODY_IM or        if ($self->{insertion_mode} & AFTER_HTML_IMS) {
           $self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) {  
2828          #          #
2829        } else {        } else {
2830          ## Generate implied end tags          ## Generate implied end tags
# Line 2596  sub _tree_construction_main ($) { Line 2880  sub _tree_construction_main ($) {
2880        redo B;        redo B;
2881      } elsif ($token->{type} == COMMENT_TOKEN) {      } elsif ($token->{type} == COMMENT_TOKEN) {
2882        my $comment = $self->{document}->create_comment ($token->{data});        my $comment = $self->{document}->create_comment ($token->{data});
2883        if ($self->{insertion_mode} == AFTER_HTML_BODY_IM or        if ($self->{insertion_mode} & AFTER_HTML_IMS) {
           $self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) {  
2884          $self->{document}->append_child ($comment);          $self->{document}->append_child ($comment);
2885        } elsif ($self->{insertion_mode} == AFTER_BODY_IM) {        } elsif ($self->{insertion_mode} == AFTER_BODY_IM) {
2886          $self->{open_elements}->[0]->[0]->append_child ($comment);          $self->{open_elements}->[0]->[0]->append_child ($comment);
# Line 2606  sub _tree_construction_main ($) { Line 2889  sub _tree_construction_main ($) {
2889        }        }
2890        !!!next-token;        !!!next-token;
2891        redo B;        redo B;
2892      } elsif ($self->{insertion_mode} == IN_HEAD_IM or      } elsif ($self->{insertion_mode} & HEAD_IMS) {
              $self->{insertion_mode} == IN_HEAD_NOSCRIPT_IM or  
              $self->{insertion_mode} == AFTER_HEAD_IM or  
              $self->{insertion_mode} == BEFORE_HEAD_IM) {  
2893        if ($token->{type} == CHARACTER_TOKEN) {        if ($token->{type} == CHARACTER_TOKEN) {
2894          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
2895            $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);            $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);
# Line 2718  sub _tree_construction_main ($) { Line 2998  sub _tree_construction_main ($) {
2998                  push @{$self->{open_elements}}, [$self->{head_element}, 'head'];                  push @{$self->{open_elements}}, [$self->{head_element}, 'head'];
2999                }                }
3000                !!!insert-element ($token->{tag_name}, $token->{attributes});                !!!insert-element ($token->{tag_name}, $token->{attributes});
3001                pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec.                my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec.
3002    
3003                unless ($self->{confident}) {                unless ($self->{confident}) {
                 my $charset;  
3004                  if ($token->{attributes}->{charset}) { ## TODO: And if supported                  if ($token->{attributes}->{charset}) { ## TODO: And if supported
3005                    $charset = $token->{attributes}->{charset}->{value};                    $self->{change_encoding}
3006                  }                        ->($self, $token->{attributes}->{charset}->{value});
3007                  if ($token->{attributes}->{'http-equiv'}) {                    
3008                      $meta_el->[0]->get_attribute_node_ns (undef, 'charset')
3009                          ->set_user_data (manakai_has_reference =>
3010                                               $token->{attributes}->{charset}
3011                                                   ->{has_reference});
3012                    } elsif ($token->{attributes}->{content}) {
3013                    ## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition.                    ## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition.
3014                    if ($token->{attributes}->{'http-equiv'}->{value}                    if ($token->{attributes}->{content}->{value}
3015                        =~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*=                        =~ /\A[^;]*;[\x09-\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt]
3016                              [\x09-\x0D\x20]*=
3017                            [\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'|                            [\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'|
3018                            ([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) {                            ([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) {
3019                      $charset = defined $1 ? $1 : defined $2 ? $2 : $3;                      $self->{change_encoding}
3020                    } ## TODO: And if supported                          ->($self, defined $1 ? $1 : defined $2 ? $2 : $3);
3021                        $meta_el->[0]->get_attribute_node_ns (undef, 'content')
3022                            ->set_user_data (manakai_has_reference =>
3023                                                 $token->{attributes}->{content}
3024                                                       ->{has_reference});
3025                      }
3026                    }
3027                  } else {
3028                    if ($token->{attributes}->{charset}) {
3029                      $meta_el->[0]->get_attribute_node_ns (undef, 'charset')
3030                          ->set_user_data (manakai_has_reference =>
3031                                               $token->{attributes}->{charset}
3032                                                   ->{has_reference});
3033                    }
3034                    if ($token->{attributes}->{content}) {
3035                      $meta_el->[0]->get_attribute_node_ns (undef, 'content')
3036                          ->set_user_data (manakai_has_reference =>
3037                                               $token->{attributes}->{content}
3038                                                   ->{has_reference});
3039                  }                  }
                 ## TODO: Change the encoding  
3040                }                }
3041    
               ## TODO: Extracting |charset| from |meta|.  
3042                pop @{$self->{open_elements}}                pop @{$self->{open_elements}}
3043                    if $self->{insertion_mode} == AFTER_HEAD_IM;                    if $self->{insertion_mode} == AFTER_HEAD_IM;
3044                !!!next-token;                !!!next-token;
# Line 2986  sub _tree_construction_main ($) { Line 3287  sub _tree_construction_main ($) {
3287            }            }
3288    
3289            ## ISSUE: An issue in the spec.            ## ISSUE: An issue in the spec.
3290      } elsif ($self->{insertion_mode} == IN_BODY_IM or      } elsif ($self->{insertion_mode} & BODY_IMS) {
              $self->{insertion_mode} == IN_CELL_IM or  
              $self->{insertion_mode} == IN_CAPTION_IM) {  
3291            if ($token->{type} == CHARACTER_TOKEN) {            if ($token->{type} == CHARACTER_TOKEN) {
3292              ## NOTE: There is a code clone of "character in body".              ## NOTE: There is a code clone of "character in body".
3293              $reconstruct_active_formatting_elements->($insert_to_current);              $reconstruct_active_formatting_elements->($insert_to_current);
# Line 3282  sub _tree_construction_main ($) { Line 3581  sub _tree_construction_main ($) {
3581              } elsif ({              } elsif ({
3582                        body => 1, col => 1, colgroup => 1, html => 1,                        body => 1, col => 1, colgroup => 1, html => 1,
3583                       }->{$token->{tag_name}}) {                       }->{$token->{tag_name}}) {
3584                if ($self->{insertion_mode} == IN_CELL_IM or                if ($self->{insertion_mode} & BODY_TABLE_IMS) {
                   $self->{insertion_mode} == IN_CAPTION_IM) {  
3585                  !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});                  !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});
3586                  ## Ignore the token                  ## Ignore the token
3587                  !!!next-token;                  !!!next-token;
# Line 3309  sub _tree_construction_main ($) { Line 3607  sub _tree_construction_main ($) {
3607    
3608        $insert = $insert_to_current;        $insert = $insert_to_current;
3609        #        #
3610      } elsif ($self->{insertion_mode} == IN_ROW_IM or      } elsif ($self->{insertion_mode} & TABLE_IMS) {
3611               $self->{insertion_mode} == IN_TABLE_BODY_IM or        if ($token->{type} == CHARACTER_TOKEN) {
              $self->{insertion_mode} == IN_TABLE_IM) {  
           if ($token->{type} == CHARACTER_TOKEN) {  
             ## NOTE: There are "character in table" code clones.  
3612              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
3613                $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);                $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);
3614                                
# Line 3370  sub _tree_construction_main ($) { Line 3665  sub _tree_construction_main ($) {
3665                            
3666              !!!next-token;              !!!next-token;
3667              redo B;              redo B;
3668            } elsif ($token->{type} == START_TAG_TOKEN) {        } elsif ($token->{type} == START_TAG_TOKEN) {
3669              if ({              if ({
3670                   tr => ($self->{insertion_mode} != IN_ROW_IM),                   tr => ($self->{insertion_mode} != IN_ROW_IM),
3671                   th => 1, td => 1,                   th => 1, td => 1,
# Line 3556  sub _tree_construction_main ($) { Line 3851  sub _tree_construction_main ($) {
3851                  die "$0: in table: <>: $token->{tag_name}";                  die "$0: in table: <>: $token->{tag_name}";
3852                }                }
3853              } elsif ($token->{tag_name} eq 'table') {              } elsif ($token->{tag_name} eq 'table') {
               ## NOTE: There are code clones for this "table in table"  
3854                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
3855    
3856                ## As if </table>                ## As if </table>
# Line 3604  sub _tree_construction_main ($) { Line 3898  sub _tree_construction_main ($) {
3898    
3899                ## reprocess                ## reprocess
3900                redo B;                redo B;
3901              } else {          } else {
3902                #            !!!parse-error (type => 'in table:'.$token->{tag_name});
3903              }  
3904            } elsif ($token->{type} == END_TAG_TOKEN) {            $insert = $insert_to_foster;
3905              #
3906            }
3907          } elsif ($token->{type} == END_TAG_TOKEN) {
3908              if ($token->{tag_name} eq 'tr' and              if ($token->{tag_name} eq 'tr' and
3909                  $self->{insertion_mode} == IN_ROW_IM) {                  $self->{insertion_mode} == IN_ROW_IM) {
3910                ## have an element in table scope                ## have an element in table scope
# Line 3766  sub _tree_construction_main ($) { Line 4063  sub _tree_construction_main ($) {
4063              } elsif ({              } elsif ({
4064                        tbody => 1, tfoot => 1, thead => 1,                        tbody => 1, tfoot => 1, thead => 1,
4065                       }->{$token->{tag_name}} and                       }->{$token->{tag_name}} and
4066                       ($self->{insertion_mode} == IN_ROW_IM or                       $self->{insertion_mode} & ROW_IMS) {
                       $self->{insertion_mode} == IN_TABLE_BODY_IM)) {  
4067                if ($self->{insertion_mode} == IN_ROW_IM) {                if ($self->{insertion_mode} == IN_ROW_IM) {
4068                  ## have an element in table scope                  ## have an element in table scope
4069                  my $i;                  my $i;
# Line 3865  sub _tree_construction_main ($) { Line 4161  sub _tree_construction_main ($) {
4161                ## Ignore the token                ## Ignore the token
4162                !!!next-token;                !!!next-token;
4163                redo B;                redo B;
4164              } else {          } else {
4165                #            !!!parse-error (type => 'in table:/'.$token->{tag_name});
             }  
           } else {  
             die "$0: $token->{type}: Unknown token type";  
           }  
   
       !!!parse-error (type => 'in table:'.$token->{tag_name});  
4166    
4167        $insert = $insert_to_foster;            $insert = $insert_to_foster;
4168        #            #
4169            }
4170          } else {
4171            die "$0: $token->{type}: Unknown token type";
4172          }
4173      } elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) {      } elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) {
4174            if ($token->{type} == CHARACTER_TOKEN) {            if ($token->{type} == CHARACTER_TOKEN) {
4175              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
# Line 3934  sub _tree_construction_main ($) { Line 4228  sub _tree_construction_main ($) {
4228              redo B;              redo B;
4229            }            }
4230      } elsif ($self->{insertion_mode} == IN_SELECT_IM) {      } elsif ($self->{insertion_mode} == IN_SELECT_IM) {
4231            if ($token->{type} == CHARACTER_TOKEN) {        if ($token->{type} == CHARACTER_TOKEN) {
4232              $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data});          $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data});
4233              !!!next-token;          !!!next-token;
4234              redo B;          redo B;
4235            } elsif ($token->{type} == START_TAG_TOKEN) {        } elsif ($token->{type} == START_TAG_TOKEN) {
4236              if ($token->{tag_name} eq 'option') {              if ($token->{tag_name} eq 'option') {
4237                if ($self->{open_elements}->[-1]->[1] eq 'option') {                if ($self->{open_elements}->[-1]->[1] eq 'option') {
4238                  ## As if </option>                  ## As if </option>
# Line 3991  sub _tree_construction_main ($) { Line 4285  sub _tree_construction_main ($) {
4285    
4286                !!!next-token;                !!!next-token;
4287                redo B;                redo B;
4288              } else {          } else {
4289                #            !!!parse-error (type => 'in select:'.$token->{tag_name});
4290              }            ## Ignore the token
4291            } elsif ($token->{type} == END_TAG_TOKEN) {            !!!next-token;
4292              redo B;
4293            }
4294          } elsif ($token->{type} == END_TAG_TOKEN) {
4295              if ($token->{tag_name} eq 'optgroup') {              if ($token->{tag_name} eq 'optgroup') {
4296                if ($self->{open_elements}->[-1]->[1] eq 'option' and                if ($self->{open_elements}->[-1]->[1] eq 'option' and
4297                    $self->{open_elements}->[-2]->[1] eq 'optgroup') {                    $self->{open_elements}->[-2]->[1] eq 'optgroup') {
# Line 4096  sub _tree_construction_main ($) { Line 4393  sub _tree_construction_main ($) {
4393    
4394                ## reprocess                ## reprocess
4395                redo B;                redo B;
4396              } else {          } else {
4397                #            !!!parse-error (type => 'in select:/'.$token->{tag_name});
             }  
           } else {  
             #  
           }  
   
           !!!parse-error (type => 'in select:'.$token->{tag_name});  
4398            ## Ignore the token            ## Ignore the token
4399            !!!next-token;            !!!next-token;
4400            redo B;            redo B;
4401      } elsif ($self->{insertion_mode} == AFTER_BODY_IM or          }
4402               $self->{insertion_mode} == AFTER_HTML_BODY_IM) {        } else {
4403            die "$0: $token->{type}: Unknown token type";
4404          }
4405        } elsif ($self->{insertion_mode} & BODY_AFTER_IMS) {
4406        if ($token->{type} == CHARACTER_TOKEN) {        if ($token->{type} == CHARACTER_TOKEN) {
4407          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
4408            my $data = $1;            my $data = $1;
# Line 4178  sub _tree_construction_main ($) { Line 4472  sub _tree_construction_main ($) {
4472        } else {        } else {
4473          die "$0: $token->{type}: Unknown token type";          die "$0: $token->{type}: Unknown token type";
4474        }        }
4475      } elsif ($self->{insertion_mode} == IN_FRAMESET_IM or      } elsif ($self->{insertion_mode} & FRAME_IMS) {
              $self->{insertion_mode} == AFTER_FRAMESET_IM or  
              $self->{insertion_mode} == AFTER_HTML_FRAMESET_IM) {  
4476        if ($token->{type} == CHARACTER_TOKEN) {        if ($token->{type} == CHARACTER_TOKEN) {
4477          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
4478            $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);            $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);
# Line 4317  sub _tree_construction_main ($) { Line 4609  sub _tree_construction_main ($) {
4609        } elsif ($token->{tag_name} eq 'meta') {        } elsif ($token->{tag_name} eq 'meta') {
4610          ## NOTE: This is an "as if in head" code clone, only "-t" differs          ## NOTE: This is an "as if in head" code clone, only "-t" differs
4611          !!!insert-element-t ($token->{tag_name}, $token->{attributes});          !!!insert-element-t ($token->{tag_name}, $token->{attributes});
4612          pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec.          my $meta_el = pop @{$self->{open_elements}}; ## ISSUE: This step is missing in the spec.
4613    
4614          unless ($self->{confident}) {          unless ($self->{confident}) {
           my $charset;  
4615            if ($token->{attributes}->{charset}) { ## TODO: And if supported            if ($token->{attributes}->{charset}) { ## TODO: And if supported
4616              $charset = $token->{attributes}->{charset}->{value};              $self->{change_encoding}
4617            }                  ->($self, $token->{attributes}->{charset}->{value});
4618            if ($token->{attributes}->{'http-equiv'}) {              
4619                $meta_el->[0]->get_attribute_node_ns (undef, 'charset')
4620                    ->set_user_data (manakai_has_reference =>
4621                                         $token->{attributes}->{charset}
4622                                             ->{has_reference});
4623              } elsif ($token->{attributes}->{content}) {
4624              ## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition.              ## ISSUE: Algorithm name in the spec was incorrect so that not linked to the definition.
4625              if ($token->{attributes}->{'http-equiv'}->{value}              if ($token->{attributes}->{content}->{value}
4626                  =~ /\A[^;]*;[\x09-\x0D\x20]*charset[\x09-\x0D\x20]*=                  =~ /\A[^;]*;[\x09-\x0D\x20]*[Cc][Hh][Aa][Rr][Ss][Ee][Tt]
4627                        [\x09-\x0D\x20]*=
4628                      [\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'|                      [\x09-\x0D\x20]*(?>"([^"]*)"|'([^']*)'|
4629                      ([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) {                      ([^"'\x09-\x0D\x20][^\x09-\x0D\x20]*))/x) {
4630                $charset = defined $1 ? $1 : defined $2 ? $2 : $3;                $self->{change_encoding}
4631              } ## TODO: And if supported                    ->($self, defined $1 ? $1 : defined $2 ? $2 : $3);
4632                  $meta_el->[0]->get_attribute_node_ns (undef, 'content')
4633                      ->set_user_data (manakai_has_reference =>
4634                                           $token->{attributes}->{content}
4635                                                 ->{has_reference});
4636                }
4637              }
4638            } else {
4639              if ($token->{attributes}->{charset}) {
4640                $meta_el->[0]->get_attribute_node_ns (undef, 'charset')
4641                    ->set_user_data (manakai_has_reference =>
4642                                         $token->{attributes}->{charset}
4643                                             ->{has_reference});
4644              }
4645              if ($token->{attributes}->{content}) {
4646                $meta_el->[0]->get_attribute_node_ns (undef, 'content')
4647                    ->set_user_data (manakai_has_reference =>
4648                                         $token->{attributes}->{content}
4649                                             ->{has_reference});
4650            }            }
           ## TODO: Change the encoding  
4651          }          }
4652    
4653          !!!next-token;          !!!next-token;
# Line 4640  sub _tree_construction_main ($) { Line 4954  sub _tree_construction_main ($) {
4954          INSCOPE: for (reverse 0..$#{$self->{open_elements}}) {          INSCOPE: for (reverse 0..$#{$self->{open_elements}}) {
4955            my $node = $self->{open_elements}->[$_];            my $node = $self->{open_elements}->[$_];
4956            if ($node->[1] eq 'nobr') {            if ($node->[1] eq 'nobr') {
4957              !!!parse-error (type => 'not closed:nobr');              !!!parse-error (type => 'in nobr:nobr');
4958              !!!back-token;              !!!back-token;
4959              $token = {type => END_TAG_TOKEN, tag_name => 'nobr'};              $token = {type => END_TAG_TOKEN, tag_name => 'nobr'};
4960              redo B;              redo B;
# Line 4845  sub _tree_construction_main ($) { Line 5159  sub _tree_construction_main ($) {
5159                  noframes => 1,                  noframes => 1,
5160                  noscript => 0, ## TODO: 1 if scripting is enabled                  noscript => 0, ## TODO: 1 if scripting is enabled
5161                 }->{$token->{tag_name}}) {                 }->{$token->{tag_name}}) {
5162          ## NOTE: There are two "as if in body" code clones.          ## NOTE: There is an "as if in body" code clone.
5163          $parse_rcdata->(CDATA_CONTENT_MODEL, $insert);          $parse_rcdata->(CDATA_CONTENT_MODEL, $insert);
5164          redo B;          redo B;
5165        } elsif ($token->{tag_name} eq 'select') {        } elsif ($token->{tag_name} eq 'select') {
# Line 5001  sub _tree_construction_main ($) { Line 5315  sub _tree_construction_main ($) {
5315          if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) {          if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) {
5316            pop @{$self->{open_elements}};            pop @{$self->{open_elements}};
5317          } else {          } else {
5318            !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);            !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});
5319          }          }
5320    
5321          undef $self->{form_element};          undef $self->{form_element};
# Line 5039  sub _tree_construction_main ($) { Line 5353  sub _tree_construction_main ($) {
5353          } # INSCOPE          } # INSCOPE
5354                    
5355          if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) {          if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) {
5356            !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);            !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});
5357          }          }
5358                    
5359          splice @{$self->{open_elements}}, $i if defined $i;          splice @{$self->{open_elements}}, $i if defined $i;
# Line 5108  sub _tree_construction_main ($) { Line 5422  sub _tree_construction_main ($) {
5422                    
5423              ## Step 2              ## Step 2
5424              if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) {              if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) {
5425                  ## NOTE: <x><y></x>
5426                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
5427              }              }
5428                            
# Line 5158  sub set_inner_html ($$$) { Line 5473  sub set_inner_html ($$$) {
5473    my $s = \$_[0];    my $s = \$_[0];
5474    my $onerror = $_[1];    my $onerror = $_[1];
5475    
5476      ## ISSUE: Should {confident} be true?
5477    
5478    my $nt = $node->node_type;    my $nt = $node->node_type;
5479    if ($nt == 9) {    if ($nt == 9) {
5480      # MUST      # MUST
# Line 5230  sub set_inner_html ($$$) { Line 5547  sub set_inner_html ($$$) {
5547      $p->_initialize_tree_constructor;      $p->_initialize_tree_constructor;
5548    
5549      ## Step 2      ## Step 2
5550      my $node_ln = $node->local_name;      my $node_ln = $node->manakai_local_name;
5551      $p->{content_model} = {      $p->{content_model} = {
5552        title => RCDATA_CONTENT_MODEL,        title => RCDATA_CONTENT_MODEL,
5553        textarea => RCDATA_CONTENT_MODEL,        textarea => RCDATA_CONTENT_MODEL,
# Line 5270  sub set_inner_html ($$$) { Line 5587  sub set_inner_html ($$$) {
5587        if ($anode->node_type == 1) {        if ($anode->node_type == 1) {
5588          my $nsuri = $anode->namespace_uri;          my $nsuri = $anode->namespace_uri;
5589          if (defined $nsuri and $nsuri eq 'http://www.w3.org/1999/xhtml') {          if (defined $nsuri and $nsuri eq 'http://www.w3.org/1999/xhtml') {
5590            if ($anode->local_name eq 'form') { ## TODO: case?            if ($anode->manakai_local_name eq 'form') {
5591              $p->{form_element} = $anode;              $p->{form_element} = $anode;
5592              last AN;              last AN;
5593            }            }
# Line 5310  sub set_inner_html ($$$) { Line 5627  sub set_inner_html ($$$) {
5627    
5628  } # tree construction stage  } # tree construction stage
5629    
5630  sub get_inner_html ($$$) {  package Whatpm::HTML::RestartParser;
5631    my (undef, $node, $on_error) = @_;  push our @ISA, 'Error';
   
   ## Step 1  
   my $s = '';  
   
   my $in_cdata;  
   my $parent = $node;  
   while (defined $parent) {  
     if ($parent->node_type == 1 and  
         $parent->namespace_uri eq 'http://www.w3.org/1999/xhtml' and  
         {  
           style => 1, script => 1, xmp => 1, iframe => 1,  
           noembed => 1, noframes => 1, noscript => 1,  
         }->{$parent->local_name}) { ## TODO: case thingy  
       $in_cdata = 1;  
     }  
     $parent = $parent->parent_node;  
   }  
   
   ## Step 2  
   my @node = @{$node->child_nodes};  
   C: while (@node) {  
     my $child = shift @node;  
     unless (ref $child) {  
       if ($child eq 'cdata-out') {  
         $in_cdata = 0;  
       } else {  
         $s .= $child; # end tag  
       }  
       next C;  
     }  
       
     my $nt = $child->node_type;  
     if ($nt == 1) { # Element  
       my $tag_name = $child->tag_name; ## TODO: manakai_tag_name  
       $s .= '<' . $tag_name;  
       ## NOTE: Non-HTML case:  
       ## <http://permalink.gmane.org/gmane.org.w3c.whatwg.discuss/11191>  
   
       my @attrs = @{$child->attributes}; # sort order MUST be stable  
       for my $attr (@attrs) { # order is implementation dependent  
         my $attr_name = $attr->name; ## TODO: manakai_name  
         $s .= ' ' . $attr_name . '="';  
         my $attr_value = $attr->value;  
         ## escape  
         $attr_value =~ s/&/&amp;/g;  
         $attr_value =~ s/</&lt;/g;  
         $attr_value =~ s/>/&gt;/g;  
         $attr_value =~ s/"/&quot;/g;  
         $s .= $attr_value . '"';  
       }  
       $s .= '>';  
         
       next C if {  
         area => 1, base => 1, basefont => 1, bgsound => 1,  
         br => 1, col => 1, embed => 1, frame => 1, hr => 1,  
         img => 1, input => 1, link => 1, meta => 1, param => 1,  
         spacer => 1, wbr => 1,  
       }->{$tag_name};  
   
       $s .= "\x0A" if $tag_name eq 'pre' or $tag_name eq 'textarea';  
   
       if (not $in_cdata and {  
         style => 1, script => 1, xmp => 1, iframe => 1,  
         noembed => 1, noframes => 1, noscript => 1,  
         plaintext => 1,  
       }->{$tag_name}) {  
         unshift @node, 'cdata-out';  
         $in_cdata = 1;  
       }  
   
       unshift @node, @{$child->child_nodes}, '</' . $tag_name . '>';  
     } elsif ($nt == 3 or $nt == 4) {  
       if ($in_cdata) {  
         $s .= $child->data;  
       } else {  
         my $value = $child->data;  
         $value =~ s/&/&amp;/g;  
         $value =~ s/</&lt;/g;  
         $value =~ s/>/&gt;/g;  
         $value =~ s/"/&quot;/g;  
         $s .= $value;  
       }  
     } elsif ($nt == 8) {  
       $s .= '<!--' . $child->data . '-->';  
     } elsif ($nt == 10) {  
       $s .= '<!DOCTYPE ' . $child->name . '>';  
     } elsif ($nt == 5) { # entrefs  
       push @node, @{$child->child_nodes};  
     } else {  
       $on_error->($child) if defined $on_error;  
     }  
     ## ISSUE: This code does not support PIs.  
   } # C  
     
   ## Step 3  
   return \$s;  
 } # get_inner_html  
5632    
5633  1;  1;
5634  # $Date$  # $Date$
Legend:
Removed from v.1.55  
changed lines
  Added in v.1.72
[email protected]
 ViewVC Help
Powered by ViewVC 1.1.24