/[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 1.12 by wakaba, Sat Jun 23 04:38:50 2007 UTC revision 1.102 by wakaba, Sun Mar 9 06:29:25 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  ## This is an early version of an HTML parser.  ## ISSUE:
7    ## var doc = implementation.createDocument (null, null, null);
8    ## doc.write ('');
9    ## alert (doc.compatMode);
10    
11    ## TODO: Control charcters and noncharacters are not allowed (HTML5 revision 1263)
12    ## TODO: 1252 parse error (revision 1264)
13    ## TODO: 8859-11 = 874 (revision 1271)
14    
15  my $permitted_slash_tag_name = {  my $permitted_slash_tag_name = {
16    base => 1,    base => 1,
# Line 10  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 18  my $permitted_slash_tag_name = { Line 26  my $permitted_slash_tag_name = {
26    input => 1,    input => 1,
27  };  };
28    
 my $entity_char = {  
   AElig => "\x{00C6}",  
   Aacute => "\x{00C1}",  
   Acirc => "\x{00C2}",  
   Agrave => "\x{00C0}",  
   Alpha => "\x{0391}",  
   Aring => "\x{00C5}",  
   Atilde => "\x{00C3}",  
   Auml => "\x{00C4}",  
   Beta => "\x{0392}",  
   Ccedil => "\x{00C7}",  
   Chi => "\x{03A7}",  
   Dagger => "\x{2021}",  
   Delta => "\x{0394}",  
   ETH => "\x{00D0}",  
   Eacute => "\x{00C9}",  
   Ecirc => "\x{00CA}",  
   Egrave => "\x{00C8}",  
   Epsilon => "\x{0395}",  
   Eta => "\x{0397}",  
   Euml => "\x{00CB}",  
   Gamma => "\x{0393}",  
   Iacute => "\x{00CD}",  
   Icirc => "\x{00CE}",  
   Igrave => "\x{00CC}",  
   Iota => "\x{0399}",  
   Iuml => "\x{00CF}",  
   Kappa => "\x{039A}",  
   Lambda => "\x{039B}",  
   Mu => "\x{039C}",  
   Ntilde => "\x{00D1}",  
   Nu => "\x{039D}",  
   OElig => "\x{0152}",  
   Oacute => "\x{00D3}",  
   Ocirc => "\x{00D4}",  
   Ograve => "\x{00D2}",  
   Omega => "\x{03A9}",  
   Omicron => "\x{039F}",  
   Oslash => "\x{00D8}",  
   Otilde => "\x{00D5}",  
   Ouml => "\x{00D6}",  
   Phi => "\x{03A6}",  
   Pi => "\x{03A0}",  
   Prime => "\x{2033}",  
   Psi => "\x{03A8}",  
   Rho => "\x{03A1}",  
   Scaron => "\x{0160}",  
   Sigma => "\x{03A3}",  
   THORN => "\x{00DE}",  
   Tau => "\x{03A4}",  
   Theta => "\x{0398}",  
   Uacute => "\x{00DA}",  
   Ucirc => "\x{00DB}",  
   Ugrave => "\x{00D9}",  
   Upsilon => "\x{03A5}",  
   Uuml => "\x{00DC}",  
   Xi => "\x{039E}",  
   Yacute => "\x{00DD}",  
   Yuml => "\x{0178}",  
   Zeta => "\x{0396}",  
   aacute => "\x{00E1}",  
   acirc => "\x{00E2}",  
   acute => "\x{00B4}",  
   aelig => "\x{00E6}",  
   agrave => "\x{00E0}",  
   alefsym => "\x{2135}",  
   alpha => "\x{03B1}",  
   amp => "\x{0026}",  
   AMP => "\x{0026}",  
   and => "\x{2227}",  
   ang => "\x{2220}",  
   apos => "\x{0027}",  
   aring => "\x{00E5}",  
   asymp => "\x{2248}",  
   atilde => "\x{00E3}",  
   auml => "\x{00E4}",  
   bdquo => "\x{201E}",  
   beta => "\x{03B2}",  
   brvbar => "\x{00A6}",  
   bull => "\x{2022}",  
   cap => "\x{2229}",  
   ccedil => "\x{00E7}",  
   cedil => "\x{00B8}",  
   cent => "\x{00A2}",  
   chi => "\x{03C7}",  
   circ => "\x{02C6}",  
   clubs => "\x{2663}",  
   cong => "\x{2245}",  
   copy => "\x{00A9}",  
   COPY => "\x{00A9}",  
   crarr => "\x{21B5}",  
   cup => "\x{222A}",  
   curren => "\x{00A4}",  
   dArr => "\x{21D3}",  
   dagger => "\x{2020}",  
   darr => "\x{2193}",  
   deg => "\x{00B0}",  
   delta => "\x{03B4}",  
   diams => "\x{2666}",  
   divide => "\x{00F7}",  
   eacute => "\x{00E9}",  
   ecirc => "\x{00EA}",  
   egrave => "\x{00E8}",  
   empty => "\x{2205}",  
   emsp => "\x{2003}",  
   ensp => "\x{2002}",  
   epsilon => "\x{03B5}",  
   equiv => "\x{2261}",  
   eta => "\x{03B7}",  
   eth => "\x{00F0}",  
   euml => "\x{00EB}",  
   euro => "\x{20AC}",  
   exist => "\x{2203}",  
   fnof => "\x{0192}",  
   forall => "\x{2200}",  
   frac12 => "\x{00BD}",  
   frac14 => "\x{00BC}",  
   frac34 => "\x{00BE}",  
   frasl => "\x{2044}",  
   gamma => "\x{03B3}",  
   ge => "\x{2265}",  
   gt => "\x{003E}",  
   GT => "\x{003E}",  
   hArr => "\x{21D4}",  
   harr => "\x{2194}",  
   hearts => "\x{2665}",  
   hellip => "\x{2026}",  
   iacute => "\x{00ED}",  
   icirc => "\x{00EE}",  
   iexcl => "\x{00A1}",  
   igrave => "\x{00EC}",  
   image => "\x{2111}",  
   infin => "\x{221E}",  
   int => "\x{222B}",  
   iota => "\x{03B9}",  
   iquest => "\x{00BF}",  
   isin => "\x{2208}",  
   iuml => "\x{00EF}",  
   kappa => "\x{03BA}",  
   lArr => "\x{21D0}",  
   lambda => "\x{03BB}",  
   lang => "\x{2329}",  
   laquo => "\x{00AB}",  
   larr => "\x{2190}",  
   lceil => "\x{2308}",  
   ldquo => "\x{201C}",  
   le => "\x{2264}",  
   lfloor => "\x{230A}",  
   lowast => "\x{2217}",  
   loz => "\x{25CA}",  
   lrm => "\x{200E}",  
   lsaquo => "\x{2039}",  
   lsquo => "\x{2018}",  
   lt => "\x{003C}",  
   LT => "\x{003C}",  
   macr => "\x{00AF}",  
   mdash => "\x{2014}",  
   micro => "\x{00B5}",  
   middot => "\x{00B7}",  
   minus => "\x{2212}",  
   mu => "\x{03BC}",  
   nabla => "\x{2207}",  
   nbsp => "\x{00A0}",  
   ndash => "\x{2013}",  
   ne => "\x{2260}",  
   ni => "\x{220B}",  
   not => "\x{00AC}",  
   notin => "\x{2209}",  
   nsub => "\x{2284}",  
   ntilde => "\x{00F1}",  
   nu => "\x{03BD}",  
   oacute => "\x{00F3}",  
   ocirc => "\x{00F4}",  
   oelig => "\x{0153}",  
   ograve => "\x{00F2}",  
   oline => "\x{203E}",  
   omega => "\x{03C9}",  
   omicron => "\x{03BF}",  
   oplus => "\x{2295}",  
   or => "\x{2228}",  
   ordf => "\x{00AA}",  
   ordm => "\x{00BA}",  
   oslash => "\x{00F8}",  
   otilde => "\x{00F5}",  
   otimes => "\x{2297}",  
   ouml => "\x{00F6}",  
   para => "\x{00B6}",  
   part => "\x{2202}",  
   permil => "\x{2030}",  
   perp => "\x{22A5}",  
   phi => "\x{03C6}",  
   pi => "\x{03C0}",  
   piv => "\x{03D6}",  
   plusmn => "\x{00B1}",  
   pound => "\x{00A3}",  
   prime => "\x{2032}",  
   prod => "\x{220F}",  
   prop => "\x{221D}",  
   psi => "\x{03C8}",  
   quot => "\x{0022}",  
   QUOT => "\x{0022}",  
   rArr => "\x{21D2}",  
   radic => "\x{221A}",  
   rang => "\x{232A}",  
   raquo => "\x{00BB}",  
   rarr => "\x{2192}",  
   rceil => "\x{2309}",  
   rdquo => "\x{201D}",  
   real => "\x{211C}",  
   reg => "\x{00AE}",  
   REG => "\x{00AE}",  
   rfloor => "\x{230B}",  
   rho => "\x{03C1}",  
   rlm => "\x{200F}",  
   rsaquo => "\x{203A}",  
   rsquo => "\x{2019}",  
   sbquo => "\x{201A}",  
   scaron => "\x{0161}",  
   sdot => "\x{22C5}",  
   sect => "\x{00A7}",  
   shy => "\x{00AD}",  
   sigma => "\x{03C3}",  
   sigmaf => "\x{03C2}",  
   sim => "\x{223C}",  
   spades => "\x{2660}",  
   sub => "\x{2282}",  
   sube => "\x{2286}",  
   sum => "\x{2211}",  
   sup => "\x{2283}",  
   sup1 => "\x{00B9}",  
   sup2 => "\x{00B2}",  
   sup3 => "\x{00B3}",  
   supe => "\x{2287}",  
   szlig => "\x{00DF}",  
   tau => "\x{03C4}",  
   there4 => "\x{2234}",  
   theta => "\x{03B8}",  
   thetasym => "\x{03D1}",  
   thinsp => "\x{2009}",  
   thorn => "\x{00FE}",  
   tilde => "\x{02DC}",  
   times => "\x{00D7}",  
   trade => "\x{2122}",  
   uArr => "\x{21D1}",  
   uacute => "\x{00FA}",  
   uarr => "\x{2191}",  
   ucirc => "\x{00FB}",  
   ugrave => "\x{00F9}",  
   uml => "\x{00A8}",  
   upsih => "\x{03D2}",  
   upsilon => "\x{03C5}",  
   uuml => "\x{00FC}",  
   weierp => "\x{2118}",  
   xi => "\x{03BE}",  
   yacute => "\x{00FD}",  
   yen => "\x{00A5}",  
   yuml => "\x{00FF}",  
   zeta => "\x{03B6}",  
   zwj => "\x{200D}",  
   zwnj => "\x{200C}",  
 }; # $entity_char  
   
29  my $c1_entity_char = {  my $c1_entity_char = {
30    0x80 => 0x20AC,    0x80 => 0x20AC,
31    0x81 => 0xFFFD,    0x81 => 0xFFFD,
# Line 337  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;
182    $self->{set_next_input_character} = sub {    $self->{set_next_char} = sub {
183      my $self = shift;      my $self = shift;
184      $self->{next_input_character} = -1 and return if $i >= length $$s;  
185      $self->{next_input_character} = ord substr $$s, $i++, 1;      pop @{$self->{prev_char}};
186        unshift @{$self->{prev_char}}, $self->{next_char};
187    
188        $self->{next_char} = -1 and return if $i >= length $$s;
189        $self->{next_char} = ord substr $$s, $i++, 1;
190      $column++;      $column++;
191            
192      if ($self->{next_input_character} == 0x000A) { # LF      if ($self->{next_char} == 0x000A) { # LF
193        $line++;        $line++;
194        $column = 0;        $column = 0;
195      } elsif ($self->{next_input_character} == 0x000D) { # CR      } elsif ($self->{next_char} == 0x000D) { # CR
196        if ($i >= length $$s) {        $i++ if substr ($$s, $i, 1) eq "\x0A";
197          #        $self->{next_char} = 0x000A; # LF # MUST
       } else {  
         my $next_char = ord substr $$s, $i++, 1;  
         if ($next_char == 0x000A) { # LF  
           #  
         } else {  
           push @{$self->{char}}, $next_char;  
         }  
       }  
       $self->{next_input_character} = 0x000A; # LF # MUST  
198        $line++;        $line++;
199        $column = 0;        $column = 0;
200      } elsif ($self->{next_input_character} > 0x10FFFF) {      } elsif ($self->{next_char} > 0x10FFFF) {
201        $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST        $self->{next_char} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
202      } elsif ($self->{next_input_character} == 0x0000) { # NULL      } elsif ($self->{next_char} == 0x0000) { # NULL
203        !!!parse-error (type => 'NULL');        !!!parse-error (type => 'NULL');
204        $self->{next_input_character} = 0xFFFD; # REPLACEMENT CHARACTER # MUST        $self->{next_char} = 0xFFFD; # REPLACEMENT CHARACTER # MUST
205      }      }
206    };    };
207      $self->{prev_char} = [-1, -1, -1];
208      $self->{next_char} = -1;
209    
210    my $onerror = $_[2] || sub {    my $onerror = $_[2] || sub {
211      my (%opt) = @_;      my (%opt) = @_;
# Line 397  sub parse_string ($$$;$) { Line 226  sub parse_string ($$$;$) {
226  sub new ($) {  sub new ($) {
227    my $class = shift;    my $class = shift;
228    my $self = bless {}, $class;    my $self = bless {}, $class;
229    $self->{set_next_input_character} = sub {    $self->{set_next_char} = sub {
230      $self->{next_input_character} = -1;      $self->{next_char} = -1;
231    };    };
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    
247    sub CM_ENTITY () { 0b001 } # & markup in data
248    sub CM_LIMITED_MARKUP () { 0b010 } # < markup in data (limited)
249    sub CM_FULL_MARKUP () { 0b100 } # < markup in data (any)
250    
251    sub PLAINTEXT_CONTENT_MODEL () { 0 }
252    sub CDATA_CONTENT_MODEL () { CM_LIMITED_MARKUP }
253    sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP }
254    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 }
292    sub COMMENT_TOKEN () { 2 }
293    sub START_TAG_TOKEN () { 3 }
294    sub END_TAG_TOKEN () { 4 }
295    sub END_OF_FILE_TOKEN () { 5 }
296    sub CHARACTER_TOKEN () { 6 }
297    
298    sub AFTER_HTML_IMS () { 0b100 }
299    sub HEAD_IMS ()       { 0b1000 }
300    sub BODY_IMS ()       { 0b10000 }
301    sub BODY_TABLE_IMS () { 0b100000 }
302    sub TABLE_IMS ()      { 0b1000000 }
303    sub ROW_IMS ()        { 0b10000000 }
304    sub BODY_AFTER_IMS () { 0b100000000 }
305    sub FRAME_IMS ()      { 0b1000000000 }
306    sub SELECT_IMS ()     { 0b10000000000 }
307    
308    ## NOTE: "initial" and "before html" insertion modes have no constants.
309    
310    ## NOTE: "after after body" insertion mode.
311    sub AFTER_HTML_BODY_IM () { AFTER_HTML_IMS | BODY_AFTER_IMS }
312    
313    ## NOTE: "after after frameset" insertion mode.
314    sub AFTER_HTML_FRAMESET_IM () { AFTER_HTML_IMS | FRAME_IMS }
315    
316    sub IN_HEAD_IM () { HEAD_IMS | 0b00 }
317    sub IN_HEAD_NOSCRIPT_IM () { HEAD_IMS | 0b01 }
318    sub AFTER_HEAD_IM () { HEAD_IMS | 0b10 }
319    sub BEFORE_HEAD_IM () { HEAD_IMS | 0b11 }
320    sub IN_BODY_IM () { BODY_IMS }
321    sub IN_CELL_IM () { BODY_IMS | BODY_TABLE_IMS | 0b01 }
322    sub IN_CAPTION_IM () { BODY_IMS | BODY_TABLE_IMS | 0b10 }
323    sub IN_ROW_IM () { TABLE_IMS | ROW_IMS | 0b01 }
324    sub IN_TABLE_BODY_IM () { TABLE_IMS | ROW_IMS | 0b10 }
325    sub IN_TABLE_IM () { TABLE_IMS }
326    sub AFTER_BODY_IM () { BODY_AFTER_IMS }
327    sub IN_FRAMESET_IM () { FRAME_IMS | 0b01 }
328    sub AFTER_FRAMESET_IM () { FRAME_IMS | 0b10 }
329    sub IN_SELECT_IM () { SELECT_IMS | 0b01 }
330    sub IN_SELECT_IN_TABLE_IM () { SELECT_IMS | 0b10 }
331    sub IN_COLUMN_GROUP_IM () { 0b10 }
332    
333  ## Implementations MUST act as if state machine in the spec  ## Implementations MUST act as if state machine in the spec
334    
335  sub _initialize_tokenizer ($) {  sub _initialize_tokenizer ($) {
336    my $self = shift;    my $self = shift;
337    $self->{state} = 'data'; # MUST    $self->{state} = DATA_STATE; # MUST
338    $self->{content_model_flag} = 'PCDATA'; # be    $self->{content_model} = PCDATA_CONTENT_MODEL; # be
339    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE
340    undef $self->{current_attribute};    undef $self->{current_attribute};
341    undef $self->{last_emitted_start_tag_name};    undef $self->{last_emitted_start_tag_name};
342    undef $self->{last_attribute_value_state};    undef $self->{last_attribute_value_state};
343    $self->{char} = [];    $self->{char} = [];
344    # $self->{next_input_character}    # $self->{next_char}
345    !!!next-input-character;    !!!next-input-character;
346    $self->{token} = [];    $self->{token} = [];
347      # $self->{escape}
348  } # _initialize_tokenizer  } # _initialize_tokenizer
349    
350  ## A token has:  ## A token has:
351  ##   ->{type} eq 'DOCTYPE', 'start tag', 'end tag', 'comment',  ##   ->{type} == DOCTYPE_TOKEN, START_TAG_TOKEN, END_TAG_TOKEN, COMMENT_TOKEN,
352  ##       'character', or 'end-of-file'  ##       CHARACTER_TOKEN, or END_OF_FILE_TOKEN
353  ##   ->{name} (DOCTYPE, start tag (tagname), end tag (tagname))  ##   ->{name} (DOCTYPE_TOKEN)
354      ## ISSUE: the spec need s/tagname/tag name/  ##   ->{tag_name} (START_TAG_TOKEN, END_TAG_TOKEN)
355  ##   ->{error} == 1 or 0 (DOCTYPE)  ##   ->{public_identifier} (DOCTYPE_TOKEN)
356  ##   ->{attributes} isa HASH (start tag, end tag)  ##   ->{system_identifier} (DOCTYPE_TOKEN)
357  ##   ->{data} (comment, character)  ##   ->{quirks} == 1 or 0 (DOCTYPE_TOKEN): "force-quirks" flag
358    ##   ->{attributes} isa HASH (START_TAG_TOKEN, END_TAG_TOKEN)
359  ## Macros  ##        ->{name}
360  ##   Macros MUST be preceded by three EXCLAMATION MARKs.  ##        ->{value}
361  ##   emit ($token)  ##        ->{has_reference} == 1 or 0
362  ##     Emits the specified token.  ##   ->{data} (COMMENT_TOKEN, CHARACTER_TOKEN)
363    
364  ## Emitted token MUST immediately be handled by the tree construction state.  ## Emitted token MUST immediately be handled by the tree construction state.
365    
# Line 444  sub _initialize_tokenizer ($) { Line 369  sub _initialize_tokenizer ($) {
369  ## has completed loading.  If one has, then it MUST be executed  ## has completed loading.  If one has, then it MUST be executed
370  ## and removed from the list.  ## and removed from the list.
371    
372    ## NOTE: HTML5 "Writing HTML documents" section, applied to
373    ## documents and not to user agents and conformance checkers,
374    ## contains some requirements that are not detected by the
375    ## parsing algorithm:
376    ## - Some requirements on character encoding declarations. ## TODO
377    ## - "Elements MUST NOT contain content that their content model disallows."
378    ##   ... Some are parse error, some are not (will be reported by c.c.).
379    ## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO
380    ## - Text (in elements, attributes, and comments) SHOULD NOT contain
381    ##   control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL?  Unicode control character?)
382    
383    ## TODO: HTML5 poses authors two SHOULD-level requirements that cannot
384    ## be detected by the HTML5 parsing algorithm:
385    ## - Text,
386    
387  sub _get_next_token ($) {  sub _get_next_token ($) {
388    my $self = shift;    my $self = shift;
389    if (@{$self->{token}}) {    if (@{$self->{token}}) {
# Line 451  sub _get_next_token ($) { Line 391  sub _get_next_token ($) {
391    }    }
392    
393    A: {    A: {
394      if ($self->{state} eq 'data') {      if ($self->{state} == DATA_STATE) {
395        if ($self->{next_input_character} == 0x0026) { # &        if ($self->{next_char} == 0x0026) { # &
396          if ($self->{content_model_flag} eq 'PCDATA' or          if ($self->{content_model} & CM_ENTITY and # PCDATA | RCDATA
397              $self->{content_model_flag} eq 'RCDATA') {              not $self->{escape}) {
398            $self->{state} = 'entity data';            !!!cp (1);
399              $self->{state} = ENTITY_DATA_STATE;
400            !!!next-input-character;            !!!next-input-character;
401            redo A;            redo A;
402          } else {          } else {
403              !!!cp (2);
404            #            #
405          }          }
406        } elsif ($self->{next_input_character} == 0x003C) { # <        } elsif ($self->{next_char} == 0x002D) { # -
407          if ($self->{content_model_flag} ne 'PLAINTEXT') {          if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
408            $self->{state} = 'tag open';            unless ($self->{escape}) {
409                if ($self->{prev_char}->[0] == 0x002D and # -
410                    $self->{prev_char}->[1] == 0x0021 and # !
411                    $self->{prev_char}->[2] == 0x003C) { # <
412                  !!!cp (3);
413                  $self->{escape} = 1;
414                } else {
415                  !!!cp (4);
416                }
417              } else {
418                !!!cp (5);
419              }
420            }
421            
422            #
423          } elsif ($self->{next_char} == 0x003C) { # <
424            if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA
425                (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA
426                 not $self->{escape})) {
427              !!!cp (6);
428              $self->{state} = TAG_OPEN_STATE;
429            !!!next-input-character;            !!!next-input-character;
430            redo A;            redo A;
431          } else {          } else {
432              !!!cp (7);
433            #            #
434          }          }
435        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == 0x003E) { # >
436          !!!emit ({type => 'end-of-file'});          if ($self->{escape} and
437                ($self->{content_model} & CM_LIMITED_MARKUP)) { # RCDATA | CDATA
438              if ($self->{prev_char}->[0] == 0x002D and # -
439                  $self->{prev_char}->[1] == 0x002D) { # -
440                !!!cp (8);
441                delete $self->{escape};
442              } else {
443                !!!cp (9);
444              }
445            } else {
446              !!!cp (10);
447            }
448            
449            #
450          } elsif ($self->{next_char} == -1) {
451            !!!cp (11);
452            !!!emit ({type => END_OF_FILE_TOKEN});
453          last A; ## TODO: ok?          last A; ## TODO: ok?
454          } else {
455            !!!cp (12);
456        }        }
457        # Anything else        # Anything else
458        my $token = {type => 'character',        my $token = {type => CHARACTER_TOKEN,
459                     data => chr $self->{next_input_character}};                     data => chr $self->{next_char}};
460        ## Stay in the data state        ## Stay in the data state
461        !!!next-input-character;        !!!next-input-character;
462    
463        !!!emit ($token);        !!!emit ($token);
464    
465        redo A;        redo A;
466      } elsif ($self->{state} eq 'entity data') {      } elsif ($self->{state} == ENTITY_DATA_STATE) {
467        ## (cannot happen in CDATA state)        ## (cannot happen in CDATA state)
468                
469        my $token = $self->_tokenize_attempt_to_consume_an_entity;        my $token = $self->_tokenize_attempt_to_consume_an_entity (0, -1);
470    
471        $self->{state} = 'data';        $self->{state} = DATA_STATE;
472        # next-input-character is already done        # next-input-character is already done
473    
474        unless (defined $token) {        unless (defined $token) {
475          !!!emit ({type => 'character', data => '&'});          !!!cp (13);
476            !!!emit ({type => CHARACTER_TOKEN, data => '&'});
477        } else {        } else {
478            !!!cp (14);
479          !!!emit ($token);          !!!emit ($token);
480        }        }
481    
482        redo A;        redo A;
483      } elsif ($self->{state} eq 'tag open') {      } elsif ($self->{state} == TAG_OPEN_STATE) {
484        if ($self->{content_model_flag} eq 'RCDATA' or        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
485            $self->{content_model_flag} eq 'CDATA') {          if ($self->{next_char} == 0x002F) { # /
486          if ($self->{next_input_character} == 0x002F) { # /            !!!cp (15);
487            !!!next-input-character;            !!!next-input-character;
488            $self->{state} = 'close tag open';            $self->{state} = CLOSE_TAG_OPEN_STATE;
489            redo A;            redo A;
490          } else {          } else {
491              !!!cp (16);
492            ## reconsume            ## reconsume
493            $self->{state} = 'data';            $self->{state} = DATA_STATE;
494    
495            !!!emit ({type => 'character', data => '<'});            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
496    
497            redo A;            redo A;
498          }          }
499        } elsif ($self->{content_model_flag} eq 'PCDATA') {        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA
500          if ($self->{next_input_character} == 0x0021) { # !          if ($self->{next_char} == 0x0021) { # !
501            $self->{state} = 'markup declaration open';            !!!cp (17);
502              $self->{state} = MARKUP_DECLARATION_OPEN_STATE;
503            !!!next-input-character;            !!!next-input-character;
504            redo A;            redo A;
505          } elsif ($self->{next_input_character} == 0x002F) { # /          } elsif ($self->{next_char} == 0x002F) { # /
506            $self->{state} = 'close tag open';            !!!cp (18);
507              $self->{state} = CLOSE_TAG_OPEN_STATE;
508            !!!next-input-character;            !!!next-input-character;
509            redo A;            redo A;
510          } elsif (0x0041 <= $self->{next_input_character} and          } elsif (0x0041 <= $self->{next_char} and
511                   $self->{next_input_character} <= 0x005A) { # A..Z                   $self->{next_char} <= 0x005A) { # A..Z
512              !!!cp (19);
513            $self->{current_token}            $self->{current_token}
514              = {type => 'start tag',              = {type => START_TAG_TOKEN,
515                 tag_name => chr ($self->{next_input_character} + 0x0020)};                 tag_name => chr ($self->{next_char} + 0x0020)};
516            $self->{state} = 'tag name';            $self->{state} = TAG_NAME_STATE;
517            !!!next-input-character;            !!!next-input-character;
518            redo A;            redo A;
519          } elsif (0x0061 <= $self->{next_input_character} and          } elsif (0x0061 <= $self->{next_char} and
520                   $self->{next_input_character} <= 0x007A) { # a..z                   $self->{next_char} <= 0x007A) { # a..z
521            $self->{current_token} = {type => 'start tag',            !!!cp (20);
522                              tag_name => chr ($self->{next_input_character})};            $self->{current_token} = {type => START_TAG_TOKEN,
523            $self->{state} = 'tag name';                              tag_name => chr ($self->{next_char})};
524              $self->{state} = TAG_NAME_STATE;
525            !!!next-input-character;            !!!next-input-character;
526            redo A;            redo A;
527          } elsif ($self->{next_input_character} == 0x003E) { # >          } elsif ($self->{next_char} == 0x003E) { # >
528              !!!cp (21);
529            !!!parse-error (type => 'empty start tag');            !!!parse-error (type => 'empty start tag');
530            $self->{state} = 'data';            $self->{state} = DATA_STATE;
531            !!!next-input-character;            !!!next-input-character;
532    
533            !!!emit ({type => 'character', data => '<>'});            !!!emit ({type => CHARACTER_TOKEN, data => '<>'});
534    
535            redo A;            redo A;
536          } elsif ($self->{next_input_character} == 0x003F) { # ?          } elsif ($self->{next_char} == 0x003F) { # ?
537              !!!cp (22);
538            !!!parse-error (type => 'pio');            !!!parse-error (type => 'pio');
539            $self->{state} = 'bogus comment';            $self->{state} = BOGUS_COMMENT_STATE;
540            ## $self->{next_input_character} is intentionally left as is            ## $self->{next_char} is intentionally left as is
541            redo A;            redo A;
542          } else {          } else {
543              !!!cp (23);
544            !!!parse-error (type => 'bare stago');            !!!parse-error (type => 'bare stago');
545            $self->{state} = 'data';            $self->{state} = DATA_STATE;
546            ## reconsume            ## reconsume
547    
548            !!!emit ({type => 'character', data => '<'});            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
549    
550            redo A;            redo A;
551          }          }
552        } else {        } else {
553          die "$0: $self->{content_model_flag}: Unknown content model flag";          die "$0: $self->{content_model} in tag open";
554        }        }
555      } elsif ($self->{state} eq 'close tag open') {      } elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) {
556        if ($self->{content_model_flag} eq 'RCDATA' or        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
557            $self->{content_model_flag} eq 'CDATA') {          if (defined $self->{last_emitted_start_tag_name}) {
558          my @next_char;            ## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>
559          TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) {            my @next_char;
560            push @next_char, $self->{next_input_character};            TAGNAME: for (my $i = 0; $i < length $self->{last_emitted_start_tag_name}; $i++) {
561            my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1);              push @next_char, $self->{next_char};
562            my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c;              my $c = ord substr ($self->{last_emitted_start_tag_name}, $i, 1);
563            if ($self->{next_input_character} == $c or $self->{next_input_character} == $C) {              my $C = 0x0061 <= $c && $c <= 0x007A ? $c - 0x0020 : $c;
564              !!!next-input-character;              if ($self->{next_char} == $c or $self->{next_char} == $C) {
565              next TAGNAME;                !!!cp (24);
566            } else {                !!!next-input-character;
567              !!!parse-error (type => 'unmatched end tag');                next TAGNAME;
568              $self->{next_input_character} = shift @next_char; # reconsume              } else {
569              !!!back-next-input-character (@next_char);                !!!cp (25);
570              $self->{state} = 'data';                $self->{next_char} = shift @next_char; # reconsume
571                  !!!back-next-input-character (@next_char);
572              !!!emit ({type => 'character', data => '</'});                $self->{state} = DATA_STATE;
573    
574                  !!!emit ({type => CHARACTER_TOKEN, data => '</'});
575      
576                  redo A;
577                }
578              }
579              push @next_char, $self->{next_char};
580          
581              unless ($self->{next_char} == 0x0009 or # HT
582                      $self->{next_char} == 0x000A or # LF
583                      $self->{next_char} == 0x000B or # VT
584                      $self->{next_char} == 0x000C or # FF
585                      $self->{next_char} == 0x0020 or # SP
586                      $self->{next_char} == 0x003E or # >
587                      $self->{next_char} == 0x002F or # /
588                      $self->{next_char} == -1) {
589                !!!cp (26);
590                $self->{next_char} = shift @next_char; # reconsume
591                !!!back-next-input-character (@next_char);
592                $self->{state} = DATA_STATE;
593                !!!emit ({type => CHARACTER_TOKEN, data => '</'});
594              redo A;              redo A;
595              } else {
596                !!!cp (27);
597                $self->{next_char} = shift @next_char;
598                !!!back-next-input-character (@next_char);
599                # and consume...
600            }            }
         }  
         push @next_char, $self->{next_input_character};  
       
         unless ($self->{next_input_character} == 0x0009 or # HT  
                 $self->{next_input_character} == 0x000A or # LF  
                 $self->{next_input_character} == 0x000B or # VT  
                 $self->{next_input_character} == 0x000C or # FF  
                 $self->{next_input_character} == 0x0020 or # SP  
                 $self->{next_input_character} == 0x003E or # >  
                 $self->{next_input_character} == 0x002F or # /  
                 $self->{next_input_character} == 0x003C or # <  
                 $self->{next_input_character} == -1) {  
           !!!parse-error (type => 'unmatched end tag');  
           $self->{next_input_character} = shift @next_char; # reconsume  
           !!!back-next-input-character (@next_char);  
           $self->{state} = 'data';  
   
           !!!emit ({type => 'character', data => '</'});  
   
           redo A;  
601          } else {          } else {
602            $self->{next_input_character} = shift @next_char;            ## No start tag token has ever been emitted
603            !!!back-next-input-character (@next_char);            !!!cp (28);
604            # and consume...            # next-input-character is already done
605              $self->{state} = DATA_STATE;
606              !!!emit ({type => CHARACTER_TOKEN, data => '</'});
607              redo A;
608          }          }
609        }        }
610                
611        if (0x0041 <= $self->{next_input_character} and        if (0x0041 <= $self->{next_char} and
612            $self->{next_input_character} <= 0x005A) { # A..Z            $self->{next_char} <= 0x005A) { # A..Z
613          $self->{current_token} = {type => 'end tag',          !!!cp (29);
614                            tag_name => chr ($self->{next_input_character} + 0x0020)};          $self->{current_token} = {type => END_TAG_TOKEN,
615          $self->{state} = 'tag name';                            tag_name => chr ($self->{next_char} + 0x0020)};
616          !!!next-input-character;          $self->{state} = TAG_NAME_STATE;
617          redo A;          !!!next-input-character;
618        } elsif (0x0061 <= $self->{next_input_character} and          redo A;
619                 $self->{next_input_character} <= 0x007A) { # a..z        } elsif (0x0061 <= $self->{next_char} and
620          $self->{current_token} = {type => 'end tag',                 $self->{next_char} <= 0x007A) { # a..z
621                            tag_name => chr ($self->{next_input_character})};          !!!cp (30);
622          $self->{state} = 'tag name';          $self->{current_token} = {type => END_TAG_TOKEN,
623                              tag_name => chr ($self->{next_char})};
624            $self->{state} = TAG_NAME_STATE;
625          !!!next-input-character;          !!!next-input-character;
626          redo A;          redo A;
627        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
628            !!!cp (31);
629          !!!parse-error (type => 'empty end tag');          !!!parse-error (type => 'empty end tag');
630          $self->{state} = 'data';          $self->{state} = DATA_STATE;
631          !!!next-input-character;          !!!next-input-character;
632          redo A;          redo A;
633        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
634            !!!cp (32);
635          !!!parse-error (type => 'bare etago');          !!!parse-error (type => 'bare etago');
636          $self->{state} = 'data';          $self->{state} = DATA_STATE;
637          # reconsume          # reconsume
638    
639          !!!emit ({type => 'character', data => '</'});          !!!emit ({type => CHARACTER_TOKEN, data => '</'});
640    
641          redo A;          redo A;
642        } else {        } else {
643            !!!cp (33);
644          !!!parse-error (type => 'bogus end tag');          !!!parse-error (type => 'bogus end tag');
645          $self->{state} = 'bogus comment';          $self->{state} = BOGUS_COMMENT_STATE;
646          ## $self->{next_input_character} is intentionally left as is          ## $self->{next_char} is intentionally left as is
647          redo A;          redo A;
648        }        }
649      } elsif ($self->{state} eq 'tag name') {      } elsif ($self->{state} == TAG_NAME_STATE) {
650        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
651            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
652            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
653            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
654            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
655          $self->{state} = 'before attribute name';          !!!cp (34);
656          !!!next-input-character;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
657          redo A;          !!!next-input-character;
658        } elsif ($self->{next_input_character} == 0x003E) { # >          redo A;
659          if ($self->{current_token}->{type} eq 'start tag') {        } elsif ($self->{next_char} == 0x003E) { # >
660            if ($self->{current_token}->{type} == START_TAG_TOKEN) {
661              !!!cp (35);
662              $self->{current_token}->{first_start_tag}
663                  = not defined $self->{last_emitted_start_tag_name};
664            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
665          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
666            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
667            if ($self->{current_token}->{attributes}) {            #if ($self->{current_token}->{attributes}) {
668              !!!parse-error (type => 'end tag attribute');            #  ## NOTE: This should never be reached.
669            }            #  !!! cp (36);
670              #  !!! parse-error (type => 'end tag attribute');
671              #} else {
672                !!!cp (37);
673              #}
674          } else {          } else {
675            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
676          }          }
677          $self->{state} = 'data';          $self->{state} = DATA_STATE;
678          !!!next-input-character;          !!!next-input-character;
679    
680          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
681    
682          redo A;          redo A;
683        } elsif (0x0041 <= $self->{next_input_character} and        } elsif (0x0041 <= $self->{next_char} and
684                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_char} <= 0x005A) { # A..Z
685          $self->{current_token}->{tag_name} .= chr ($self->{next_input_character} + 0x0020);          !!!cp (38);
686            $self->{current_token}->{tag_name} .= chr ($self->{next_char} + 0x0020);
687            # start tag or end tag            # start tag or end tag
688          ## Stay in this state          ## Stay in this state
689          !!!next-input-character;          !!!next-input-character;
690          redo A;          redo A;
691        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
692          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
693          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
694              !!!cp (39);
695              $self->{current_token}->{first_start_tag}
696                  = not defined $self->{last_emitted_start_tag_name};
697            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
698          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
699            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
700            if ($self->{current_token}->{attributes}) {            #if ($self->{current_token}->{attributes}) {
701              !!!parse-error (type => 'end tag attribute');            #  ## NOTE: This state should never be reached.
702            }            #  !!! cp (40);
703              #  !!! parse-error (type => 'end tag attribute');
704              #} else {
705                !!!cp (41);
706              #}
707          } else {          } else {
708            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
709          }          }
710          $self->{state} = 'data';          $self->{state} = DATA_STATE;
711          # reconsume          # reconsume
712    
713          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
714    
715          redo A;          redo A;
716        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_char} == 0x002F) { # /
717          !!!next-input-character;          !!!next-input-character;
718          if ($self->{next_input_character} == 0x003E and # >          if ($self->{next_char} == 0x003E and # >
719              $self->{current_token}->{type} eq 'start tag' and              $self->{current_token}->{type} == START_TAG_TOKEN and
720              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
721            # permitted slash            # permitted slash
722              !!!cp (42);
723            #            #
724          } else {          } else {
725              !!!cp (43);
726            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
727          }          }
728          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
729          # next-input-character is already done          # next-input-character is already done
730          redo A;          redo A;
731        } else {        } else {
732          $self->{current_token}->{tag_name} .= chr $self->{next_input_character};          !!!cp (44);
733            $self->{current_token}->{tag_name} .= chr $self->{next_char};
734            # start tag or end tag            # start tag or end tag
735          ## Stay in the state          ## Stay in the state
736          !!!next-input-character;          !!!next-input-character;
737          redo A;          redo A;
738        }        }
739      } elsif ($self->{state} eq 'before attribute name') {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) {
740        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
741            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
742            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
743            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
744            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
745            !!!cp (45);
746          ## Stay in the state          ## Stay in the state
747          !!!next-input-character;          !!!next-input-character;
748          redo A;          redo A;
749        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
750          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
751              !!!cp (46);
752              $self->{current_token}->{first_start_tag}
753                  = not defined $self->{last_emitted_start_tag_name};
754            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
755          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
756            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
757            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
758                !!!cp (47);
759              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
760              } else {
761                !!!cp (48);
762            }            }
763          } else {          } else {
764            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
765          }          }
766          $self->{state} = 'data';          $self->{state} = DATA_STATE;
767          !!!next-input-character;          !!!next-input-character;
768    
769          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
770    
771          redo A;          redo A;
772        } elsif (0x0041 <= $self->{next_input_character} and        } elsif (0x0041 <= $self->{next_char} and
773                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_char} <= 0x005A) { # A..Z
774          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),          !!!cp (49);
775            $self->{current_attribute} = {name => chr ($self->{next_char} + 0x0020),
776                                value => ''};                                value => ''};
777          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
778          !!!next-input-character;          !!!next-input-character;
779          redo A;          redo A;
780        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_char} == 0x002F) { # /
781          !!!next-input-character;          !!!next-input-character;
782          if ($self->{next_input_character} == 0x003E and # >          if ($self->{next_char} == 0x003E and # >
783              $self->{current_token}->{type} eq 'start tag' and              $self->{current_token}->{type} == START_TAG_TOKEN and
784              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
785            # permitted slash            # permitted slash
786              !!!cp (50);
787            #            #
788          } else {          } else {
789              !!!cp (51);
790            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
791          }          }
792          ## Stay in the state          ## Stay in the state
793          # next-input-character is already done          # next-input-character is already done
794          redo A;          redo A;
795        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
796          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
797          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
798              !!!cp (52);
799              $self->{current_token}->{first_start_tag}
800                  = not defined $self->{last_emitted_start_tag_name};
801            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
802          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
803            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
804            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
805                !!!cp (53);
806              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
807              } else {
808                !!!cp (54);
809            }            }
810          } else {          } else {
811            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
812          }          }
813          $self->{state} = 'data';          $self->{state} = DATA_STATE;
814          # reconsume          # reconsume
815    
816          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
817    
818          redo A;          redo A;
819        } else {        } else {
820          $self->{current_attribute} = {name => chr ($self->{next_input_character}),          if ({
821                 0x0022 => 1, # "
822                 0x0027 => 1, # '
823                 0x003D => 1, # =
824                }->{$self->{next_char}}) {
825              !!!cp (55);
826              !!!parse-error (type => 'bad attribute name');
827            } else {
828              !!!cp (56);
829            }
830            $self->{current_attribute} = {name => chr ($self->{next_char}),
831                                value => ''};                                value => ''};
832          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
833          !!!next-input-character;          !!!next-input-character;
834          redo A;          redo A;
835        }        }
836      } elsif ($self->{state} eq 'attribute name') {      } elsif ($self->{state} == ATTRIBUTE_NAME_STATE) {
837        my $before_leave = sub {        my $before_leave = sub {
838          if (exists $self->{current_token}->{attributes} # start tag or end tag          if (exists $self->{current_token}->{attributes} # start tag or end tag
839              ->{$self->{current_attribute}->{name}}) { # MUST              ->{$self->{current_attribute}->{name}}) { # MUST
840            !!!parse-error (type => 'dupulicate attribute');            !!!cp (57);
841              !!!parse-error (type => 'duplicate attribute:'.$self->{current_attribute}->{name});
842            ## Discard $self->{current_attribute} # MUST            ## Discard $self->{current_attribute} # MUST
843          } else {          } else {
844              !!!cp (58);
845            $self->{current_token}->{attributes}->{$self->{current_attribute}->{name}}            $self->{current_token}->{attributes}->{$self->{current_attribute}->{name}}
846              = $self->{current_attribute};              = $self->{current_attribute};
847          }          }
848        }; # $before_leave        }; # $before_leave
849    
850        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
851            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
852            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
853            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
854            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
855            !!!cp (59);
856          $before_leave->();          $before_leave->();
857          $self->{state} = 'after attribute name';          $self->{state} = AFTER_ATTRIBUTE_NAME_STATE;
858          !!!next-input-character;          !!!next-input-character;
859          redo A;          redo A;
860        } elsif ($self->{next_input_character} == 0x003D) { # =        } elsif ($self->{next_char} == 0x003D) { # =
861            !!!cp (60);
862          $before_leave->();          $before_leave->();
863          $self->{state} = 'before attribute value';          $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
864          !!!next-input-character;          !!!next-input-character;
865          redo A;          redo A;
866        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
867          $before_leave->();          $before_leave->();
868          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
869              !!!cp (61);
870              $self->{current_token}->{first_start_tag}
871                  = not defined $self->{last_emitted_start_tag_name};
872            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
873          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
874            $self->{content_model_flag} = 'PCDATA'; # MUST            !!!cp (62);
875              $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
876            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
877              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
878            }            }
879          } else {          } else {
880            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
881          }          }
882          $self->{state} = 'data';          $self->{state} = DATA_STATE;
883          !!!next-input-character;          !!!next-input-character;
884    
885          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
886    
887          redo A;          redo A;
888        } elsif (0x0041 <= $self->{next_input_character} and        } elsif (0x0041 <= $self->{next_char} and
889                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_char} <= 0x005A) { # A..Z
890          $self->{current_attribute}->{name} .= chr ($self->{next_input_character} + 0x0020);          !!!cp (63);
891            $self->{current_attribute}->{name} .= chr ($self->{next_char} + 0x0020);
892          ## Stay in the state          ## Stay in the state
893          !!!next-input-character;          !!!next-input-character;
894          redo A;          redo A;
895        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_char} == 0x002F) { # /
896          $before_leave->();          $before_leave->();
897          !!!next-input-character;          !!!next-input-character;
898          if ($self->{next_input_character} == 0x003E and # >          if ($self->{next_char} == 0x003E and # >
899              $self->{current_token}->{type} eq 'start tag' and              $self->{current_token}->{type} == START_TAG_TOKEN and
900              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
901            # permitted slash            # permitted slash
902              !!!cp (64);
903            #            #
904          } else {          } else {
905              !!!cp (65);
906            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
907          }          }
908          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
909          # next-input-character is already done          # next-input-character is already done
910          redo A;          redo A;
911        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
912          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
913          $before_leave->();          $before_leave->();
914          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
915              !!!cp (66);
916              $self->{current_token}->{first_start_tag}
917                  = not defined $self->{last_emitted_start_tag_name};
918            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
919          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
920            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
921            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
922                !!!cp (67);
923              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
924              } else {
925                ## NOTE: This state should never be reached.
926                !!!cp (68);
927            }            }
928          } else {          } else {
929            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
930          }          }
931          $self->{state} = 'data';          $self->{state} = DATA_STATE;
932          # reconsume          # reconsume
933    
934          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
935    
936          redo A;          redo A;
937        } else {        } else {
938          $self->{current_attribute}->{name} .= chr ($self->{next_input_character});          if ($self->{next_char} == 0x0022 or # "
939                $self->{next_char} == 0x0027) { # '
940              !!!cp (69);
941              !!!parse-error (type => 'bad attribute name');
942            } else {
943              !!!cp (70);
944            }
945            $self->{current_attribute}->{name} .= chr ($self->{next_char});
946          ## Stay in the state          ## Stay in the state
947          !!!next-input-character;          !!!next-input-character;
948          redo A;          redo A;
949        }        }
950      } elsif ($self->{state} eq 'after attribute name') {      } elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) {
951        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
952            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
953            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
954            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
955            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
956            !!!cp (71);
957          ## Stay in the state          ## Stay in the state
958          !!!next-input-character;          !!!next-input-character;
959          redo A;          redo A;
960        } elsif ($self->{next_input_character} == 0x003D) { # =        } elsif ($self->{next_char} == 0x003D) { # =
961          $self->{state} = 'before attribute value';          !!!cp (72);
962            $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
963          !!!next-input-character;          !!!next-input-character;
964          redo A;          redo A;
965        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
966          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
967              !!!cp (73);
968              $self->{current_token}->{first_start_tag}
969                  = not defined $self->{last_emitted_start_tag_name};
970            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
971          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
972            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
973            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
974                !!!cp (74);
975              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
976              } else {
977                ## NOTE: This state should never be reached.
978                !!!cp (75);
979            }            }
980          } else {          } else {
981            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
982          }          }
983          $self->{state} = 'data';          $self->{state} = DATA_STATE;
984          !!!next-input-character;          !!!next-input-character;
985    
986          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
987    
988          redo A;          redo A;
989        } elsif (0x0041 <= $self->{next_input_character} and        } elsif (0x0041 <= $self->{next_char} and
990                 $self->{next_input_character} <= 0x005A) { # A..Z                 $self->{next_char} <= 0x005A) { # A..Z
991          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),          !!!cp (76);
992            $self->{current_attribute} = {name => chr ($self->{next_char} + 0x0020),
993                                value => ''};                                value => ''};
994          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
995          !!!next-input-character;          !!!next-input-character;
996          redo A;          redo A;
997        } elsif ($self->{next_input_character} == 0x002F) { # /        } elsif ($self->{next_char} == 0x002F) { # /
998          !!!next-input-character;          !!!next-input-character;
999          if ($self->{next_input_character} == 0x003E and # >          if ($self->{next_char} == 0x003E and # >
1000              $self->{current_token}->{type} eq 'start tag' and              $self->{current_token}->{type} == START_TAG_TOKEN and
1001              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {              $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
1002            # permitted slash            # permitted slash
1003              !!!cp (77);
1004            #            #
1005          } else {          } else {
1006              !!!cp (78);
1007            !!!parse-error (type => 'nestc');            !!!parse-error (type => 'nestc');
1008              ## TODO: Different error type for <aa / bb> than <aa/>
1009          }          }
1010          $self->{state} = 'before attribute name';          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1011          # next-input-character is already done          # next-input-character is already done
1012          redo A;          redo A;
1013        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
1014          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1015          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1016              !!!cp (79);
1017              $self->{current_token}->{first_start_tag}
1018                  = not defined $self->{last_emitted_start_tag_name};
1019            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1020          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1021            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1022            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1023                !!!cp (80);
1024              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1025              } else {
1026                ## NOTE: This state should never be reached.
1027                !!!cp (81);
1028            }            }
1029          } else {          } else {
1030            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1031          }          }
1032          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1033          # reconsume          # reconsume
1034    
1035          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1036    
1037          redo A;          redo A;
1038        } else {        } else {
1039          $self->{current_attribute} = {name => chr ($self->{next_input_character}),          !!!cp (82);
1040            $self->{current_attribute} = {name => chr ($self->{next_char}),
1041                                value => ''};                                value => ''};
1042          $self->{state} = 'attribute name';          $self->{state} = ATTRIBUTE_NAME_STATE;
1043          !!!next-input-character;          !!!next-input-character;
1044          redo A;                  redo A;        
1045        }        }
1046      } elsif ($self->{state} eq 'before attribute value') {      } elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) {
1047        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
1048            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
1049            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
1050            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
1051            $self->{next_input_character} == 0x0020) { # SP                  $self->{next_char} == 0x0020) { # SP      
1052            !!!cp (83);
1053          ## Stay in the state          ## Stay in the state
1054          !!!next-input-character;          !!!next-input-character;
1055          redo A;          redo A;
1056        } elsif ($self->{next_input_character} == 0x0022) { # "        } elsif ($self->{next_char} == 0x0022) { # "
1057          $self->{state} = 'attribute value (double-quoted)';          !!!cp (84);
1058            $self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE;
1059          !!!next-input-character;          !!!next-input-character;
1060          redo A;          redo A;
1061        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_char} == 0x0026) { # &
1062          $self->{state} = 'attribute value (unquoted)';          !!!cp (85);
1063            $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
1064          ## reconsume          ## reconsume
1065          redo A;          redo A;
1066        } elsif ($self->{next_input_character} == 0x0027) { # '        } elsif ($self->{next_char} == 0x0027) { # '
1067          $self->{state} = 'attribute value (single-quoted)';          !!!cp (86);
1068            $self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE;
1069          !!!next-input-character;          !!!next-input-character;
1070          redo A;          redo A;
1071        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
1072          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1073              !!!cp (87);
1074              $self->{current_token}->{first_start_tag}
1075                  = not defined $self->{last_emitted_start_tag_name};
1076            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1077          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1078            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1079            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1080                !!!cp (88);
1081              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1082              } else {
1083                ## NOTE: This state should never be reached.
1084                !!!cp (89);
1085            }            }
1086          } else {          } else {
1087            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1088          }          }
1089          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1090          !!!next-input-character;          !!!next-input-character;
1091    
1092          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1093    
1094          redo A;          redo A;
1095        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
1096          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1097          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1098              !!!cp (90);
1099              $self->{current_token}->{first_start_tag}
1100                  = not defined $self->{last_emitted_start_tag_name};
1101            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1102          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1103            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1104            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1105                !!!cp (91);
1106              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1107              } else {
1108                ## NOTE: This state should never be reached.
1109                !!!cp (92);
1110            }            }
1111          } else {          } else {
1112            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1113          }          }
1114          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1115          ## reconsume          ## reconsume
1116    
1117          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1118    
1119          redo A;          redo A;
1120        } else {        } else {
1121          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          if ($self->{next_char} == 0x003D) { # =
1122          $self->{state} = 'attribute value (unquoted)';            !!!cp (93);
1123              !!!parse-error (type => 'bad attribute value');
1124            } else {
1125              !!!cp (94);
1126            }
1127            $self->{current_attribute}->{value} .= chr ($self->{next_char});
1128            $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
1129          !!!next-input-character;          !!!next-input-character;
1130          redo A;          redo A;
1131        }        }
1132      } elsif ($self->{state} eq 'attribute value (double-quoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) {
1133        if ($self->{next_input_character} == 0x0022) { # "        if ($self->{next_char} == 0x0022) { # "
1134          $self->{state} = 'before attribute name';          !!!cp (95);
1135            $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1136          !!!next-input-character;          !!!next-input-character;
1137          redo A;          redo A;
1138        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_char} == 0x0026) { # &
1139          $self->{last_attribute_value_state} = 'attribute value (double-quoted)';          !!!cp (96);
1140          $self->{state} = 'entity in attribute value';          $self->{last_attribute_value_state} = $self->{state};
1141            $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1142          !!!next-input-character;          !!!next-input-character;
1143          redo A;          redo A;
1144        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1145          !!!parse-error (type => 'unclosed attribute value');          !!!parse-error (type => 'unclosed attribute value');
1146          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1147              !!!cp (97);
1148              $self->{current_token}->{first_start_tag}
1149                  = not defined $self->{last_emitted_start_tag_name};
1150            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1151          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1152            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1153            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1154                !!!cp (98);
1155              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1156              } else {
1157                ## NOTE: This state should never be reached.
1158                !!!cp (99);
1159            }            }
1160          } else {          } else {
1161            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1162          }          }
1163          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1164          ## reconsume          ## reconsume
1165    
1166          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1167    
1168          redo A;          redo A;
1169        } else {        } else {
1170          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          !!!cp (100);
1171            $self->{current_attribute}->{value} .= chr ($self->{next_char});
1172          ## Stay in the state          ## Stay in the state
1173          !!!next-input-character;          !!!next-input-character;
1174          redo A;          redo A;
1175        }        }
1176      } elsif ($self->{state} eq 'attribute value (single-quoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) {
1177        if ($self->{next_input_character} == 0x0027) { # '        if ($self->{next_char} == 0x0027) { # '
1178          $self->{state} = 'before attribute name';          !!!cp (101);
1179            $self->{state} = AFTER_ATTRIBUTE_VALUE_QUOTED_STATE;
1180          !!!next-input-character;          !!!next-input-character;
1181          redo A;          redo A;
1182        } elsif ($self->{next_input_character} == 0x0026) { # &        } elsif ($self->{next_char} == 0x0026) { # &
1183          $self->{last_attribute_value_state} = 'attribute value (single-quoted)';          !!!cp (102);
1184          $self->{state} = 'entity in attribute value';          $self->{last_attribute_value_state} = $self->{state};
1185            $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1186          !!!next-input-character;          !!!next-input-character;
1187          redo A;          redo A;
1188        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1189          !!!parse-error (type => 'unclosed attribute value');          !!!parse-error (type => 'unclosed attribute value');
1190          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1191              !!!cp (103);
1192              $self->{current_token}->{first_start_tag}
1193                  = not defined $self->{last_emitted_start_tag_name};
1194            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1195          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1196            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1197            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1198                !!!cp (104);
1199              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1200              } else {
1201                ## NOTE: This state should never be reached.
1202                !!!cp (105);
1203            }            }
1204          } else {          } else {
1205            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1206          }          }
1207          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1208          ## reconsume          ## reconsume
1209    
1210          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1211    
1212          redo A;          redo A;
1213        } else {        } else {
1214          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          !!!cp (106);
1215            $self->{current_attribute}->{value} .= chr ($self->{next_char});
1216          ## Stay in the state          ## Stay in the state
1217          !!!next-input-character;          !!!next-input-character;
1218          redo A;          redo A;
1219        }        }
1220      } elsif ($self->{state} eq 'attribute value (unquoted)') {      } elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) {
1221        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
1222            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
1223            $self->{next_input_character} == 0x000B or # HT            $self->{next_char} == 0x000B or # HT
1224            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
1225            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
1226          $self->{state} = 'before attribute name';          !!!cp (107);
1227          !!!next-input-character;          $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1228          redo A;          !!!next-input-character;
1229        } elsif ($self->{next_input_character} == 0x0026) { # &          redo A;
1230          $self->{last_attribute_value_state} = 'attribute value (unquoted)';        } elsif ($self->{next_char} == 0x0026) { # &
1231          $self->{state} = 'entity in attribute value';          !!!cp (108);
1232          !!!next-input-character;          $self->{last_attribute_value_state} = $self->{state};
1233          redo A;          $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
1234        } elsif ($self->{next_input_character} == 0x003E) { # >          !!!next-input-character;
1235          if ($self->{current_token}->{type} eq 'start tag') {          redo A;
1236          } elsif ($self->{next_char} == 0x003E) { # >
1237            if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1238              !!!cp (109);
1239              $self->{current_token}->{first_start_tag}
1240                  = not defined $self->{last_emitted_start_tag_name};
1241            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1242          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1243            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1244            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1245                !!!cp (110);
1246              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1247              } else {
1248                ## NOTE: This state should never be reached.
1249                !!!cp (111);
1250            }            }
1251          } else {          } else {
1252            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1253          }          }
1254          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1255          !!!next-input-character;          !!!next-input-character;
1256    
1257          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1258    
1259          redo A;          redo A;
1260        } elsif ($self->{next_input_character} == 0x003C or # <        } elsif ($self->{next_char} == -1) {
                $self->{next_input_character} == -1) {  
1261          !!!parse-error (type => 'unclosed tag');          !!!parse-error (type => 'unclosed tag');
1262          if ($self->{current_token}->{type} eq 'start tag') {          if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1263              !!!cp (112);
1264              $self->{current_token}->{first_start_tag}
1265                  = not defined $self->{last_emitted_start_tag_name};
1266            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};            $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1267          } elsif ($self->{current_token}->{type} eq 'end tag') {          } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1268            $self->{content_model_flag} = 'PCDATA'; # MUST            $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1269            if ($self->{current_token}->{attributes}) {            if ($self->{current_token}->{attributes}) {
1270                !!!cp (113);
1271              !!!parse-error (type => 'end tag attribute');              !!!parse-error (type => 'end tag attribute');
1272              } else {
1273                ## NOTE: This state should never be reached.
1274                !!!cp (114);
1275            }            }
1276          } else {          } else {
1277            die "$0: $self->{current_token}->{type}: Unknown token type";            die "$0: $self->{current_token}->{type}: Unknown token type";
1278          }          }
1279          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1280          ## reconsume          ## reconsume
1281    
1282          !!!emit ($self->{current_token}); # start tag or end tag          !!!emit ($self->{current_token}); # start tag or end tag
         undef $self->{current_token};  
1283    
1284          redo A;          redo A;
1285        } else {        } else {
1286          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});          if ({
1287                 0x0022 => 1, # "
1288                 0x0027 => 1, # '
1289                 0x003D => 1, # =
1290                }->{$self->{next_char}}) {
1291              !!!cp (115);
1292              !!!parse-error (type => 'bad attribute value');
1293            } else {
1294              !!!cp (116);
1295            }
1296            $self->{current_attribute}->{value} .= chr ($self->{next_char});
1297          ## Stay in the state          ## Stay in the state
1298          !!!next-input-character;          !!!next-input-character;
1299          redo A;          redo A;
1300        }        }
1301      } elsif ($self->{state} eq 'entity in attribute value') {      } elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) {
1302        my $token = $self->_tokenize_attempt_to_consume_an_entity;        my $token = $self->_tokenize_attempt_to_consume_an_entity
1303              (1,
1304               $self->{last_attribute_value_state}
1305                 == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE ? 0x0022 : # "
1306               $self->{last_attribute_value_state}
1307                 == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE ? 0x0027 : # '
1308               -1);
1309    
1310        unless (defined $token) {        unless (defined $token) {
1311            !!!cp (117);
1312          $self->{current_attribute}->{value} .= '&';          $self->{current_attribute}->{value} .= '&';
1313        } else {        } else {
1314            !!!cp (118);
1315          $self->{current_attribute}->{value} .= $token->{data};          $self->{current_attribute}->{value} .= $token->{data};
1316            $self->{current_attribute}->{has_reference} = $token->{has_reference};
1317          ## 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"
1318        }        }
1319    
1320        $self->{state} = $self->{last_attribute_value_state};        $self->{state} = $self->{last_attribute_value_state};
1321        # next-input-character is already done        # next-input-character is already done
1322        redo A;        redo A;
1323      } elsif ($self->{state} eq 'bogus comment') {      } elsif ($self->{state} == AFTER_ATTRIBUTE_VALUE_QUOTED_STATE) {
1324          if ($self->{next_char} == 0x0009 or # HT
1325              $self->{next_char} == 0x000A or # LF
1326              $self->{next_char} == 0x000B or # VT
1327              $self->{next_char} == 0x000C or # FF
1328              $self->{next_char} == 0x0020) { # SP
1329            !!!cp (118);
1330            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1331            !!!next-input-character;
1332            redo A;
1333          } elsif ($self->{next_char} == 0x003E) { # >
1334            if ($self->{current_token}->{type} == START_TAG_TOKEN) {
1335              !!!cp (119);
1336              $self->{current_token}->{first_start_tag}
1337                  = not defined $self->{last_emitted_start_tag_name};
1338              $self->{last_emitted_start_tag_name} = $self->{current_token}->{tag_name};
1339            } elsif ($self->{current_token}->{type} == END_TAG_TOKEN) {
1340              $self->{content_model} = PCDATA_CONTENT_MODEL; # MUST
1341              if ($self->{current_token}->{attributes}) {
1342                !!!cp (120);
1343                !!!parse-error (type => 'end tag attribute');
1344              } else {
1345                ## NOTE: This state should never be reached.
1346                !!!cp (121);
1347              }
1348            } else {
1349              die "$0: $self->{current_token}->{type}: Unknown token type";
1350            }
1351            $self->{state} = DATA_STATE;
1352            !!!next-input-character;
1353    
1354            !!!emit ($self->{current_token}); # start tag or end tag
1355    
1356            redo A;
1357          } elsif ($self->{next_char} == 0x002F) { # /
1358            !!!next-input-character;
1359            if ($self->{next_char} == 0x003E and # >
1360                $self->{current_token}->{type} == START_TAG_TOKEN and
1361                $permitted_slash_tag_name->{$self->{current_token}->{tag_name}}) {
1362              # permitted slash
1363              !!!cp (122);
1364              #
1365            } else {
1366              !!!cp (123);
1367              !!!parse-error (type => 'nestc');
1368            }
1369            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1370            # next-input-character is already done
1371            redo A;
1372          } else {
1373            !!!cp (124);
1374            !!!parse-error (type => 'no space between attributes');
1375            $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
1376            ## reconsume
1377            redo A;
1378          }
1379        } elsif ($self->{state} == BOGUS_COMMENT_STATE) {
1380        ## (only happen if PCDATA state)        ## (only happen if PCDATA state)
1381                
1382        my $token = {type => 'comment', data => ''};        my $token = {type => COMMENT_TOKEN, data => ''};
1383    
1384        BC: {        BC: {
1385          if ($self->{next_input_character} == 0x003E) { # >          if ($self->{next_char} == 0x003E) { # >
1386            $self->{state} = 'data';            !!!cp (124);
1387              $self->{state} = DATA_STATE;
1388            !!!next-input-character;            !!!next-input-character;
1389    
1390            !!!emit ($token);            !!!emit ($token);
1391    
1392            redo A;            redo A;
1393          } elsif ($self->{next_input_character} == -1) {          } elsif ($self->{next_char} == -1) {
1394            $self->{state} = 'data';            !!!cp (125);
1395              $self->{state} = DATA_STATE;
1396            ## reconsume            ## reconsume
1397    
1398            !!!emit ($token);            !!!emit ($token);
1399    
1400            redo A;            redo A;
1401          } else {          } else {
1402            $token->{data} .= chr ($self->{next_input_character});            !!!cp (126);
1403              $token->{data} .= chr ($self->{next_char});
1404            !!!next-input-character;            !!!next-input-character;
1405            redo BC;            redo BC;
1406          }          }
1407        } # BC        } # BC
1408      } elsif ($self->{state} eq 'markup declaration open') {  
1409          die "$0: _get_next_token: unexpected case [BC]";
1410        } elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) {
1411        ## (only happen if PCDATA state)        ## (only happen if PCDATA state)
1412    
1413        my @next_char;        my @next_char;
1414        push @next_char, $self->{next_input_character};        push @next_char, $self->{next_char};
1415                
1416        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_char} == 0x002D) { # -
1417          !!!next-input-character;          !!!next-input-character;
1418          push @next_char, $self->{next_input_character};          push @next_char, $self->{next_char};
1419          if ($self->{next_input_character} == 0x002D) { # -          if ($self->{next_char} == 0x002D) { # -
1420            $self->{current_token} = {type => 'comment', data => ''};            !!!cp (127);
1421            $self->{state} = 'comment';            $self->{current_token} = {type => COMMENT_TOKEN, data => ''};
1422              $self->{state} = COMMENT_START_STATE;
1423            !!!next-input-character;            !!!next-input-character;
1424            redo A;            redo A;
1425            } else {
1426              !!!cp (128);
1427          }          }
1428        } elsif ($self->{next_input_character} == 0x0044 or # D        } elsif ($self->{next_char} == 0x0044 or # D
1429                 $self->{next_input_character} == 0x0064) { # d                 $self->{next_char} == 0x0064) { # d
1430          !!!next-input-character;          !!!next-input-character;
1431          push @next_char, $self->{next_input_character};          push @next_char, $self->{next_char};
1432          if ($self->{next_input_character} == 0x004F or # O          if ($self->{next_char} == 0x004F or # O
1433              $self->{next_input_character} == 0x006F) { # o              $self->{next_char} == 0x006F) { # o
1434            !!!next-input-character;            !!!next-input-character;
1435            push @next_char, $self->{next_input_character};            push @next_char, $self->{next_char};
1436            if ($self->{next_input_character} == 0x0043 or # C            if ($self->{next_char} == 0x0043 or # C
1437                $self->{next_input_character} == 0x0063) { # c                $self->{next_char} == 0x0063) { # c
1438              !!!next-input-character;              !!!next-input-character;
1439              push @next_char, $self->{next_input_character};              push @next_char, $self->{next_char};
1440              if ($self->{next_input_character} == 0x0054 or # T              if ($self->{next_char} == 0x0054 or # T
1441                  $self->{next_input_character} == 0x0074) { # t                  $self->{next_char} == 0x0074) { # t
1442                !!!next-input-character;                !!!next-input-character;
1443                push @next_char, $self->{next_input_character};                push @next_char, $self->{next_char};
1444                if ($self->{next_input_character} == 0x0059 or # Y                if ($self->{next_char} == 0x0059 or # Y
1445                    $self->{next_input_character} == 0x0079) { # y                    $self->{next_char} == 0x0079) { # y
1446                  !!!next-input-character;                  !!!next-input-character;
1447                  push @next_char, $self->{next_input_character};                  push @next_char, $self->{next_char};
1448                  if ($self->{next_input_character} == 0x0050 or # P                  if ($self->{next_char} == 0x0050 or # P
1449                      $self->{next_input_character} == 0x0070) { # p                      $self->{next_char} == 0x0070) { # p
1450                    !!!next-input-character;                    !!!next-input-character;
1451                    push @next_char, $self->{next_input_character};                    push @next_char, $self->{next_char};
1452                    if ($self->{next_input_character} == 0x0045 or # E                    if ($self->{next_char} == 0x0045 or # E
1453                        $self->{next_input_character} == 0x0065) { # e                        $self->{next_char} == 0x0065) { # e
1454                      ## ISSUE: What a stupid code this is!                      !!!cp (129);
1455                      $self->{state} = 'DOCTYPE';                      ## TODO: What a stupid code this is!
1456                        $self->{state} = DOCTYPE_STATE;
1457                      !!!next-input-character;                      !!!next-input-character;
1458                      redo A;                      redo A;
1459                      } else {
1460                        !!!cp (130);
1461                    }                    }
1462                    } else {
1463                      !!!cp (131);
1464                  }                  }
1465                  } else {
1466                    !!!cp (132);
1467                }                }
1468                } else {
1469                  !!!cp (133);
1470              }              }
1471              } else {
1472                !!!cp (134);
1473            }            }
1474            } else {
1475              !!!cp (135);
1476          }          }
1477          } else {
1478            !!!cp (136);
1479        }        }
1480    
1481        !!!parse-error (type => 'bogus comment open');        !!!parse-error (type => 'bogus comment');
1482        $self->{next_input_character} = shift @next_char;        $self->{next_char} = shift @next_char;
1483        !!!back-next-input-character (@next_char);        !!!back-next-input-character (@next_char);
1484        $self->{state} = 'bogus comment';        $self->{state} = BOGUS_COMMENT_STATE;
1485        redo A;        redo A;
1486                
1487        ## ISSUE: typos in spec: chacacters, is is a parse error        ## ISSUE: typos in spec: chacacters, is is a parse error
1488        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?
1489      } elsif ($self->{state} eq 'comment') {      } elsif ($self->{state} == COMMENT_START_STATE) {
1490        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_char} == 0x002D) { # -
1491          $self->{state} = 'comment dash';          !!!cp (137);
1492            $self->{state} = COMMENT_START_DASH_STATE;
1493          !!!next-input-character;          !!!next-input-character;
1494          redo A;          redo A;
1495        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == 0x003E) { # >
1496            !!!cp (138);
1497            !!!parse-error (type => 'bogus comment');
1498            $self->{state} = DATA_STATE;
1499            !!!next-input-character;
1500    
1501            !!!emit ($self->{current_token}); # comment
1502    
1503            redo A;
1504          } elsif ($self->{next_char} == -1) {
1505            !!!cp (139);
1506          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1507          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1508          ## reconsume          ## reconsume
1509    
1510          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
         undef $self->{current_token};  
1511    
1512          redo A;          redo A;
1513        } else {        } else {
1514          $self->{current_token}->{data} .= chr ($self->{next_input_character}); # comment          !!!cp (140);
1515            $self->{current_token}->{data} # comment
1516                .= chr ($self->{next_char});
1517            $self->{state} = COMMENT_STATE;
1518            !!!next-input-character;
1519            redo A;
1520          }
1521        } elsif ($self->{state} == COMMENT_START_DASH_STATE) {
1522          if ($self->{next_char} == 0x002D) { # -
1523            !!!cp (141);
1524            $self->{state} = COMMENT_END_STATE;
1525            !!!next-input-character;
1526            redo A;
1527          } elsif ($self->{next_char} == 0x003E) { # >
1528            !!!cp (142);
1529            !!!parse-error (type => 'bogus comment');
1530            $self->{state} = DATA_STATE;
1531            !!!next-input-character;
1532    
1533            !!!emit ($self->{current_token}); # comment
1534    
1535            redo A;
1536          } elsif ($self->{next_char} == -1) {
1537            !!!cp (143);
1538            !!!parse-error (type => 'unclosed comment');
1539            $self->{state} = DATA_STATE;
1540            ## reconsume
1541    
1542            !!!emit ($self->{current_token}); # comment
1543    
1544            redo A;
1545          } else {
1546            !!!cp (144);
1547            $self->{current_token}->{data} # comment
1548                .= '-' . chr ($self->{next_char});
1549            $self->{state} = COMMENT_STATE;
1550            !!!next-input-character;
1551            redo A;
1552          }
1553        } elsif ($self->{state} == COMMENT_STATE) {
1554          if ($self->{next_char} == 0x002D) { # -
1555            !!!cp (145);
1556            $self->{state} = COMMENT_END_DASH_STATE;
1557            !!!next-input-character;
1558            redo A;
1559          } elsif ($self->{next_char} == -1) {
1560            !!!cp (146);
1561            !!!parse-error (type => 'unclosed comment');
1562            $self->{state} = DATA_STATE;
1563            ## reconsume
1564    
1565            !!!emit ($self->{current_token}); # comment
1566    
1567            redo A;
1568          } else {
1569            !!!cp (147);
1570            $self->{current_token}->{data} .= chr ($self->{next_char}); # comment
1571          ## Stay in the state          ## Stay in the state
1572          !!!next-input-character;          !!!next-input-character;
1573          redo A;          redo A;
1574        }        }
1575      } elsif ($self->{state} eq 'comment dash') {      } elsif ($self->{state} == COMMENT_END_DASH_STATE) {
1576        if ($self->{next_input_character} == 0x002D) { # -        if ($self->{next_char} == 0x002D) { # -
1577          $self->{state} = 'comment end';          !!!cp (148);
1578            $self->{state} = COMMENT_END_STATE;
1579          !!!next-input-character;          !!!next-input-character;
1580          redo A;          redo A;
1581        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1582            !!!cp (149);
1583          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1584          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1585          ## reconsume          ## reconsume
1586    
1587          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
         undef $self->{current_token};  
1588    
1589          redo A;          redo A;
1590        } else {        } else {
1591          $self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment          !!!cp (150);
1592          $self->{state} = 'comment';          $self->{current_token}->{data} .= '-' . chr ($self->{next_char}); # comment
1593            $self->{state} = COMMENT_STATE;
1594          !!!next-input-character;          !!!next-input-character;
1595          redo A;          redo A;
1596        }        }
1597      } elsif ($self->{state} eq 'comment end') {      } elsif ($self->{state} == COMMENT_END_STATE) {
1598        if ($self->{next_input_character} == 0x003E) { # >        if ($self->{next_char} == 0x003E) { # >
1599          $self->{state} = 'data';          !!!cp (151);
1600            $self->{state} = DATA_STATE;
1601          !!!next-input-character;          !!!next-input-character;
1602    
1603          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
         undef $self->{current_token};  
1604    
1605          redo A;          redo A;
1606        } elsif ($self->{next_input_character} == 0x002D) { # -        } elsif ($self->{next_char} == 0x002D) { # -
1607            !!!cp (152);
1608          !!!parse-error (type => 'dash in comment');          !!!parse-error (type => 'dash in comment');
1609          $self->{current_token}->{data} .= '-'; # comment          $self->{current_token}->{data} .= '-'; # comment
1610          ## Stay in the state          ## Stay in the state
1611          !!!next-input-character;          !!!next-input-character;
1612          redo A;          redo A;
1613        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1614            !!!cp (153);
1615          !!!parse-error (type => 'unclosed comment');          !!!parse-error (type => 'unclosed comment');
1616          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1617          ## reconsume          ## reconsume
1618    
1619          !!!emit ($self->{current_token}); # comment          !!!emit ($self->{current_token}); # comment
         undef $self->{current_token};  
1620    
1621          redo A;          redo A;
1622        } else {        } else {
1623            !!!cp (154);
1624          !!!parse-error (type => 'dash in comment');          !!!parse-error (type => 'dash in comment');
1625          $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment          $self->{current_token}->{data} .= '--' . chr ($self->{next_char}); # comment
1626          $self->{state} = 'comment';          $self->{state} = COMMENT_STATE;
1627          !!!next-input-character;          !!!next-input-character;
1628          redo A;          redo A;
1629        }        }
1630      } elsif ($self->{state} eq 'DOCTYPE') {      } elsif ($self->{state} == DOCTYPE_STATE) {
1631        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
1632            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
1633            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
1634            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
1635            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
1636          $self->{state} = 'before DOCTYPE name';          !!!cp (155);
1637            $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
1638          !!!next-input-character;          !!!next-input-character;
1639          redo A;          redo A;
1640        } else {        } else {
1641            !!!cp (156);
1642          !!!parse-error (type => 'no space before DOCTYPE name');          !!!parse-error (type => 'no space before DOCTYPE name');
1643          $self->{state} = 'before DOCTYPE name';          $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
1644          ## reconsume          ## reconsume
1645          redo A;          redo A;
1646        }        }
1647      } elsif ($self->{state} eq 'before DOCTYPE name') {      } elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) {
1648        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0009 or # HT
1649            $self->{next_input_character} == 0x000A or # LF            $self->{next_char} == 0x000A or # LF
1650            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000B or # VT
1651            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000C or # FF
1652            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x0020) { # SP
1653            !!!cp (157);
1654          ## Stay in the state          ## Stay in the state
1655          !!!next-input-character;          !!!next-input-character;
1656          redo A;          redo A;
1657        } elsif (0x0061 <= $self->{next_input_character} and        } elsif ($self->{next_char} == 0x003E) { # >
1658                 $self->{next_input_character} <= 0x007A) { # a..z          !!!cp (158);
 ## ISSUE: "Set the token's name name to the" in the spec  
         $self->{current_token} = {type => 'DOCTYPE',  
                           name => chr ($self->{next_input_character} - 0x0020),  
                           error => 1};  
         $self->{state} = 'DOCTYPE name';  
         !!!next-input-character;  
         redo A;  
       } elsif ($self->{next_input_character} == 0x003E) { # >  
1659          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
1660          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1661          !!!next-input-character;          !!!next-input-character;
1662    
1663          !!!emit ({type => 'DOCTYPE', name => '', error => 1});          !!!emit ({type => DOCTYPE_TOKEN, quirks => 1});
1664    
1665          redo A;          redo A;
1666        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1667            !!!cp (159);
1668          !!!parse-error (type => 'no DOCTYPE name');          !!!parse-error (type => 'no DOCTYPE name');
1669          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1670          ## reconsume          ## reconsume
1671    
1672          !!!emit ({type => 'DOCTYPE', name => '', error => 1});          !!!emit ({type => DOCTYPE_TOKEN, quirks => 1});
1673    
1674          redo A;          redo A;
1675        } else {        } else {
1676          $self->{current_token} = {type => 'DOCTYPE',          !!!cp (160);
1677                            name => chr ($self->{next_input_character}),          $self->{current_token}
1678                            error => 1};              = {type => DOCTYPE_TOKEN,
1679                   name => chr ($self->{next_char}),
1680                   #quirks => 0,
1681                  };
1682  ## ISSUE: "Set the token's name name to the" in the spec  ## ISSUE: "Set the token's name name to the" in the spec
1683          $self->{state} = 'DOCTYPE name';          $self->{state} = DOCTYPE_NAME_STATE;
1684          !!!next-input-character;          !!!next-input-character;
1685          redo A;          redo A;
1686        }        }
1687      } elsif ($self->{state} eq 'DOCTYPE name') {      } elsif ($self->{state} == DOCTYPE_NAME_STATE) {
1688        if ($self->{next_input_character} == 0x0009 or # HT  ## ISSUE: Redundant "First," in the spec.
1689            $self->{next_input_character} == 0x000A or # LF        if ($self->{next_char} == 0x0009 or # HT
1690            $self->{next_input_character} == 0x000B or # VT            $self->{next_char} == 0x000A or # LF
1691            $self->{next_input_character} == 0x000C or # FF            $self->{next_char} == 0x000B or # VT
1692            $self->{next_input_character} == 0x0020) { # SP            $self->{next_char} == 0x000C or # FF
1693          $self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE            $self->{next_char} == 0x0020) { # SP
1694          $self->{state} = 'after DOCTYPE name';          !!!cp (161);
1695            $self->{state} = AFTER_DOCTYPE_NAME_STATE;
1696          !!!next-input-character;          !!!next-input-character;
1697          redo A;          redo A;
1698        } elsif ($self->{next_input_character} == 0x003E) { # >        } elsif ($self->{next_char} == 0x003E) { # >
1699          $self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE          !!!cp (162);
1700          $self->{state} = 'data';          $self->{state} = DATA_STATE;
1701          !!!next-input-character;          !!!next-input-character;
1702    
1703          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
         undef $self->{current_token};  
1704    
1705          redo A;          redo A;
1706        } elsif (0x0061 <= $self->{next_input_character} and        } elsif ($self->{next_char} == -1) {
1707                 $self->{next_input_character} <= 0x007A) { # a..z          !!!cp (163);
1708          $self->{current_token}->{name} .= chr ($self->{next_input_character} - 0x0020); # DOCTYPE          !!!parse-error (type => 'unclosed DOCTYPE');
1709          #$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML');          $self->{state} = DATA_STATE;
1710            ## reconsume
1711    
1712            $self->{current_token}->{quirks} = 1;
1713            !!!emit ($self->{current_token}); # DOCTYPE
1714    
1715            redo A;
1716          } else {
1717            !!!cp (164);
1718            $self->{current_token}->{name}
1719              .= chr ($self->{next_char}); # DOCTYPE
1720            ## Stay in the state
1721            !!!next-input-character;
1722            redo A;
1723          }
1724        } elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) {
1725          if ($self->{next_char} == 0x0009 or # HT
1726              $self->{next_char} == 0x000A or # LF
1727              $self->{next_char} == 0x000B or # VT
1728              $self->{next_char} == 0x000C or # FF
1729              $self->{next_char} == 0x0020) { # SP
1730            !!!cp (165);
1731          ## Stay in the state          ## Stay in the state
1732          !!!next-input-character;          !!!next-input-character;
1733          redo A;          redo A;
1734        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == 0x003E) { # >
1735            !!!cp (166);
1736            $self->{state} = DATA_STATE;
1737            !!!next-input-character;
1738    
1739            !!!emit ($self->{current_token}); # DOCTYPE
1740    
1741            redo A;
1742          } elsif ($self->{next_char} == -1) {
1743            !!!cp (167);
1744          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1745          $self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML'); # DOCTYPE          $self->{state} = DATA_STATE;
         $self->{state} = 'data';  
1746          ## reconsume          ## reconsume
1747    
1748          !!!emit ($self->{current_token});          $self->{current_token}->{quirks} = 1;
1749          undef $self->{current_token};          !!!emit ($self->{current_token}); # DOCTYPE
1750    
1751          redo A;          redo A;
1752          } elsif ($self->{next_char} == 0x0050 or # P
1753                   $self->{next_char} == 0x0070) { # p
1754            !!!next-input-character;
1755            if ($self->{next_char} == 0x0055 or # U
1756                $self->{next_char} == 0x0075) { # u
1757              !!!next-input-character;
1758              if ($self->{next_char} == 0x0042 or # B
1759                  $self->{next_char} == 0x0062) { # b
1760                !!!next-input-character;
1761                if ($self->{next_char} == 0x004C or # L
1762                    $self->{next_char} == 0x006C) { # l
1763                  !!!next-input-character;
1764                  if ($self->{next_char} == 0x0049 or # I
1765                      $self->{next_char} == 0x0069) { # i
1766                    !!!next-input-character;
1767                    if ($self->{next_char} == 0x0043 or # C
1768                        $self->{next_char} == 0x0063) { # c
1769                      !!!cp (168);
1770                      $self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1771                      !!!next-input-character;
1772                      redo A;
1773                    } else {
1774                      !!!cp (169);
1775                    }
1776                  } else {
1777                    !!!cp (170);
1778                  }
1779                } else {
1780                  !!!cp (171);
1781                }
1782              } else {
1783                !!!cp (172);
1784              }
1785            } else {
1786              !!!cp (173);
1787            }
1788    
1789            #
1790          } elsif ($self->{next_char} == 0x0053 or # S
1791                   $self->{next_char} == 0x0073) { # s
1792            !!!next-input-character;
1793            if ($self->{next_char} == 0x0059 or # Y
1794                $self->{next_char} == 0x0079) { # y
1795              !!!next-input-character;
1796              if ($self->{next_char} == 0x0053 or # S
1797                  $self->{next_char} == 0x0073) { # s
1798                !!!next-input-character;
1799                if ($self->{next_char} == 0x0054 or # T
1800                    $self->{next_char} == 0x0074) { # t
1801                  !!!next-input-character;
1802                  if ($self->{next_char} == 0x0045 or # E
1803                      $self->{next_char} == 0x0065) { # e
1804                    !!!next-input-character;
1805                    if ($self->{next_char} == 0x004D or # M
1806                        $self->{next_char} == 0x006D) { # m
1807                      !!!cp (174);
1808                      $self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
1809                      !!!next-input-character;
1810                      redo A;
1811                    } else {
1812                      !!!cp (175);
1813                    }
1814                  } else {
1815                    !!!cp (176);
1816                  }
1817                } else {
1818                  !!!cp (177);
1819                }
1820              } else {
1821                !!!cp (178);
1822              }
1823            } else {
1824              !!!cp (179);
1825            }
1826    
1827            #
1828        } else {        } else {
1829          $self->{current_token}->{name}          !!!cp (180);
1830            .= chr ($self->{next_input_character}); # DOCTYPE          !!!next-input-character;
1831          #$self->{current_token}->{error} = ($self->{current_token}->{name} ne 'HTML');          #
1832          }
1833    
1834          !!!parse-error (type => 'string after DOCTYPE name');
1835          $self->{current_token}->{quirks} = 1;
1836    
1837          $self->{state} = BOGUS_DOCTYPE_STATE;
1838          # next-input-character is already done
1839          redo A;
1840        } elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
1841          if ({
1842                0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1843                #0x000D => 1, # HT, LF, VT, FF, SP, CR
1844              }->{$self->{next_char}}) {
1845            !!!cp (181);
1846          ## Stay in the state          ## Stay in the state
1847          !!!next-input-character;          !!!next-input-character;
1848          redo A;          redo A;
1849          } elsif ($self->{next_char} eq 0x0022) { # "
1850            !!!cp (182);
1851            $self->{current_token}->{public_identifier} = ''; # DOCTYPE
1852            $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE;
1853            !!!next-input-character;
1854            redo A;
1855          } elsif ($self->{next_char} eq 0x0027) { # '
1856            !!!cp (183);
1857            $self->{current_token}->{public_identifier} = ''; # DOCTYPE
1858            $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE;
1859            !!!next-input-character;
1860            redo A;
1861          } elsif ($self->{next_char} eq 0x003E) { # >
1862            !!!cp (184);
1863            !!!parse-error (type => 'no PUBLIC literal');
1864    
1865            $self->{state} = DATA_STATE;
1866            !!!next-input-character;
1867    
1868            $self->{current_token}->{quirks} = 1;
1869            !!!emit ($self->{current_token}); # DOCTYPE
1870    
1871            redo A;
1872          } elsif ($self->{next_char} == -1) {
1873            !!!cp (185);
1874            !!!parse-error (type => 'unclosed DOCTYPE');
1875    
1876            $self->{state} = DATA_STATE;
1877            ## reconsume
1878    
1879            $self->{current_token}->{quirks} = 1;
1880            !!!emit ($self->{current_token}); # DOCTYPE
1881    
1882            redo A;
1883          } else {
1884            !!!cp (186);
1885            !!!parse-error (type => 'string after PUBLIC');
1886            $self->{current_token}->{quirks} = 1;
1887    
1888            $self->{state} = BOGUS_DOCTYPE_STATE;
1889            !!!next-input-character;
1890            redo A;
1891        }        }
1892      } elsif ($self->{state} eq 'after DOCTYPE name') {      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) {
1893        if ($self->{next_input_character} == 0x0009 or # HT        if ($self->{next_char} == 0x0022) { # "
1894            $self->{next_input_character} == 0x000A or # LF          !!!cp (187);
1895            $self->{next_input_character} == 0x000B or # VT          $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1896            $self->{next_input_character} == 0x000C or # FF          !!!next-input-character;
1897            $self->{next_input_character} == 0x0020) { # SP          redo A;
1898          } elsif ($self->{next_char} == 0x003E) { # >
1899            !!!cp (188);
1900            !!!parse-error (type => 'unclosed PUBLIC literal');
1901    
1902            $self->{state} = DATA_STATE;
1903            !!!next-input-character;
1904    
1905            $self->{current_token}->{quirks} = 1;
1906            !!!emit ($self->{current_token}); # DOCTYPE
1907    
1908            redo A;
1909          } elsif ($self->{next_char} == -1) {
1910            !!!cp (189);
1911            !!!parse-error (type => 'unclosed PUBLIC literal');
1912    
1913            $self->{state} = DATA_STATE;
1914            ## reconsume
1915    
1916            $self->{current_token}->{quirks} = 1;
1917            !!!emit ($self->{current_token}); # DOCTYPE
1918    
1919            redo A;
1920          } else {
1921            !!!cp (190);
1922            $self->{current_token}->{public_identifier} # DOCTYPE
1923                .= chr $self->{next_char};
1924          ## Stay in the state          ## Stay in the state
1925          !!!next-input-character;          !!!next-input-character;
1926          redo A;          redo A;
1927        } elsif ($self->{next_input_character} == 0x003E) { # >        }
1928          $self->{state} = 'data';      } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) {
1929          if ($self->{next_char} == 0x0027) { # '
1930            !!!cp (191);
1931            $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
1932            !!!next-input-character;
1933            redo A;
1934          } elsif ($self->{next_char} == 0x003E) { # >
1935            !!!cp (192);
1936            !!!parse-error (type => 'unclosed PUBLIC literal');
1937    
1938            $self->{state} = DATA_STATE;
1939            !!!next-input-character;
1940    
1941            $self->{current_token}->{quirks} = 1;
1942            !!!emit ($self->{current_token}); # DOCTYPE
1943    
1944            redo A;
1945          } elsif ($self->{next_char} == -1) {
1946            !!!cp (193);
1947            !!!parse-error (type => 'unclosed PUBLIC literal');
1948    
1949            $self->{state} = DATA_STATE;
1950            ## reconsume
1951    
1952            $self->{current_token}->{quirks} = 1;
1953            !!!emit ($self->{current_token}); # DOCTYPE
1954    
1955            redo A;
1956          } else {
1957            !!!cp (194);
1958            $self->{current_token}->{public_identifier} # DOCTYPE
1959                .= chr $self->{next_char};
1960            ## Stay in the state
1961            !!!next-input-character;
1962            redo A;
1963          }
1964        } elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
1965          if ({
1966                0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
1967                #0x000D => 1, # HT, LF, VT, FF, SP, CR
1968              }->{$self->{next_char}}) {
1969            !!!cp (195);
1970            ## Stay in the state
1971            !!!next-input-character;
1972            redo A;
1973          } elsif ($self->{next_char} == 0x0022) { # "
1974            !!!cp (196);
1975            $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1976            $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
1977            !!!next-input-character;
1978            redo A;
1979          } elsif ($self->{next_char} == 0x0027) { # '
1980            !!!cp (197);
1981            $self->{current_token}->{system_identifier} = ''; # DOCTYPE
1982            $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
1983            !!!next-input-character;
1984            redo A;
1985          } elsif ($self->{next_char} == 0x003E) { # >
1986            !!!cp (198);
1987            $self->{state} = DATA_STATE;
1988          !!!next-input-character;          !!!next-input-character;
1989    
1990          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
         undef $self->{current_token};  
1991    
1992          redo A;          redo A;
1993        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
1994            !!!cp (199);
1995          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
1996          $self->{state} = 'data';  
1997            $self->{state} = DATA_STATE;
1998            ## reconsume
1999    
2000            $self->{current_token}->{quirks} = 1;
2001            !!!emit ($self->{current_token}); # DOCTYPE
2002    
2003            redo A;
2004          } else {
2005            !!!cp (200);
2006            !!!parse-error (type => 'string after PUBLIC literal');
2007            $self->{current_token}->{quirks} = 1;
2008    
2009            $self->{state} = BOGUS_DOCTYPE_STATE;
2010            !!!next-input-character;
2011            redo A;
2012          }
2013        } elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
2014          if ({
2015                0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
2016                #0x000D => 1, # HT, LF, VT, FF, SP, CR
2017              }->{$self->{next_char}}) {
2018            !!!cp (201);
2019            ## Stay in the state
2020            !!!next-input-character;
2021            redo A;
2022          } elsif ($self->{next_char} == 0x0022) { # "
2023            !!!cp (202);
2024            $self->{current_token}->{system_identifier} = ''; # DOCTYPE
2025            $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
2026            !!!next-input-character;
2027            redo A;
2028          } elsif ($self->{next_char} == 0x0027) { # '
2029            !!!cp (203);
2030            $self->{current_token}->{system_identifier} = ''; # DOCTYPE
2031            $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
2032            !!!next-input-character;
2033            redo A;
2034          } elsif ($self->{next_char} == 0x003E) { # >
2035            !!!cp (204);
2036            !!!parse-error (type => 'no SYSTEM literal');
2037            $self->{state} = DATA_STATE;
2038            !!!next-input-character;
2039    
2040            $self->{current_token}->{quirks} = 1;
2041            !!!emit ($self->{current_token}); # DOCTYPE
2042    
2043            redo A;
2044          } elsif ($self->{next_char} == -1) {
2045            !!!cp (205);
2046            !!!parse-error (type => 'unclosed DOCTYPE');
2047    
2048            $self->{state} = DATA_STATE;
2049            ## reconsume
2050    
2051            $self->{current_token}->{quirks} = 1;
2052            !!!emit ($self->{current_token}); # DOCTYPE
2053    
2054            redo A;
2055          } else {
2056            !!!cp (206);
2057            !!!parse-error (type => 'string after SYSTEM');
2058            $self->{current_token}->{quirks} = 1;
2059    
2060            $self->{state} = BOGUS_DOCTYPE_STATE;
2061            !!!next-input-character;
2062            redo A;
2063          }
2064        } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) {
2065          if ($self->{next_char} == 0x0022) { # "
2066            !!!cp (207);
2067            $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
2068            !!!next-input-character;
2069            redo A;
2070          } elsif ($self->{next_char} == 0x003E) { # >
2071            !!!cp (208);
2072            !!!parse-error (type => 'unclosed PUBLIC literal');
2073    
2074            $self->{state} = DATA_STATE;
2075            !!!next-input-character;
2076    
2077            $self->{current_token}->{quirks} = 1;
2078            !!!emit ($self->{current_token}); # DOCTYPE
2079    
2080            redo A;
2081          } elsif ($self->{next_char} == -1) {
2082            !!!cp (209);
2083            !!!parse-error (type => 'unclosed SYSTEM literal');
2084    
2085            $self->{state} = DATA_STATE;
2086            ## reconsume
2087    
2088            $self->{current_token}->{quirks} = 1;
2089            !!!emit ($self->{current_token}); # DOCTYPE
2090    
2091            redo A;
2092          } else {
2093            !!!cp (210);
2094            $self->{current_token}->{system_identifier} # DOCTYPE
2095                .= chr $self->{next_char};
2096            ## Stay in the state
2097            !!!next-input-character;
2098            redo A;
2099          }
2100        } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) {
2101          if ($self->{next_char} == 0x0027) { # '
2102            !!!cp (211);
2103            $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
2104            !!!next-input-character;
2105            redo A;
2106          } elsif ($self->{next_char} == 0x003E) { # >
2107            !!!cp (212);
2108            !!!parse-error (type => 'unclosed PUBLIC literal');
2109    
2110            $self->{state} = DATA_STATE;
2111            !!!next-input-character;
2112    
2113            $self->{current_token}->{quirks} = 1;
2114            !!!emit ($self->{current_token}); # DOCTYPE
2115    
2116            redo A;
2117          } elsif ($self->{next_char} == -1) {
2118            !!!cp (213);
2119            !!!parse-error (type => 'unclosed SYSTEM literal');
2120    
2121            $self->{state} = DATA_STATE;
2122          ## reconsume          ## reconsume
2123    
2124            $self->{current_token}->{quirks} = 1;
2125          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
         undef $self->{current_token};  
2126    
2127          redo A;          redo A;
2128        } else {        } else {
2129          !!!parse-error (type => 'string after DOCTYPE name');          !!!cp (214);
2130          $self->{current_token}->{error} = 1; # DOCTYPE          $self->{current_token}->{system_identifier} # DOCTYPE
2131          $self->{state} = 'bogus DOCTYPE';              .= chr $self->{next_char};
2132            ## Stay in the state
2133          !!!next-input-character;          !!!next-input-character;
2134          redo A;          redo A;
2135        }        }
2136      } elsif ($self->{state} eq 'bogus DOCTYPE') {      } elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
2137        if ($self->{next_input_character} == 0x003E) { # >        if ({
2138          $self->{state} = 'data';              0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
2139                #0x000D => 1, # HT, LF, VT, FF, SP, CR
2140              }->{$self->{next_char}}) {
2141            !!!cp (215);
2142            ## Stay in the state
2143            !!!next-input-character;
2144            redo A;
2145          } elsif ($self->{next_char} == 0x003E) { # >
2146            !!!cp (216);
2147            $self->{state} = DATA_STATE;
2148          !!!next-input-character;          !!!next-input-character;
2149    
2150          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
         undef $self->{current_token};  
2151    
2152          redo A;          redo A;
2153        } elsif ($self->{next_input_character} == -1) {        } elsif ($self->{next_char} == -1) {
2154            !!!cp (217);
2155          !!!parse-error (type => 'unclosed DOCTYPE');          !!!parse-error (type => 'unclosed DOCTYPE');
2156          $self->{state} = 'data';  
2157            $self->{state} = DATA_STATE;
2158          ## reconsume          ## reconsume
2159    
2160            $self->{current_token}->{quirks} = 1;
2161          !!!emit ($self->{current_token}); # DOCTYPE          !!!emit ($self->{current_token}); # DOCTYPE
         undef $self->{current_token};  
2162    
2163          redo A;          redo A;
2164        } else {        } else {
2165            !!!cp (218);
2166            !!!parse-error (type => 'string after SYSTEM literal');
2167            #$self->{current_token}->{quirks} = 1;
2168    
2169            $self->{state} = BOGUS_DOCTYPE_STATE;
2170            !!!next-input-character;
2171            redo A;
2172          }
2173        } elsif ($self->{state} == BOGUS_DOCTYPE_STATE) {
2174          if ($self->{next_char} == 0x003E) { # >
2175            !!!cp (219);
2176            $self->{state} = DATA_STATE;
2177            !!!next-input-character;
2178    
2179            !!!emit ($self->{current_token}); # DOCTYPE
2180    
2181            redo A;
2182          } elsif ($self->{next_char} == -1) {
2183            !!!cp (220);
2184            !!!parse-error (type => 'unclosed DOCTYPE');
2185            $self->{state} = DATA_STATE;
2186            ## reconsume
2187    
2188            !!!emit ($self->{current_token}); # DOCTYPE
2189    
2190            redo A;
2191          } else {
2192            !!!cp (221);
2193          ## Stay in the state          ## Stay in the state
2194          !!!next-input-character;          !!!next-input-character;
2195          redo A;          redo A;
# Line 1490  sub _get_next_token ($) { Line 2202  sub _get_next_token ($) {
2202    die "$0: _get_next_token: unexpected case";    die "$0: _get_next_token: unexpected case";
2203  } # _get_next_token  } # _get_next_token
2204    
2205  sub _tokenize_attempt_to_consume_an_entity ($) {  sub _tokenize_attempt_to_consume_an_entity ($$$) {
2206    my $self = shift;    my ($self, $in_attr, $additional) = @_;
2207      
2208    if ($self->{next_input_character} == 0x0023) { # #    if ({
2209           0x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, # HT, LF, VT, FF,
2210           0x0020 => 1, 0x003C => 1, 0x0026 => 1, -1 => 1, # SP, <, & # 0x000D # CR
2211           $additional => 1,
2212          }->{$self->{next_char}}) {
2213        !!!cp (1001);
2214        ## Don't consume
2215        ## No error
2216        return undef;
2217      } elsif ($self->{next_char} == 0x0023) { # #
2218      !!!next-input-character;      !!!next-input-character;
2219      if ($self->{next_input_character} == 0x0078 or # x      if ($self->{next_char} == 0x0078 or # x
2220          $self->{next_input_character} == 0x0058) { # X          $self->{next_char} == 0x0058) { # X
2221        my $num;        my $code;
2222        X: {        X: {
2223          my $x_char = $self->{next_input_character};          my $x_char = $self->{next_char};
2224          !!!next-input-character;          !!!next-input-character;
2225          if (0x0030 <= $self->{next_input_character} and          if (0x0030 <= $self->{next_char} and
2226              $self->{next_input_character} <= 0x0039) { # 0..9              $self->{next_char} <= 0x0039) { # 0..9
2227            $num ||= 0;            !!!cp (1002);
2228            $num *= 0x10;            $code ||= 0;
2229            $num += $self->{next_input_character} - 0x0030;            $code *= 0x10;
2230              $code += $self->{next_char} - 0x0030;
2231            redo X;            redo X;
2232          } elsif (0x0061 <= $self->{next_input_character} and          } elsif (0x0061 <= $self->{next_char} and
2233                   $self->{next_input_character} <= 0x0066) { # a..f                   $self->{next_char} <= 0x0066) { # a..f
2234            ## ISSUE: the spec says U+0078, which is apparently incorrect            !!!cp (1003);
2235            $num ||= 0;            $code ||= 0;
2236            $num *= 0x10;            $code *= 0x10;
2237            $num += $self->{next_input_character} - 0x0060 + 9;            $code += $self->{next_char} - 0x0060 + 9;
2238            redo X;            redo X;
2239          } elsif (0x0041 <= $self->{next_input_character} and          } elsif (0x0041 <= $self->{next_char} and
2240                   $self->{next_input_character} <= 0x0046) { # A..F                   $self->{next_char} <= 0x0046) { # A..F
2241            ## ISSUE: the spec says U+0058, which is apparently incorrect            !!!cp (1004);
2242            $num ||= 0;            $code ||= 0;
2243            $num *= 0x10;            $code *= 0x10;
2244            $num += $self->{next_input_character} - 0x0040 + 9;            $code += $self->{next_char} - 0x0040 + 9;
2245            redo X;            redo X;
2246          } elsif (not defined $num) { # no hexadecimal digit          } elsif (not defined $code) { # no hexadecimal digit
2247              !!!cp (1005);
2248            !!!parse-error (type => 'bare hcro');            !!!parse-error (type => 'bare hcro');
2249            $self->{next_input_character} = 0x0023; # #            !!!back-next-input-character ($x_char, $self->{next_char});
2250            !!!back-next-input-character ($x_char);            $self->{next_char} = 0x0023; # #
2251            return undef;            return undef;
2252          } elsif ($self->{next_input_character} == 0x003B) { # ;          } elsif ($self->{next_char} == 0x003B) { # ;
2253              !!!cp (1006);
2254            !!!next-input-character;            !!!next-input-character;
2255          } else {          } else {
2256              !!!cp (1007);
2257            !!!parse-error (type => 'no refc');            !!!parse-error (type => 'no refc');
2258          }          }
2259    
2260          ## TODO: check the definition for |a valid Unicode character|.          if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) {
2261          ## <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2006-December/thread.html#8189>            !!!cp (1008);
2262          if ($num > 1114111 or $num == 0) {            !!!parse-error (type => sprintf 'invalid character reference:U+%04X', $code);
2263            $num = 0xFFFD; # REPLACEMENT CHARACTER            $code = 0xFFFD;
2264            ## ISSUE: Why this is not an error?          } elsif ($code > 0x10FFFF) {
2265          } elsif (0x80 <= $num and $num <= 0x9F) {            !!!cp (1009);
2266            !!!parse-error (type => sprintf 'c1 entity:U+%04X', $num);            !!!parse-error (type => sprintf 'invalid character reference:U-%08X', $code);
2267            $num = $c1_entity_char->{$num};            $code = 0xFFFD;
2268            } elsif ($code == 0x000D) {
2269              !!!cp (1010);
2270              !!!parse-error (type => 'CR character reference');
2271              $code = 0x000A;
2272            } elsif (0x80 <= $code and $code <= 0x9F) {
2273              !!!cp (1011);
2274              !!!parse-error (type => sprintf 'C1 character reference:U+%04X', $code);
2275              $code = $c1_entity_char->{$code};
2276          }          }
2277    
2278          return {type => 'character', data => chr $num};          return {type => CHARACTER_TOKEN, data => chr $code,
2279                    has_reference => 1};
2280        } # X        } # X
2281      } elsif (0x0030 <= $self->{next_input_character} and      } elsif (0x0030 <= $self->{next_char} and
2282               $self->{next_input_character} <= 0x0039) { # 0..9               $self->{next_char} <= 0x0039) { # 0..9
2283        my $code = $self->{next_input_character} - 0x0030;        my $code = $self->{next_char} - 0x0030;
2284        !!!next-input-character;        !!!next-input-character;
2285                
2286        while (0x0030 <= $self->{next_input_character} and        while (0x0030 <= $self->{next_char} and
2287                  $self->{next_input_character} <= 0x0039) { # 0..9                  $self->{next_char} <= 0x0039) { # 0..9
2288            !!!cp (1012);
2289          $code *= 10;          $code *= 10;
2290          $code += $self->{next_input_character} - 0x0030;          $code += $self->{next_char} - 0x0030;
2291                    
2292          !!!next-input-character;          !!!next-input-character;
2293        }        }
2294    
2295        if ($self->{next_input_character} == 0x003B) { # ;        if ($self->{next_char} == 0x003B) { # ;
2296            !!!cp (1013);
2297          !!!next-input-character;          !!!next-input-character;
2298        } else {        } else {
2299            !!!cp (1014);
2300          !!!parse-error (type => 'no refc');          !!!parse-error (type => 'no refc');
2301        }        }
2302    
2303        ## TODO: check the definition for |a valid Unicode character|.        if ($code == 0 or (0xD800 <= $code and $code <= 0xDFFF)) {
2304        if ($code > 1114111 or $code == 0) {          !!!cp (1015);
2305          $code = 0xFFFD; # REPLACEMENT CHARACTER          !!!parse-error (type => sprintf 'invalid character reference:U+%04X', $code);
2306          ## ISSUE: Why this is not an error?          $code = 0xFFFD;
2307          } elsif ($code > 0x10FFFF) {
2308            !!!cp (1016);
2309            !!!parse-error (type => sprintf 'invalid character reference:U-%08X', $code);
2310            $code = 0xFFFD;
2311          } elsif ($code == 0x000D) {
2312            !!!cp (1017);
2313            !!!parse-error (type => 'CR character reference');
2314            $code = 0x000A;
2315        } elsif (0x80 <= $code and $code <= 0x9F) {        } elsif (0x80 <= $code and $code <= 0x9F) {
2316          !!!parse-error (type => sprintf 'c1 entity:U+%04X', $code);          !!!cp (1018);
2317            !!!parse-error (type => sprintf 'C1 character reference:U+%04X', $code);
2318          $code = $c1_entity_char->{$code};          $code = $c1_entity_char->{$code};
2319        }        }
2320                
2321        return {type => 'character', data => chr $code};        return {type => CHARACTER_TOKEN, data => chr $code, has_reference => 1};
2322      } else {      } else {
2323          !!!cp (1019);
2324        !!!parse-error (type => 'bare nero');        !!!parse-error (type => 'bare nero');
2325        !!!back-next-input-character ($self->{next_input_character});        !!!back-next-input-character ($self->{next_char});
2326        $self->{next_input_character} = 0x0023; # #        $self->{next_char} = 0x0023; # #
2327        return undef;        return undef;
2328      }      }
2329    } elsif ((0x0041 <= $self->{next_input_character} and    } elsif ((0x0041 <= $self->{next_char} and
2330              $self->{next_input_character} <= 0x005A) or              $self->{next_char} <= 0x005A) or
2331             (0x0061 <= $self->{next_input_character} and             (0x0061 <= $self->{next_char} and
2332              $self->{next_input_character} <= 0x007A)) {              $self->{next_char} <= 0x007A)) {
2333      my $entity_name = chr $self->{next_input_character};      my $entity_name = chr $self->{next_char};
2334      !!!next-input-character;      !!!next-input-character;
2335    
2336      my $value = $entity_name;      my $value = $entity_name;
2337      my $match;      my $match = 0;
2338        require Whatpm::_NamedEntityList;
2339        our $EntityChar;
2340    
2341      while (length $entity_name < 10 and      while (length $entity_name < 10 and
2342             ## NOTE: Some number greater than the maximum length of entity name             ## NOTE: Some number greater than the maximum length of entity name
2343             ((0x0041 <= $self->{next_input_character} and             ((0x0041 <= $self->{next_char} and # a
2344               $self->{next_input_character} <= 0x005A) or               $self->{next_char} <= 0x005A) or # x
2345              (0x0061 <= $self->{next_input_character} and              (0x0061 <= $self->{next_char} and # a
2346               $self->{next_input_character} <= 0x007A) or               $self->{next_char} <= 0x007A) or # z
2347              (0x0030 <= $self->{next_input_character} and              (0x0030 <= $self->{next_char} and # 0
2348               $self->{next_input_character} <= 0x0039))) {               $self->{next_char} <= 0x0039) or # 9
2349        $entity_name .= chr $self->{next_input_character};              $self->{next_char} == 0x003B)) { # ;
2350        if (defined $entity_char->{$entity_name}) {        $entity_name .= chr $self->{next_char};
2351          $value = $entity_char->{$entity_name};        if (defined $EntityChar->{$entity_name}) {
2352          $match = 1;          if ($self->{next_char} == 0x003B) { # ;
2353              !!!cp (1020);
2354              $value = $EntityChar->{$entity_name};
2355              $match = 1;
2356              !!!next-input-character;
2357              last;
2358            } else {
2359              !!!cp (1021);
2360              $value = $EntityChar->{$entity_name};
2361              $match = -1;
2362              !!!next-input-character;
2363            }
2364        } else {        } else {
2365          $value .= chr $self->{next_input_character};          !!!cp (1022);
2366            $value .= chr $self->{next_char};
2367            $match *= 2;
2368            !!!next-input-character;
2369        }        }
       !!!next-input-character;  
2370      }      }
2371            
2372      if ($match) {      if ($match > 0) {
2373        if ($self->{next_input_character} == 0x003B) { # ;        !!!cp (1023);
2374          !!!next-input-character;        return {type => CHARACTER_TOKEN, data => $value, has_reference => 1};
2375        } elsif ($match < 0) {
2376          !!!parse-error (type => 'no refc');
2377          if ($in_attr and $match < -1) {
2378            !!!cp (1024);
2379            return {type => CHARACTER_TOKEN, data => '&'.$entity_name};
2380        } else {        } else {
2381          !!!parse-error (type => 'refc');          !!!cp (1025);
2382            return {type => CHARACTER_TOKEN, data => $value, has_reference => 1};
2383        }        }
   
       return {type => 'character', data => $value};  
2384      } else {      } else {
2385          !!!cp (1026);
2386        !!!parse-error (type => 'bare ero');        !!!parse-error (type => 'bare ero');
2387        ## NOTE: No characters are consumed in the spec.        ## NOTE: "No characters are consumed" in the spec.
2388        !!!back-token ({type => 'character', data => $value});        return {type => CHARACTER_TOKEN, data => '&'.$value};
       return undef;  
2389      }      }
2390    } else {    } else {
2391        !!!cp (1027);
2392      ## no characters are consumed      ## no characters are consumed
2393      !!!parse-error (type => 'bare ero');      !!!parse-error (type => 'bare ero');
2394      return undef;      return undef;
# Line 1634  sub _initialize_tree_constructor ($) { Line 2401  sub _initialize_tree_constructor ($) {
2401    $self->{document}->strict_error_checking (0);    $self->{document}->strict_error_checking (0);
2402    ## TODO: Turn mutation events off # MUST    ## TODO: Turn mutation events off # MUST
2403    ## TODO: Turn loose Document option (manakai extension) on    ## TODO: Turn loose Document option (manakai extension) on
2404    ## TODO: Mark the Document as an HTML document # MUST    $self->{document}->manakai_is_html (1); # MUST
2405  } # _initialize_tree_constructor  } # _initialize_tree_constructor
2406    
2407  sub _terminate_tree_constructor ($) {  sub _terminate_tree_constructor ($) {
# Line 1661  sub _construct_tree ($) { Line 2428  sub _construct_tree ($) {
2428        
2429    !!!next-token;    !!!next-token;
2430    
   $self->{insertion_mode} = 'before head';  
2431    undef $self->{form_element};    undef $self->{form_element};
2432    undef $self->{head_element};    undef $self->{head_element};
2433    $self->{open_elements} = [];    $self->{open_elements} = [];
2434    undef $self->{inner_html_node};    undef $self->{inner_html_node};
2435    
2436      ## NOTE: The "initial" insertion mode.
2437    $self->_tree_construction_initial; # MUST    $self->_tree_construction_initial; # MUST
2438    
2439      ## NOTE: The "before html" insertion mode.
2440    $self->_tree_construction_root_element;    $self->_tree_construction_root_element;
2441      $self->{insertion_mode} = BEFORE_HEAD_IM;
2442    
2443      ## NOTE: The "before head" insertion mode and so on.
2444    $self->_tree_construction_main;    $self->_tree_construction_main;
2445  } # _construct_tree  } # _construct_tree
2446    
2447  sub _tree_construction_initial ($) {  sub _tree_construction_initial ($) {
2448    my $self = shift;    my $self = shift;
2449    B: {  
2450        if ($token->{type} eq 'DOCTYPE') {    ## NOTE: "initial" insertion mode
2451          if ($token->{error}) {  
2452            ## ISSUE: Spec currently left this case undefined.    INITIAL: {
2453            !!!parse-error (type => 'bogus DOCTYPE');      if ($token->{type} == DOCTYPE_TOKEN) {
2454          }        ## NOTE: Conformance checkers MAY, instead of reporting "not HTML5"
2455          my $doctype = $self->{document}->create_document_type_definition        ## error, switch to a conformance checking mode for another
2456            ($token->{name});        ## language.
2457          $self->{document}->append_child ($doctype);        my $doctype_name = $token->{name};
2458          #$phase = 'root element';        $doctype_name = '' unless defined $doctype_name;
2459          !!!next-token;        $doctype_name =~ tr/a-z/A-Z/;
2460          #redo B;        if (not defined $token->{name} or # <!DOCTYPE>
2461          return;            defined $token->{public_identifier} or
2462        } elsif ({            defined $token->{system_identifier}) {
2463                  comment => 1,          !!!cp ('t1');
2464                  'start tag' => 1,          !!!parse-error (type => 'not HTML5');
2465                  'end tag' => 1,        } elsif ($doctype_name ne 'HTML') {
2466                  'end-of-file' => 1,          !!!cp ('t2');
2467                 }->{$token->{type}}) {          ## ISSUE: ASCII case-insensitive? (in fact it does not matter)
2468          ## ISSUE: Spec currently left this case undefined.          !!!parse-error (type => 'not HTML5');
2469          !!!parse-error (type => 'missing DOCTYPE');        } else {
2470          #$phase = 'root element';          !!!cp ('t3');
2471          ## reprocess        }
2472          #redo B;        
2473          return;        my $doctype = $self->{document}->create_document_type_definition
2474        } elsif ($token->{type} eq 'character') {          ($token->{name}); ## ISSUE: If name is missing (e.g. <!DOCTYPE>)?
2475          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {        $doctype->public_id ($token->{public_identifier})
2476            $self->{document}->manakai_append_text ($1);            if defined $token->{public_identifier};
2477            ## ISSUE: DOM3 Core does not allow Document > Text        $doctype->system_id ($token->{system_identifier})
2478            unless (length $token->{data}) {            if defined $token->{system_identifier};
2479              ## Stay in the phase        ## NOTE: Other DocumentType attributes are null or empty lists.
2480              !!!next-token;        ## ISSUE: internalSubset = null??
2481              redo B;        $self->{document}->append_child ($doctype);
2482          
2483          if ($token->{quirks} or $doctype_name ne 'HTML') {
2484            !!!cp ('t4');
2485            $self->{document}->manakai_compat_mode ('quirks');
2486          } elsif (defined $token->{public_identifier}) {
2487            my $pubid = $token->{public_identifier};
2488            $pubid =~ tr/a-z/A-z/;
2489            if ({
2490              "+//SILMARIL//DTD HTML PRO V0R11 19970101//EN" => 1,
2491              "-//ADVASOFT LTD//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1,
2492              "-//AS//DTD HTML 3.0 ASWEDIT + EXTENSIONS//EN" => 1,
2493              "-//IETF//DTD HTML 2.0 LEVEL 1//EN" => 1,
2494              "-//IETF//DTD HTML 2.0 LEVEL 2//EN" => 1,
2495              "-//IETF//DTD HTML 2.0 STRICT LEVEL 1//EN" => 1,
2496              "-//IETF//DTD HTML 2.0 STRICT LEVEL 2//EN" => 1,
2497              "-//IETF//DTD HTML 2.0 STRICT//EN" => 1,
2498              "-//IETF//DTD HTML 2.0//EN" => 1,
2499              "-//IETF//DTD HTML 2.1E//EN" => 1,
2500              "-//IETF//DTD HTML 3.0//EN" => 1,
2501              "-//IETF//DTD HTML 3.0//EN//" => 1,
2502              "-//IETF//DTD HTML 3.2 FINAL//EN" => 1,
2503              "-//IETF//DTD HTML 3.2//EN" => 1,
2504              "-//IETF//DTD HTML 3//EN" => 1,
2505              "-//IETF//DTD HTML LEVEL 0//EN" => 1,
2506              "-//IETF//DTD HTML LEVEL 0//EN//2.0" => 1,
2507              "-//IETF//DTD HTML LEVEL 1//EN" => 1,
2508              "-//IETF//DTD HTML LEVEL 1//EN//2.0" => 1,
2509              "-//IETF//DTD HTML LEVEL 2//EN" => 1,
2510              "-//IETF//DTD HTML LEVEL 2//EN//2.0" => 1,
2511              "-//IETF//DTD HTML LEVEL 3//EN" => 1,
2512              "-//IETF//DTD HTML LEVEL 3//EN//3.0" => 1,
2513              "-//IETF//DTD HTML STRICT LEVEL 0//EN" => 1,
2514              "-//IETF//DTD HTML STRICT LEVEL 0//EN//2.0" => 1,
2515              "-//IETF//DTD HTML STRICT LEVEL 1//EN" => 1,
2516              "-//IETF//DTD HTML STRICT LEVEL 1//EN//2.0" => 1,
2517              "-//IETF//DTD HTML STRICT LEVEL 2//EN" => 1,
2518              "-//IETF//DTD HTML STRICT LEVEL 2//EN//2.0" => 1,
2519              "-//IETF//DTD HTML STRICT LEVEL 3//EN" => 1,
2520              "-//IETF//DTD HTML STRICT LEVEL 3//EN//3.0" => 1,
2521              "-//IETF//DTD HTML STRICT//EN" => 1,
2522              "-//IETF//DTD HTML STRICT//EN//2.0" => 1,
2523              "-//IETF//DTD HTML STRICT//EN//3.0" => 1,
2524              "-//IETF//DTD HTML//EN" => 1,
2525              "-//IETF//DTD HTML//EN//2.0" => 1,
2526              "-//IETF//DTD HTML//EN//3.0" => 1,
2527              "-//METRIUS//DTD METRIUS PRESENTATIONAL//EN" => 1,
2528              "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML STRICT//EN" => 1,
2529              "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 HTML//EN" => 1,
2530              "-//MICROSOFT//DTD INTERNET EXPLORER 2.0 TABLES//EN" => 1,
2531              "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML STRICT//EN" => 1,
2532              "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 HTML//EN" => 1,
2533              "-//MICROSOFT//DTD INTERNET EXPLORER 3.0 TABLES//EN" => 1,
2534              "-//NETSCAPE COMM. CORP.//DTD HTML//EN" => 1,
2535              "-//NETSCAPE COMM. CORP.//DTD STRICT HTML//EN" => 1,
2536              "-//O'REILLY AND ASSOCIATES//DTD HTML 2.0//EN" => 1,
2537              "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED 1.0//EN" => 1,
2538              "-//O'REILLY AND ASSOCIATES//DTD HTML EXTENDED RELAXED 1.0//EN" => 1,
2539              "-//SOFTQUAD SOFTWARE//DTD HOTMETAL PRO 6.0::19990601::EXTENSIONS TO HTML 4.0//EN" => 1,
2540              "-//SOFTQUAD//DTD HOTMETAL PRO 4.0::19971010::EXTENSIONS TO HTML 4.0//EN" => 1,
2541              "-//SPYGLASS//DTD HTML 2.0 EXTENDED//EN" => 1,
2542              "-//SQ//DTD HTML 2.0 HOTMETAL + EXTENSIONS//EN" => 1,
2543              "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA HTML//EN" => 1,
2544              "-//SUN MICROSYSTEMS CORP.//DTD HOTJAVA STRICT HTML//EN" => 1,
2545              "-//W3C//DTD HTML 3 1995-03-24//EN" => 1,
2546              "-//W3C//DTD HTML 3.2 DRAFT//EN" => 1,
2547              "-//W3C//DTD HTML 3.2 FINAL//EN" => 1,
2548              "-//W3C//DTD HTML 3.2//EN" => 1,
2549              "-//W3C//DTD HTML 3.2S DRAFT//EN" => 1,
2550              "-//W3C//DTD HTML 4.0 FRAMESET//EN" => 1,
2551              "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN" => 1,
2552              "-//W3C//DTD HTML EXPERIMETNAL 19960712//EN" => 1,
2553              "-//W3C//DTD HTML EXPERIMENTAL 970421//EN" => 1,
2554              "-//W3C//DTD W3 HTML//EN" => 1,
2555              "-//W3O//DTD W3 HTML 3.0//EN" => 1,
2556              "-//W3O//DTD W3 HTML 3.0//EN//" => 1,
2557              "-//W3O//DTD W3 HTML STRICT 3.0//EN//" => 1,
2558              "-//WEBTECHS//DTD MOZILLA HTML 2.0//EN" => 1,
2559              "-//WEBTECHS//DTD MOZILLA HTML//EN" => 1,
2560              "-/W3C/DTD HTML 4.0 TRANSITIONAL/EN" => 1,
2561              "HTML" => 1,
2562            }->{$pubid}) {
2563              !!!cp ('t5');
2564              $self->{document}->manakai_compat_mode ('quirks');
2565            } elsif ($pubid eq "-//W3C//DTD HTML 4.01 FRAMESET//EN" or
2566                     $pubid eq "-//W3C//DTD HTML 4.01 TRANSITIONAL//EN") {
2567              if (defined $token->{system_identifier}) {
2568                !!!cp ('t6');
2569                $self->{document}->manakai_compat_mode ('quirks');
2570              } else {
2571                !!!cp ('t7');
2572                $self->{document}->manakai_compat_mode ('limited quirks');
2573            }            }
2574            } elsif ($pubid eq "-//W3C//DTD XHTML 1.0 FRAMESET//EN" or
2575                     $pubid eq "-//W3C//DTD XHTML 1.0 TRANSITIONAL//EN") {
2576              !!!cp ('t8');
2577              $self->{document}->manakai_compat_mode ('limited quirks');
2578            } else {
2579              !!!cp ('t9');
2580            }
2581          } else {
2582            !!!cp ('t10');
2583          }
2584          if (defined $token->{system_identifier}) {
2585            my $sysid = $token->{system_identifier};
2586            $sysid =~ tr/A-Z/a-z/;
2587            if ($sysid eq "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd") {
2588              ## TODO: Check the spec: PUBLIC "(limited quirks)" "(quirks)"
2589              $self->{document}->manakai_compat_mode ('quirks');
2590              !!!cp ('t11');
2591            } else {
2592              !!!cp ('t12');
2593          }          }
         ## ISSUE: Spec currently left this case undefined.  
         !!!parse-error (type => 'missing DOCTYPE');  
         #$phase = 'root element';  
         ## reprocess  
         #redo B;  
         return;  
2594        } else {        } else {
2595          die "$0: $token->{type}: Unknown token";          !!!cp ('t13');
2596        }        }
2597      } # B        
2598          ## Go to the "before html" insertion mode.
2599          !!!next-token;
2600          return;
2601        } elsif ({
2602                  START_TAG_TOKEN, 1,
2603                  END_TAG_TOKEN, 1,
2604                  END_OF_FILE_TOKEN, 1,
2605                 }->{$token->{type}}) {
2606          !!!cp ('t14');
2607          !!!parse-error (type => 'no DOCTYPE');
2608          $self->{document}->manakai_compat_mode ('quirks');
2609          ## Go to the "before html" insertion mode.
2610          ## reprocess
2611          return;
2612        } elsif ($token->{type} == CHARACTER_TOKEN) {
2613          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D
2614            ## Ignore the token
2615    
2616            unless (length $token->{data}) {
2617              !!!cp ('t15');
2618              ## Stay in the insertion mode.
2619              !!!next-token;
2620              redo INITIAL;
2621            } else {
2622              !!!cp ('t16');
2623            }
2624          } else {
2625            !!!cp ('t17');
2626          }
2627    
2628          !!!parse-error (type => 'no DOCTYPE');
2629          $self->{document}->manakai_compat_mode ('quirks');
2630          ## Go to the "before html" insertion mode.
2631          ## reprocess
2632          return;
2633        } elsif ($token->{type} == COMMENT_TOKEN) {
2634          !!!cp ('t18');
2635          my $comment = $self->{document}->create_comment ($token->{data});
2636          $self->{document}->append_child ($comment);
2637          
2638          ## Stay in the insertion mode.
2639          !!!next-token;
2640          redo INITIAL;
2641        } else {
2642          die "$0: $token->{type}: Unknown token type";
2643        }
2644      } # INITIAL
2645    
2646      die "$0: _tree_construction_initial: This should be never reached";
2647  } # _tree_construction_initial  } # _tree_construction_initial
2648    
2649  sub _tree_construction_root_element ($) {  sub _tree_construction_root_element ($) {
2650    my $self = shift;    my $self = shift;
2651    
2652      ## NOTE: "before html" insertion mode.
2653        
2654    B: {    B: {
2655        if ($token->{type} eq 'DOCTYPE') {        if ($token->{type} == DOCTYPE_TOKEN) {
2656            !!!cp ('t19');
2657          !!!parse-error (type => 'in html:#DOCTYPE');          !!!parse-error (type => 'in html:#DOCTYPE');
2658          ## Ignore the token          ## Ignore the token
2659          ## Stay in the phase          ## Stay in the insertion mode.
2660          !!!next-token;          !!!next-token;
2661          redo B;          redo B;
2662        } elsif ($token->{type} eq 'comment') {        } elsif ($token->{type} == COMMENT_TOKEN) {
2663            !!!cp ('t20');
2664          my $comment = $self->{document}->create_comment ($token->{data});          my $comment = $self->{document}->create_comment ($token->{data});
2665          $self->{document}->append_child ($comment);          $self->{document}->append_child ($comment);
2666          ## Stay in the phase          ## Stay in the insertion mode.
2667          !!!next-token;          !!!next-token;
2668          redo B;          redo B;
2669        } elsif ($token->{type} eq 'character') {        } elsif ($token->{type} == CHARACTER_TOKEN) {
2670          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) { # \x0D
2671            $self->{document}->manakai_append_text ($1);            ## Ignore the token.
2672            ## ISSUE: DOM3 Core does not allow Document > Text  
2673            unless (length $token->{data}) {            unless (length $token->{data}) {
2674              ## Stay in the phase              !!!cp ('t21');
2675                ## Stay in the insertion mode.
2676              !!!next-token;              !!!next-token;
2677              redo B;              redo B;
2678              } else {
2679                !!!cp ('t22');
2680            }            }
2681            } else {
2682              !!!cp ('t23');
2683          }          }
2684    
2685            $self->{application_cache_selection}->(undef);
2686    
2687          #          #
2688          } elsif ($token->{type} == START_TAG_TOKEN) {
2689            if ($token->{tag_name} eq 'html') {
2690              my $root_element;
2691              !!!create-element ($root_element, $token->{tag_name}, $token->{attributes});
2692              $self->{document}->append_child ($root_element);
2693              push @{$self->{open_elements}}, [$root_element, 'html'];
2694    
2695              if ($token->{attributes}->{manifest}) {
2696                !!!cp ('t24');
2697                $self->{application_cache_selection}
2698                    ->($token->{attributes}->{manifest}->{value});
2699                ## ISSUE: No relative reference resolution?
2700              } else {
2701                !!!cp ('t25');
2702                $self->{application_cache_selection}->(undef);
2703              }
2704    
2705              !!!next-token;
2706              return; ## Go to the "before head" insertion mode.
2707            } else {
2708              !!!cp ('t25.1');
2709              #
2710            }
2711        } elsif ({        } elsif ({
2712                  'start tag' => 1,                  END_TAG_TOKEN, 1,
2713                  'end tag' => 1,                  END_OF_FILE_TOKEN, 1,
                 'end-of-file' => 1,  
2714                 }->{$token->{type}}) {                 }->{$token->{type}}) {
2715          ## ISSUE: There is an issue in the spec          !!!cp ('t26');
2716          #          #
2717        } else {        } else {
2718          die "$0: $token->{type}: Unknown token";          die "$0: $token->{type}: Unknown token type";
2719        }        }
2720        my $root_element; !!!create-element ($root_element, 'html');  
2721        $self->{document}->append_child ($root_element);      my $root_element; !!!create-element ($root_element, 'html');
2722        push @{$self->{open_elements}}, [$root_element, 'html'];      $self->{document}->append_child ($root_element);
2723        #$phase = 'main';      push @{$self->{open_elements}}, [$root_element, 'html'];
2724        ## reprocess  
2725        #redo B;      $self->{application_cache_selection}->(undef);
2726        return;  
2727        ## NOTE: Reprocess the token.
2728        return; ## Go to the "before head" insertion mode.
2729    
2730        ## ISSUE: There is an issue in the spec
2731    } # B    } # B
2732    
2733      die "$0: _tree_construction_root_element: This should never be reached";
2734  } # _tree_construction_root_element  } # _tree_construction_root_element
2735    
2736  sub _reset_insertion_mode ($) {  sub _reset_insertion_mode ($) {
# Line 1780  sub _reset_insertion_mode ($) { Line 2745  sub _reset_insertion_mode ($) {
2745            
2746      ## Step 3      ## Step 3
2747      S3: {      S3: {
2748        $last = 1 if $self->{open_elements}->[0]->[0] eq $node->[0];        if ($self->{open_elements}->[0]->[0] eq $node->[0]) {
2749        if (defined $self->{inner_html_node}) {          $last = 1;
2750          if ($self->{inner_html_node}->[1] eq 'td' or          if (defined $self->{inner_html_node}) {
2751              $self->{inner_html_node}->[1] eq 'th') {            if ($self->{inner_html_node}->[1] eq 'td' or
2752            #                $self->{inner_html_node}->[1] eq 'th') {
2753          } else {              !!!cp ('t27');
2754            $node = $self->{inner_html_node};              #
2755              } else {
2756                !!!cp ('t28');
2757                $node = $self->{inner_html_node};
2758              }
2759          }          }
2760        }        }
2761            
2762        ## Step 4..13        ## Step 4..13
2763        my $new_mode = {        my $new_mode = {
2764                        select => 'in select',                        select => IN_SELECT_IM,
2765                        td => 'in cell',                        ## NOTE: |option| and |optgroup| do not set
2766                        th => 'in cell',                        ## insertion mode to "in select" by themselves.
2767                        tr => 'in row',                        td => IN_CELL_IM,
2768                        tbody => 'in table body',                        th => IN_CELL_IM,
2769                        thead => 'in table head',                        tr => IN_ROW_IM,
2770                        tfoot => 'in table foot',                        tbody => IN_TABLE_BODY_IM,
2771                        caption => 'in caption',                        thead => IN_TABLE_BODY_IM,
2772                        colgroup => 'in column group',                        tfoot => IN_TABLE_BODY_IM,
2773                        table => 'in table',                        caption => IN_CAPTION_IM,
2774                        head => 'in body', # not in head!                        colgroup => IN_COLUMN_GROUP_IM,
2775                        body => 'in body',                        table => IN_TABLE_IM,
2776                        frameset => 'in frameset',                        head => IN_BODY_IM, # not in head!
2777                          body => IN_BODY_IM,
2778                          frameset => IN_FRAMESET_IM,
2779                       }->{$node->[1]};                       }->{$node->[1]};
2780        $self->{insertion_mode} = $new_mode and return if defined $new_mode;        $self->{insertion_mode} = $new_mode and return if defined $new_mode;
2781                
2782        ## Step 14        ## Step 14
2783        if ($node->[1] eq 'html') {        if ($node->[1] eq 'html') {
2784          unless (defined $self->{head_element}) {          unless (defined $self->{head_element}) {
2785            $self->{insertion_mode} = 'before head';            !!!cp ('t29');
2786              $self->{insertion_mode} = BEFORE_HEAD_IM;
2787          } else {          } else {
2788            $self->{insertion_mode} = 'after head';            ## ISSUE: Can this state be reached?
2789              !!!cp ('t30');
2790              $self->{insertion_mode} = AFTER_HEAD_IM;
2791          }          }
2792          return;          return;
2793          } else {
2794            !!!cp ('t31');
2795        }        }
2796                
2797        ## Step 15        ## Step 15
2798        $self->{insertion_mode} = 'in body' and return if $last;        $self->{insertion_mode} = IN_BODY_IM and return if $last;
2799                
2800        ## Step 16        ## Step 16
2801        $i--;        $i--;
# Line 1828  sub _reset_insertion_mode ($) { Line 2804  sub _reset_insertion_mode ($) {
2804        ## Step 17        ## Step 17
2805        redo S3;        redo S3;
2806      } # S3      } # S3
2807    
2808      die "$0: _reset_insertion_mode: This line should never be reached";
2809  } # _reset_insertion_mode  } # _reset_insertion_mode
2810    
2811  sub _tree_construction_main ($) {  sub _tree_construction_main ($) {
2812    my $self = shift;    my $self = shift;
2813    
   my $phase = 'main';  
   
2814    my $active_formatting_elements = [];    my $active_formatting_elements = [];
2815    
2816    my $reconstruct_active_formatting_elements = sub { # MUST    my $reconstruct_active_formatting_elements = sub { # MUST
# Line 1851  sub _tree_construction_main ($) { Line 2827  sub _tree_construction_main ($) {
2827      return if $entry->[0] eq '#marker';      return if $entry->[0] eq '#marker';
2828      for (@{$self->{open_elements}}) {      for (@{$self->{open_elements}}) {
2829        if ($entry->[0] eq $_->[0]) {        if ($entry->[0] eq $_->[0]) {
2830            !!!cp ('t32');
2831          return;          return;
2832        }        }
2833      }      }
# Line 1865  sub _tree_construction_main ($) { Line 2842  sub _tree_construction_main ($) {
2842    
2843        ## Step 6        ## Step 6
2844        if ($entry->[0] eq '#marker') {        if ($entry->[0] eq '#marker') {
2845            !!!cp ('t33_1');
2846          #          #
2847        } else {        } else {
2848          my $in_open_elements;          my $in_open_elements;
2849          OE: for (@{$self->{open_elements}}) {          OE: for (@{$self->{open_elements}}) {
2850            if ($entry->[0] eq $_->[0]) {            if ($entry->[0] eq $_->[0]) {
2851                !!!cp ('t33');
2852              $in_open_elements = 1;              $in_open_elements = 1;
2853              last OE;              last OE;
2854            }            }
2855          }          }
2856          if ($in_open_elements) {          if ($in_open_elements) {
2857              !!!cp ('t34');
2858            #            #
2859          } else {          } else {
2860              ## NOTE: <!DOCTYPE HTML><p><b><i><u></p> <p>X
2861              !!!cp ('t35');
2862            redo S4;            redo S4;
2863          }          }
2864        }        }
# Line 1899  sub _tree_construction_main ($) { Line 2881  sub _tree_construction_main ($) {
2881    
2882        ## Step 11        ## Step 11
2883        unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) {        unless ($clone->[0] eq $active_formatting_elements->[-1]->[0]) {
2884            !!!cp ('t36');
2885          ## Step 7'          ## Step 7'
2886          $i++;          $i++;
2887          $entry = $active_formatting_elements->[$i];          $entry = $active_formatting_elements->[$i];
2888                    
2889          redo S7;          redo S7;
2890        }        }
2891    
2892          !!!cp ('t37');
2893      } # S7      } # S7
2894    }; # $reconstruct_active_formatting_elements    }; # $reconstruct_active_formatting_elements
2895    
2896    my $clear_up_to_marker = sub {    my $clear_up_to_marker = sub {
2897      for (reverse 0..$#$active_formatting_elements) {      for (reverse 0..$#$active_formatting_elements) {
2898        if ($active_formatting_elements->[$_]->[0] eq '#marker') {        if ($active_formatting_elements->[$_]->[0] eq '#marker') {
2899            !!!cp ('t38');
2900          splice @$active_formatting_elements, $_;          splice @$active_formatting_elements, $_;
2901          return;          return;
2902        }        }
2903      }      }
2904    
2905        !!!cp ('t39');
2906    }; # $clear_up_to_marker    }; # $clear_up_to_marker
2907    
2908    my $style_start_tag = sub {    my $insert;
2909      my $style_el; !!!create-element ($style_el, 'style', $token->{attributes});  
2910      ## $self->{insertion_mode} eq 'in head' and ... (always true)    my $parse_rcdata = sub ($) {
2911      (($self->{insertion_mode} eq 'in head' and defined $self->{head_element})      my ($content_model_flag) = @_;
2912       ? $self->{head_element} : $self->{open_elements}->[-1]->[0])  
2913        ->append_child ($style_el);      ## Step 1
2914      $self->{content_model_flag} = 'CDATA';      my $start_tag_name = $token->{tag_name};
2915                      my $el;
2916        !!!create-element ($el, $start_tag_name, $token->{attributes});
2917    
2918        ## Step 2
2919        $insert->($el);
2920    
2921        ## Step 3
2922        $self->{content_model} = $content_model_flag; # CDATA or RCDATA
2923        delete $self->{escape}; # MUST
2924    
2925        ## Step 4
2926      my $text = '';      my $text = '';
2927      !!!next-token;      !!!next-token;
2928      while ($token->{type} eq 'character') {      while ($token->{type} == CHARACTER_TOKEN) { # or until stop tokenizing
2929          !!!cp ('t40');
2930        $text .= $token->{data};        $text .= $token->{data};
2931        !!!next-token;        !!!next-token;
2932      } # stop if non-character token or tokenizer stops tokenising      }
2933    
2934        ## Step 5
2935      if (length $text) {      if (length $text) {
2936        $style_el->manakai_append_text ($text);        !!!cp ('t41');
2937          my $text = $self->{document}->create_text_node ($text);
2938          $el->append_child ($text);
2939      }      }
2940        
2941      $self->{content_model_flag} = 'PCDATA';      ## Step 6
2942                      $self->{content_model} = PCDATA_CONTENT_MODEL;
2943      if ($token->{type} eq 'end tag' and $token->{tag_name} eq 'style') {  
2944        ## Step 7
2945        if ($token->{type} == END_TAG_TOKEN and
2946            $token->{tag_name} eq $start_tag_name) {
2947          !!!cp ('t42');
2948        ## Ignore the token        ## Ignore the token
2949      } else {      } else {
2950        !!!parse-error (type => 'in CDATA:#'.$token->{type});        ## NOTE: An end-of-file token.
2951        ## ISSUE: And ignore?        if ($content_model_flag == CDATA_CONTENT_MODEL) {
2952            !!!cp ('t43');
2953            !!!parse-error (type => 'in CDATA:#'.$token->{type});
2954          } elsif ($content_model_flag == RCDATA_CONTENT_MODEL) {
2955            !!!cp ('t44');
2956            !!!parse-error (type => 'in RCDATA:#'.$token->{type});
2957          } else {