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

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

Parent Directory | Revision Log | View Patch Patch

-revision 1.56 by wakaba,
Sat Aug 11 07:19:18 2007 UTC
+revision 1.60 by wakaba,
Sun Oct 14 09:21:46 2007 UTC
 Line 159 
 sub CDATA_CONTENT_MODEL () { CM_LIMITED_
  sub RCDATA_CONTENT_MODEL () { CM_ENTITY | CM_LIMITED_MARKUP }
  sub PCDATA_CONTENT_MODEL () { CM_ENTITY | CM_FULL_MARKUP }
+ sub DATA_STATE () { 0 }
+ sub ENTITY_DATA_STATE () { 1 }
+ sub TAG_OPEN_STATE () { 2 }
+ sub CLOSE_TAG_OPEN_STATE () { 3 }
+ sub TAG_NAME_STATE () { 4 }
+ sub BEFORE_ATTRIBUTE_NAME_STATE () { 5 }
+ sub ATTRIBUTE_NAME_STATE () { 6 }
+ sub AFTER_ATTRIBUTE_NAME_STATE () { 7 }
+ sub BEFORE_ATTRIBUTE_VALUE_STATE () { 8 }
+ sub ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE () { 9 }
+ sub ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE () { 10 }
+ sub ATTRIBUTE_VALUE_UNQUOTED_STATE () { 11 }
+ sub ENTITY_IN_ATTRIBUTE_VALUE_STATE () { 12 }
+ sub MARKUP_DECLARATION_OPEN_STATE () { 13 }
+ sub COMMENT_START_STATE () { 14 }
+ sub COMMENT_START_DASH_STATE () { 15 }
+ sub COMMENT_STATE () { 16 }
+ sub COMMENT_END_STATE () { 17 }
+ sub COMMENT_END_DASH_STATE () { 18 }
+ sub BOGUS_COMMENT_STATE () { 19 }
+ sub DOCTYPE_STATE () { 20 }
+ sub BEFORE_DOCTYPE_NAME_STATE () { 21 }
+ sub DOCTYPE_NAME_STATE () { 22 }
+ sub AFTER_DOCTYPE_NAME_STATE () { 23 }
+ sub BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 24 }
+ sub DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE () { 25 }
+ sub DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE () { 26 }
+ sub AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE () { 27 }
+ sub BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 28 }
+ sub DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE () { 29 }
+ sub DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE () { 30 }
+ sub AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE () { 31 }
+ sub BOGUS_DOCTYPE_STATE () { 32 }
  sub DOCTYPE_TOKEN () { 1 }
  sub COMMENT_TOKEN () { 2 }
  sub START_TAG_TOKEN () { 3 }
-Line 197 
 sub IN_COLUMN_GROUP_IM () { 0b10 }
+Line 231 
 sub IN_COLUMN_GROUP_IM () { 0b10 }
  sub _initialize_tokenizer ($) {
    my $self = shift;
-   $self->{state} = 'data'; # MUST
+   $self->{state} = DATA_STATE; # MUST
    $self->{content_model} = PCDATA_CONTENT_MODEL; # be
    undef $self->{current_token}; # start tag, end tag, comment, or DOCTYPE
    undef $self->{current_attribute};
-Line 229 
 sub _initialize_tokenizer ($) {
+Line 263 
 sub _initialize_tokenizer ($) {
  ## has completed loading.  If one has, then it MUST be executed
  ## and removed from the list.
+ ## NOTE: HTML5 "Writing HTML documents" section, applied to
+ ## documents and not to user agents and conformance checkers,
+ ## contains some requirements that are not detected by the
+ ## parsing algorithm:
+ ## - Some requirements on character encoding declarations. ## TODO
+ ## - "Elements MUST NOT contain content that their content model disallows."
+ ##   ... Some are parse error, some are not (will be reported by c.c.).
+ ## - Polytheistic slash SHOULD NOT be used. (Applied only to atheists.) ## TODO
+ ## - Text (in elements, attributes, and comments) SHOULD NOT contain
+ ##   control characters other than space characters. ## TODO: (what is control character? C0, C1 and DEL?  Unicode control character?)
+ ## TODO: HTML5 poses authors two SHOULD-level requirements that cannot
+ ## be detected by the HTML5 parsing algorithm:
+ ## - Text,
  sub _get_next_token ($) {
    my $self = shift;
    if (@{$self->{token}}) {
-Line 236 
 sub _get_next_token ($) {
+Line 285 
 sub _get_next_token ($) {
    }
    A: {
-     if ($self->{state} eq 'data') {
+     if ($self->{state} == DATA_STATE) {
        if ($self->{next_input_character} == 0x0026) { # &
          if ($self->{content_model} & CM_ENTITY) { # PCDATA | RCDATA
-           $self->{state} = 'entity data';
+           $self->{state} = ENTITY_DATA_STATE;
            !!!next-input-character;
            redo A;
          } else {
-Line 261 
 sub _get_next_token ($) {
+Line 310 
 sub _get_next_token ($) {
          if ($self->{content_model} & CM_FULL_MARKUP or # PCDATA
              (($self->{content_model} & CM_LIMITED_MARKUP) and # CDATA | RCDATA
               not $self->{escape})) {
-           $self->{state} = 'tag open';
+           $self->{state} = TAG_OPEN_STATE;
            !!!next-input-character;
            redo A;
          } else {
-Line 290 
 sub _get_next_token ($) {
+Line 339 
 sub _get_next_token ($) {
        !!!emit ($token);
        redo A;
-     } elsif ($self->{state} eq 'entity data') {
+     } elsif ($self->{state} == ENTITY_DATA_STATE) {
        ## (cannot happen in CDATA state)
        my $token = $self->_tokenize_attempt_to_consume_an_entity (0);
-       $self->{state} = 'data';
+       $self->{state} = DATA_STATE;
        # next-input-character is already done
        unless (defined $token) {
-Line 305 
 sub _get_next_token ($) {
+Line 354 
 sub _get_next_token ($) {
        }
        redo A;
-     } elsif ($self->{state} eq 'tag open') {
+     } elsif ($self->{state} == TAG_OPEN_STATE) {
        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
          if ($self->{next_input_character} == 0x002F) { # /
            !!!next-input-character;
-           $self->{state} = 'close tag open';
+           $self->{state} = CLOSE_TAG_OPEN_STATE;
            redo A;
          } else {
            ## reconsume
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
-Line 321 
 sub _get_next_token ($) {
+Line 370 
 sub _get_next_token ($) {
          }
        } elsif ($self->{content_model} & CM_FULL_MARKUP) { # PCDATA
          if ($self->{next_input_character} == 0x0021) { # !
-           $self->{state} = 'markup declaration open';
+           $self->{state} = MARKUP_DECLARATION_OPEN_STATE;
            !!!next-input-character;
            redo A;
          } elsif ($self->{next_input_character} == 0x002F) { # /
-           $self->{state} = 'close tag open';
+           $self->{state} = CLOSE_TAG_OPEN_STATE;
            !!!next-input-character;
            redo A;
          } elsif (0x0041 <= $self->{next_input_character} and
-Line 333 
 sub _get_next_token ($) {
+Line 382 
 sub _get_next_token ($) {
            $self->{current_token}
              = {type => START_TAG_TOKEN,
                 tag_name => chr ($self->{next_input_character} + 0x0020)};
-           $self->{state} = 'tag name';
+           $self->{state} = TAG_NAME_STATE;
            !!!next-input-character;
            redo A;
          } elsif (0x0061 <= $self->{next_input_character} and
                   $self->{next_input_character} <= 0x007A) { # a..z
            $self->{current_token} = {type => START_TAG_TOKEN,
                              tag_name => chr ($self->{next_input_character})};
-           $self->{state} = 'tag name';
+           $self->{state} = TAG_NAME_STATE;
            !!!next-input-character;
            redo A;
          } elsif ($self->{next_input_character} == 0x003E) { # >
            !!!parse-error (type => 'empty start tag');
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            !!!next-input-character;
            !!!emit ({type => CHARACTER_TOKEN, data => '<>'});
-Line 353 
 sub _get_next_token ($) {
+Line 402 
 sub _get_next_token ($) {
            redo A;
          } elsif ($self->{next_input_character} == 0x003F) { # ?
            !!!parse-error (type => 'pio');
-           $self->{state} = 'bogus comment';
+           $self->{state} = BOGUS_COMMENT_STATE;
            ## $self->{next_input_character} is intentionally left as is
            redo A;
          } else {
            !!!parse-error (type => 'bare stago');
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            ## reconsume
            !!!emit ({type => CHARACTER_TOKEN, data => '<'});
-Line 368 
 sub _get_next_token ($) {
+Line 417 
 sub _get_next_token ($) {
        } else {
          die "$0: $self->{content_model} in tag open";
        }
-     } elsif ($self->{state} eq 'close tag open') {
+     } elsif ($self->{state} == CLOSE_TAG_OPEN_STATE) {
        if ($self->{content_model} & CM_LIMITED_MARKUP) { # RCDATA | CDATA
          if (defined $self->{last_emitted_start_tag_name}) {
            ## NOTE: <http://krijnhoetmer.nl/irc-logs/whatwg/20070626#l-564>
-Line 383 
 sub _get_next_token ($) {
+Line 432 
 sub _get_next_token ($) {
              } else {
                $self->{next_input_character} = shift @next_char; # reconsume
                !!!back-next-input-character (@next_char);
-               $self->{state} = 'data';
+               $self->{state} = DATA_STATE;
                !!!emit ({type => CHARACTER_TOKEN, data => '</'});
-Line 402 
 sub _get_next_token ($) {
+Line 451 
 sub _get_next_token ($) {
                    $self->{next_input_character} == -1) {
              $self->{next_input_character} = shift @next_char; # reconsume
              !!!back-next-input-character (@next_char);
-             $self->{state} = 'data';
+             $self->{state} = DATA_STATE;
              !!!emit ({type => CHARACTER_TOKEN, data => '</'});
              redo A;
            } else {
-Line 413 
 sub _get_next_token ($) {
+Line 462 
 sub _get_next_token ($) {
          } else {
            ## No start tag token has ever been emitted
            # next-input-character is already done
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            !!!emit ({type => CHARACTER_TOKEN, data => '</'});
            redo A;
          }
-Line 423 
 sub _get_next_token ($) {
+Line 472 
 sub _get_next_token ($) {
            $self->{next_input_character} <= 0x005A) { # A..Z
          $self->{current_token} = {type => END_TAG_TOKEN,
                            tag_name => chr ($self->{next_input_character} + 0x0020)};
-         $self->{state} = 'tag name';
+         $self->{state} = TAG_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif (0x0061 <= $self->{next_input_character} and
                 $self->{next_input_character} <= 0x007A) { # a..z
          $self->{current_token} = {type => END_TAG_TOKEN,
                            tag_name => chr ($self->{next_input_character})};
-         $self->{state} = 'tag name';
+         $self->{state} = TAG_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
          !!!parse-error (type => 'empty end tag');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'bare etago');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          # reconsume
          !!!emit ({type => CHARACTER_TOKEN, data => '</'});
-Line 448 
 sub _get_next_token ($) {
+Line 497 
 sub _get_next_token ($) {
          redo A;
        } else {
          !!!parse-error (type => 'bogus end tag');
-         $self->{state} = 'bogus comment';
+         $self->{state} = BOGUS_COMMENT_STATE;
          ## $self->{next_input_character} is intentionally left as is
          redo A;
        }
-     } elsif ($self->{state} eq 'tag name') {
+     } elsif ($self->{state} == TAG_NAME_STATE) {
        if ($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) { # SP
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-Line 474 
 sub _get_next_token ($) {
+Line 523 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 501 
 sub _get_next_token ($) {
+Line 550 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          # reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 517 
 sub _get_next_token ($) {
+Line 566 
 sub _get_next_token ($) {
          } else {
            !!!parse-error (type => 'nestc');
          }
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          # next-input-character is already done
          redo A;
        } else {
-Line 527 
 sub _get_next_token ($) {
+Line 576 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'before attribute name') {
+     } elsif ($self->{state} == BEFORE_ATTRIBUTE_NAME_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # VT
-Line 549 
 sub _get_next_token ($) {
+Line 598 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 559 
 sub _get_next_token ($) {
+Line 608 
 sub _get_next_token ($) {
                 $self->{next_input_character} <= 0x005A) { # A..Z
          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),
                                value => ''};
-         $self->{state} = 'attribute name';
+         $self->{state} = ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x002F) { # /
-Line 589 
 sub _get_next_token ($) {
+Line 638 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          # reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 598 
 sub _get_next_token ($) {
+Line 647 
 sub _get_next_token ($) {
        } else {
          $self->{current_attribute} = {name => chr ($self->{next_input_character}),
                                value => ''};
-         $self->{state} = 'attribute name';
+         $self->{state} = ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'attribute name') {
+     } elsif ($self->{state} == ATTRIBUTE_NAME_STATE) {
        my $before_leave = sub {
          if (exists $self->{current_token}->{attributes} # start tag or end tag
              ->{$self->{current_attribute}->{name}}) { # MUST
-Line 620 
 sub _get_next_token ($) {
+Line 669 
 sub _get_next_token ($) {
            $self->{next_input_character} == 0x000C or # FF
            $self->{next_input_character} == 0x0020) { # SP
          $before_leave->();
-         $self->{state} = 'after attribute name';
+         $self->{state} = AFTER_ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003D) { # =
          $before_leave->();
-         $self->{state} = 'before attribute value';
+         $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-Line 642 
 sub _get_next_token ($) {
+Line 691 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 665 
 sub _get_next_token ($) {
+Line 714 
 sub _get_next_token ($) {
          } else {
            !!!parse-error (type => 'nestc');
          }
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          # next-input-character is already done
          redo A;
        } elsif ($self->{next_input_character} == -1) {
-Line 683 
 sub _get_next_token ($) {
+Line 732 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          # reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 695 
 sub _get_next_token ($) {
+Line 744 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'after attribute name') {
+     } elsif ($self->{state} == AFTER_ATTRIBUTE_NAME_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # VT
-Line 705 
 sub _get_next_token ($) {
+Line 754 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003D) { # =
-         $self->{state} = 'before attribute value';
+         $self->{state} = BEFORE_ATTRIBUTE_VALUE_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-Line 721 
 sub _get_next_token ($) {
+Line 770 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 731 
 sub _get_next_token ($) {
+Line 780 
 sub _get_next_token ($) {
                 $self->{next_input_character} <= 0x005A) { # A..Z
          $self->{current_attribute} = {name => chr ($self->{next_input_character} + 0x0020),
                                value => ''};
-         $self->{state} = 'attribute name';
+         $self->{state} = ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x002F) { # /
-Line 745 
 sub _get_next_token ($) {
+Line 794 
 sub _get_next_token ($) {
            !!!parse-error (type => 'nestc');
            ## TODO: Different error type for <aa / bb> than <aa/>
          }
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          # next-input-character is already done
          redo A;
        } elsif ($self->{next_input_character} == -1) {
-Line 762 
 sub _get_next_token ($) {
+Line 811 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          # reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 771 
 sub _get_next_token ($) {
+Line 820 
 sub _get_next_token ($) {
        } else {
          $self->{current_attribute} = {name => chr ($self->{next_input_character}),
                                value => ''};
-         $self->{state} = 'attribute name';
+         $self->{state} = ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'before attribute value') {
+     } elsif ($self->{state} == BEFORE_ATTRIBUTE_VALUE_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # VT
-Line 785 
 sub _get_next_token ($) {
+Line 834 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0022) { # "
-         $self->{state} = 'attribute value (double-quoted)';
+         $self->{state} = ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0026) { # &
-         $self->{state} = 'attribute value (unquoted)';
+         $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
          ## reconsume
          redo A;
        } elsif ($self->{next_input_character} == 0x0027) { # '
-         $self->{state} = 'attribute value (single-quoted)';
+         $self->{state} = ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-Line 809 
 sub _get_next_token ($) {
+Line 858 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 829 
 sub _get_next_token ($) {
+Line 878 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 837 
 sub _get_next_token ($) {
+Line 886 
 sub _get_next_token ($) {
          redo A;
        } else {
          $self->{current_attribute}->{value} .= chr ($self->{next_input_character});
-         $self->{state} = 'attribute value (unquoted)';
+         $self->{state} = ATTRIBUTE_VALUE_UNQUOTED_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'attribute value (double-quoted)') {
+     } elsif ($self->{state} == ATTRIBUTE_VALUE_DOUBLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0022) { # "
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0026) { # &
-         $self->{last_attribute_value_state} = 'attribute value (double-quoted)';
+         $self->{last_attribute_value_state} = $self->{state};
-         $self->{state} = 'entity in attribute value';
+         $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
-Line 865 
 sub _get_next_token ($) {
+Line 914 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 877 
 sub _get_next_token ($) {
+Line 926 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'attribute value (single-quoted)') {
+     } elsif ($self->{state} == ATTRIBUTE_VALUE_SINGLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0027) { # '
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0026) { # &
-         $self->{last_attribute_value_state} = 'attribute value (single-quoted)';
+         $self->{last_attribute_value_state} = $self->{state};
-         $self->{state} = 'entity in attribute value';
+         $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
-Line 901 
 sub _get_next_token ($) {
+Line 950 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 913 
 sub _get_next_token ($) {
+Line 962 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'attribute value (unquoted)') {
+     } elsif ($self->{state} == ATTRIBUTE_VALUE_UNQUOTED_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # HT
            $self->{next_input_character} == 0x000C or # FF
            $self->{next_input_character} == 0x0020) { # SP
-         $self->{state} = 'before attribute name';
+         $self->{state} = BEFORE_ATTRIBUTE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0026) { # &
-         $self->{last_attribute_value_state} = 'attribute value (unquoted)';
+         $self->{last_attribute_value_state} = $self->{state};
-         $self->{state} = 'entity in attribute value';
+         $self->{state} = ENTITY_IN_ATTRIBUTE_VALUE_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-Line 940 
 sub _get_next_token ($) {
+Line 989 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 960 
 sub _get_next_token ($) {
+Line 1009 
 sub _get_next_token ($) {
          } else {
            die "$0: $self->{current_token}->{type}: Unknown token type";
          }
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # start tag or end tag
-Line 972 
 sub _get_next_token ($) {
+Line 1021 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'entity in attribute value') {
+     } elsif ($self->{state} == ENTITY_IN_ATTRIBUTE_VALUE_STATE) {
        my $token = $self->_tokenize_attempt_to_consume_an_entity (1);
        unless (defined $token) {
-Line 985 
 sub _get_next_token ($) {
+Line 1034 
 sub _get_next_token ($) {
        $self->{state} = $self->{last_attribute_value_state};
        # next-input-character is already done
        redo A;
-     } elsif ($self->{state} eq 'bogus comment') {
+     } elsif ($self->{state} == BOGUS_COMMENT_STATE) {
        ## (only happen if PCDATA state)
        my $token = {type => COMMENT_TOKEN, data => ''};
        BC: {
          if ($self->{next_input_character} == 0x003E) { # >
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            !!!next-input-character;
            !!!emit ($token);
            redo A;
          } elsif ($self->{next_input_character} == -1) {
-           $self->{state} = 'data';
+           $self->{state} = DATA_STATE;
            ## reconsume
            !!!emit ($token);
-Line 1011 
 sub _get_next_token ($) {
+Line 1060 
 sub _get_next_token ($) {
            redo BC;
          }
        } # BC
-     } elsif ($self->{state} eq 'markup declaration open') {
+     } elsif ($self->{state} == MARKUP_DECLARATION_OPEN_STATE) {
        ## (only happen if PCDATA state)
        my @next_char;
-Line 1022 
 sub _get_next_token ($) {
+Line 1071 
 sub _get_next_token ($) {
          push @next_char, $self->{next_input_character};
          if ($self->{next_input_character} == 0x002D) { # -
            $self->{current_token} = {type => COMMENT_TOKEN, data => ''};
-           $self->{state} = 'comment start';
+           $self->{state} = COMMENT_START_STATE;
            !!!next-input-character;
            redo A;
          }
-Line 1053 
 sub _get_next_token ($) {
+Line 1102 
 sub _get_next_token ($) {
                    if ($self->{next_input_character} == 0x0045 or # E
                        $self->{next_input_character} == 0x0065) { # e
                      ## ISSUE: What a stupid code this is!
-                     $self->{state} = 'DOCTYPE';
+                     $self->{state} = DOCTYPE_STATE;
                      !!!next-input-character;
                      redo A;
                    }
-Line 1067 
 sub _get_next_token ($) {
+Line 1116 
 sub _get_next_token ($) {
        !!!parse-error (type => 'bogus comment');
        $self->{next_input_character} = shift @next_char;
        !!!back-next-input-character (@next_char);
-       $self->{state} = 'bogus comment';
+       $self->{state} = BOGUS_COMMENT_STATE;
        redo A;
        ## ISSUE: typos in spec: chacacters, is is a parse error
        ## ISSUE: spec is somewhat unclear on "is the first character that will be in the comment"; what is "that will be in the comment" is what the algorithm defines, isn't it?
-     } elsif ($self->{state} eq 'comment start') {
+     } elsif ($self->{state} == COMMENT_START_STATE) {
        if ($self->{next_input_character} == 0x002D) { # -
-         $self->{state} = 'comment start dash';
+         $self->{state} = COMMENT_START_DASH_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
          !!!parse-error (type => 'bogus comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # comment
-Line 1087 
 sub _get_next_token ($) {
+Line 1136 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # comment
-Line 1096 
 sub _get_next_token ($) {
+Line 1145 
 sub _get_next_token ($) {
        } else {
          $self->{current_token}->{data} # comment
              .= chr ($self->{next_input_character});
-         $self->{state} = 'comment';
+         $self->{state} = COMMENT_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'comment start dash') {
+     } elsif ($self->{state} == COMMENT_START_DASH_STATE) {
        if ($self->{next_input_character} == 0x002D) { # -
-         $self->{state} = 'comment end';
+         $self->{state} = COMMENT_END_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
          !!!parse-error (type => 'bogus comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # comment
-Line 1115 
 sub _get_next_token ($) {
+Line 1164 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # comment
-Line 1124 
 sub _get_next_token ($) {
+Line 1173 
 sub _get_next_token ($) {
        } else {
          $self->{current_token}->{data} # comment
              .= '-' . chr ($self->{next_input_character});
-         $self->{state} = 'comment';
+         $self->{state} = COMMENT_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'comment') {
+     } elsif ($self->{state} == COMMENT_STATE) {
        if ($self->{next_input_character} == 0x002D) { # -
-         $self->{state} = 'comment end dash';
+         $self->{state} = COMMENT_END_DASH_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # comment
-Line 1147 
 sub _get_next_token ($) {
+Line 1196 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'comment end dash') {
+     } elsif ($self->{state} == COMMENT_END_DASH_STATE) {
        if ($self->{next_input_character} == 0x002D) { # -
-         $self->{state} = 'comment end';
+         $self->{state} = COMMENT_END_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # comment
-Line 1162 
 sub _get_next_token ($) {
+Line 1211 
 sub _get_next_token ($) {
          redo A;
        } else {
          $self->{current_token}->{data} .= '-' . chr ($self->{next_input_character}); # comment
-         $self->{state} = 'comment';
+         $self->{state} = COMMENT_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'comment end') {
+     } elsif ($self->{state} == COMMENT_END_STATE) {
        if ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # comment
-Line 1182 
 sub _get_next_token ($) {
+Line 1231 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed comment');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ($self->{current_token}); # comment
-Line 1191 
 sub _get_next_token ($) {
+Line 1240 
 sub _get_next_token ($) {
        } else {
          !!!parse-error (type => 'dash in comment');
          $self->{current_token}->{data} .= '--' . chr ($self->{next_input_character}); # comment
-         $self->{state} = 'comment';
+         $self->{state} = COMMENT_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE') {
+     } elsif ($self->{state} == DOCTYPE_STATE) {
        if ($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) { # SP
-         $self->{state} = 'before DOCTYPE name';
+         $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } else {
          !!!parse-error (type => 'no space before DOCTYPE name');
-         $self->{state} = 'before DOCTYPE name';
+         $self->{state} = BEFORE_DOCTYPE_NAME_STATE;
          ## reconsume
          redo A;
        }
-     } elsif ($self->{state} eq 'before DOCTYPE name') {
+     } elsif ($self->{state} == BEFORE_DOCTYPE_NAME_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # VT
-Line 1221 
 sub _get_next_token ($) {
+Line 1270 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
          !!!parse-error (type => 'no DOCTYPE name');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect
-Line 1229 
 sub _get_next_token ($) {
+Line 1278 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'no DOCTYPE name');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          !!!emit ({type => DOCTYPE_TOKEN}); # incorrect
-Line 1241 
 sub _get_next_token ($) {
+Line 1290 
 sub _get_next_token ($) {
                 name => chr ($self->{next_input_character}),
                 correct => 1};
  ## ISSUE: "Set the token's name name to the" in the spec
-         $self->{state} = 'DOCTYPE name';
+         $self->{state} = DOCTYPE_NAME_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE name') {
+     } elsif ($self->{state} == DOCTYPE_NAME_STATE) {
  ## ISSUE: Redundant "First," in the spec.
        if ($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) { # SP
-         $self->{state} = 'after DOCTYPE name';
+         $self->{state} = AFTER_DOCTYPE_NAME_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # DOCTYPE
-Line 1264 
 sub _get_next_token ($) {
+Line 1313 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1278 
 sub _get_next_token ($) {
+Line 1327 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'after DOCTYPE name') {
+     } elsif ($self->{state} == AFTER_DOCTYPE_NAME_STATE) {
        if ($self->{next_input_character} == 0x0009 or # HT
            $self->{next_input_character} == 0x000A or # LF
            $self->{next_input_character} == 0x000B or # VT
-Line 1288 
 sub _get_next_token ($) {
+Line 1337 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # DOCTYPE
-Line 1296 
 sub _get_next_token ($) {
+Line 1345 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1320 
 sub _get_next_token ($) {
+Line 1369 
 sub _get_next_token ($) {
                  !!!next-input-character;
                  if ($self->{next_input_character} == 0x0043 or # C
                      $self->{next_input_character} == 0x0063) { # c
-                   $self->{state} = 'before DOCTYPE public identifier';
+                   $self->{state} = BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
                    !!!next-input-character;
                    redo A;
                  }
-Line 1347 
 sub _get_next_token ($) {
+Line 1396 
 sub _get_next_token ($) {
                  !!!next-input-character;
                  if ($self->{next_input_character} == 0x004D or # M
                      $self->{next_input_character} == 0x006D) { # m
-                   $self->{state} = 'before DOCTYPE system identifier';
+                   $self->{state} = BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
                    !!!next-input-character;
                    redo A;
                  }
-Line 1363 
 sub _get_next_token ($) {
+Line 1412 
 sub _get_next_token ($) {
        }
        !!!parse-error (type => 'string after DOCTYPE name');
-       $self->{state} = 'bogus DOCTYPE';
+       $self->{state} = BOGUS_DOCTYPE_STATE;
        # next-input-character is already done
        redo A;
-     } elsif ($self->{state} eq 'before DOCTYPE public identifier') {
+     } elsif ($self->{state} == BEFORE_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
        if ({
 x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
              #0x000D => 1, # HT, LF, VT, FF, SP, CR
-Line 1376 
 sub _get_next_token ($) {
+Line 1425 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} eq 0x0022) { # "
          $self->{current_token}->{public_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE public identifier (double-quoted)';
+         $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} eq 0x0027) { # '
          $self->{current_token}->{public_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE public identifier (single-quoted)';
+         $self->{state} = DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} eq 0x003E) { # >
          !!!parse-error (type => 'no PUBLIC literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          delete $self->{current_token}->{correct};
-Line 1397 
 sub _get_next_token ($) {
+Line 1446 
 sub _get_next_token ($) {
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1406 
 sub _get_next_token ($) {
+Line 1455 
 sub _get_next_token ($) {
          redo A;
        } else {
          !!!parse-error (type => 'string after PUBLIC');
-         $self->{state} = 'bogus DOCTYPE';
+         $self->{state} = BOGUS_DOCTYPE_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE public identifier (double-quoted)') {
+     } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_DOUBLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0022) { # "
-         $self->{state} = 'after DOCTYPE public identifier';
+         $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed PUBLIC literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1432 
 sub _get_next_token ($) {
+Line 1481 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE public identifier (single-quoted)') {
+     } elsif ($self->{state} == DOCTYPE_PUBLIC_IDENTIFIER_SINGLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0027) { # '
-         $self->{state} = 'after DOCTYPE public identifier';
+         $self->{state} = AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed PUBLIC literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1454 
 sub _get_next_token ($) {
+Line 1503 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'after DOCTYPE public identifier') {
+     } elsif ($self->{state} == AFTER_DOCTYPE_PUBLIC_IDENTIFIER_STATE) {
        if ({
 x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
              #0x000D => 1, # HT, LF, VT, FF, SP, CR
-Line 1464 
 sub _get_next_token ($) {
+Line 1513 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == 0x0022) { # "
          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE system identifier (double-quoted)';
+         $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0027) { # '
          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE system identifier (single-quoted)';
+         $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # DOCTYPE
-Line 1482 
 sub _get_next_token ($) {
+Line 1531 
 sub _get_next_token ($) {
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1491 
 sub _get_next_token ($) {
+Line 1540 
 sub _get_next_token ($) {
          redo A;
        } else {
          !!!parse-error (type => 'string after PUBLIC literal');
-         $self->{state} = 'bogus DOCTYPE';
+         $self->{state} = BOGUS_DOCTYPE_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'before DOCTYPE system identifier') {
+     } elsif ($self->{state} == BEFORE_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
        if ({
 x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
              #0x000D => 1, # HT, LF, VT, FF, SP, CR
-Line 1505 
 sub _get_next_token ($) {
+Line 1554 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == 0x0022) { # "
          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE system identifier (double-quoted)';
+         $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x0027) { # '
          $self->{current_token}->{system_identifier} = ''; # DOCTYPE
-         $self->{state} = 'DOCTYPE system identifier (single-quoted)';
+         $self->{state} = DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
          !!!parse-error (type => 'no SYSTEM literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          delete $self->{current_token}->{correct};
-Line 1525 
 sub _get_next_token ($) {
+Line 1574 
 sub _get_next_token ($) {
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1534 
 sub _get_next_token ($) {
+Line 1583 
 sub _get_next_token ($) {
          redo A;
        } else {
          !!!parse-error (type => 'string after SYSTEM');
-         $self->{state} = 'bogus DOCTYPE';
+         $self->{state} = BOGUS_DOCTYPE_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE system identifier (double-quoted)') {
+     } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_DOUBLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0022) { # "
-         $self->{state} = 'after DOCTYPE system identifier';
+         $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed SYSTEM literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1560 
 sub _get_next_token ($) {
+Line 1609 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'DOCTYPE system identifier (single-quoted)') {
+     } elsif ($self->{state} == DOCTYPE_SYSTEM_IDENTIFIER_SINGLE_QUOTED_STATE) {
        if ($self->{next_input_character} == 0x0027) { # '
-         $self->{state} = 'after DOCTYPE system identifier';
+         $self->{state} = AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE;
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed SYSTEM literal');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1582 
 sub _get_next_token ($) {
+Line 1631 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'after DOCTYPE system identifier') {
+     } elsif ($self->{state} == AFTER_DOCTYPE_SYSTEM_IDENTIFIER_STATE) {
        if ({
 x0009 => 1, 0x000A => 1, 0x000B => 1, 0x000C => 1, 0x0020 => 1,
              #0x000D => 1, # HT, LF, VT, FF, SP, CR
-Line 1591 
 sub _get_next_token ($) {
+Line 1640 
 sub _get_next_token ($) {
          !!!next-input-character;
          redo A;
        } elsif ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          !!!emit ($self->{current_token}); # DOCTYPE
-Line 1600 
 sub _get_next_token ($) {
+Line 1649 
 sub _get_next_token ($) {
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 1609 
 sub _get_next_token ($) {
+Line 1658 
 sub _get_next_token ($) {
          redo A;
        } else {
          !!!parse-error (type => 'string after SYSTEM literal');
-         $self->{state} = 'bogus DOCTYPE';
+         $self->{state} = BOGUS_DOCTYPE_STATE;
          !!!next-input-character;
          redo A;
        }
-     } elsif ($self->{state} eq 'bogus DOCTYPE') {
+     } elsif ($self->{state} == BOGUS_DOCTYPE_STATE) {
        if ($self->{next_input_character} == 0x003E) { # >
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          !!!next-input-character;
          delete $self->{current_token}->{correct};
-Line 1624 
 sub _get_next_token ($) {
+Line 1673 
 sub _get_next_token ($) {
          redo A;
        } elsif ($self->{next_input_character} == -1) {
          !!!parse-error (type => 'unclosed DOCTYPE');
-         $self->{state} = 'data';
+         $self->{state} = DATA_STATE;
          ## reconsume
          delete $self->{current_token}->{correct};
-Line 2057 
 sub _tree_construction_root_element ($)
+Line 2106 
 sub _tree_construction_root_element ($)
        } else {
          die "$0: $token->{type}: Unknown token type";
        }
+       ## TODO: application cache selection algorithm
        my $root_element; !!!create-element ($root_element, 'html');
        $self->{document}->append_child ($root_element);
        push @{$self->{open_elements}}, [$root_element, 'html'];
-Line 3302 
 sub _tree_construction_main ($) {
+Line 3352 
 sub _tree_construction_main ($) {
        $insert = $insert_to_current;
        #
      } elsif ($self->{insertion_mode} & TABLE_IMS) {
-           if ($token->{type} == CHARACTER_TOKEN) {
+       if ($token->{type} == CHARACTER_TOKEN) {
-             ## NOTE: There are "character in table" code clones.
              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
                $self->{open_elements}->[-1]->[0]->manakai_append_text ($1);
-Line 3360 
 sub _tree_construction_main ($) {
+Line 3409 
 sub _tree_construction_main ($) {
              !!!next-token;
              redo B;
-           } elsif ($token->{type} == START_TAG_TOKEN) {
+       } elsif ($token->{type} == START_TAG_TOKEN) {
              if ({
                   tr => ($self->{insertion_mode} != IN_ROW_IM),
                   th => 1, td => 1,
-Line 3546 
 sub _tree_construction_main ($) {
+Line 3595 
 sub _tree_construction_main ($) {
                  die "$0: in table: <>: $token->{tag_name}";
                }
              } elsif ($token->{tag_name} eq 'table') {
-               ## NOTE: There are code clones for this "table in table"
                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
                ## As if </table>
-Line 3594 
 sub _tree_construction_main ($) {
+Line 3642 
 sub _tree_construction_main ($) {
                ## reprocess
                redo B;
-             } else {
+         } else {
-               #
+           !!!parse-error (type => 'in table:'.$token->{tag_name});
-             }
-           } elsif ($token->{type} == END_TAG_TOKEN) {
+           $insert = $insert_to_foster;
+           #
+         }
+       } elsif ($token->{type} == END_TAG_TOKEN) {
              if ($token->{tag_name} eq 'tr' and
                  $self->{insertion_mode} == IN_ROW_IM) {
                ## have an element in table scope
-Line 3854 
 sub _tree_construction_main ($) {
+Line 3905 
 sub _tree_construction_main ($) {
                ## Ignore the token
                !!!next-token;
                redo B;
-             } else {
+         } else {
-               #
+           !!!parse-error (type => 'in table:/'.$token->{tag_name});
-             }
-           } else {
-             die "$0: $token->{type}: Unknown token type";
-           }
-       !!!parse-error (type => 'in table:'.$token->{tag_name});
-       $insert = $insert_to_foster;
+           $insert = $insert_to_foster;
-       #
+           #
+         }
+       } else {
+         die "$0: $token->{type}: Unknown token type";
+       }
      } elsif ($self->{insertion_mode} == IN_COLUMN_GROUP_IM) {
            if ($token->{type} == CHARACTER_TOKEN) {
              if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
-Line 3923 
 sub _tree_construction_main ($) {
+Line 3972 
 sub _tree_construction_main ($) {
              redo B;
            }
      } elsif ($self->{insertion_mode} == IN_SELECT_IM) {
-           if ($token->{type} == CHARACTER_TOKEN) {
+       if ($token->{type} == CHARACTER_TOKEN) {
-             $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data});
+         $self->{open_elements}->[-1]->[0]->manakai_append_text ($token->{data});
-             !!!next-token;
+         !!!next-token;
-             redo B;
+         redo B;
-           } elsif ($token->{type} == START_TAG_TOKEN) {
+       } elsif ($token->{type} == START_TAG_TOKEN) {
              if ($token->{tag_name} eq 'option') {
                if ($self->{open_elements}->[-1]->[1] eq 'option') {
                  ## As if </option>
-Line 3980 
 sub _tree_construction_main ($) {
+Line 4029 
 sub _tree_construction_main ($) {
                !!!next-token;
                redo B;
-             } else {
+         } else {
-               #
+           !!!parse-error (type => 'in select:'.$token->{tag_name});
-             }
+           ## Ignore the token
-           } elsif ($token->{type} == END_TAG_TOKEN) {
+           !!!next-token;
+           redo B;
+         }
+       } elsif ($token->{type} == END_TAG_TOKEN) {
              if ($token->{tag_name} eq 'optgroup') {
                if ($self->{open_elements}->[-1]->[1] eq 'option' and
                    $self->{open_elements}->[-2]->[1] eq 'optgroup') {
-Line 4085 
 sub _tree_construction_main ($) {
+Line 4137 
 sub _tree_construction_main ($) {
                ## reprocess
                redo B;
-             } else {
+         } else {
-               #
+           !!!parse-error (type => 'in select:/'.$token->{tag_name});
-             }
-           } else {
-             #
-           }
-           !!!parse-error (type => 'in select:'.$token->{tag_name});
            ## Ignore the token
            !!!next-token;
            redo B;
+         }
+       } else {
+         die "$0: $token->{type}: Unknown token type";
+       }
      } elsif ($self->{insertion_mode} & BODY_AFTER_IMS) {
        if ($token->{type} == CHARACTER_TOKEN) {
          if ($token->{data} =~ s/^([\x09\x0A\x0B\x0C\x20]+)//) {
-Line 4626 
 sub _tree_construction_main ($) {
+Line 4676 
 sub _tree_construction_main ($) {
          INSCOPE: for (reverse 0..$#{$self->{open_elements}}) {
            my $node = $self->{open_elements}->[$_];
            if ($node->[1] eq 'nobr') {
-             !!!parse-error (type => 'not closed:nobr');
+             !!!parse-error (type => 'in nobr:nobr');
              !!!back-token;
              $token = {type => END_TAG_TOKEN, tag_name => 'nobr'};
              redo B;
-Line 4831 
 sub _tree_construction_main ($) {
+Line 4881 
 sub _tree_construction_main ($) {
                  noframes => 1,
                  noscript => 0, ## TODO: 1 if scripting is enabled
                 }->{$token->{tag_name}}) {
-         ## NOTE: There are two "as if in body" code clones.
+         ## NOTE: There is an "as if in body" code clone.
          $parse_rcdata->(CDATA_CONTENT_MODEL, $insert);
          redo B;
        } elsif ($token->{tag_name} eq 'select') {
-Line 4987 
 sub _tree_construction_main ($) {
+Line 5037 
 sub _tree_construction_main ($) {
          if ($self->{open_elements}->[-1]->[1] eq $token->{tag_name}) {
            pop @{$self->{open_elements}};
          } else {
-           !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
+           !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});
          }
          undef $self->{form_element};
-Line 5025 
 sub _tree_construction_main ($) {
+Line 5075 
 sub _tree_construction_main ($) {
          } # INSCOPE
          if ($self->{open_elements}->[-1]->[1] ne $token->{tag_name}) {
-           !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
+           !!!parse-error (type => 'unmatched end tag:'.$token->{tag_name});
          }
          splice @{$self->{open_elements}}, $i if defined $i;
-Line 5094 
 sub _tree_construction_main ($) {
+Line 5144 
 sub _tree_construction_main ($) {
              ## Step 2
              if ($token->{tag_name} ne $self->{open_elements}->[-1]->[1]) {
+               ## NOTE: <x><y></x>
                !!!parse-error (type => 'not closed:'.$self->{open_elements}->[-1]->[1]);
              }

 Legend:



Removed from v.1.56
 


changed lines


 
Added in v.1.60
 Legend:



Removed from v.1.56
 


changed lines


 
Added in v.1.60
-Removed from v.1.56
+Added in v.1.60

[email protected]	ViewVC Help
Powered by ViewVC 1.1.24