| 1 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 2 |
|
|
| 3 |
|
* HTML.pm.src: Make a "bare ero" error for unknown |
| 4 |
|
entities point the "&" character. |
| 5 |
|
|
| 6 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 7 |
|
|
| 8 |
|
* HTML.pm.src: It turns out that U+FFFD don't have to |
| 9 |
|
be added to the list of excluded characters. |
| 10 |
|
|
| 11 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 12 |
|
|
| 13 |
|
* HTML.pm.src ($char_onerror): Have character decoder's |line| |
| 14 |
|
and |column| a higher priority than the one set by the |
| 15 |
|
tokenizer's input handler. |
| 16 |
|
($self->{read_until}): Exclude U+FFFD (but this might |
| 17 |
|
not be necessary, since now we do line/column fixup in |
| 18 |
|
the character decode handle). |
| 19 |
|
|
| 20 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 21 |
|
|
| 22 |
|
* HTML.pm.src: Use |{read_until}| where possible. |
| 23 |
|
|
| 24 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 25 |
|
|
| 26 |
|
* HTML.pm.src: Change |{getc_until}| to |{read_until}| |
| 27 |
|
and |manakai_getc_until| to |manakai_read_until| to |
| 28 |
|
reduce the number of string copies. |
| 29 |
|
|
| 30 |
|
2008-09-14 Wakaba <wakaba@suika.fam.cx> |
| 31 |
|
|
| 32 |
|
* HTML.pm.src (parse_char_string): Use newly created |
| 33 |
|
|Whatpm::Charset::DecodeHandle::CharString| instead of Perl's |
| 34 |
|
standard feature to |open| a string as a filehandle, |
| 35 |
|
since Perl's string filehandle seems not supporting |ungetc| |
| 36 |
|
method correctly. |
| 37 |
|
(parse_char_stream): Define |{getc_until}| method. |
| 38 |
|
(DATA_STATE): Experimental support for |getc_until| feature. |
| 39 |
|
|
| 40 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 41 |
|
|
| 42 |
|
* HTML.pm.src: Check points added to newly added branches. |
| 43 |
|
|
| 44 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 45 |
|
|
| 46 |
|
* HTML.pm.src: Remove |{char}|, which is no longer used. |
| 47 |
|
Remove |{entity_in_attr}| and |{last_attribute_value_state}| |
| 48 |
|
and replaced by |{prev_state}|. |
| 49 |
|
|
| 50 |
|
* mkhtmlparser.pl: Remove |{char}| feature. |
| 51 |
|
Remove |!!!back-next-input-character;| macro. |
| 52 |
|
|
| 53 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 54 |
|
|
| 55 |
|
* HTML.pm.src: Finally we get rid of all the inner loops. Remove |
| 56 |
|
entity related tokenizer states in favor of new states |
| 57 |
|
implementing the consume character reference algorithm. |
| 58 |
|
|
| 59 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 60 |
|
|
| 61 |
|
* HTML.pm.src: "Consume a character reference" algorithm is |
| 62 |
|
now implemented as a tokenizer's state, rather than |
| 63 |
|
a method, with minimum changes (more changes will |
| 64 |
|
be made, in due course). "Bogus comment state"'s inner |
| 65 |
|
loop gets removed. |
| 66 |
|
|
| 67 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 68 |
|
|
| 69 |
|
* HTML.pm.src: Make |PUBLIC| and |SYSTEM| keyword tokenizing |
| 70 |
|
into their own tokenizer states. |
| 71 |
|
|
| 72 |
|
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 73 |
|
|
| 74 |
|
* HTML.pm.src: |CDATA_SECTION_STATE| (formally |CDATA_BLOCK_STATE| |
| 75 |
|
is split into three states. |
| 76 |
|
|
| 77 |
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
2008-09-13 Wakaba <wakaba@suika.fam.cx> |
| 78 |
|
|
| 79 |
* HTML.pm.src: |CLOSE_TAG_OPEN_STATE| is broken into |
* HTML.pm.src: |CLOSE_TAG_OPEN_STATE| is broken into |
Parent Directory
| 
Revision Log
| 
Patch