diff options
| author | Hye-Shik Chang <hyeshik@gmail.com> | 2006-09-05 12:07:09 (GMT) |
|---|---|---|
| committer | Hye-Shik Chang <hyeshik@gmail.com> | 2006-09-05 12:07:09 (GMT) |
| commit | 199f1db1fa75e1f1b7cfe73259d6a811aa494a36 (patch) | |
| tree | dafe1e12cea693de74824077b7871a3fdc4df5a2 /Modules/cjkcodecs/_codecs_iso2022.c | |
| parent | d042132268d75461e157bd87a3f4358ce603d311 (diff) | |
| download | cpython-199f1db1fa75e1f1b7cfe73259d6a811aa494a36.zip cpython-199f1db1fa75e1f1b7cfe73259d6a811aa494a36.tar.gz cpython-199f1db1fa75e1f1b7cfe73259d6a811aa494a36.tar.bz2 | |
Fix a few bugs on cjkcodecs found by Oren Tirosh:
- gbk and gb18030 codec now handle U+30FB KATAKANA MIDDLE DOT correctly.
- iso2022_jp_2 codec now encodes into G0 for KS X 1001, GB2312
codepoints to conform the standard.
- iso2022_jp_3 and iso2022_jp_2004 codec can encode JIS X 2013:2
codepoints now.
Diffstat (limited to 'Modules/cjkcodecs/_codecs_iso2022.c')
| -rw-r--r-- | Modules/cjkcodecs/_codecs_iso2022.c | 15 |
1 files changed, 9 insertions, 6 deletions
diff --git a/Modules/cjkcodecs/_codecs_iso2022.c b/Modules/cjkcodecs/_codecs_iso2022.c index 8a2ab7e..2a11e9a 100644 --- a/Modules/cjkcodecs/_codecs_iso2022.c +++ b/Modules/cjkcodecs/_codecs_iso2022.c @@ -854,7 +854,7 @@ jisx0213_2000_2_encoder(const ucs4_t *data, Py_ssize_t *length) if (coded == MAP_UNMAPPABLE || coded == MAP_MULTIPLE_AVAIL) return coded; else if (coded & 0x8000) - return coded; + return coded & 0x7fff; else return MAP_UNMAPPABLE; } @@ -901,7 +901,7 @@ jisx0213_2004_2_encoder(const ucs4_t *data, Py_ssize_t *length) if (coded == MAP_UNMAPPABLE || coded == MAP_MULTIPLE_AVAIL) return coded; else if (coded & 0x8000) - return coded; + return coded & 0x7fff; else return MAP_UNMAPPABLE; } @@ -992,7 +992,10 @@ dummy_encoder(const ucs4_t *data, Py_ssize_t *length) /*-*- registry tables -*-*/ -#define REGISTRY_KSX1001 { CHARSET_KSX1001, 1, 2, \ +#define REGISTRY_KSX1001_G0 { CHARSET_KSX1001, 0, 2, \ + ksx1001_init, \ + ksx1001_decoder, ksx1001_encoder } +#define REGISTRY_KSX1001_G1 { CHARSET_KSX1001, 1, 2, \ ksx1001_init, \ ksx1001_decoder, ksx1001_encoder } #define REGISTRY_JISX0201_R { CHARSET_JISX0201_R, 0, 1, \ @@ -1034,7 +1037,7 @@ dummy_encoder(const ucs4_t *data, Py_ssize_t *length) jisx0213_init, \ jisx0213_2004_2_decoder, \ jisx0213_2004_2_encoder } -#define REGISTRY_GB2312 { CHARSET_GB2312, 1, 2, \ +#define REGISTRY_GB2312 { CHARSET_GB2312, 0, 2, \ gb2312_init, \ gb2312_decoder, gb2312_encoder } #define REGISTRY_CNS11643_1 { CHARSET_CNS11643_1, 1, 2, \ @@ -1054,7 +1057,7 @@ dummy_encoder(const ucs4_t *data, Py_ssize_t *length) }; static const struct iso2022_designation iso2022_kr_designations[] = { - REGISTRY_KSX1001, REGISTRY_SENTINEL + REGISTRY_KSX1001_G1, REGISTRY_SENTINEL }; CONFIGDEF(kr, 0) @@ -1071,7 +1074,7 @@ static const struct iso2022_designation iso2022_jp_1_designations[] = { CONFIGDEF(jp_1, NO_SHIFT | USE_JISX0208_EXT) static const struct iso2022_designation iso2022_jp_2_designations[] = { - REGISTRY_JISX0208, REGISTRY_JISX0212, REGISTRY_KSX1001, + REGISTRY_JISX0208, REGISTRY_JISX0212, REGISTRY_KSX1001_G0, REGISTRY_GB2312, REGISTRY_JISX0201_R, REGISTRY_JISX0208_O, REGISTRY_ISO8859_1, REGISTRY_ISO8859_7, REGISTRY_SENTINEL }; |