diff --git a/ChangeLog b/ChangeLog index 9d249fe148..98cf3c2d44 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,8 @@ +Mon Jan 21 19:42:42 2008 Martin Duerst + + * transcode.c, enc/trans/utf_16_32.c, test/ruby/test_transcode.rb: + added UTF-32BE and UTF-32LE conversions. + Mon Jan 21 14:36:14 2008 Nobuyoshi Nakada * transcode.c (str_transcode): initialize transcoder in diff --git a/enc/trans/utf_16_32.c b/enc/trans/utf_16_32.c index 0247e5dce6..d44b2bfa43 100644 --- a/enc/trans/utf_16_32.c +++ b/enc/trans/utf_16_32.c @@ -114,12 +114,82 @@ fun_so_to_utf_16le(const unsigned char* s, unsigned char* o) return 4; } } + +static int +fun_so_from_utf_32be(const unsigned char* s, unsigned char* o) +{ + if (!s[1]) { + if (s[2]==0 && s[3]<0x80) { + o[0] = s[3]; + return 1; + } + else if (s[2]<0x08) { + o[0] = 0xC0 | (s[2]<<2) | (s[3]>>6); + o[1] = 0x80 | s[3]&0x3F; + return 2; + } + else { + o[0] = 0xE0 | s[2]>>4; + o[1] = 0x80 | ((s[2]&0x0F)<<2) | (s[3]>>6); + o[2] = 0x80 | s[3]&0x3F; + return 3; + } + } + else { + o[0] = 0xF0 | (s[1]>>2); + o[1] = 0x80 | ((s[1]&0x03)<<4) | (s[2]>>4); + o[2] = 0x80 | ((s[2]&0x0F)<<2) | (s[3]>>6); + o[3] = 0x80 | s[3]&0x3F; + return 4; + } +} + +static int +fun_so_to_utf_32be(const unsigned char* s, unsigned char* o) +{ + o[0] = 0; + if (!(s[0]&0x80)) { + o[1] = o[2] = 0x00; + o[3] = s[0]; + } + else if ((s[0]&0xE0)==0xC0) { + o[1] = 0x00; + o[2] = (s[0]>>2)&0x07; + o[3] = ((s[0]&0x03)<<6) | s[1]&0x3F; + } + else if ((s[0]&0xF0)==0xE0) { + o[1] = 0x00; + o[2] = (s[0]<<4) | (s[1]>>2)^0x20; + o[3] = (s[1]<<6) | s[2]^0x80; + } + else { + o[1] = ((s[0]&0x07)<<2) | (s[1]>>4)&0x03; + o[2] = ((s[1]&0x0F)<<4) | (s[2]>>2)&0x0F; + o[3] = ((s[2]&0x03)<<6) | s[3]&0x3F; + } + return 4; +} + +static int +fun_so_from_utf_32le(const unsigned char* s, unsigned char* o) +{ + return 1; +} + +static int +fun_so_to_utf_32le(const unsigned char* s, unsigned char* o) +{ + return 4; +} static const unsigned char from_UTF_16BE_00_offsets[256] = { /* used by from_UTF_16BE_00 */ + /* used by from_UTF_32BE_00_00_D8 */ + /* used by from_UTF_32BE_00_01 */ + /* used by from_UTF_32BE_00_11 */ /* used by from_UTF_16BE_D8 */ - /* used by from_UTF_16BE_D8_00_00 */ /* used by from_UTF_16LE */ + /* used by from_UTF_32LE */ /* used by from_UTF_16LE_00_D8 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -141,27 +211,35 @@ from_UTF_16BE_00_offsets[256] = { static const struct byte_lookup* const from_UTF_16BE_00_infos[1] = { /* used by from_UTF_16BE_00 */ - /* used by to_UTF_16BE_82 */ + /* used by to_UTF_32BE_82 */ FUNso, }; static const BYTE_LOOKUP from_UTF_16BE_00 = { /* used as from_UTF_16BE */ + /* used as from_UTF_32BE_00_00 */ + /* used as from_UTF_32BE_00_01 */ /* used as from_UTF_16BE_D8_00 */ + /* used as from_UTF_32LE_00_00 */ + /* used as from_UTF_32LE_00_D8 */ from_UTF_16BE_00_offsets, from_UTF_16BE_00_infos }; static const struct byte_lookup* const -from_UTF_16BE_D8_00_00_infos[1] = { +from_UTF_32BE_00_00_D8_infos[1] = { INVALID, }; static const BYTE_LOOKUP -from_UTF_16BE_D8_00_00 = { +from_UTF_32BE_00_00_D8 = { + /* used as from_UTF_32BE_00_00 */ + /* used as from_UTF_32BE_00_11 */ /* used as from_UTF_16BE_D8_00 */ /* used as from_UTF_16BE */ + /* used as from_UTF_32LE_00_00 */ + /* used as from_UTF_32LE_00_D8 */ from_UTF_16BE_00_offsets, - from_UTF_16BE_D8_00_00_infos + from_UTF_32BE_00_00_D8_infos }; static const unsigned char @@ -187,7 +265,9 @@ from_UTF_16BE_D8_00_offsets[256] = { }; static const struct byte_lookup* const from_UTF_16BE_D8_00_infos[2] = { - &from_UTF_16BE_D8_00_00, &from_UTF_16BE_00, + /* used by from_UTF_16BE_D8_00 */ + /* used by from_UTF_32LE_00_D8 */ + &from_UTF_32BE_00_00_D8, &from_UTF_16BE_00, }; static const BYTE_LOOKUP from_UTF_16BE_D8_00 = { @@ -229,7 +309,7 @@ from_UTF_16BE_offsets[256] = { static const struct byte_lookup* const from_UTF_16BE_infos[3] = { &from_UTF_16BE_00, &from_UTF_16BE_D8, - &from_UTF_16BE_D8_00_00, + &from_UTF_32BE_00_00_D8, }; static const BYTE_LOOKUP from_UTF_16BE = { @@ -244,133 +324,165 @@ rb_from_UTF_16BE = { }; static const unsigned char -to_UTF_16BE_82_offsets[64] = { - /* used by to_UTF_16BE_82 */ - /* used by to_UTF_16BE_E1 */ - /* used by to_UTF_16BE_F1 */ +to_UTF_32BE_82_offsets[64] = { + /* used by to_UTF_32BE_82 */ + /* used by to_UTF_32BE_E1 */ + /* used by to_UTF_32BE_F1 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; static const BYTE_LOOKUP -to_UTF_16BE_82 = { +to_UTF_32BE_82 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ + /* used as to_UTF_32BE_E0 */ /* used as to_UTF_16BE_E0 */ /* used as to_UTF_16BE_E1 */ + /* used as to_UTF_32BE_E1 */ + /* used as to_UTF_32BE_ED */ /* used as to_UTF_16BE_ED */ /* used as to_UTF_16BE_F0_90 */ + /* used as to_UTF_32BE_F0_90 */ /* used as to_UTF_16BE_F1_80 */ + /* used as to_UTF_32BE_F1_80 */ + /* used as to_UTF_32BE_F4_80 */ /* used as to_UTF_16BE_F4_80 */ /* used as to_UTF_16LE */ + /* used as to_UTF_32LE */ /* used as to_UTF_16LE_E0 */ + /* used as to_UTF_32LE_E0 */ + /* used as to_UTF_32LE_E1 */ /* used as to_UTF_16LE_E1 */ + /* used as to_UTF_32LE_ED */ /* used as to_UTF_16LE_ED */ /* used as to_UTF_16LE_F0_90 */ + /* used as to_UTF_32LE_F0_90 */ /* used as to_UTF_16LE_F1_80 */ + /* used as to_UTF_32LE_F1_80 */ /* used as to_UTF_16LE_F4_80 */ - to_UTF_16BE_82_offsets, + /* used as to_UTF_32LE_F4_80 */ + to_UTF_32BE_82_offsets, from_UTF_16BE_00_infos }; static const unsigned char -to_UTF_16BE_E0_offsets[64] = { - /* used by to_UTF_16BE_E0 */ - /* used by to_UTF_16BE_ED */ +to_UTF_32BE_E0_offsets[64] = { + /* used by to_UTF_32BE_E0 */ + /* used by to_UTF_32BE_ED */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, }; static const struct byte_lookup* const -to_UTF_16BE_E0_infos[2] = { - INVALID, &to_UTF_16BE_82, +to_UTF_32BE_E0_infos[2] = { + INVALID, &to_UTF_32BE_82, }; static const BYTE_LOOKUP -to_UTF_16BE_E0 = { +to_UTF_32BE_E0 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_E0_offsets, - to_UTF_16BE_E0_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_E0_offsets, + to_UTF_32BE_E0_infos }; static const struct byte_lookup* const -to_UTF_16BE_E1_infos[1] = { - &to_UTF_16BE_82, +to_UTF_32BE_E1_infos[1] = { + &to_UTF_32BE_82, }; static const BYTE_LOOKUP -to_UTF_16BE_E1 = { +to_UTF_32BE_E1 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ + /* used as to_UTF_32BE_F0 */ /* used as to_UTF_16BE_F0 */ /* used as to_UTF_16BE_F1 */ + /* used as to_UTF_32BE_F1 */ /* used as to_UTF_16BE_F4 */ + /* used as to_UTF_32BE_F4 */ + /* used as to_UTF_32LE */ /* used as to_UTF_16LE */ /* used as to_UTF_16LE_F0 */ + /* used as to_UTF_32LE_F0 */ /* used as to_UTF_16LE_F1 */ + /* used as to_UTF_32LE_F1 */ /* used as to_UTF_16LE_F4 */ - to_UTF_16BE_82_offsets, - to_UTF_16BE_E1_infos + /* used as to_UTF_32LE_F4 */ + to_UTF_32BE_82_offsets, + to_UTF_32BE_E1_infos }; static const struct byte_lookup* const -to_UTF_16BE_ED_infos[2] = { - &to_UTF_16BE_82, INVALID, +to_UTF_32BE_ED_infos[2] = { + &to_UTF_32BE_82, INVALID, }; static const BYTE_LOOKUP -to_UTF_16BE_ED = { +to_UTF_32BE_ED = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_E0_offsets, - to_UTF_16BE_ED_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_E0_offsets, + to_UTF_32BE_ED_infos }; static const unsigned char -to_UTF_16BE_F0_offsets[64] = { - /* used by to_UTF_16BE_F0 */ - /* used by to_UTF_16BE_F4 */ +to_UTF_32BE_F0_offsets[64] = { + /* used by to_UTF_32BE_F0 */ + /* used by to_UTF_32BE_F4 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, }; static const struct byte_lookup* const -to_UTF_16BE_F0_infos[2] = { - INVALID, &to_UTF_16BE_E1, +to_UTF_32BE_F0_infos[2] = { + INVALID, &to_UTF_32BE_E1, }; static const BYTE_LOOKUP -to_UTF_16BE_F0 = { +to_UTF_32BE_F0 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_F0_offsets, - to_UTF_16BE_F0_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_F0_offsets, + to_UTF_32BE_F0_infos }; static const struct byte_lookup* const -to_UTF_16BE_F1_infos[1] = { - &to_UTF_16BE_E1, +to_UTF_32BE_F1_infos[1] = { + &to_UTF_32BE_E1, }; static const BYTE_LOOKUP -to_UTF_16BE_F1 = { +to_UTF_32BE_F1 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_82_offsets, - to_UTF_16BE_F1_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_82_offsets, + to_UTF_32BE_F1_infos }; static const struct byte_lookup* const -to_UTF_16BE_F4_infos[2] = { - &to_UTF_16BE_E1, INVALID, +to_UTF_32BE_F4_infos[2] = { + &to_UTF_32BE_E1, INVALID, }; static const BYTE_LOOKUP -to_UTF_16BE_F4 = { +to_UTF_32BE_F4 = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_F0_offsets, - to_UTF_16BE_F4_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_F0_offsets, + to_UTF_32BE_F4_infos }; static const unsigned char -to_UTF_16BE_offsets[256] = { +to_UTF_32BE_offsets[256] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -389,22 +501,24 @@ to_UTF_16BE_offsets[256] = { 6, 7, 7, 7, 8, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, }; static const struct byte_lookup* const -to_UTF_16BE_infos[9] = { - FUNso, INVALID, &to_UTF_16BE_82, &to_UTF_16BE_E0, - &to_UTF_16BE_E1, &to_UTF_16BE_ED, &to_UTF_16BE_F0, &to_UTF_16BE_F1, - &to_UTF_16BE_F4, +to_UTF_32BE_infos[9] = { + FUNso, INVALID, &to_UTF_32BE_82, &to_UTF_32BE_E0, + &to_UTF_32BE_E1, &to_UTF_32BE_ED, &to_UTF_32BE_F0, &to_UTF_32BE_F1, + &to_UTF_32BE_F4, }; static const BYTE_LOOKUP -to_UTF_16BE = { +to_UTF_32BE = { + /* used as to_UTF_32BE */ /* used as to_UTF_16BE */ /* used as to_UTF_16LE */ - to_UTF_16BE_offsets, - to_UTF_16BE_infos + /* used as to_UTF_32LE */ + to_UTF_32BE_offsets, + to_UTF_32BE_infos }; static rb_transcoder rb_to_UTF_16BE = { - "UTF-8", "UTF-16BE", &to_UTF_16BE, 4, 1, + "UTF-8", "UTF-16BE", &to_UTF_32BE, 4, 1, NULL, NULL, NULL, NULL, NULL, &fun_so_to_utf_16be }; @@ -457,10 +571,216 @@ rb_from_UTF_16LE = { static rb_transcoder rb_to_UTF_16LE = { - "UTF-8", "UTF-16LE", &to_UTF_16BE, 4, 1, + "UTF-8", "UTF-16LE", &to_UTF_32BE, 4, 1, NULL, NULL, NULL, NULL, NULL, &fun_so_to_utf_16le }; +static const unsigned char +from_UTF_32BE_00_00_offsets[256] = { + /* used by from_UTF_32BE_00_00 */ + /* used by from_UTF_32LE_00 */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, +}; +static const struct byte_lookup* const +from_UTF_32BE_00_00_infos[2] = { + /* used by from_UTF_32BE_00_00 */ + /* used by from_UTF_32LE_00_00 */ + &from_UTF_16BE_00, &from_UTF_32BE_00_00_D8, +}; +static const BYTE_LOOKUP +from_UTF_32BE_00_00 = { + from_UTF_32BE_00_00_offsets, + from_UTF_32BE_00_00_infos +}; + +static const struct byte_lookup* const +from_UTF_32BE_00_01_infos[1] = { + &from_UTF_16BE_00, +}; +static const BYTE_LOOKUP +from_UTF_32BE_00_01 = { + from_UTF_16BE_00_offsets, + from_UTF_32BE_00_01_infos +}; + +static const struct byte_lookup* const +from_UTF_32BE_00_11_infos[1] = { + &from_UTF_32BE_00_00_D8, +}; +static const BYTE_LOOKUP +from_UTF_32BE_00_11 = { + from_UTF_16BE_00_offsets, + from_UTF_32BE_00_11_infos +}; + +static const unsigned char +from_UTF_32BE_00_offsets[256] = { + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, +}; +static const struct byte_lookup* const +from_UTF_32BE_00_infos[3] = { + &from_UTF_32BE_00_00, &from_UTF_32BE_00_01, + &from_UTF_32BE_00_11, +}; +static const BYTE_LOOKUP +from_UTF_32BE_00 = { + from_UTF_32BE_00_offsets, + from_UTF_32BE_00_infos +}; + +static const unsigned char +from_UTF_32BE_offsets[256] = { + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, +}; +static const struct byte_lookup* const +from_UTF_32BE_infos[2] = { + &from_UTF_32BE_00, INVALID, +}; +static const BYTE_LOOKUP +from_UTF_32BE = { + from_UTF_32BE_offsets, + from_UTF_32BE_infos +}; + +static rb_transcoder +rb_from_UTF_32BE = { + "UTF-32BE", "UTF-8", &from_UTF_32BE, 4, 0, + NULL, NULL, NULL, NULL, NULL, &fun_so_from_utf_32be +}; + +static rb_transcoder +rb_to_UTF_32BE = { + "UTF-8", "UTF-32BE", &to_UTF_32BE, 4, 1, + NULL, NULL, NULL, NULL, NULL, &fun_so_to_utf_32be +}; + +static const unsigned char +from_UTF_32LE_00_00_offsets[256] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, +}; +static const BYTE_LOOKUP +from_UTF_32LE_00_00 = { + from_UTF_32LE_00_00_offsets, + from_UTF_32BE_00_00_infos +}; + +static const unsigned char +from_UTF_32LE_00_D8_offsets[256] = { + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, +}; +static const BYTE_LOOKUP +from_UTF_32LE_00_D8 = { + from_UTF_32LE_00_D8_offsets, + from_UTF_16BE_D8_00_infos +}; + +static const struct byte_lookup* const +from_UTF_32LE_00_infos[2] = { + &from_UTF_32LE_00_00, &from_UTF_32LE_00_D8, +}; +static const BYTE_LOOKUP +from_UTF_32LE_00 = { + from_UTF_32BE_00_00_offsets, + from_UTF_32LE_00_infos +}; + +static const struct byte_lookup* const +from_UTF_32LE_infos[1] = { + &from_UTF_32LE_00, +}; +static const BYTE_LOOKUP +from_UTF_32LE = { + from_UTF_16BE_00_offsets, + from_UTF_32LE_infos +}; + +static rb_transcoder +rb_from_UTF_32LE = { + "UTF-32LE", "UTF-8", &from_UTF_32LE, 4, 0, + NULL, NULL, NULL, NULL, NULL, &fun_so_from_utf_32le +}; + +static rb_transcoder +rb_to_UTF_32LE = { + "UTF-8", "UTF-32LE", &to_UTF_32BE, 4, 1, + NULL, NULL, NULL, NULL, NULL, &fun_so_to_utf_32le +}; + void Init_utf_16_32(void) { @@ -468,5 +788,9 @@ Init_utf_16_32(void) rb_register_transcoder(&rb_to_UTF_16BE); rb_register_transcoder(&rb_from_UTF_16LE); rb_register_transcoder(&rb_to_UTF_16LE); + rb_register_transcoder(&rb_from_UTF_32BE); + rb_register_transcoder(&rb_to_UTF_32BE); + rb_register_transcoder(&rb_from_UTF_32LE); + rb_register_transcoder(&rb_to_UTF_32LE); } -/* Footprint (bytes): gross: 6036, saved: 4548, net: 1488 */ +/* Footprint (bytes): gross: 13684, saved: 10796, net: 2888 */ diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb index cbb942f45f..aa4a1cd3ad 100644 --- a/test/ruby/test_transcode.rb +++ b/test/ruby/test_transcode.rb @@ -178,4 +178,63 @@ class TestTranscode < Test::Unit::TestCase check_utf_16_both_ways("\u{8FF00}", "\xD9\xFF\xDF\x00") check_utf_16_both_ways("\u{F00FF}", "\xDB\x80\xDC\xFF") end + + def check_utf_32_both_ways(utf8, raw) + copy = raw.dup + 0.step(copy.length-1, 4) do |i| + copy[i+3], copy[i+2], copy[i+1], copy[i] = copy[i], copy[i+1], copy[i+2], copy[i+3] + end + check_both_ways(utf8, raw, 'utf-32be') + #check_both_ways(utf8, copy, 'utf-32le') + end + + def test_utf_32 + check_utf_32_both_ways("abc", "\x00\x00\x00a\x00\x00\x00b\x00\x00\x00c") + check_utf_32_both_ways("\u00E9", "\x00\x00\x00\xE9"); + check_utf_32_both_ways("\u00E9\u0070\u00E9\u0065", + "\x00\x00\x00\xE9\x00\x00\x00\x70\x00\x00\x00\xE9\x00\x00\x00\x65") # épée + check_utf_32_both_ways("\u677E\u672C\u884C\u5F18", + "\x00\x00\x67\x7E\x00\x00\x67\x2C\x00\x00\x88\x4C\x00\x00\x5F\x18") # 松本行弘 + check_utf_32_both_ways("\u9752\u5C71\u5B66\u9662\u5927\u5B66", + "\x00\x00\x97\x52\x00\x00\x5C\x71\x00\x00\x5B\x66\x00\x00\x96\x62\x00\x00\x59\x27\x00\x00\x5B\x66") # 青山学院大学 + check_utf_32_both_ways("Martin D\u00FCrst", + "\x00\x00\x00M\x00\x00\x00a\x00\x00\x00r\x00\x00\x00t\x00\x00\x00i\x00\x00\x00n\x00\x00\x00 \x00\x00\x00D\x00\x00\x00\xFC\x00\x00\x00r\x00\x00\x00s\x00\x00\x00t") # Martin Dürst + # BMP + check_utf_32_both_ways("\u0000", "\x00\x00\x00\x00") + check_utf_32_both_ways("\u007F", "\x00\x00\x00\x7F") + check_utf_32_both_ways("\u0080", "\x00\x00\x00\x80") + check_utf_32_both_ways("\u0555", "\x00\x00\x05\x55") + check_utf_32_both_ways("\u04AA", "\x00\x00\x04\xAA") + check_utf_32_both_ways("\u0333", "\x00\x00\x03\x33") + check_utf_32_both_ways("\u04CC", "\x00\x00\x04\xCC") + check_utf_32_both_ways("\u00F0", "\x00\x00\x00\xF0") + check_utf_32_both_ways("\u070F", "\x00\x00\x07\x0F") + check_utf_32_both_ways("\u07FF", "\x00\x00\x07\xFF") + check_utf_32_both_ways("\u0800", "\x00\x00\x08\x00") + check_utf_32_both_ways("\uD7FF", "\x00\x00\xD7\xFF") + check_utf_32_both_ways("\uE000", "\x00\x00\xE0\x00") + check_utf_32_both_ways("\uFFFF", "\x00\x00\xFF\xFF") + check_utf_32_both_ways("\u5555", "\x00\x00\x55\x55") + check_utf_32_both_ways("\uAAAA", "\x00\x00\xAA\xAA") + check_utf_32_both_ways("\u3333", "\x00\x00\x33\x33") + check_utf_32_both_ways("\uCCCC", "\x00\x00\xCC\xCC") + check_utf_32_both_ways("\uF0F0", "\x00\x00\xF0\xF0") + check_utf_32_both_ways("\u0F0F", "\x00\x00\x0F\x0F") + check_utf_32_both_ways("\uFF00", "\x00\x00\xFF\x00") + check_utf_32_both_ways("\u00FF", "\x00\x00\x00\xFF") + # outer planes + check_utf_32_both_ways("\u{10000}", "\x00\x01\x00\x00") + check_utf_32_both_ways("\u{FFFFF}", "\x00\x0F\xFF\xFF") + check_utf_32_both_ways("\u{100000}","\x00\x10\x00\x00") + check_utf_32_both_ways("\u{10FFFF}","\x00\x10\xFF\xFF") + check_utf_32_both_ways("\u{105555}","\x00\x10\x55\x55") + check_utf_32_both_ways("\u{55555}", "\x00\x05\x55\x55") + check_utf_32_both_ways("\u{AAAAA}", "\x00\x0A\xAA\xAA") + check_utf_32_both_ways("\u{33333}", "\x00\x03\x33\x33") + check_utf_32_both_ways("\u{CCCCC}", "\x00\x0C\xCC\xCC") + check_utf_32_both_ways("\u{8F0F0}", "\x00\x08\xF0\xF0") + check_utf_32_both_ways("\u{F0F0F}", "\x00\x0F\x0F\x0F") + check_utf_32_both_ways("\u{8FF00}", "\x00\x08\xFF\x00") + check_utf_32_both_ways("\u{F00FF}", "\x00\x0F\x00\xFF") + end end diff --git a/transcode.c b/transcode.c index 1b8e8ee9e5..7daca3694b 100644 --- a/transcode.c +++ b/transcode.c @@ -106,6 +106,8 @@ init_transcoder_table(void) rb_declare_transcoder("ISO-2022-JP", "UTF-8", "japanese"); rb_declare_transcoder("UTF-16BE", "UTF-8", "utf_16_32"); rb_declare_transcoder("UTF-16LE", "UTF-8", "utf_16_32"); + rb_declare_transcoder("UTF-32BE", "UTF-8", "utf_16_32"); + rb_declare_transcoder("UTF-32LE", "UTF-8", "utf_16_32"); } #define encoding_equal(enc1, enc2) (STRCASECMP(enc1, enc2) == 0)