From 3d0c7bea4d2ad108889d0c4d81d41c4ff03f2a77 Mon Sep 17 00:00:00 2001 From: duerst Date: Sun, 20 Jan 2008 06:12:48 +0000 Subject: [PATCH] Sun Jan 20 15:08:08 2008 Martin Duerst * enc/trans/utf_16_32.c: new file, currently implementing UTF-16BE conversions only. * test/ruby/test_transcode.rb: Added tests for UTF-16BE; made check_both_ways() use force_encoding differently. * transcode_data.h, transcode.c: Support for more conversion functions. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@15142 b2dd03c8-39d4-4d8f-98ff-823fe69b080e --- enc/trans/utf_16_32.c | 332 ++++++++++++++++++++++++++++++++++++ test/ruby/test_transcode.rb | 55 +++++- transcode.c | 13 ++ transcode_data.h | 8 +- 4 files changed, 404 insertions(+), 4 deletions(-) create mode 100644 enc/trans/utf_16_32.c diff --git a/enc/trans/utf_16_32.c b/enc/trans/utf_16_32.c new file mode 100644 index 0000000000..c0f5a7ab70 --- /dev/null +++ b/enc/trans/utf_16_32.c @@ -0,0 +1,332 @@ +/* Autogenerated, do not change */ +/* Report bugs to Martin Duerst (duerst@it.aoyama.ac.jp) */ + +#include "transcode_data.h" + +static int +fun_so_from_utf_16be(const unsigned char* s, unsigned char* o) +{ + if (!s[0] && s[1]<0x80) { + o[0] = s[1]; + return 1; + } + else if (s[0]<0x08) { + o[0] = 0xC0 | (s[0]<<2) | (s[1]>>6); + o[1] = 0x80 | s[1]&0x3F; + return 2; + } + else if ((s[0]&0xF8)!=0xD8) { + o[0] = 0xE0 | s[0]>>4; + o[1] = 0x80 | ((s[0]&0x0F)<<2) | (s[1]>>6); + o[2] = 0x80 | s[1]&0x3F; + return 3; + } + else { + unsigned int u = (((s[0]&0x03)<<2)|(s[1]>>6)) + 1; + o[0] = 0xF0 | u>>2; + o[1] = 0x80 | ((u&0x03)<<4) | (s[1]>>2)&0x0F; + o[2] = 0x80 | ((s[1]&0x03)<<4) | ((s[2]&0x03)<<2) | (s[3]>>6); + o[3] = 0x80 | s[3]&0x3F; + return 4; + } +} + +static int +fun_so_to_utf_16be(const unsigned char* s, unsigned char* o) +{ + if (!(s[0]&0x80)) { + o[0] = 0x00; + o[1] = s[0]; + return 2; + } + else if ((s[0]&0xE0)==0xC0) { + o[0] = (s[0]>>2)&0x07; + o[1] = ((s[0]&0x03)<<6) | s[1]&0x3F; + return 2; + } + else if ((s[0]&0xF0)==0xE0) { + o[0] = (s[0]<<4) | (s[1]>>2)^0x20; + o[1] = (s[1]<<6) | s[2]^0x80; + return 2; + } + else { + int w = (((s[0]&0x07)<<2) | (s[1]>>4)&0x03) - 1; + o[0] = 0xD8 | (w>>2); + o[1] = (w<<6) | ((s[1]&0x0F)<<2) | ((s[2]>>4)-8); + o[2] = 0xDC | ((s[2]>>2)&0x03); + o[3] = (s[2]<<6) | (s[3]&~0x80); + return 4; + } +} +static const unsigned char +from_UTF_16BE_00_offsets[256] = { + /* used by from_UTF_16BE_00 */ + /* used by from_UTF_16BE_D8 */ + /* used by from_UTF_16BE_D8_00_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, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 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_16BE_00_infos[1] = { + /* used by from_UTF_16BE_00 */ + /* used by to_UTF_16BE_82 */ + FUNso, +}; +static const BYTE_LOOKUP +from_UTF_16BE_00 = { + /* used as from_UTF_16BE */ + /* used as from_UTF_16BE_D8_00 */ + from_UTF_16BE_00_offsets, + from_UTF_16BE_00_infos +}; + +static const struct byte_lookup* const +from_UTF_16BE_D8_00_00_infos[1] = { + INVALID, +}; +static const BYTE_LOOKUP +from_UTF_16BE_D8_00_00 = { + /* used as from_UTF_16BE_D8_00 */ + /* used as from_UTF_16BE */ + from_UTF_16BE_00_offsets, + from_UTF_16BE_D8_00_00_infos +}; + +static const unsigned char +from_UTF_16BE_D8_00_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, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 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, + 0, 0, 0, 0, 0, 0, 0, 0, 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_16BE_D8_00_infos[2] = { + &from_UTF_16BE_D8_00_00, &from_UTF_16BE_00, +}; +static const BYTE_LOOKUP +from_UTF_16BE_D8_00 = { + from_UTF_16BE_D8_00_offsets, + from_UTF_16BE_D8_00_infos +}; + +static const struct byte_lookup* const +from_UTF_16BE_D8_infos[1] = { + &from_UTF_16BE_D8_00, +}; +static const BYTE_LOOKUP +from_UTF_16BE_D8 = { + from_UTF_16BE_00_offsets, + from_UTF_16BE_D8_infos +}; + +static const unsigned char +from_UTF_16BE_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, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 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, 2, 2, 2, 2, + 0, 0, 0, 0, 0, 0, 0, 0, 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_16BE_infos[3] = { + &from_UTF_16BE_00, &from_UTF_16BE_D8, + &from_UTF_16BE_D8_00_00, +}; +static const BYTE_LOOKUP +from_UTF_16BE = { + from_UTF_16BE_offsets, + from_UTF_16BE_infos +}; + +static rb_transcoder +rb_from_UTF_16BE = { + "UTF-16BE", "UTF-8", &from_UTF_16BE, 4, 0, + NULL, NULL, NULL, NULL, NULL, &fun_so_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 */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 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 = { + /* used as to_UTF_16BE */ + /* used as to_UTF_16BE_E0 */ + /* used as to_UTF_16BE_E1 */ + /* used as to_UTF_16BE_ED */ + /* used as to_UTF_16BE_F0_90 */ + /* used as to_UTF_16BE_F1_80 */ + /* used as to_UTF_16BE_F4_80 */ + to_UTF_16BE_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 */ + 0, 0, 0, 0, 0, 0, 0, 0, 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, +}; +static const BYTE_LOOKUP +to_UTF_16BE_E0 = { + to_UTF_16BE_E0_offsets, + to_UTF_16BE_E0_infos +}; + +static const struct byte_lookup* const +to_UTF_16BE_E1_infos[1] = { + &to_UTF_16BE_82, +}; +static const BYTE_LOOKUP +to_UTF_16BE_E1 = { + /* used as to_UTF_16BE */ + /* used as to_UTF_16BE_F0 */ + /* used as to_UTF_16BE_F1 */ + /* used as to_UTF_16BE_F4 */ + to_UTF_16BE_82_offsets, + to_UTF_16BE_E1_infos +}; + +static const struct byte_lookup* const +to_UTF_16BE_ED_infos[2] = { + &to_UTF_16BE_82, INVALID, +}; +static const BYTE_LOOKUP +to_UTF_16BE_ED = { + to_UTF_16BE_E0_offsets, + to_UTF_16BE_ED_infos +}; + +static const unsigned char +to_UTF_16BE_F0_offsets[64] = { + /* used by to_UTF_16BE_F0 */ + /* used by to_UTF_16BE_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, +}; +static const BYTE_LOOKUP +to_UTF_16BE_F0 = { + to_UTF_16BE_F0_offsets, + to_UTF_16BE_F0_infos +}; + +static const struct byte_lookup* const +to_UTF_16BE_F1_infos[1] = { + &to_UTF_16BE_E1, +}; +static const BYTE_LOOKUP +to_UTF_16BE_F1 = { + to_UTF_16BE_82_offsets, + to_UTF_16BE_F1_infos +}; + +static const struct byte_lookup* const +to_UTF_16BE_F4_infos[2] = { + &to_UTF_16BE_E1, INVALID, +}; +static const BYTE_LOOKUP +to_UTF_16BE_F4 = { + to_UTF_16BE_F0_offsets, + to_UTF_16BE_F4_infos +}; + +static const unsigned char +to_UTF_16BE_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, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 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, 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, + 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 5, 4, 4, + 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, +}; +static const BYTE_LOOKUP +to_UTF_16BE = { + to_UTF_16BE_offsets, + to_UTF_16BE_infos +}; + +static rb_transcoder +rb_to_UTF_16BE = { + "UTF-8", "UTF-16BE", &to_UTF_16BE, 4, 1, + NULL, NULL, NULL, NULL, NULL, &fun_so_to_utf_16be +}; + +void +Init_utf_16_32(void) +{ + rb_register_transcoder(&rb_from_UTF_16BE); + rb_register_transcoder(&rb_to_UTF_16BE); +} +/* Footprint (bytes): gross: 3420, saved: 1992, net: 1428 */ diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb index 4b3e1e77f3..cf5600926d 100644 --- a/test/ruby/test_transcode.rb +++ b/test/ruby/test_transcode.rb @@ -10,7 +10,8 @@ class TestTranscode < Test::Unit::TestCase 'ISO-8859-7', 'ISO-8859-8', 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-11', 'ISO-8859-13', - 'ISO-8859-14', 'ISO-8859-15' + 'ISO-8859-14', 'ISO-8859-15', + 'UTF-16BE' ] all_encodings.each do |enc| 'abc'.encode(enc, 'UTF-8') @@ -54,8 +55,8 @@ class TestTranscode < Test::Unit::TestCase end def check_both_ways(utf8, raw, encoding) - assert_equal(utf8, raw.encode('utf-8', encoding)) - assert_equal(raw, utf8.encode(encoding).force_encoding('ASCII-8BIT')) + assert_equal(utf8.force_encoding('utf-8'), raw.encode('utf-8', encoding)) + assert_equal(raw.force_encoding(encoding), utf8.encode(encoding, 'utf-8')) end def test_encodings @@ -124,4 +125,52 @@ class TestTranscode < Test::Unit::TestCase assert_equal(test_start, test_start.encode('UTF-8',enc).encode(enc).force_encoding('ASCII-8BIT')) end end + + def test_utf_16be + check_both_ways("abc".force_encoding('UTF-8'), "\x00a\x00b\x00c", 'utf-16be') + check_both_ways("\u00E9", "\x00\xE9", 'utf-16be'); + check_both_ways("\u00E9\u0070\u00E9\u0065", "\x00\xE9\x00\x70\x00\xE9\x00\x65", 'utf-16be') # épée + check_both_ways("\u677E\u672C\u884C\u5F18", "\x67\x7E\x67\x2C\x88\x4C\x5F\x18", 'utf-16be') # 松本行弘 + check_both_ways("\u9752\u5C71\u5B66\u9662\u5927\u5B66", "\x97\x52\x5C\x71\x5B\x66\x96\x62\x59\x27\x5B\x66", + 'utf-16be') # 青山学院大学 + check_both_ways("Martin D\u00FCrst", "\x00M\x00a\x00r\x00t\x00i\x00n\x00 \x00D\x00\xFC\x00r\x00s\x00t", + 'utf-16be') # Martin Dürst + # BMP + check_both_ways("\u0000", "\x00\x00", 'utf-16be') + check_both_ways("\u007F", "\x00\x7F", 'utf-16be') + check_both_ways("\u0080", "\x00\x80", 'utf-16be') + check_both_ways("\u0555", "\x05\x55", 'utf-16be') + check_both_ways("\u04AA", "\x04\xAA", 'utf-16be') + check_both_ways("\u0333", "\x03\x33", 'utf-16be') + check_both_ways("\u04CC", "\x04\xCC", 'utf-16be') + check_both_ways("\u00F0", "\x00\xF0", 'utf-16be') + check_both_ways("\u070F", "\x07\x0F", 'utf-16be') + check_both_ways("\u07FF", "\x07\xFF", 'utf-16be') + check_both_ways("\u0800", "\x08\x00", 'utf-16be') + check_both_ways("\uD7FF", "\xD7\xFF", 'utf-16be') + check_both_ways("\uE000", "\xE0\x00", 'utf-16be') + check_both_ways("\uFFFF", "\xFF\xFF", 'utf-16be') + check_both_ways("\u5555", "\x55\x55", 'utf-16be') + check_both_ways("\uAAAA", "\xAA\xAA", 'utf-16be') + check_both_ways("\u3333", "\x33\x33", 'utf-16be') + check_both_ways("\uCCCC", "\xCC\xCC", 'utf-16be') + check_both_ways("\uF0F0", "\xF0\xF0", 'utf-16be') + check_both_ways("\u0F0F", "\x0F\x0F", 'utf-16be') + check_both_ways("\uFF00", "\xFF\x00", 'utf-16be') + check_both_ways("\u00FF", "\x00\xFF", 'utf-16be') + # outer planes + check_both_ways("\u{10000}", "\xD8\x00\xDC\x00", 'utf-16be') + check_both_ways("\u{FFFFF}", "\xDB\xBF\xDF\xFF", 'utf-16be') + check_both_ways("\u{100000}", "\xDB\xC0\xDC\x00", 'utf-16be') + check_both_ways("\u{10FFFF}", "\xDB\xFF\xDF\xFF", 'utf-16be') + check_both_ways("\u{105555}", "\xDB\xD5\xDD\x55", 'utf-16be') + check_both_ways("\u{55555}", "\xD9\x15\xDD\x55", 'utf-16be') + check_both_ways("\u{AAAAA}", "\xDA\x6A\xDE\xAA", 'utf-16be') + check_both_ways("\u{33333}", "\xD8\x8C\xDF\x33", 'utf-16be') + check_both_ways("\u{CCCCC}", "\xDA\xF3\xDC\xCC", 'utf-16be') + check_both_ways("\u{8F0F0}", "\xD9\xFC\xDC\xF0", 'utf-16be') + check_both_ways("\u{F0F0F}", "\xDB\x83\xDF\x0F", 'utf-16be') + check_both_ways("\u{8FF00}", "\xD9\xFF\xDF\x00", 'utf-16be') + check_both_ways("\u{F00FF}", "\xDB\x80\xDC\xFF", 'utf-16be') + end end diff --git a/transcode.c b/transcode.c index 8c69185414..1a6efc6100 100644 --- a/transcode.c +++ b/transcode.c @@ -104,6 +104,7 @@ init_transcoder_table(void) rb_declare_transcoder("SHIFT_JIS", "UTF-8", "japanese"); rb_declare_transcoder("EUC-JP", "UTF-8", "japanese"); rb_declare_transcoder("ISO-2022-JP", "UTF-8", "japanese"); + rb_declare_transcoder("UTF-16BE", "UTF-8", "utf_16_32"); } #define encoding_equal(enc1, enc2) (STRCASECMP(enc1, enc2) == 0) @@ -153,12 +154,14 @@ transcode_loop(char **in_pos, char **out_pos, char *in_p = *in_pos, *out_p = *out_pos; const BYTE_LOOKUP *conv_tree_start = my_transcoder->conv_tree_start; const BYTE_LOOKUP *next_table; + char *char_start; unsigned int next_offset; VALUE next_info; unsigned char next_byte; int from_utf8 = my_transcoder->from_utf8; char *out_s = out_stop - my_transcoder->max_output + 1; while (in_p < in_stop) { + char_start = in_p; next_table = conv_tree_start; if (out_p >= out_s) { int len = (out_p - *out_pos); @@ -212,6 +215,16 @@ transcode_loop(char **in_pos, char **out_pos, case FUNii: next_info = (VALUE)(*my_transcoder->func_ii)(next_info); goto follow_info; + case FUNsi: + next_info = (VALUE)(*my_transcoder->func_si)(char_start); + goto follow_info; + break; + case FUNio: + out_p += (VALUE)(*my_transcoder->func_io)(next_info, out_p); + break; + case FUNso: + out_p += (VALUE)(*my_transcoder->func_so)(char_start, out_p); + break; case INVALID: goto invalid; case UNDEF: diff --git a/transcode_data.h b/transcode_data.h index 65929390b9..8d0c8ccf33 100644 --- a/transcode_data.h +++ b/transcode_data.h @@ -35,6 +35,9 @@ typedef struct byte_lookup { #define UNDEF (PType 0x09) /* legal but undefined */ #define ZERObt (PType 0x0A) /* zero bytes of payload, i.e. remove */ #define FUNii (PType 0x0B) /* function from info to info */ +#define FUNsi (PType 0x0D) /* function from start to info */ +#define FUNio (PType 0x0E) /* function from info to output */ +#define FUNso (PType 0x0F) /* function from start to output */ #define o1(b1) (PType((((unsigned char)(b1))<<8)|ONEbt)) #define o2(b1,b2) (PType((((unsigned char)(b1))<<8)|(((unsigned char)(b2))<<16)|TWObt)) @@ -72,7 +75,10 @@ typedef struct rb_transcoder { struct rb_transcoder *, struct rb_transcoding *); void (*postprocessor)(char**, char**, char*, char*, struct rb_transcoder *, struct rb_transcoding *); - VALUE (*func_ii)(VALUE); /* function from info to info */ + VALUE (*func_ii)(VALUE); /* info -> info */ + VALUE (*func_si)(const unsigned char* const); /* start -> info */ + int (*func_io)(VALUE, const unsigned char*); /* info -> output */ + int (*func_so)(const unsigned char*, unsigned char*); /* start -> output */ } rb_transcoder; void rb_declare_transcoder(const char *enc1, const char *enc2, const char *lib);