From fcbf63e62c627deae76c1b8cb8c0876c536ed811 Mon Sep 17 00:00:00 2001 From: Jari Vetoniemi Date: Mon, 16 Mar 2020 18:49:26 +0900 Subject: Fresh start --- jni/ruby/enc/trans/utf_16_32.trans | 556 +++++++++++++++++++++++++++++++++++++ 1 file changed, 556 insertions(+) create mode 100644 jni/ruby/enc/trans/utf_16_32.trans (limited to 'jni/ruby/enc/trans/utf_16_32.trans') diff --git a/jni/ruby/enc/trans/utf_16_32.trans b/jni/ruby/enc/trans/utf_16_32.trans new file mode 100644 index 0000000..632c880 --- /dev/null +++ b/jni/ruby/enc/trans/utf_16_32.trans @@ -0,0 +1,556 @@ +#include "transcode_data.h" + +<% + map = {} + map["{00-ff}{00-d7,e0-ff}"] = :func_so + map["{00-ff}{d8-db}{00-ff}{dc-df}"] = :func_so + transcode_generate_node(ActionMap.parse(map), "from_UTF_16LE") + + map = {} + map["{00-ff}{00-d7,e0-ff}0000"] = :func_so + map["{00-ff}{00-ff}{01-10}00"] = :func_so + transcode_generate_node(ActionMap.parse(map), "from_UTF_32LE") + + map = {} + map["{00-d7,e0-ff}{00-ff}"] = :func_so + map["{d8-db}{00-ff}{dc-df}{00-ff}"] = :func_so + transcode_generate_node(ActionMap.parse(map), "from_UTF_16BE") + + map = {} + map["0000{00-d7,e0-ff}{00-ff}"] = :func_so + map["00{01-10}{00-ff}{00-ff}"] = :func_so + transcode_generate_node(ActionMap.parse(map), "from_UTF_32BE") + + map = {} + map["{00-ff}{00-ff}"] = :func_si + transcode_generate_node(ActionMap.parse(map), "from_UTF_16") + + map = {} + map["{00-ff}{00-ff}{00-ff}{00-ff}"] = :func_si + transcode_generate_node(ActionMap.parse(map), "from_UTF_32") + + map = {} + map["{00-7f}"] = :func_so + map["{c2-df}{80-bf}"] = :func_so + map["e0{a0-bf}{80-bf}"] = :func_so + map["{e1-ec}{80-bf}{80-bf}"] = :func_so + map["ed{80-9f}{80-bf}"] = :func_so + map["{ee-ef}{80-bf}{80-bf}"] = :func_so + map["f0{90-bf}{80-bf}{80-bf}"] = :func_so + map["{f1-f3}{80-bf}{80-bf}{80-bf}"] = :func_so + map["f4{80-8f}{80-bf}{80-bf}"] = :func_so + am = ActionMap.parse(map) + transcode_generate_node(am, "from_UTF_8") +%> + +<%= transcode_generated_code %> + +static ssize_t +fun_so_from_utf_16be(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + 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 ssize_t +fun_so_to_utf_16be(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + 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 ssize_t +fun_so_from_utf_16le(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + if (!s[1] && s[0]<0x80) { + o[0] = s[0]; + return 1; + } + else if (s[1]<0x08) { + o[0] = 0xC0 | (s[1]<<2) | (s[0]>>6); + o[1] = 0x80 | (s[0]&0x3F); + return 2; + } + else if ((s[1]&0xF8)!=0xD8) { + o[0] = 0xE0 | (s[1]>>4); + o[1] = 0x80 | ((s[1]&0x0F)<<2) | (s[0]>>6); + o[2] = 0x80 | (s[0]&0x3F); + return 3; + } + else { + unsigned int u = (((s[1]&0x03)<<2)|(s[0]>>6)) + 1; + o[0] = 0xF0 | u>>2; + o[1] = 0x80 | ((u&0x03)<<4) | ((s[0]>>2)&0x0F); + o[2] = 0x80 | ((s[0]&0x03)<<4) | ((s[3]&0x03)<<2) | (s[2]>>6); + o[3] = 0x80 | (s[2]&0x3F); + return 4; + } +} + +static ssize_t +fun_so_to_utf_16le(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + if (!(s[0]&0x80)) { + o[1] = 0x00; + o[0] = s[0]; + return 2; + } + else if ((s[0]&0xE0)==0xC0) { + o[1] = (s[0]>>2)&0x07; + o[0] = ((s[0]&0x03)<<6) | (s[1]&0x3F); + return 2; + } + else if ((s[0]&0xF0)==0xE0) { + o[1] = (s[0]<<4) | ((s[1]>>2)^0x20); + o[0] = (s[1]<<6) | (s[2]^0x80); + return 2; + } + else { + int w = (((s[0]&0x07)<<2) | ((s[1]>>4)&0x03)) - 1; + o[1] = 0xD8 | (w>>2); + o[0] = (w<<6) | ((s[1]&0x0F)<<2) | ((s[2]>>4)-8); + o[3] = 0xDC | ((s[2]>>2)&0x03); + o[2] = (s[2]<<6) | (s[3]&~0x80); + return 4; + } +} + +static ssize_t +fun_so_from_utf_32be(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + 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 ssize_t +fun_so_to_utf_32be(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + 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 ssize_t +fun_so_from_utf_32le(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + if (!s[2]) { + if (s[1]==0 && s[0]<0x80) { + o[0] = s[0]; + return 1; + } + else if (s[1]<0x08) { + o[0] = 0xC0 | (s[1]<<2) | (s[0]>>6); + o[1] = 0x80 | (s[0]&0x3F); + return 2; + } + else { + o[0] = 0xE0 | (s[1]>>4); + o[1] = 0x80 | ((s[1]&0x0F)<<2) | (s[0]>>6); + o[2] = 0x80 | (s[0]&0x3F); + return 3; + } + } + else { + o[0] = 0xF0 | (s[2]>>2); + o[1] = 0x80 | ((s[2]&0x03)<<4) | (s[1]>>4); + o[2] = 0x80 | ((s[1]&0x0F)<<2) | (s[0]>>6); + o[3] = 0x80 | (s[0]&0x3F); + return 4; + } +} + +static ssize_t +fun_so_to_utf_32le(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + o[3] = 0; + if (!(s[0]&0x80)) { + o[2] = o[1] = 0x00; + o[0] = s[0]; + } + else if ((s[0]&0xE0)==0xC0) { + o[2] = 0x00; + o[1] = (s[0]>>2)&0x07; + o[0] = ((s[0]&0x03)<<6) | (s[1]&0x3F); + } + else if ((s[0]&0xF0)==0xE0) { + o[2] = 0x00; + o[1] = (s[0]<<4) | ((s[1]>>2)^0x20); + o[0] = (s[1]<<6) | (s[2]^0x80); + } + else { + o[2] = ((s[0]&0x07)<<2) | ((s[1]>>4)&0x03); + o[1] = ((s[1]&0x0F)<<4) | ((s[2]>>2)&0x0F); + o[0] = ((s[2]&0x03)<<6) | (s[3]&0x3F); + } + return 4; +} + +static int +state_init(void *statep) +{ + unsigned char *sp = statep; + *sp = 0; + return 0; +} + +static VALUE +fun_si_from_utf_16(void *statep, const unsigned char *s, size_t l) +{ + #define BE 1 + #define LE 2 + unsigned char *sp = statep; + switch (*sp) { + case 0: + if (s[0] == 0xFE && s[1] == 0xFF) { + *sp = BE; + return ZERObt; + } + else if (s[0] == 0xFF && s[1] == 0xFE) { + *sp = LE; + return ZERObt; + } + break; + case BE: + if (s[0] < 0xD8 || 0xDF < s[0]) { + return (VALUE)FUNso; + } + else if (s[0] <= 0xDB) { + return (VALUE)from_UTF_16BE_D8toDB_00toFF; + } + break; + case LE: + if (s[1] < 0xD8 || 0xDF < s[1]) { + return (VALUE)FUNso; + } + else if (s[1] <= 0xDB) { + return (VALUE)from_UTF_16LE_00toFF_D8toDB; + } + break; + } + return (VALUE)INVALID; +} + +static ssize_t +fun_so_from_utf_16(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + unsigned char *sp = statep; + switch (*sp) { + case BE: + return fun_so_from_utf_16be(statep, s, l, o, osize); + case LE: + return fun_so_from_utf_16le(statep, s, l, o, osize); + } + return 0; +} + +static VALUE +fun_si_from_utf_32(void *statep, const unsigned char *s, size_t l) +{ + unsigned char *sp = statep; + switch (*sp) { + case 0: + if (s[0] == 0 && s[1] == 0 && s[2] == 0xFE && s[3] == 0xFF) { + *sp = BE; + return ZERObt; + } + else if (s[0] == 0xFF && s[1] == 0xFE && s[2] == 0 && s[3] == 0) { + *sp = LE; + return ZERObt; + } + break; + case BE: + if (s[0] == 0 && ((0 < s[1] && s[1] <= 0x10) || + (s[1] == 0 && (s[2] < 0xD8 || 0xDF < s[2])))) + return (VALUE)FUNso; + break; + case LE: + if (s[3] == 0 && ((0 < s[2] && s[2] <= 0x10) || + (s[2] == 0 && (s[1] < 0xD8 || 0xDF < s[1])))) + return (VALUE)FUNso; + break; + } + return (VALUE)INVALID; +} + +static ssize_t +fun_so_from_utf_32(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + unsigned char *sp = statep; + switch (*sp) { + case BE: + return fun_so_from_utf_32be(statep, s, l, o, osize); + case LE: + return fun_so_from_utf_32le(statep, s, l, o, osize); + } + return 0; +} + +static ssize_t +fun_so_to_utf_16(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + unsigned char *sp = statep; + if (*sp == 0) { + *o++ = 0xFE; + *o++ = 0xFF; + *sp = 1; + return 2 + fun_so_to_utf_16be(statep, s, l, o, osize); + } + return fun_so_to_utf_16be(statep, s, l, o, osize); +} + +static ssize_t +fun_so_to_utf_32(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + unsigned char *sp = statep; + if (*sp == 0) { + *o++ = 0x00; + *o++ = 0x00; + *o++ = 0xFE; + *o++ = 0xFF; + *sp = 1; + return 4 + fun_so_to_utf_32be(statep, s, l, o, osize); + } + return fun_so_to_utf_32be(statep, s, l, o, osize); +} + +static const rb_transcoder +rb_from_UTF_16BE = { + "UTF-16BE", "UTF-8", from_UTF_16BE, + TRANSCODE_TABLE_INFO, + 2, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_from_utf_16be +}; + +static const rb_transcoder +rb_to_UTF_16BE = { + "UTF-8", "UTF-16BE", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_16be +}; + +static const rb_transcoder +rb_from_UTF_16LE = { + "UTF-16LE", "UTF-8", from_UTF_16LE, + TRANSCODE_TABLE_INFO, + 2, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_from_utf_16le +}; + +static const rb_transcoder +rb_to_UTF_16LE = { + "UTF-8", "UTF-16LE", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_16le +}; + +static const rb_transcoder +rb_from_UTF_32BE = { + "UTF-32BE", "UTF-8", from_UTF_32BE, + TRANSCODE_TABLE_INFO, + 4, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_from_utf_32be +}; + +static const rb_transcoder +rb_to_UTF_32BE = { + "UTF-8", "UTF-32BE", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_32be +}; + +static const rb_transcoder +rb_from_UTF_32LE = { + "UTF-32LE", "UTF-8", from_UTF_32LE, + TRANSCODE_TABLE_INFO, + 4, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_from_utf_32le +}; + +static const rb_transcoder +rb_to_UTF_32LE = { + "UTF-8", "UTF-32LE", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 0, NULL, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_32le +}; + +static const rb_transcoder +rb_from_UTF_16 = { + "UTF-16", "UTF-8", from_UTF_16, + TRANSCODE_TABLE_INFO, + 2, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 1, state_init, NULL, /* state_size, state_init, state_fini */ + NULL, fun_si_from_utf_16, NULL, fun_so_from_utf_16 +}; + +static const rb_transcoder +rb_from_UTF_32 = { + "UTF-32", "UTF-8", from_UTF_32, + TRANSCODE_TABLE_INFO, + 4, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 1, state_init, NULL, /* state_size, state_init, state_fini */ + NULL, fun_si_from_utf_32, NULL, fun_so_from_utf_32 +}; + +static const rb_transcoder +rb_to_UTF_16 = { + "UTF-8", "UTF-16", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 1, state_init, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_16 +}; + +static const rb_transcoder +rb_to_UTF_32 = { + "UTF-8", "UTF-32", from_UTF_8, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 4, /* max_input */ + 4, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 1, state_init, NULL, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_to_utf_32 +}; + +TRANS_INIT(utf_16_32) +{ + rb_register_transcoder(&rb_from_UTF_16BE); + 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); + rb_register_transcoder(&rb_from_UTF_16); + rb_register_transcoder(&rb_to_UTF_16); + rb_register_transcoder(&rb_from_UTF_32); + rb_register_transcoder(&rb_to_UTF_32); +} -- cgit v1.2.3