ruby-changes:11668
From: naruse <ko1@a...>
Date: Wed, 29 Apr 2009 21:51:12 +0900 (JST)
Subject: [ruby-changes:11668] Ruby:r23307 (trunk): Add new transcoder: CP51932 <-> CP50221.
naruse 2009-04-29 21:50:57 +0900 (Wed, 29 Apr 2009) New Revision: 23307 http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=rev&revision=23307 Log: Add new transcoder: CP51932 <-> CP50221. Modified files: trunk/ChangeLog trunk/enc/iso_2022_jp.h trunk/enc/trans/iso2022.trans Index: ChangeLog =================================================================== --- ChangeLog (revision 23306) +++ ChangeLog (revision 23307) @@ -1,3 +1,9 @@ +Wed Apr 29 21:23:40 2009 NARUSE, Yui <naruse@r...> + + * enc/iso_2022_jp.h: add CP50221. + + * enc/trans/iso2022.trans: add converter for CP50221. + Wed Apr 29 15:22:26 2009 Nobuyoshi Nakada <nobu@r...> * file.c (rb_file_join): recursive array has no meaning as path Index: enc/trans/iso2022.trans =================================================================== --- enc/trans/iso2022.trans (revision 23306) +++ enc/trans/iso2022.trans (revision 23307) @@ -38,6 +38,26 @@ } transcode_generate_node(ActionMap.parse(map), "eucjp_to_stateless_iso2022jp") + map = { + "1b2842" => :func_so, # designate US-ASCII to G0. "ESC ( B" + "1b2849" => :func_so, # designate JIS X 0201 katakana to G0. "ESC ( I" + "1b284a" => :func_so, # designate JIS X 0201 latin to G0. "ESC ( J" + "1b2440" => :func_so, # designate JIS X 0208 1978 to G0. "ESC $ @" + "1b2442" => :func_so, # designate JIS X 0208 1983 to G0. "ESC $ B" + "0e" => :func_so, # designate JIS X 0201 katakana to G0. "SO" + "0f" => :func_so, # designate US-ASCII to G0. "SI" + "{00-0d,10-1a,1c-92}" => :func_si, + } + transcode_generate_node(ActionMap.parse(map), "cp50221_decoder") + + map = { + "{0e,0f,1b}" => :undef, + "{00-0d,10-1a,1c-7f}" => :func_so, + "{a1-fe}{a1-fe}" => :func_so, + "8e{a1-fe}" => :func_so, + } + transcode_generate_node(ActionMap.parse(map), "cp50221_encoder") + %> <%= transcode_generated_code %> @@ -46,6 +66,7 @@ /* ignore JIS X 0201 latin */ #define G0_JISX0208_1978 1 #define G0_JISX0208_1983 2 +#define G0_JISX0201_KATAKANA 3 #define EMACS_MULE_LEADING_CODE_JISX0208_1978 0220 #define EMACS_MULE_LEADING_CODE_JISX0208_1983 0222 @@ -244,6 +265,166 @@ NULL, NULL, NULL, fun_so_eucjp_to_stateless_iso2022jp, }; +static VALUE +fun_si_cp50221_decoder(void *statep, const unsigned char *s, size_t l) +{ + unsigned char *sp = statep; + switch (*sp) { + case G0_ASCII: + return (VALUE)NOMAP; + case G0_JISX0201_KATAKANA: + if (0x21 <= s[0] && s[0] <= 0x5f) + return (VALUE)FUNso; + break; + case G0_JISX0208_1978: + if (0x21 <= s[0] && s[0] <= 0x28 || 0x30 <= s[0] && s[0] <= 0x74) + return (VALUE)iso2022jp_decoder_jisx0208_rest; + break; + case G0_JISX0208_1983: + if (0x21 <= s[0] && s[0] <= 0x28 || + s[0] == 0x2D || + 0x30 <= s[0] && s[0] <= 0x74 || + 0x79 <= s[0] && s[0] <= 0x7C) + /* 0x7F <= s[0] && s[0] <= 0x92) */ + return (VALUE)iso2022jp_decoder_jisx0208_rest; + break; + } + return (VALUE)INVALID; +} + +static ssize_t +fun_so_cp50221_decoder(void *statep, const unsigned char *s, size_t l, unsigned char* o, size_t osize) +{ + unsigned char *sp = statep; + switch (s[0]) { + case 0x1b: + if (s[1] == '(') { + switch (s[l-1]) { + case 'B': + case 'J': + *sp = G0_ASCII; + break; + case 'I': + *sp = G0_JISX0201_KATAKANA; + break; + } + } + else { + switch (s[l-1]) { + case '@': + *sp = G0_JISX0208_1978; + break; + case 'B': + *sp = G0_JISX0208_1983; + break; + } + } + return 0; + case 0x0E: + *sp = G0_JISX0201_KATAKANA; + return 0; + case 0x0F: + *sp = G0_ASCII; + return 0; + default: + if (*sp == G0_JISX0201_KATAKANA) { + o[0] = 0x8E; + o[1] = s[0] | 0x80; + } + /* else if (0x7F == s[0] && s[0] <= 0x88) { */ + /* User Defined Characters */ + /* o[n++] = s[0] | 0xE0; */ + /* o[n++] = s[1] | 0x80; */ + /* else if (0x89 <= s[0] && s[0] <= 0x92) { */ + /* User Defined Characters 2 */ + /* o[n++] = 0x8f; */ + /* o[n++] = s[0] + 0x6C; */ + /* o[n++] = s[1] | 0x80; */ + /* } */ + else { + /* JIS X 0208 */ + /* NEC Special Characters */ + /* NEC-selected IBM extended Characters */ + o[0] = s[0] | 0x80; + o[1] = s[1] | 0x80; + } + return 2; + } +} + +static const rb_transcoder +rb_cp50221_decoder = { + "CP50221", "cp51932", cp50221_decoder, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 3, /* max_input */ + 3, /* max_output */ + asciicompat_decoder, /* asciicompat_type */ + 1, iso2022jp_init, iso2022jp_init, /* state_size, state_init, state_fini */ + NULL, fun_si_cp50221_decoder, NULL, fun_so_cp50221_decoder +}; + +static ssize_t +fun_so_cp50221_encoder(void *statep, const unsigned char *s, size_t l, unsigned char *o, size_t osize) +{ + unsigned char *sp = statep; + unsigned char *output0 = o; + int newstate; + + if (l == 1) + newstate = G0_ASCII; + else if (s[0] == 0x8E) { + s++; + l = 1; + newstate = G0_JISX0201_KATAKANA; + } + else + newstate = G0_JISX0208_1983; + + if (*sp != newstate) { + if (newstate == G0_ASCII) { + *o++ = 0x1b; + *o++ = '('; + *o++ = 'B'; + } + else if (newstate == G0_JISX0201_KATAKANA) { + *o++ = 0x1b; + *o++ = '('; + *o++ = 'I'; + } + else { + *o++ = 0x1b; + *o++ = '$'; + *o++ = 'B'; + } + *sp = newstate; + } + + if (l == 1) { + *o++ = s[0] & 0x7f; + } + else { + *o++ = s[0] & 0x7f; + *o++ = s[1] & 0x7f; + } + + return o - output0; +} + +static const rb_transcoder +rb_cp50221_encoder = { + "CP51932", "CP50221", cp50221_encoder, + TRANSCODE_TABLE_INFO, + 1, /* input_unit_length */ + 3, /* max_input */ + 5, /* max_output */ + asciicompat_encoder, /* asciicompat_type */ + 1, iso2022jp_init, iso2022jp_init, /* state_size, state_init, state_fini */ + NULL, NULL, NULL, fun_so_cp50221_encoder, + finish_iso2022jp_encoder, + iso2022jp_encoder_reset_sequence_size, finish_iso2022jp_encoder +}; + void Init_iso2022(void) { @@ -251,5 +432,7 @@ rb_register_transcoder(&rb_iso2022jp_encoder); rb_register_transcoder(&rb_stateless_iso2022jp_to_eucjp); rb_register_transcoder(&rb_eucjp_to_stateless_iso2022jp); + rb_register_transcoder(&rb_cp50221_decoder); + rb_register_transcoder(&rb_cp50221_encoder); } Index: enc/iso_2022_jp.h =================================================================== --- enc/iso_2022_jp.h (revision 23306) +++ enc/iso_2022_jp.h (revision 23307) @@ -4,3 +4,21 @@ ENC_ALIAS("ISO2022-JP", "ISO-2022-JP"); ENC_REPLICATE("ISO-2022-JP-2", "ISO-2022-JP"); ENC_ALIAS("ISO2022-JP2", "ISO-2022-JP-2"); + +/* Windows Codepage 50221 + * a ISO-2022-JP variant. + * This includes + * * US-ASCII + * * JIS X 0201 Katakana + * * JIS X 0208 + * * NEC special characters + * * NEC selected IBM extended characters + * and this implementation doesn't include + * * JIS X 0201 Latin + * * User Defined Characters + * + * So this CP50221 has the same characters of CP51932. + * + * See http://legacy-encoding.sourceforge.jp/wiki/index.php?cp50221 + */ +ENC_REPLICATE("CP50221", "ISO-2022-JP"); -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/