diff options
Diffstat (limited to 'gl/mbrtowc.c')
| -rw-r--r-- | gl/mbrtowc.c | 474 |
1 files changed, 237 insertions, 237 deletions
diff --git a/gl/mbrtowc.c b/gl/mbrtowc.c index 0fec5f1c..5c2650e9 100644 --- a/gl/mbrtowc.c +++ b/gl/mbrtowc.c | |||
| @@ -1,5 +1,5 @@ | |||
| 1 | /* Convert multibyte character to wide character. | 1 | /* Convert multibyte character to wide character. |
| 2 | Copyright (C) 1999-2002, 2005-2009 Free Software Foundation, Inc. | 2 | Copyright (C) 1999-2002, 2005-2010 Free Software Foundation, Inc. |
| 3 | Written by Bruno Haible <bruno@clisp.org>, 2008. | 3 | Written by Bruno Haible <bruno@clisp.org>, 2008. |
| 4 | 4 | ||
| 5 | This program is free software: you can redistribute it and/or modify | 5 | This program is free software: you can redistribute it and/or modify |
| @@ -63,30 +63,30 @@ mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) | |||
| 63 | switch (nstate) | 63 | switch (nstate) |
| 64 | { | 64 | { |
| 65 | case 0: | 65 | case 0: |
| 66 | p = s; | 66 | p = s; |
| 67 | m = n; | 67 | m = n; |
| 68 | break; | 68 | break; |
| 69 | case 3: | 69 | case 3: |
| 70 | buf[2] = pstate[3]; | 70 | buf[2] = pstate[3]; |
| 71 | /*FALLTHROUGH*/ | 71 | /*FALLTHROUGH*/ |
| 72 | case 2: | 72 | case 2: |
| 73 | buf[1] = pstate[2]; | 73 | buf[1] = pstate[2]; |
| 74 | /*FALLTHROUGH*/ | 74 | /*FALLTHROUGH*/ |
| 75 | case 1: | 75 | case 1: |
| 76 | buf[0] = pstate[1]; | 76 | buf[0] = pstate[1]; |
| 77 | p = buf; | 77 | p = buf; |
| 78 | m = nstate; | 78 | m = nstate; |
| 79 | buf[m++] = s[0]; | 79 | buf[m++] = s[0]; |
| 80 | if (n >= 2 && m < 4) | 80 | if (n >= 2 && m < 4) |
| 81 | { | 81 | { |
| 82 | buf[m++] = s[1]; | 82 | buf[m++] = s[1]; |
| 83 | if (n >= 3 && m < 4) | 83 | if (n >= 3 && m < 4) |
| 84 | buf[m++] = s[2]; | 84 | buf[m++] = s[2]; |
| 85 | } | 85 | } |
| 86 | break; | 86 | break; |
| 87 | default: | 87 | default: |
| 88 | errno = EINVAL; | 88 | errno = EINVAL; |
| 89 | return (size_t)(-1); | 89 | return (size_t)(-1); |
| 90 | } | 90 | } |
| 91 | 91 | ||
| 92 | /* Here m > 0. */ | 92 | /* Here m > 0. */ |
| @@ -99,208 +99,208 @@ mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) | |||
| 99 | int res = mbtowc (pwc, p, m); | 99 | int res = mbtowc (pwc, p, m); |
| 100 | 100 | ||
| 101 | if (res >= 0) | 101 | if (res >= 0) |
| 102 | { | 102 | { |
| 103 | if (pwc != NULL && ((*pwc == 0) != (res == 0))) | 103 | if (pwc != NULL && ((*pwc == 0) != (res == 0))) |
| 104 | abort (); | 104 | abort (); |
| 105 | if (nstate >= (res > 0 ? res : 1)) | 105 | if (nstate >= (res > 0 ? res : 1)) |
| 106 | abort (); | 106 | abort (); |
| 107 | res -= nstate; | 107 | res -= nstate; |
| 108 | pstate[0] = 0; | 108 | pstate[0] = 0; |
| 109 | return res; | 109 | return res; |
| 110 | } | 110 | } |
| 111 | 111 | ||
| 112 | /* mbtowc does not distinguish between invalid and incomplete multibyte | 112 | /* mbtowc does not distinguish between invalid and incomplete multibyte |
| 113 | sequences. But mbrtowc needs to make this distinction. | 113 | sequences. But mbrtowc needs to make this distinction. |
| 114 | There are two possible approaches: | 114 | There are two possible approaches: |
| 115 | - Use iconv() and its return value. | 115 | - Use iconv() and its return value. |
| 116 | - Use built-in knowledge about the possible encodings. | 116 | - Use built-in knowledge about the possible encodings. |
| 117 | Given the low quality of implementation of iconv() on the systems that | 117 | Given the low quality of implementation of iconv() on the systems that |
| 118 | lack mbrtowc(), we use the second approach. | 118 | lack mbrtowc(), we use the second approach. |
| 119 | The possible encodings are: | 119 | The possible encodings are: |
| 120 | - 8-bit encodings, | 120 | - 8-bit encodings, |
| 121 | - EUC-JP, EUC-KR, GB2312, EUC-TW, BIG5, GB18030, SJIS, | 121 | - EUC-JP, EUC-KR, GB2312, EUC-TW, BIG5, GB18030, SJIS, |
| 122 | - UTF-8. | 122 | - UTF-8. |
| 123 | Use specialized code for each. */ | 123 | Use specialized code for each. */ |
| 124 | if (m >= 4 || m >= MB_CUR_MAX) | 124 | if (m >= 4 || m >= MB_CUR_MAX) |
| 125 | goto invalid; | 125 | goto invalid; |
| 126 | /* Here MB_CUR_MAX > 1 and 0 < m < 4. */ | 126 | /* Here MB_CUR_MAX > 1 and 0 < m < 4. */ |
| 127 | { | 127 | { |
| 128 | const char *encoding = locale_charset (); | 128 | const char *encoding = locale_charset (); |
| 129 | 129 | ||
| 130 | if (STREQ (encoding, "UTF-8", 'U', 'T', 'F', '-', '8', 0, 0, 0, 0)) | 130 | if (STREQ (encoding, "UTF-8", 'U', 'T', 'F', '-', '8', 0, 0, 0, 0)) |
| 131 | { | 131 | { |
| 132 | /* Cf. unistr/u8-mblen.c. */ | 132 | /* Cf. unistr/u8-mblen.c. */ |
| 133 | unsigned char c = (unsigned char) p[0]; | 133 | unsigned char c = (unsigned char) p[0]; |
| 134 | 134 | ||
| 135 | if (c >= 0xc2) | 135 | if (c >= 0xc2) |
| 136 | { | 136 | { |
| 137 | if (c < 0xe0) | 137 | if (c < 0xe0) |
| 138 | { | 138 | { |
| 139 | if (m == 1) | 139 | if (m == 1) |
| 140 | goto incomplete; | 140 | goto incomplete; |
| 141 | } | 141 | } |
| 142 | else if (c < 0xf0) | 142 | else if (c < 0xf0) |
| 143 | { | 143 | { |
| 144 | if (m == 1) | 144 | if (m == 1) |
| 145 | goto incomplete; | 145 | goto incomplete; |
| 146 | if (m == 2) | 146 | if (m == 2) |
| 147 | { | 147 | { |
| 148 | unsigned char c2 = (unsigned char) p[1]; | 148 | unsigned char c2 = (unsigned char) p[1]; |
| 149 | 149 | ||
| 150 | if ((c2 ^ 0x80) < 0x40 | 150 | if ((c2 ^ 0x80) < 0x40 |
| 151 | && (c >= 0xe1 || c2 >= 0xa0) | 151 | && (c >= 0xe1 || c2 >= 0xa0) |
| 152 | && (c != 0xed || c2 < 0xa0)) | 152 | && (c != 0xed || c2 < 0xa0)) |
| 153 | goto incomplete; | 153 | goto incomplete; |
| 154 | } | 154 | } |
| 155 | } | 155 | } |
| 156 | else if (c <= 0xf4) | 156 | else if (c <= 0xf4) |
| 157 | { | 157 | { |
| 158 | if (m == 1) | 158 | if (m == 1) |
| 159 | goto incomplete; | 159 | goto incomplete; |
| 160 | else /* m == 2 || m == 3 */ | 160 | else /* m == 2 || m == 3 */ |
| 161 | { | 161 | { |
| 162 | unsigned char c2 = (unsigned char) p[1]; | 162 | unsigned char c2 = (unsigned char) p[1]; |
| 163 | 163 | ||
| 164 | if ((c2 ^ 0x80) < 0x40 | 164 | if ((c2 ^ 0x80) < 0x40 |
| 165 | && (c >= 0xf1 || c2 >= 0x90) | 165 | && (c >= 0xf1 || c2 >= 0x90) |
| 166 | && (c < 0xf4 || (c == 0xf4 && c2 < 0x90))) | 166 | && (c < 0xf4 || (c == 0xf4 && c2 < 0x90))) |
| 167 | { | 167 | { |
| 168 | if (m == 2) | 168 | if (m == 2) |
| 169 | goto incomplete; | 169 | goto incomplete; |
| 170 | else /* m == 3 */ | 170 | else /* m == 3 */ |
| 171 | { | 171 | { |
| 172 | unsigned char c3 = (unsigned char) p[2]; | 172 | unsigned char c3 = (unsigned char) p[2]; |
| 173 | 173 | ||
| 174 | if ((c3 ^ 0x80) < 0x40) | 174 | if ((c3 ^ 0x80) < 0x40) |
| 175 | goto incomplete; | 175 | goto incomplete; |
| 176 | } | 176 | } |
| 177 | } | 177 | } |
| 178 | } | 178 | } |
| 179 | } | 179 | } |
| 180 | } | 180 | } |
| 181 | goto invalid; | 181 | goto invalid; |
| 182 | } | 182 | } |
| 183 | 183 | ||
| 184 | /* As a reference for this code, you can use the GNU libiconv | 184 | /* As a reference for this code, you can use the GNU libiconv |
| 185 | implementation. Look for uses of the RET_TOOFEW macro. */ | 185 | implementation. Look for uses of the RET_TOOFEW macro. */ |
| 186 | 186 | ||
| 187 | if (STREQ (encoding, "EUC-JP", 'E', 'U', 'C', '-', 'J', 'P', 0, 0, 0)) | 187 | if (STREQ (encoding, "EUC-JP", 'E', 'U', 'C', '-', 'J', 'P', 0, 0, 0)) |
| 188 | { | 188 | { |
| 189 | if (m == 1) | 189 | if (m == 1) |
| 190 | { | 190 | { |
| 191 | unsigned char c = (unsigned char) p[0]; | 191 | unsigned char c = (unsigned char) p[0]; |
| 192 | 192 | ||
| 193 | if ((c >= 0xa1 && c < 0xff) || c == 0x8e || c == 0x8f) | 193 | if ((c >= 0xa1 && c < 0xff) || c == 0x8e || c == 0x8f) |
| 194 | goto incomplete; | 194 | goto incomplete; |
| 195 | } | 195 | } |
| 196 | if (m == 2) | 196 | if (m == 2) |
| 197 | { | 197 | { |
| 198 | unsigned char c = (unsigned char) p[0]; | 198 | unsigned char c = (unsigned char) p[0]; |
| 199 | 199 | ||
| 200 | if (c == 0x8f) | 200 | if (c == 0x8f) |
| 201 | { | 201 | { |
| 202 | unsigned char c2 = (unsigned char) p[1]; | 202 | unsigned char c2 = (unsigned char) p[1]; |
| 203 | 203 | ||
| 204 | if (c2 >= 0xa1 && c2 < 0xff) | 204 | if (c2 >= 0xa1 && c2 < 0xff) |
| 205 | goto incomplete; | 205 | goto incomplete; |
| 206 | } | 206 | } |
| 207 | } | 207 | } |
| 208 | goto invalid; | 208 | goto invalid; |
| 209 | } | 209 | } |
| 210 | if (STREQ (encoding, "EUC-KR", 'E', 'U', 'C', '-', 'K', 'R', 0, 0, 0) | 210 | if (STREQ (encoding, "EUC-KR", 'E', 'U', 'C', '-', 'K', 'R', 0, 0, 0) |
| 211 | || STREQ (encoding, "GB2312", 'G', 'B', '2', '3', '1', '2', 0, 0, 0) | 211 | || STREQ (encoding, "GB2312", 'G', 'B', '2', '3', '1', '2', 0, 0, 0) |
| 212 | || STREQ (encoding, "BIG5", 'B', 'I', 'G', '5', 0, 0, 0, 0, 0)) | 212 | || STREQ (encoding, "BIG5", 'B', 'I', 'G', '5', 0, 0, 0, 0, 0)) |
| 213 | { | 213 | { |
| 214 | if (m == 1) | 214 | if (m == 1) |
| 215 | { | 215 | { |
| 216 | unsigned char c = (unsigned char) p[0]; | 216 | unsigned char c = (unsigned char) p[0]; |
| 217 | 217 | ||
| 218 | if (c >= 0xa1 && c < 0xff) | 218 | if (c >= 0xa1 && c < 0xff) |
| 219 | goto incomplete; | 219 | goto incomplete; |
| 220 | } | 220 | } |
| 221 | goto invalid; | 221 | goto invalid; |
| 222 | } | 222 | } |
| 223 | if (STREQ (encoding, "EUC-TW", 'E', 'U', 'C', '-', 'T', 'W', 0, 0, 0)) | 223 | if (STREQ (encoding, "EUC-TW", 'E', 'U', 'C', '-', 'T', 'W', 0, 0, 0)) |
| 224 | { | 224 | { |
| 225 | if (m == 1) | 225 | if (m == 1) |
| 226 | { | 226 | { |
| 227 | unsigned char c = (unsigned char) p[0]; | 227 | unsigned char c = (unsigned char) p[0]; |
| 228 | 228 | ||
| 229 | if ((c >= 0xa1 && c < 0xff) || c == 0x8e) | 229 | if ((c >= 0xa1 && c < 0xff) || c == 0x8e) |
| 230 | goto incomplete; | 230 | goto incomplete; |
| 231 | } | 231 | } |
| 232 | else /* m == 2 || m == 3 */ | 232 | else /* m == 2 || m == 3 */ |
| 233 | { | 233 | { |
| 234 | unsigned char c = (unsigned char) p[0]; | 234 | unsigned char c = (unsigned char) p[0]; |
| 235 | 235 | ||
| 236 | if (c == 0x8e) | 236 | if (c == 0x8e) |
| 237 | goto incomplete; | 237 | goto incomplete; |
| 238 | } | 238 | } |
| 239 | goto invalid; | 239 | goto invalid; |
| 240 | } | 240 | } |
| 241 | if (STREQ (encoding, "GB18030", 'G', 'B', '1', '8', '0', '3', '0', 0, 0)) | 241 | if (STREQ (encoding, "GB18030", 'G', 'B', '1', '8', '0', '3', '0', 0, 0)) |
| 242 | { | 242 | { |
| 243 | if (m == 1) | 243 | if (m == 1) |
| 244 | { | 244 | { |
| 245 | unsigned char c = (unsigned char) p[0]; | 245 | unsigned char c = (unsigned char) p[0]; |
| 246 | 246 | ||
| 247 | if ((c >= 0x90 && c <= 0xe3) || (c >= 0xf8 && c <= 0xfe)) | 247 | if ((c >= 0x90 && c <= 0xe3) || (c >= 0xf8 && c <= 0xfe)) |
| 248 | goto incomplete; | 248 | goto incomplete; |
| 249 | } | 249 | } |
| 250 | else /* m == 2 || m == 3 */ | 250 | else /* m == 2 || m == 3 */ |
| 251 | { | 251 | { |
| 252 | unsigned char c = (unsigned char) p[0]; | 252 | unsigned char c = (unsigned char) p[0]; |
| 253 | 253 | ||
| 254 | if (c >= 0x90 && c <= 0xe3) | 254 | if (c >= 0x90 && c <= 0xe3) |
| 255 | { | 255 | { |
| 256 | unsigned char c2 = (unsigned char) p[1]; | 256 | unsigned char c2 = (unsigned char) p[1]; |
| 257 | 257 | ||
| 258 | if (c2 >= 0x30 && c2 <= 0x39) | 258 | if (c2 >= 0x30 && c2 <= 0x39) |
| 259 | { | 259 | { |
| 260 | if (m == 2) | 260 | if (m == 2) |
| 261 | goto incomplete; | 261 | goto incomplete; |
| 262 | else /* m == 3 */ | 262 | else /* m == 3 */ |
| 263 | { | 263 | { |
| 264 | unsigned char c3 = (unsigned char) p[2]; | 264 | unsigned char c3 = (unsigned char) p[2]; |
| 265 | 265 | ||
| 266 | if (c3 >= 0x81 && c3 <= 0xfe) | 266 | if (c3 >= 0x81 && c3 <= 0xfe) |
| 267 | goto incomplete; | 267 | goto incomplete; |
| 268 | } | 268 | } |
| 269 | } | 269 | } |
| 270 | } | 270 | } |
| 271 | } | 271 | } |
| 272 | goto invalid; | 272 | goto invalid; |
| 273 | } | 273 | } |
| 274 | if (STREQ (encoding, "SJIS", 'S', 'J', 'I', 'S', 0, 0, 0, 0, 0)) | 274 | if (STREQ (encoding, "SJIS", 'S', 'J', 'I', 'S', 0, 0, 0, 0, 0)) |
| 275 | { | 275 | { |
| 276 | if (m == 1) | 276 | if (m == 1) |
| 277 | { | 277 | { |
| 278 | unsigned char c = (unsigned char) p[0]; | 278 | unsigned char c = (unsigned char) p[0]; |
| 279 | 279 | ||
| 280 | if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xea) | 280 | if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xea) |
| 281 | || (c >= 0xf0 && c <= 0xf9)) | 281 | || (c >= 0xf0 && c <= 0xf9)) |
| 282 | goto incomplete; | 282 | goto incomplete; |
| 283 | } | 283 | } |
| 284 | goto invalid; | 284 | goto invalid; |
| 285 | } | 285 | } |
| 286 | 286 | ||
| 287 | /* An unknown multibyte encoding. */ | 287 | /* An unknown multibyte encoding. */ |
| 288 | goto incomplete; | 288 | goto incomplete; |
| 289 | } | 289 | } |
| 290 | 290 | ||
| 291 | incomplete: | 291 | incomplete: |
| 292 | { | 292 | { |
| 293 | size_t k = nstate; | 293 | size_t k = nstate; |
| 294 | /* Here 0 <= k < m < 4. */ | 294 | /* Here 0 <= k < m < 4. */ |
| 295 | pstate[++k] = s[0]; | 295 | pstate[++k] = s[0]; |
| 296 | if (k < m) | 296 | if (k < m) |
| 297 | { | 297 | { |
| 298 | pstate[++k] = s[1]; | 298 | pstate[++k] = s[1]; |
| 299 | if (k < m) | 299 | if (k < m) |
| 300 | pstate[++k] = s[2]; | 300 | pstate[++k] = s[2]; |
| 301 | } | 301 | } |
| 302 | if (k != m) | 302 | if (k != m) |
| 303 | abort (); | 303 | abort (); |
| 304 | } | 304 | } |
| 305 | pstate[0] = m; | 305 | pstate[0] = m; |
| 306 | return (size_t)(-2); | 306 | return (size_t)(-2); |
| @@ -341,25 +341,25 @@ rpl_mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) | |||
| 341 | 341 | ||
| 342 | if (!mbsinit (ps)) | 342 | if (!mbsinit (ps)) |
| 343 | { | 343 | { |
| 344 | /* Parse the rest of the multibyte character byte for byte. */ | 344 | /* Parse the rest of the multibyte character byte for byte. */ |
| 345 | size_t count = 0; | 345 | size_t count = 0; |
| 346 | for (; n > 0; s++, n--) | 346 | for (; n > 0; s++, n--) |
| 347 | { | 347 | { |
| 348 | wchar_t wc; | 348 | wchar_t wc; |
| 349 | size_t ret = mbrtowc (&wc, s, 1, ps); | 349 | size_t ret = mbrtowc (&wc, s, 1, ps); |
| 350 | 350 | ||
| 351 | if (ret == (size_t)(-1)) | 351 | if (ret == (size_t)(-1)) |
| 352 | return (size_t)(-1); | 352 | return (size_t)(-1); |
| 353 | count++; | 353 | count++; |
| 354 | if (ret != (size_t)(-2)) | 354 | if (ret != (size_t)(-2)) |
| 355 | { | 355 | { |
| 356 | /* The multibyte character has been completed. */ | 356 | /* The multibyte character has been completed. */ |
| 357 | if (pwc != NULL) | 357 | if (pwc != NULL) |
| 358 | *pwc = wc; | 358 | *pwc = wc; |
| 359 | return (wc == 0 ? 0 : count); | 359 | return (wc == 0 ? 0 : count); |
| 360 | } | 360 | } |
| 361 | } | 361 | } |
| 362 | return (size_t)(-2); | 362 | return (size_t)(-2); |
| 363 | } | 363 | } |
| 364 | } | 364 | } |
| 365 | # endif | 365 | # endif |
| @@ -371,10 +371,10 @@ rpl_mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps) | |||
| 371 | 371 | ||
| 372 | if (ret != (size_t)(-1) && ret != (size_t)(-2)) | 372 | if (ret != (size_t)(-1) && ret != (size_t)(-2)) |
| 373 | { | 373 | { |
| 374 | if (pwc != NULL) | 374 | if (pwc != NULL) |
| 375 | *pwc = wc; | 375 | *pwc = wc; |
| 376 | if (wc == 0) | 376 | if (wc == 0) |
| 377 | ret = 0; | 377 | ret = 0; |
| 378 | } | 378 | } |
| 379 | return ret; | 379 | return ret; |
| 380 | } | 380 | } |
