src/japanese.c

   1 /*!
   2  *  @file japanese.c
   3  *  @brief 日本語処理関数
   4  *  @date 2014/07/07
   5  */
   6
   7 #include "angband.h"
   8
   9 #ifdef JP
  10
  11 typedef struct convert_key convert_key;
  12
  13 struct convert_key
  14 {
  15         cptr key1;
  16         cptr key2;
  17 };
  18
  19 static const convert_key s2j_table[] = {
  20         {"mb","nb"}, {"mp","np"}, {"mv","nv"}, {"mm","nm"},
  21         {"x","ks"},
  22         /* sindar:シンダール  parantir:パランティア  feanor:フェアノール */
  23         {"ar$","a-ru$"}, {"ir$","ia$"}, {"or$","o-ru$"},
  24         {"ra","ラ"}, {"ri","リ"}, {"ru","ル"}, {"re","レ"}, {"ro","ロ"},
  25         {"ir","ia"}, {"ur","ua"}, {"er","ea"}, {"ar","aル"},
  26         {"sha","シャ"}, {"shi","シ"}, {"shu","シュ"}, {"she","シェ"}, {"sho","ショ"},
  27         {"tha","サ"}, {"thi","シ"}, {"thu","ス"}, {"the","セ"}, {"tho","ソ"},
  28         {"cha","ハ"}, {"chi","ヒ"}, {"chu","フ"}, {"che","ヘ"}, {"cho","ホ"},
  29         {"dha","ザ"}, {"dhi","ジ"}, {"dhu","ズ"}, {"dhe","ゼ"}, {"dho","ゾ"},
  30         {"ba","バ"}, {"bi","ビ"}, {"bu","ブ"}, {"be","ベ"}, {"bo","ボ"},
  31         {"ca","カ"}, {"ci","キ"}, {"cu","ク"}, {"ce","ケ"}, {"co","コ"},
  32         {"da","ダ"}, {"di","ディ"}, {"du","ドゥ"}, {"de","デ"}, {"do","ド"},
  33         {"fa","ファ"}, {"fi","フィ"}, {"fu","フ"}, {"fe","フェ"}, {"fo","フォ"},
  34         {"ga","ガ"}, {"gi","ギ"}, {"gu","グ"}, {"ge","ゲ"}, {"go","ゴ"},
  35         {"ha","ハ"}, {"hi","ヒ"}, {"hu","フ"}, {"he","ヘ"}, {"ho","ホ"},
  36         {"ja","ジャ"}, {"ji","ジ"}, {"ju","ジュ"}, {"je","ジェ"}, {"jo","ジョ"},
  37         {"ka","カ"}, {"ki","キ"}, {"ku","ク"}, {"ke","ケ"}, {"ko","コ"},
  38         {"la","ラ"}, {"li","リ"}, {"lu","ル"}, {"le","レ"}, {"lo","ロ"},
  39         {"ma","マ"}, {"mi","ミ"}, {"mu","ム"}, {"me","メ"}, {"mo","モ"},
  40         {"na","ナ"}, {"ni","ニ"}, {"nu","ヌ"}, {"ne","ネ"}, {"no","ノ"},
  41         {"pa","パ"}, {"pi","ピ"}, {"pu","プ"}, {"pe","ペ"}, {"po","ポ"},
  42         {"qu","ク"},
  43         {"sa","サ"}, {"si","シ"}, {"su","ス"}, {"se","セ"}, {"so","ソ"},
  44         {"ta","タ"}, {"ti","ティ"}, {"tu","トゥ"}, {"te","テ"}, {"to","ト"},
  45         {"va","ヴァ"}, {"vi","ヴィ"}, {"vu","ヴ"}, {"ve","ヴェ"}, {"vo","ヴォ"},
  46         {"wa","ワ"}, {"wi","ウィ"}, {"wu","ウ"}, {"we","ウェ"}, {"wo","ウォ"},
  47         {"ya","ヤ"}, {"yu","ユ"}, {"yo","ヨ"},
  48         {"za","ザ"}, {"zi","ジ"}, {"zu","ズ"}, {"ze","ゼ"}, {"zo","ゾ"},
  49         {"dh","ズ"}, {"ch","フ"}, {"th","ス"},
  50         {"b","ブ"}, {"c","ク"}, {"d","ド"}, {"f","フ"}, {"g","グ"},
  51         {"h","フ"}, {"j","ジュ"}, {"k","ク"}, {"l","ル"}, {"m","ム"},
  52         {"n","ン"}, {"p","プ"}, {"q","ク"}, {"r","ル"}, {"s","ス"},
  53         {"t","ト"}, {"v","ヴ"}, {"w","ウ"}, {"y","イ"},
  54         {"a","ア"}, {"i","イ"}, {"u","ウ"}, {"e","エ"}, {"o","オ"},
  55         {"-","ー"},
  56         {NULL,NULL}
  57 };
  58
  59 /*!
  60  * @brief シンダリンを日本語の読みに変換する
  61  * @param kana 変換後の日本語文字列ポインタ
  62  * @param sindarin 変換前のシンダリン文字列ポインタ
  63  * @return なし
  64  * @details
  65  */
  66 void sindarin_to_kana(char *kana, const char *sindarin)
  67 {
  68         char buf[256];
  69         int idx;
  70
  71         sprintf(kana, "%s$", sindarin);
  72         for (idx = 0; kana[idx]; idx++)
  73                 if (isupper(kana[idx])) kana[idx] = (char)tolower(kana[idx]);
  74
  75         for (idx = 0; s2j_table[idx].key1 != NULL; idx++)
  76         {
  77                 cptr pat1 = s2j_table[idx].key1;
  78                 cptr pat2 = s2j_table[idx].key2;
  79                 int len = strlen(pat1);
  80                 char *src = kana;
  81                 char *dest = buf;
  82
  83                 while (*src)
  84                 {
  85                         if (strncmp(src, pat1, len) == 0)
  86                         {
  87                                 strcpy(dest, pat2);
  88                                 src += len;
  89                                 dest += strlen(pat2);
  90                         }
  91                         else
  92                         {
  93                                 if (iskanji(*src))
  94                                 {
  95                                         *dest = *src;
  96                                         src++;
  97                                         dest++;
  98                                 }
  99                                 *dest = *src;
 100                                 src++;
 101                                 dest++;
 102                         }
 103                 }
 104
 105                 *dest = 0;
 106                 strcpy(kana, buf);
 107         }
 108
 109         idx = 0;
 110
 111         while (kana[idx] != '$') idx++;
 112
 113         kana[idx] = '\0';
 114 }
 115
 116
 117 /*! 日本語動詞活用 (打つ＞打って,打ち etc)
 118  * JVERB_AND: 殴る,蹴る > 殴り,蹴る
 119  * JVERB_TO:  殴る,蹴る > 殴って蹴る
 120  * JVERB_OR:  殴る,蹴る > 殴ったり蹴ったり */
 121 static const struct jverb_table_t {
 122         const char* from;
 123         const char* to[3];
 124 } jverb_table[] = {
 125         { "する", {"し", "して", "した"}},
 126         { "いる", {"いて", "いて", "いた"}},
 127
 128         { "える", {"え", "えて", "えた"}},
 129         { "ける", {"け", "けて", "けた"}},
 130         { "げる", {"げ", "えて", "げた"}},
 131         { "せる", {"せ", "せて", "せた"}},
 132         { "ぜる", {"ぜ", "ぜて", "ぜた"}},
 133         { "てる", {"て", "てって", "てった"}},
 134         { "でる", {"で", "でて", "でた"}},
 135         { "ねる", {"ね", "ねて", "ねた"}},
 136         { "へる", {"へ", "へて", "へた"}},
 137         { "べる", {"べ", "べて", "べた"}},
 138         { "める", {"め", "めて", "めた"}},
 139         { "れる", {"れ", "れて", "れた"}},
 140
 141         { "う", {"い", "って", "った"}},
 142         { "く", {"き", "いて", "いた"}},
 143         { "ぐ", {"ぎ", "いで", "いだ"}},
 144         { "す", {"し", "して", "した"}},
 145         { "ず", {"じ", "じて", "じた"}},
 146         { "つ", {"ち", "って", "った"}},
 147         { "づ", {"ぢ", "って", "った"}},
 148         { "ぬ", {"に", "ねて", "ねた"}},
 149         { "ふ", {"ひ", "へて", "へた"}},
 150         { "ぶ", {"び", "んで", "んだ"}},
 151         { "む", {"み", "んで", "んだ"}},
 152         { "る", {"り", "って", "った"}},
 153         { NULL, {"そして", "ことにより", "ことや"}},
 154 };
 155
 156 /*!
 157  * @brief jverb_table_tに従って動詞を活用する
 158  * @param in 変換元文字列ポインタ
 159  * @param out 変換先文字列ポインタ
 160  * @param flag 変換種類を指定(JVERB_AND/JVERB_TO/JVERB_OR)
 161  * @return なし
 162  * @details
 163  */
 164 void jverb(const char *in, char *out, int flag)
 165 {
 166         const struct jverb_table_t * p;
 167         int in_len = strlen(in);
 168
 169         strcpy(out, in);
 170
 171         for (p = jverb_table; p->from; p++) {
 172                 int from_len = strlen(p->from);
 173                 if (strncmp(&in[in_len-from_len], p->from, from_len) == 0) {
 174                         strcpy(&out[in_len - from_len], p->to[flag - 1]);
 175                         break;
 176                 }
 177         }
 178
 179         if (p->from == NULL)
 180                 strcpy(&out[in_len], p->to[flag - 1]);
 181 }
 182
 183 /*!
 184  * @brief 文字コードをSJISからEUCに変換する / Convert SJIS string to EUC string
 185  * @param str 変換する文字列のポインタ
 186  * @return なし
 187  * @details
 188  */
 189 void sjis2euc(char *str)
 190 {
 191         int i;
 192         unsigned char c1, c2;
 193         unsigned char *tmp;
 194
 195         int len = strlen(str);
 196
 197         C_MAKE(tmp, len+1, byte);
 198
 199         for (i = 0; i < len; i++)
 200         {
 201                 c1 = str[i];
 202                 if (c1 & 0x80)
 203                 {
 204                         i++;
 205                         c2 = str[i];
 206                         if (c2 >= 0x9f)
 207                         {
 208                                 c1 = c1 * 2 - (c1 >= 0xe0 ? 0xe0 : 0x60);
 209                                 c2 += 2;
 210                         }
 211                         else
 212                         {
 213                                 c1 = c1 * 2 - (c1 >= 0xe0 ? 0xe1 : 0x61);
 214                                 c2 += 0x60 + (c2 < 0x7f);
 215                         }
 216                         tmp[i - 1] = c1;
 217                         tmp[i] = c2;
 218                 }
 219                 else
 220                         tmp[i] = c1;
 221         }
 222         tmp[len] = 0;
 223         strcpy(str, (char *)tmp);
 224
 225         C_KILL(tmp, len+1, byte);
 226 }
 227
 228
 229 /*!
 230  * @brief 文字コードをEUCからSJISに変換する / Convert EUC string to SJIS string
 231  * @param str 変換する文字列のポインタ
 232  * @return なし
 233  * @details
 234  */
 235 void euc2sjis(char *str)
 236 {
 237         int i;
 238         unsigned char c1, c2;
 239         unsigned char *tmp;
 240
 241         int len = strlen(str);
 242
 243         C_MAKE(tmp, len+1, byte);
 244
 245         for (i = 0; i < len; i++)
 246         {
 247                 c1 = str[i];
 248                 if (c1 & 0x80)
 249                 {
 250                         i++;
 251                         c2 = str[i];
 252                         if (c1 % 2)
 253                         {
 254                                 c1 = (c1 >> 1) + (c1 < 0xdf ? 0x31 : 0x71);
 255                                 c2 -= 0x60 + (c2 < 0xe0);
 256                         }
 257                         else
 258                         {
 259                                 c1 = (c1 >> 1) + (c1 < 0xdf ? 0x30 : 0x70);
 260                                 c2 -= 2;
 261                         }
 262
 263                         tmp[i - 1] = c1;
 264                         tmp[i] = c2;
 265                 }
 266                 else
 267                         tmp[i] = c1;
 268         }
 269         tmp[len] = 0;
 270         strcpy(str, (char *)tmp);
 271
 272         C_KILL(tmp, len+1, byte);
 273 }
 274
 275
 276 /*!
 277  * @brief strを環境に合った文字コードに変換し、変換前の文字コードを返す。strの長さに制限はない。
 278  * @param str 変換する文字列のポインタ
 279  * @return
 280  * 0: Unknown<br>
 281  * 1: ASCII (Never known to be ASCII in this function.)<br>
 282  * 2: EUC<br>
 283  * 3: SJIS<br>
 284  */
 285 byte codeconv(char *str)
 286 {
 287         byte code = 0;
 288         int i;
 289
 290         for (i = 0; str[i]; i++)
 291         {
 292                 unsigned char c1;
 293                 unsigned char c2;
 294
 295                 /* First byte */
 296                 c1 = str[i];
 297
 298                 /* ASCII? */
 299                 if (!(c1 & 0x80)) continue;
 300
 301                 /* Second byte */
 302                 i++;
 303                 c2 = str[i];
 304
 305                 if (((0xa1 <= c1 && c1 <= 0xdf) || (0xfd <= c1 && c1 <= 0xfe)) &&
 306                     (0xa1 <= c2 && c2 <= 0xfe))
 307                 {
 308                         /* Only EUC is allowed */
 309                         if (!code)
 310                         {
 311                                 /* EUC */
 312                                 code = 2;
 313                         }
 314
 315                         /* Broken string? */
 316                         else if (code != 2)
 317                         {
 318                                 /* No conversion */
 319                                 return 0;
 320                         }
 321                 }
 322
 323                 else if (((0x81 <= c1 && c1 <= 0x9f) &&
 324                           ((0x40 <= c2 && c2 <= 0x7e) || (0x80 <= c2 && c2 <= 0xfc))) ||
 325                          ((0xe0 <= c1 && c1 <= 0xfc) &&
 326                           (0x40 <= c2 && c2 <= 0x7e)))
 327                 {
 328                         /* Only SJIS is allowed */
 329                         if (!code)
 330                         {
 331                                 /* SJIS */
 332                                 code = 3;
 333                         }
 334
 335                         /* Broken string? */
 336                         else if (code != 3)
 337                         {
 338                                 /* No conversion */
 339                                 return 0;
 340                         }
 341                 }
 342         }
 343
 344
 345         switch (code)
 346         {
 347 #ifdef EUC
 348         case 3:
 349                 /* SJIS -> EUC */
 350                 sjis2euc(str);
 351                 break;
 352 #endif
 353
 354 #ifdef SJIS
 355         case 2:
 356                 /* EUC -> SJIS */
 357                 euc2sjis(str);
 358
 359                 break;
 360 #endif
 361         }
 362
 363         /* Return kanji code */
 364         return code;
 365 }
 366
 367 /*!
 368  * @brief 文字列sのxバイト目が漢字の1バイト目かどうか判定する
 369  * @param s 判定する文字列のポインタ
 370  * @param x 判定する位置(バイト)
 371  * @return 漢字の1バイト目ならばTRUE
 372  */
 373 bool iskanji2(cptr s, int x)
 374 {
 375         int i;
 376
 377         for (i = 0; i < x; i++)
 378         {
 379                 if (iskanji(s[i])) i++;
 380         }
 381         if ((x == i) && iskanji(s[x])) return TRUE;
 382
 383         return FALSE;
 384 }
 385
 386 /*!
 387  * @brief 文字列の文字コードがASCIIかどうかを判定する
 388  * @param str 判定する文字列へのポインタ
 389  * @return 文字列の文字コードがASCIIならTRUE、そうでなければFALSE
 390  */
 391 static bool is_ascii_str(cptr str)
 392 {
 393         for (;*str; str++) {
 394                 if (!(0x00 < *str && *str <= 0x7f))
 395                         return FALSE;
 396         }
 397         return TRUE;
 398 }
 399
 400 /*!
 401  * @brief 文字列の文字コードがUTF-8かどうかを判定する
 402  * @param str 判定する文字列へのポインタ
 403  * @return 文字列の文字コードがUTF-8ならTRUE、そうでなければFALSE
 404  */
 405 static bool is_utf8_str(cptr str)
 406 {
 407         const unsigned char* p;
 408         for (p = (const unsigned char*)str; *p; p++) {
 409                 int subseq_num = 0;
 410                 if (0x00 < *p && *p <= 0x7f) continue;
 411
 412                 if ((*p & 0xe0) == 0xc0) subseq_num = 1;
 413                 if ((*p & 0xf0) == 0xe0) subseq_num = 2;
 414                 if ((*p & 0xf8) == 0xf0) subseq_num = 3;
 415
 416                 if (subseq_num == 0) return FALSE;
 417                 while (subseq_num--) {
 418                         p++;
 419                         if (!*p || (*p & 0xc0) != 0x80) return FALSE;
 420                 }
 421         }
 422         return TRUE;
 423 }
 424
 425 #if defined(EUC)
 426 #include <iconv.h>
 427
 428 static const struct ms_to_jis_unicode_conv_t {
 429         char from[3];
 430         char to[3];
 431 } ms_to_jis_unicode_conv[] = {
 432         {{0xef, 0xbd, 0x9e}, {0xe3, 0x80, 0x9c}}, /* FULLWIDTH TILDE -> WAVE DASH */
 433         {{0xef, 0xbc, 0x8d}, {0xe2, 0x88, 0x92}}, /* FULLWIDTH HYPHEN-MINUS -> MINUS SIGN */
 434 };
 435
 436 /*!
 437  * @brief EUCがシステムコードである環境下向けにUTF-8から変換処理を行うサブルーチン
 438  * @param str 変換する文字列のポインタ
 439  * @return なし
 440  */
 441 static void ms_to_jis_unicode(char* str)
 442 {
 443         unsigned char* p;
 444         for (p = (unsigned char*)str; *p; p++) {
 445                 int subseq_num = 0;
 446                 if (0x00 < *p && *p <= 0x7f) continue;
 447
 448                 if ((*p & 0xe0) == 0xc0) subseq_num = 1;
 449                 if ((*p & 0xf0) == 0xe0) {
 450                         int i;
 451                         for (i = 0; i < sizeof(ms_to_jis_unicode_conv) / sizeof(ms_to_jis_unicode_conv[0]); ++ i) {
 452                                 const struct ms_to_jis_unicode_conv_t *c = &ms_to_jis_unicode_conv[i];
 453                                 if (memcmp(p, c->from, 3) == 0) {
 454                                         memcpy(p, c->to, 3);
 455                                 }
 456                         }
 457                         subseq_num = 2;
 458                 }
 459                 if ((*p & 0xf8) == 0xf0) subseq_num = 3;
 460
 461                 p += subseq_num;
 462         }
 463 }
 464
 465 #elif defined(SJIS) && defined(WINDOWS)
 466 #include <Windows.h>
 467 #endif
 468 /*!
 469  * @brief 文字コードがUTF-8の文字列をシステムの文字コードに変換する
 470  * @param utf8_str 変換するUTF-8の文字列へのポインタ
 471  * @param sys_str_buffer 変換したシステムの文字コードの文字列を格納するバッファへのポインタ
 472  * @param sys_str_buflen 変換したシステムの文字コードの文字列を格納するバッファの長さ
 473  * @return 変換に成功した場合TRUE、失敗した場合FALSEを返す
 474  */
 475 static bool utf8_to_sys(char* utf8_str, char* sys_str_buffer, size_t sys_str_buflen)
 476 {
 477 #if defined(EUC)
 478
 479         iconv_t cd = iconv_open("EUC-JP", "UTF-8");
 480         size_t utf8_len = strlen(utf8_str) + 1; /* include termination character */
 481         char *from = utf8_str;
 482         int ret;
 483
 484         ms_to_jis_unicode(utf8_str);
 485         ret = iconv(cd, &from, &utf8_len, &sys_str_buffer, &sys_str_buflen);
 486         iconv_close(cd);
 487         return (ret >= 0);
 488
 489 #elif defined(SJIS) && defined(WINDOWS)
 490
 491         LPWSTR utf16buf;
 492         int input_len = strlen(utf8_str) + 1; /* include termination character */
 493
 494         C_MAKE(utf16buf, input_len, WCHAR);
 495
 496         /* UTF-8 -> UTF-16 */
 497         if (MultiByteToWideChar( CP_UTF8, 0, utf8_str, input_len, utf16buf, input_len) == 0) {
 498                 C_KILL(utf16buf, input_len, WCHAR);
 499                 return FALSE;
 500         }
 501
 502         /* UTF-8 -> SJIS(CP932) */
 503         if (WideCharToMultiByte( CP_ACP, 0, utf16buf, -1, sys_str_buffer, sys_str_buflen, NULL, NULL ) == 0) {
 504                 C_KILL(utf16buf, input_len, WCHAR);
 505                 return FALSE;
 506         }
 507
 508         C_KILL(utf16buf, input_len, WCHAR);
 509         return TRUE;
 510
 511 #endif
 512 }
 513
 514 /*!
 515  * @brief 受け取った文字列の文字コードを推定し、システムの文字コードへ変換する
 516  * @param strbuf 変換する文字列を格納したバッファへのポインタ。
 517  *               バッファは変換した文字列で上書きされる。
 518  *               UTF-8からSJISもしくはEUCへの変換を想定しているのでバッファの長さが足りなくなることはない。
 519  * @param buflen バッファの長さ。
 520  * @return なし
 521  */
 522 void guess_convert_to_system_encoding(char* strbuf, int buflen)
 523 {
 524         if (is_ascii_str(strbuf)) return;
 525
 526         if (is_utf8_str(strbuf)) {
 527                 char* work;
 528                 C_MAKE(work, buflen, char);
 529                 my_strcpy(work, strbuf, buflen);
 530                 if (!utf8_to_sys(work, strbuf, buflen)) {
 531                         msg_print("警告:文字コードの変換に失敗しました");
 532                         msg_print(NULL);
 533                 }
 534                 C_KILL(work, buflen, char);
 535         }
 536 }
 537
 538 #endif /* JP */