Line data Source code
1 : /*-------------------------------------------------------------------------
2 : *
3 : * encnames.c
4 : * Encoding names and routines for working with them.
5 : *
6 : * Portions Copyright (c) 2001-2025, PostgreSQL Global Development Group
7 : *
8 : * IDENTIFICATION
9 : * src/common/encnames.c
10 : *
11 : *-------------------------------------------------------------------------
12 : */
13 : #include "c.h"
14 :
15 : #include <ctype.h>
16 : #include <unistd.h>
17 :
18 : #include "mb/pg_wchar.h"
19 :
20 :
21 : /* ----------
22 : * All encoding names, sorted: *** A L P H A B E T I C ***
23 : *
24 : * All names must be without irrelevant chars, search routines use
25 : * isalnum() chars only. It means ISO-8859-1, iso_8859-1 and Iso8859_1
26 : * are always converted to 'iso88591'. All must be lower case.
27 : *
28 : * The table doesn't contain 'cs' aliases (like csISOLatin1). It's needed?
29 : *
30 : * Karel Zak, Aug 2001
31 : * ----------
32 : */
33 : typedef struct pg_encname
34 : {
35 : const char *name;
36 : pg_enc encoding;
37 : } pg_encname;
38 :
39 : static const pg_encname pg_encname_tbl[] =
40 : {
41 : {
42 : "abc", PG_WIN1258
43 : }, /* alias for WIN1258 */
44 : {
45 : "alt", PG_WIN866
46 : }, /* IBM866 */
47 : {
48 : "big5", PG_BIG5
49 : }, /* Big5; Chinese for Taiwan multibyte set */
50 : {
51 : "euccn", PG_EUC_CN
52 : }, /* EUC-CN; Extended Unix Code for simplified
53 : * Chinese */
54 : {
55 : "eucjis2004", PG_EUC_JIS_2004
56 : }, /* EUC-JIS-2004; Extended UNIX Code fixed
57 : * Width for Japanese, standard JIS X 0213 */
58 : {
59 : "eucjp", PG_EUC_JP
60 : }, /* EUC-JP; Extended UNIX Code fixed Width for
61 : * Japanese, standard OSF */
62 : {
63 : "euckr", PG_EUC_KR
64 : }, /* EUC-KR; Extended Unix Code for Korean , KS
65 : * X 1001 standard */
66 : {
67 : "euctw", PG_EUC_TW
68 : }, /* EUC-TW; Extended Unix Code for
69 : *
70 : * traditional Chinese */
71 : {
72 : "gb18030", PG_GB18030
73 : }, /* GB18030;GB18030 */
74 : {
75 : "gbk", PG_GBK
76 : }, /* GBK; Chinese Windows CodePage 936
77 : * simplified Chinese */
78 : {
79 : "iso88591", PG_LATIN1
80 : }, /* ISO-8859-1; RFC1345,KXS2 */
81 : {
82 : "iso885910", PG_LATIN6
83 : }, /* ISO-8859-10; RFC1345,KXS2 */
84 : {
85 : "iso885913", PG_LATIN7
86 : }, /* ISO-8859-13; RFC1345,KXS2 */
87 : {
88 : "iso885914", PG_LATIN8
89 : }, /* ISO-8859-14; RFC1345,KXS2 */
90 : {
91 : "iso885915", PG_LATIN9
92 : }, /* ISO-8859-15; RFC1345,KXS2 */
93 : {
94 : "iso885916", PG_LATIN10
95 : }, /* ISO-8859-16; RFC1345,KXS2 */
96 : {
97 : "iso88592", PG_LATIN2
98 : }, /* ISO-8859-2; RFC1345,KXS2 */
99 : {
100 : "iso88593", PG_LATIN3
101 : }, /* ISO-8859-3; RFC1345,KXS2 */
102 : {
103 : "iso88594", PG_LATIN4
104 : }, /* ISO-8859-4; RFC1345,KXS2 */
105 : {
106 : "iso88595", PG_ISO_8859_5
107 : }, /* ISO-8859-5; RFC1345,KXS2 */
108 : {
109 : "iso88596", PG_ISO_8859_6
110 : }, /* ISO-8859-6; RFC1345,KXS2 */
111 : {
112 : "iso88597", PG_ISO_8859_7
113 : }, /* ISO-8859-7; RFC1345,KXS2 */
114 : {
115 : "iso88598", PG_ISO_8859_8
116 : }, /* ISO-8859-8; RFC1345,KXS2 */
117 : {
118 : "iso88599", PG_LATIN5
119 : }, /* ISO-8859-9; RFC1345,KXS2 */
120 : {
121 : "johab", PG_JOHAB
122 : }, /* JOHAB; Extended Unix Code for simplified
123 : * Chinese */
124 : {
125 : "koi8", PG_KOI8R
126 : }, /* _dirty_ alias for KOI8-R (backward
127 : * compatibility) */
128 : {
129 : "koi8r", PG_KOI8R
130 : }, /* KOI8-R; RFC1489 */
131 : {
132 : "koi8u", PG_KOI8U
133 : }, /* KOI8-U; RFC2319 */
134 : {
135 : "latin1", PG_LATIN1
136 : }, /* alias for ISO-8859-1 */
137 : {
138 : "latin10", PG_LATIN10
139 : }, /* alias for ISO-8859-16 */
140 : {
141 : "latin2", PG_LATIN2
142 : }, /* alias for ISO-8859-2 */
143 : {
144 : "latin3", PG_LATIN3
145 : }, /* alias for ISO-8859-3 */
146 : {
147 : "latin4", PG_LATIN4
148 : }, /* alias for ISO-8859-4 */
149 : {
150 : "latin5", PG_LATIN5
151 : }, /* alias for ISO-8859-9 */
152 : {
153 : "latin6", PG_LATIN6
154 : }, /* alias for ISO-8859-10 */
155 : {
156 : "latin7", PG_LATIN7
157 : }, /* alias for ISO-8859-13 */
158 : {
159 : "latin8", PG_LATIN8
160 : }, /* alias for ISO-8859-14 */
161 : {
162 : "latin9", PG_LATIN9
163 : }, /* alias for ISO-8859-15 */
164 : {
165 : "mskanji", PG_SJIS
166 : }, /* alias for Shift_JIS */
167 : {
168 : "muleinternal", PG_MULE_INTERNAL
169 : },
170 : {
171 : "shiftjis", PG_SJIS
172 : }, /* Shift_JIS; JIS X 0202-1991 */
173 :
174 : {
175 : "shiftjis2004", PG_SHIFT_JIS_2004
176 : }, /* SHIFT-JIS-2004; Shift JIS for Japanese,
177 : * standard JIS X 0213 */
178 : {
179 : "sjis", PG_SJIS
180 : }, /* alias for Shift_JIS */
181 : {
182 : "sqlascii", PG_SQL_ASCII
183 : },
184 : {
185 : "tcvn", PG_WIN1258
186 : }, /* alias for WIN1258 */
187 : {
188 : "tcvn5712", PG_WIN1258
189 : }, /* alias for WIN1258 */
190 : {
191 : "uhc", PG_UHC
192 : }, /* UHC; Korean Windows CodePage 949 */
193 : {
194 : "unicode", PG_UTF8
195 : }, /* alias for UTF8 */
196 : {
197 : "utf8", PG_UTF8
198 : }, /* alias for UTF8 */
199 : {
200 : "vscii", PG_WIN1258
201 : }, /* alias for WIN1258 */
202 : {
203 : "win", PG_WIN1251
204 : }, /* _dirty_ alias for windows-1251 (backward
205 : * compatibility) */
206 : {
207 : "win1250", PG_WIN1250
208 : }, /* alias for Windows-1250 */
209 : {
210 : "win1251", PG_WIN1251
211 : }, /* alias for Windows-1251 */
212 : {
213 : "win1252", PG_WIN1252
214 : }, /* alias for Windows-1252 */
215 : {
216 : "win1253", PG_WIN1253
217 : }, /* alias for Windows-1253 */
218 : {
219 : "win1254", PG_WIN1254
220 : }, /* alias for Windows-1254 */
221 : {
222 : "win1255", PG_WIN1255
223 : }, /* alias for Windows-1255 */
224 : {
225 : "win1256", PG_WIN1256
226 : }, /* alias for Windows-1256 */
227 : {
228 : "win1257", PG_WIN1257
229 : }, /* alias for Windows-1257 */
230 : {
231 : "win1258", PG_WIN1258
232 : }, /* alias for Windows-1258 */
233 : {
234 : "win866", PG_WIN866
235 : }, /* IBM866 */
236 : {
237 : "win874", PG_WIN874
238 : }, /* alias for Windows-874 */
239 : {
240 : "win932", PG_SJIS
241 : }, /* alias for Shift_JIS */
242 : {
243 : "win936", PG_GBK
244 : }, /* alias for GBK */
245 : {
246 : "win949", PG_UHC
247 : }, /* alias for UHC */
248 : {
249 : "win950", PG_BIG5
250 : }, /* alias for BIG5 */
251 : {
252 : "windows1250", PG_WIN1250
253 : }, /* Windows-1251; Microsoft */
254 : {
255 : "windows1251", PG_WIN1251
256 : }, /* Windows-1251; Microsoft */
257 : {
258 : "windows1252", PG_WIN1252
259 : }, /* Windows-1252; Microsoft */
260 : {
261 : "windows1253", PG_WIN1253
262 : }, /* Windows-1253; Microsoft */
263 : {
264 : "windows1254", PG_WIN1254
265 : }, /* Windows-1254; Microsoft */
266 : {
267 : "windows1255", PG_WIN1255
268 : }, /* Windows-1255; Microsoft */
269 : {
270 : "windows1256", PG_WIN1256
271 : }, /* Windows-1256; Microsoft */
272 : {
273 : "windows1257", PG_WIN1257
274 : }, /* Windows-1257; Microsoft */
275 : {
276 : "windows1258", PG_WIN1258
277 : }, /* Windows-1258; Microsoft */
278 : {
279 : "windows866", PG_WIN866
280 : }, /* IBM866 */
281 : {
282 : "windows874", PG_WIN874
283 : }, /* Windows-874; Microsoft */
284 : {
285 : "windows932", PG_SJIS
286 : }, /* alias for Shift_JIS */
287 : {
288 : "windows936", PG_GBK
289 : }, /* alias for GBK */
290 : {
291 : "windows949", PG_UHC
292 : }, /* alias for UHC */
293 : {
294 : "windows950", PG_BIG5
295 : } /* alias for BIG5 */
296 : };
297 :
298 : /* ----------
299 : * These are "official" encoding names.
300 : * ----------
301 : */
302 : #ifndef WIN32
303 : #define DEF_ENC2NAME(name, codepage) { #name, PG_##name }
304 : #else
305 : #define DEF_ENC2NAME(name, codepage) { #name, PG_##name, codepage }
306 : #endif
307 :
308 : const pg_enc2name pg_enc2name_tbl[] =
309 : {
310 : [PG_SQL_ASCII] = DEF_ENC2NAME(SQL_ASCII, 0),
311 : [PG_EUC_JP] = DEF_ENC2NAME(EUC_JP, 20932),
312 : [PG_EUC_CN] = DEF_ENC2NAME(EUC_CN, 20936),
313 : [PG_EUC_KR] = DEF_ENC2NAME(EUC_KR, 51949),
314 : [PG_EUC_TW] = DEF_ENC2NAME(EUC_TW, 0),
315 : [PG_EUC_JIS_2004] = DEF_ENC2NAME(EUC_JIS_2004, 20932),
316 : [PG_UTF8] = DEF_ENC2NAME(UTF8, 65001),
317 : [PG_MULE_INTERNAL] = DEF_ENC2NAME(MULE_INTERNAL, 0),
318 : [PG_LATIN1] = DEF_ENC2NAME(LATIN1, 28591),
319 : [PG_LATIN2] = DEF_ENC2NAME(LATIN2, 28592),
320 : [PG_LATIN3] = DEF_ENC2NAME(LATIN3, 28593),
321 : [PG_LATIN4] = DEF_ENC2NAME(LATIN4, 28594),
322 : [PG_LATIN5] = DEF_ENC2NAME(LATIN5, 28599),
323 : [PG_LATIN6] = DEF_ENC2NAME(LATIN6, 0),
324 : [PG_LATIN7] = DEF_ENC2NAME(LATIN7, 0),
325 : [PG_LATIN8] = DEF_ENC2NAME(LATIN8, 0),
326 : [PG_LATIN9] = DEF_ENC2NAME(LATIN9, 28605),
327 : [PG_LATIN10] = DEF_ENC2NAME(LATIN10, 0),
328 : [PG_WIN1256] = DEF_ENC2NAME(WIN1256, 1256),
329 : [PG_WIN1258] = DEF_ENC2NAME(WIN1258, 1258),
330 : [PG_WIN866] = DEF_ENC2NAME(WIN866, 866),
331 : [PG_WIN874] = DEF_ENC2NAME(WIN874, 874),
332 : [PG_KOI8R] = DEF_ENC2NAME(KOI8R, 20866),
333 : [PG_WIN1251] = DEF_ENC2NAME(WIN1251, 1251),
334 : [PG_WIN1252] = DEF_ENC2NAME(WIN1252, 1252),
335 : [PG_ISO_8859_5] = DEF_ENC2NAME(ISO_8859_5, 28595),
336 : [PG_ISO_8859_6] = DEF_ENC2NAME(ISO_8859_6, 28596),
337 : [PG_ISO_8859_7] = DEF_ENC2NAME(ISO_8859_7, 28597),
338 : [PG_ISO_8859_8] = DEF_ENC2NAME(ISO_8859_8, 28598),
339 : [PG_WIN1250] = DEF_ENC2NAME(WIN1250, 1250),
340 : [PG_WIN1253] = DEF_ENC2NAME(WIN1253, 1253),
341 : [PG_WIN1254] = DEF_ENC2NAME(WIN1254, 1254),
342 : [PG_WIN1255] = DEF_ENC2NAME(WIN1255, 1255),
343 : [PG_WIN1257] = DEF_ENC2NAME(WIN1257, 1257),
344 : [PG_KOI8U] = DEF_ENC2NAME(KOI8U, 21866),
345 : [PG_SJIS] = DEF_ENC2NAME(SJIS, 932),
346 : [PG_BIG5] = DEF_ENC2NAME(BIG5, 950),
347 : [PG_GBK] = DEF_ENC2NAME(GBK, 936),
348 : [PG_UHC] = DEF_ENC2NAME(UHC, 949),
349 : [PG_GB18030] = DEF_ENC2NAME(GB18030, 54936),
350 : [PG_JOHAB] = DEF_ENC2NAME(JOHAB, 0),
351 : [PG_SHIFT_JIS_2004] = DEF_ENC2NAME(SHIFT_JIS_2004, 932),
352 : };
353 :
354 : /* ----------
355 : * These are encoding names for gettext.
356 : *
357 : * This covers all encodings except MULE_INTERNAL, which is alien to gettext.
358 : * ----------
359 : */
360 : const char *pg_enc2gettext_tbl[] =
361 : {
362 : [PG_SQL_ASCII] = "US-ASCII",
363 : [PG_UTF8] = "UTF-8",
364 : [PG_MULE_INTERNAL] = NULL,
365 : [PG_LATIN1] = "LATIN1",
366 : [PG_LATIN2] = "LATIN2",
367 : [PG_LATIN3] = "LATIN3",
368 : [PG_LATIN4] = "LATIN4",
369 : [PG_ISO_8859_5] = "ISO-8859-5",
370 : [PG_ISO_8859_6] = "ISO_8859-6",
371 : [PG_ISO_8859_7] = "ISO-8859-7",
372 : [PG_ISO_8859_8] = "ISO-8859-8",
373 : [PG_LATIN5] = "LATIN5",
374 : [PG_LATIN6] = "LATIN6",
375 : [PG_LATIN7] = "LATIN7",
376 : [PG_LATIN8] = "LATIN8",
377 : [PG_LATIN9] = "LATIN-9",
378 : [PG_LATIN10] = "LATIN10",
379 : [PG_KOI8R] = "KOI8-R",
380 : [PG_KOI8U] = "KOI8-U",
381 : [PG_WIN1250] = "CP1250",
382 : [PG_WIN1251] = "CP1251",
383 : [PG_WIN1252] = "CP1252",
384 : [PG_WIN1253] = "CP1253",
385 : [PG_WIN1254] = "CP1254",
386 : [PG_WIN1255] = "CP1255",
387 : [PG_WIN1256] = "CP1256",
388 : [PG_WIN1257] = "CP1257",
389 : [PG_WIN1258] = "CP1258",
390 : [PG_WIN866] = "CP866",
391 : [PG_WIN874] = "CP874",
392 : [PG_EUC_CN] = "EUC-CN",
393 : [PG_EUC_JP] = "EUC-JP",
394 : [PG_EUC_KR] = "EUC-KR",
395 : [PG_EUC_TW] = "EUC-TW",
396 : [PG_EUC_JIS_2004] = "EUC-JP",
397 : [PG_SJIS] = "SHIFT-JIS",
398 : [PG_BIG5] = "BIG5",
399 : [PG_GBK] = "GBK",
400 : [PG_UHC] = "UHC",
401 : [PG_GB18030] = "GB18030",
402 : [PG_JOHAB] = "JOHAB",
403 : [PG_SHIFT_JIS_2004] = "SHIFT_JISX0213",
404 : };
405 :
406 :
407 : /*
408 : * Table of encoding names for ICU (currently covers backend encodings only)
409 : *
410 : * Reference: <https://ssl.icu-project.org/icu-bin/convexp>
411 : *
412 : * NULL entries are not supported by ICU, or their mapping is unclear.
413 : */
414 : static const char *const pg_enc2icu_tbl[] =
415 : {
416 : [PG_SQL_ASCII] = NULL,
417 : [PG_EUC_JP] = "EUC-JP",
418 : [PG_EUC_CN] = "EUC-CN",
419 : [PG_EUC_KR] = "EUC-KR",
420 : [PG_EUC_TW] = "EUC-TW",
421 : [PG_EUC_JIS_2004] = NULL,
422 : [PG_UTF8] = "UTF-8",
423 : [PG_MULE_INTERNAL] = NULL,
424 : [PG_LATIN1] = "ISO-8859-1",
425 : [PG_LATIN2] = "ISO-8859-2",
426 : [PG_LATIN3] = "ISO-8859-3",
427 : [PG_LATIN4] = "ISO-8859-4",
428 : [PG_LATIN5] = "ISO-8859-9",
429 : [PG_LATIN6] = "ISO-8859-10",
430 : [PG_LATIN7] = "ISO-8859-13",
431 : [PG_LATIN8] = "ISO-8859-14",
432 : [PG_LATIN9] = "ISO-8859-15",
433 : [PG_LATIN10] = NULL,
434 : [PG_WIN1256] = "CP1256",
435 : [PG_WIN1258] = "CP1258",
436 : [PG_WIN866] = "CP866",
437 : [PG_WIN874] = NULL,
438 : [PG_KOI8R] = "KOI8-R",
439 : [PG_WIN1251] = "CP1251",
440 : [PG_WIN1252] = "CP1252",
441 : [PG_ISO_8859_5] = "ISO-8859-5",
442 : [PG_ISO_8859_6] = "ISO-8859-6",
443 : [PG_ISO_8859_7] = "ISO-8859-7",
444 : [PG_ISO_8859_8] = "ISO-8859-8",
445 : [PG_WIN1250] = "CP1250",
446 : [PG_WIN1253] = "CP1253",
447 : [PG_WIN1254] = "CP1254",
448 : [PG_WIN1255] = "CP1255",
449 : [PG_WIN1257] = "CP1257",
450 : [PG_KOI8U] = "KOI8-U",
451 : };
452 :
453 : StaticAssertDecl(lengthof(pg_enc2icu_tbl) == PG_ENCODING_BE_LAST + 1,
454 : "pg_enc2icu_tbl incomplete");
455 :
456 :
457 : /*
458 : * Is this encoding supported by ICU?
459 : */
460 : bool
461 1520 : is_encoding_supported_by_icu(int encoding)
462 : {
463 1520 : if (!PG_VALID_BE_ENCODING(encoding))
464 0 : return false;
465 1520 : return (pg_enc2icu_tbl[encoding] != NULL);
466 : }
467 :
468 : /*
469 : * Returns ICU's name for encoding, or NULL if not supported
470 : */
471 : const char *
472 6 : get_encoding_name_for_icu(int encoding)
473 : {
474 6 : if (!PG_VALID_BE_ENCODING(encoding))
475 0 : return NULL;
476 6 : return pg_enc2icu_tbl[encoding];
477 : }
478 :
479 :
480 : /* ----------
481 : * Encoding checks, for error returns -1 else encoding id
482 : * ----------
483 : */
484 : int
485 43222 : pg_valid_client_encoding(const char *name)
486 : {
487 : int enc;
488 :
489 43222 : if ((enc = pg_char_to_encoding(name)) < 0)
490 0 : return -1;
491 :
492 43222 : if (!PG_VALID_FE_ENCODING(enc))
493 0 : return -1;
494 :
495 43222 : return enc;
496 : }
497 :
498 : int
499 132 : pg_valid_server_encoding(const char *name)
500 : {
501 : int enc;
502 :
503 132 : if ((enc = pg_char_to_encoding(name)) < 0)
504 0 : return -1;
505 :
506 132 : if (!PG_VALID_BE_ENCODING(enc))
507 0 : return -1;
508 :
509 132 : return enc;
510 : }
511 :
512 : int
513 658642 : pg_valid_server_encoding_id(int encoding)
514 : {
515 658642 : return PG_VALID_BE_ENCODING(encoding);
516 : }
517 :
518 : /*
519 : * Remove irrelevant chars from encoding name, store at *newkey
520 : *
521 : * (Caller's responsibility to provide a large enough buffer)
522 : */
523 : static char *
524 108918 : clean_encoding_name(const char *key, char *newkey)
525 : {
526 : const char *p;
527 : char *np;
528 :
529 647790 : for (p = key, np = newkey; *p != '\0'; p++)
530 : {
531 538872 : if (isalnum((unsigned char) *p))
532 : {
533 522948 : if (*p >= 'A' && *p <= 'Z')
534 273864 : *np++ = *p + 'a' - 'A';
535 : else
536 249084 : *np++ = *p;
537 : }
538 : }
539 108918 : *np = '\0';
540 108918 : return newkey;
541 : }
542 :
543 : /*
544 : * Search encoding by encoding name
545 : *
546 : * Returns encoding ID, or -1 if not recognized
547 : */
548 : int
549 108918 : pg_char_to_encoding(const char *name)
550 : {
551 108918 : unsigned int nel = lengthof(pg_encname_tbl);
552 108918 : const pg_encname *base = pg_encname_tbl,
553 108918 : *last = base + nel - 1,
554 : *position;
555 : int result;
556 : char buff[NAMEDATALEN],
557 : *key;
558 :
559 108918 : if (name == NULL || *name == '\0')
560 0 : return -1;
561 :
562 108918 : if (strlen(name) >= NAMEDATALEN)
563 0 : return -1; /* it's certainly not in the table */
564 :
565 108918 : key = clean_encoding_name(name, buff);
566 :
567 650090 : while (last >= base)
568 : {
569 650088 : position = base + ((last - base) >> 1);
570 650088 : result = key[0] - position->name[0];
571 :
572 650088 : if (result == 0)
573 : {
574 226604 : result = strcmp(key, position->name);
575 226604 : if (result == 0)
576 108916 : return position->encoding;
577 : }
578 541172 : if (result < 0)
579 230778 : last = position - 1;
580 : else
581 310394 : base = position + 1;
582 : }
583 2 : return -1;
584 : }
585 :
586 : const char *
587 69768 : pg_encoding_to_char(int encoding)
588 : {
589 69768 : if (PG_VALID_ENCODING(encoding))
590 : {
591 69768 : const pg_enc2name *p = &pg_enc2name_tbl[encoding];
592 :
593 : Assert(encoding == p->encoding);
594 69768 : return p->name;
595 : }
596 0 : return "";
597 : }
|
