conv.c 36.3 KB
Newer Older
B
Bruce Momjian 已提交
1 2 3 4
/*
 * conversion between client encoding and server internal encoding
 * (currently mule internal code (mic) is used)
 * Tatsuo Ishii
T
Tatsuo Ishii 已提交
5
 *
6
 * WIN1250 client encoding support contributed by Pavel Behal
T
Tatsuo Ishii 已提交
7
 * SJIS UDC (NEC selection IBM kanji) support contributed by Eiji Tokuya
8
 *
9
 * $Id: conv.c,v 1.25 2001/04/29 07:27:38 ishii Exp $
10 11
 *
 *
B
Bruce Momjian 已提交
12
 */
13
#include "postgres.h"
M
 
Marc G. Fournier 已提交
14
#include "mb/pg_wchar.h"
B
Bruce Momjian 已提交
15

16
#ifdef UNICODE_CONVERSION
T
Tatsuo Ishii 已提交
17

18 19 20
/*
 * for Unicode (UTF-8) support
 */
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
#include "Unicode/utf8_to_iso8859_2.map"
#include "Unicode/utf8_to_iso8859_3.map"
#include "Unicode/utf8_to_iso8859_4.map"
#include "Unicode/utf8_to_iso8859_5.map"
#include "Unicode/iso8859_2_to_utf8.map"
#include "Unicode/iso8859_3_to_utf8.map"
#include "Unicode/iso8859_4_to_utf8.map"
#include "Unicode/iso8859_5_to_utf8.map"
#include "Unicode/utf8_to_euc_jp.map"
#include "Unicode/euc_jp_to_utf8.map"
#include "Unicode/utf8_to_euc_cn.map"
#include "Unicode/euc_cn_to_utf8.map"
#include "Unicode/utf8_to_euc_kr.map"
#include "Unicode/euc_kr_to_utf8.map"
#include "Unicode/utf8_to_euc_tw.map"
#include "Unicode/euc_tw_to_utf8.map"
#include "Unicode/utf8_to_sjis.map"
#include "Unicode/sjis_to_utf8.map"
#include "Unicode/utf8_to_big5.map"
#include "Unicode/big5_to_utf8.map"
41 42 43 44 45 46 47 48
/* Cyrillic charset conversion */
#include "Unicode/ALT_to_utf8.map"
#include "Unicode/KOI8_to_utf8.map"
#include "Unicode/WIN_to_utf8.map"
#include "Unicode/utf8_to_ALT.map"
#include "Unicode/utf8_to_KOI8.map"
#include "Unicode/utf8_to_WIN.map"

49

B
Bruce Momjian 已提交
50
#endif	 /* UNICODE_CONVERSION */
51

52 53 54 55 56 57 58 59 60 61
/*
 * SJIS alternative code.
 * this code is used if a mapping EUC -> SJIS is not defined.
 */
#define PGSJISALTCODE 0x81ac
#define PGEUCALTCODE 0xa2ae

/*
 * conversion table between SJIS UDC (IBM kanji) and EUC_JP
 */
62 63
#include "sjis.map"

B
Bruce Momjian 已提交
64
/*
65 66
 * convert bogus chars that cannot be represented in the current
encoding
B
Bruce Momjian 已提交
67 68
 * system.
 */
69 70
static void
printBogusChar(unsigned char **mic, unsigned char **p)
B
Bruce Momjian 已提交
71
{
72 73 74 75 76 77 78 79 80 81 82
	char		strbuf[16];
	int			l = pg_mic_mblen(*mic);

	*(*p)++ = '(';
	while (l--)
	{
		sprintf(strbuf, "%02x", *(*mic)++);
		*(*p)++ = strbuf[0];
		*(*p)++ = strbuf[1];
	}
	*(*p)++ = ')';
B
Bruce Momjian 已提交
83 84 85 86 87
}

/*
 * SJIS ---> MIC
 */
88 89
static void
sjis2mic(unsigned char *sjis, unsigned char *p, int len)
B
Bruce Momjian 已提交
90
{
91
	int			c1,
92
				c2,
T
Tatsuo Ishii 已提交
93 94 95 96
/* Eiji Tokuya patched begin */
				i,
				k,
				k2;
B
Bruce Momjian 已提交
97

T
Tatsuo Ishii 已提交
98
/* Eiji Tokuya patched end */
99 100 101
	while (len > 0 && (c1 = *sjis++))
	{
		if (c1 >= 0xa1 && c1 <= 0xdf)
102 103
		{
			/* JIS X0201 (1 byte kana) */
104 105 106 107 108
			len--;
			*p++ = LC_JISX0201K;
			*p++ = c1;
		}
		else if (c1 > 0x7f)
109
		{
110 111 112 113

			/*
			 * JIS X0208, X0212, user defined extended characters
			 */
114
			c2 = *sjis++;
115
			k = (c1 << 8) + c2;
T
Tatsuo Ishii 已提交
116 117 118 119 120 121 122 123 124 125 126 127 128
/* Eiji Tokuya patched begin */
			if (k >= 0xed40 && k < 0xf040)
			{
				/* NEC selection IBM kanji */
				for (i = 0;; i++)
				{
					k2 = ibmkanji[i].nec;
					if (k2 == 0xffff)
						break;
					if (k2 == k)
					{
						k = ibmkanji[i].sjis;
						c1 = (k >> 8) & 0xff;
B
Bruce Momjian 已提交
129
						c2 = k & 0xff;
T
Tatsuo Ishii 已提交
130 131 132
					}
				}
			}
133

T
Tatsuo Ishii 已提交
134 135
			if (k < 0xeb3f)
/* Eiji Tokuya patched end */
136 137 138 139 140 141 142
			{
				/* JIS X0208 */
				len -= 2;
				*p++ = LC_JISX0208;
				*p++ = ((c1 & 0x3f) << 1) + 0x9f + (c2 > 0x9e);
				*p++ = c2 + ((c2 > 0x9e) ? 2 : 0x60) + (c2 < 0x80);
			}
T
Tatsuo Ishii 已提交
143
/* Eiji Tokuya patched begin */
B
Bruce Momjian 已提交
144
			else if ((k >= 0xeb40 && k < 0xf040) || (k >= 0xfc4c && k <= 0xfcfc))
145
			{
T
Tatsuo Ishii 已提交
146 147
				/* NEC selection IBM kanji - Other undecided justice */
/* Eiji Tokuya patched end */
148 149
				*p++ = LC_JISX0208;
				*p++ = PGEUCALTCODE >> 8;
150
				*p++ = PGEUCALTCODE & 0xff;
151 152 153
			}
			else if (k >= 0xf040 && k < 0xf540)
			{
154 155 156 157 158

				/*
				 * UDC1 mapping to X0208 85 ku - 94 ku JIS code 0x7521 -
				 * 0x7e7e EUC 0xf5a1 - 0xfefe
				 */
159 160 161 162 163 164 165 166
				len -= 2;
				*p++ = LC_JISX0208;
				c1 -= 0x6f;
				*p++ = ((c1 & 0x3f) << 1) + 0xf3 + (c2 > 0x9e);
				*p++ = c2 + ((c2 > 0x9e) ? 2 : 0x60) + (c2 < 0x80);
			}
			else if (k >= 0xf540 && k < 0xfa40)
			{
167 168 169 170 171

				/*
				 * UDC2 mapping to X0212 85 ku - 94 ku JIS code 0x7521 -
				 * 0x7e7e EUC 0x8ff5a1 - 0x8ffefe
				 */
172 173 174 175 176 177 178 179
				len -= 2;
				*p++ = LC_JISX0212;
				c1 -= 0x74;
				*p++ = ((c1 & 0x3f) << 1) + 0xf3 + (c2 > 0x9e);
				*p++ = c2 + ((c2 > 0x9e) ? 2 : 0x60) + (c2 < 0x80);
			}
			else if (k >= 0xfa40)
			{
180 181 182

				/*
				 * mapping IBM kanji to X0208 and X0212
183 184 185
				 *
				 */
				len -= 2;
186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
				for (i = 0;; i++)
				{
					k2 = ibmkanji[i].sjis;
					if (k2 == 0xffff)
						break;
					if (k2 == k)
					{
						k = ibmkanji[i].euc;
						if (k >= 0x8f0000)
						{
							*p++ = LC_JISX0212;
							*p++ = 0x80 | ((k & 0xff00) >> 8);
							*p++ = 0x80 | (k & 0xff);
						}
						else
						{
							*p++ = LC_JISX0208;
							*p++ = 0x80 | (k >> 8);
							*p++ = 0x80 | (k & 0xff);
						}
					}
207 208
				}
			}
209 210 211 212 213 214 215 216
		}
		else
		{						/* should be ASCII */
			len--;
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
217 218 219 220 221
}

/*
 * MIC ---> SJIS
 */
222 223
static void
mic2sjis(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
224
{
225
	int			c1,
226 227
				c2,
				k;
228 229 230 231 232 233 234 235 236 237 238

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

		if (c1 == LC_JISX0201K)
			*p++ = *mic++;
		else if (c1 == LC_JISX0208)
		{
			c1 = *mic++;
			c2 = *mic++;
239 240 241 242 243 244 245 246 247
			k = (c1 << 8) | (c2 & 0xff);
			if (k >= 0xf5a1)
			{
				/* UDC1 */
				c1 -= 0x54;
				*p++ = ((c1 - 0xa1) >> 1) + ((c1 < 0xdf) ? 0x81 : 0xc1) + 0x6f;
			}
			else
				*p++ = ((c1 - 0xa1) >> 1) + ((c1 < 0xdf) ? 0x81 : 0xc1);
248 249
			*p++ = c2 - ((c1 & 1) ? ((c2 < 0xe0) ? 0x61 : 0x60) : 2);
		}
250 251
		else if (c1 == LC_JISX0212)
		{
252 253 254
			int			i,
						k2;

255 256 257 258 259 260 261 262 263 264 265 266
			c1 = *mic++;
			c2 = *mic++;
			k = c1 << 8 | c2;
			if (k >= 0xf5a1)
			{
				/* UDC2 */
				c1 -= 0x54;
				*p++ = ((c1 - 0xa1) >> 1) + ((c1 < 0xdf) ? 0x81 : 0xc1) + 0x74;
				*p++ = c2 - ((c1 & 1) ? ((c2 < 0xe0) ? 0x61 : 0x60) : 2);
			}
			else
			{
267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
				/* IBM kanji */
				for (i = 0;; i++)
				{
					k2 = ibmkanji[i].euc & 0xffff;
					if (k2 == 0xffff)
					{
						*p++ = PGSJISALTCODE >> 8;
						*p++ = PGSJISALTCODE & 0xff;
						break;
					}
					if (k2 == k)
					{
						k = ibmkanji[i].sjis;
						*p++ = k >> 8;
						*p++ = k & 0xff;
						break;
					}
				}
285 286
			}
		}
287
		else if (c1 > 0x7f)
288 289 290 291
		{
			/* cannot convert to SJIS! */
			*p++ = PGSJISALTCODE >> 8;
			*p++ = PGSJISALTCODE & 0xff;
292 293 294 295 296 297 298
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
299 300 301 302 303
}

/*
 * EUC_JP ---> MIC
 */
304 305
static void
euc_jp2mic(unsigned char *euc, unsigned char *p, int len)
B
Bruce Momjian 已提交
306
{
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337
	int			c1;

	while (len > 0 && (c1 = *euc++))
	{
		if (c1 == SS2)
		{						/* 1 byte kana? */
			len -= 2;
			*p++ = LC_JISX0201K;
			*p++ = *euc++;
		}
		else if (c1 == SS3)
		{						/* JIS X0212 kanji? */
			len -= 3;
			*p++ = LC_JISX0212;
			*p++ = *euc++;
			*p++ = *euc++;
		}
		else if (c1 & 0x80)
		{						/* kanji? */
			len -= 2;
			*p++ = LC_JISX0208;
			*p++ = c1;
			*p++ = *euc++;
		}
		else
		{						/* should be ASCII */
			len--;
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
338 339 340 341 342
}

/*
 * MIC ---> EUC_JP
 */
343 344
static void
mic2euc_jp(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
345
{
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
	int			c1;

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

		if (c1 == LC_JISX0201K)
		{
			*p++ = SS2;
			*p++ = *mic++;
		}
		else if (c1 == LC_JISX0212)
		{
			*p++ = SS3;
			*p++ = *mic++;
			*p++ = *mic++;
		}
		else if (c1 == LC_JISX0208)
		{
			*p++ = *mic++;
			*p++ = *mic++;
		}
		else if (c1 > 0x7f)
		{						/* cannot convert to EUC_JP! */
			mic--;
			printBogusChar(&mic, &p);
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
379 380 381 382 383
}

/*
 * EUC_KR ---> MIC
 */
384 385
static void
euc_kr2mic(unsigned char *euc, unsigned char *p, int len)
B
Bruce Momjian 已提交
386
{
387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
	int			c1;

	while (len > 0 && (c1 = *euc++))
	{
		if (c1 & 0x80)
		{
			len -= 2;
			*p++ = LC_KS5601;
			*p++ = c1;
			*p++ = *euc++;
		}
		else
		{						/* should be ASCII */
			len--;
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
405 406 407 408 409
}

/*
 * MIC ---> EUC_KR
 */
410 411
static void
mic2euc_kr(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
412
{
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
	int			c1;

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

		if (c1 == LC_KS5601)
		{
			*p++ = *mic++;
			*p++ = *mic++;
		}
		else if (c1 > 0x7f)
		{						/* cannot convert to EUC_KR! */
			mic--;
			printBogusChar(&mic, &p);
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
435 436 437 438 439
}

/*
 * EUC_CN ---> MIC
 */
440 441
static void
euc_cn2mic(unsigned char *euc, unsigned char *p, int len)
B
Bruce Momjian 已提交
442
{
443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460
	int			c1;

	while (len > 0 && (c1 = *euc++))
	{
		if (c1 & 0x80)
		{
			len -= 2;
			*p++ = LC_GB2312_80;
			*p++ = c1;
			*p++ = *euc++;
		}
		else
		{						/* should be ASCII */
			len--;
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
461 462 463 464 465
}

/*
 * MIC ---> EUC_CN
 */
466 467
static void
mic2euc_cn(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
468
{
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
	int			c1;

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

		if (c1 == LC_GB2312_80)
		{
			*p++ = *mic++;
			*p++ = *mic++;
		}
		else if (c1 > 0x7f)
		{						/* cannot convert to EUC_CN! */
			mic--;
			printBogusChar(&mic, &p);
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
491 492 493 494 495
}

/*
 * EUC_TW ---> MIC
 */
496 497
static void
euc_tw2mic(unsigned char *euc, unsigned char *p, int len)
B
Bruce Momjian 已提交
498
{
499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532
	int			c1;

	while (len > 0 && (c1 = *euc++))
	{
		if (c1 == SS2)
		{
			len -= 4;
			c1 = *euc++;		/* plane No. */
			if (c1 == 0xa1)
				*p++ = LC_CNS11643_1;
			else if (c1 == 0xa2)
				*p++ = LC_CNS11643_2;
			else
			{
				*p++ = 0x9d;	/* LCPRV2 */
				*p++ = 0xa3 - c1 + LC_CNS11643_3;
			}
			*p++ = *euc++;
			*p++ = *euc++;
		}
		else if (c1 & 0x80)
		{						/* CNS11643-1 */
			len -= 2;
			*p++ = LC_CNS11643_1;
			*p++ = c1;
			*p++ = *euc++;
		}
		else
		{						/* should be ASCII */
			len--;
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
533 534 535 536 537
}

/*
 * MIC ---> EUC_TW
 */
538 539
static void
mic2euc_tw(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
540
{
541 542 543 544 545 546
	int			c1;

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

547
		if (c1 == LC_CNS11643_1)
548 549 550 551
		{
			*p++ = *mic++;
			*p++ = *mic++;
		}
552 553 554 555 556 557 558
		else if (c1 == LC_CNS11643_2)
		{
			*p++ = SS2;
			*p++ = 0xa2;
			*p++ = *mic++;
			*p++ = *mic++;
		}
559 560 561
		else if (c1 == 0x9d)
		{						/* LCPRV2? */
			*p++ = SS2;
562
			*p++ = *mic++ - LC_CNS11643_3 + 0xa3;
563 564 565 566 567 568 569 570 571 572 573 574 575 576
			*p++ = *mic++;
			*p++ = *mic++;
		}
		else if (c1 > 0x7f)
		{						/* cannot convert to EUC_TW! */
			mic--;
			printBogusChar(&mic, &p);
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
577 578
}

579 580 581 582 583 584
/*
 * Big5 ---> MIC
 */
static void
big52mic(unsigned char *big5, unsigned char *p, int len)
{
B
Bruce Momjian 已提交
585 586 587 588
	unsigned short c1;
	unsigned short big5buf,
				cnsBuf;
	unsigned char lc;
589
	char		bogusBuf[3];
B
Bruce Momjian 已提交
590 591 592 593
	int			i;

	while (len > 0 && (c1 = *big5++))
	{
594
		if (c1 <= 0x7fU)
B
Bruce Momjian 已提交
595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628
		{						/* ASCII */
			len--;
			*p++ = c1;
		}
		else
		{
			len -= 2;
			big5buf = c1 << 8;
			c1 = *big5++;
			big5buf |= c1;
			cnsBuf = BIG5toCNS(big5buf, &lc);
			if (lc != 0)
			{
				if (lc == LC_CNS11643_3 || lc == LC_CNS11643_4)
				{
					*p++ = 0x9d;/* LCPRV2 */
				}
				*p++ = lc;		/* Plane No. */
				*p++ = (cnsBuf >> 8) & 0x00ff;
				*p++ = cnsBuf & 0x00ff;
			}
			else
			{					/* cannot convert */
				big5 -= 2;
				*p++ = '(';
				for (i = 0; i < 2; i++)
				{
					sprintf(bogusBuf, "%02x", *big5++);
					*p++ = bogusBuf[0];
					*p++ = bogusBuf[1];
				}
				*p++ = ')';
			}
		}
629
	}
B
Bruce Momjian 已提交
630
	*p = '\0';
631 632 633 634 635 636 637 638
}

/*
 * MIC ---> Big5
 */
static void
mic2big5(unsigned char *mic, unsigned char *p, int len)
{
B
Bruce Momjian 已提交
639 640 641 642
	int			l;
	unsigned short c1;
	unsigned short big5buf,
				cnsBuf;
643

B
Bruce Momjian 已提交
644
	while (len > 0 && (c1 = *mic))
645
	{
B
Bruce Momjian 已提交
646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676
		l = pg_mic_mblen(mic++);
		len -= l;

		/* 0x9d means LCPRV2 */
		if (c1 == LC_CNS11643_1 || c1 == LC_CNS11643_2 || c1 == 0x9d)
		{
			if (c1 == 0x9d)
			{
				c1 = *mic++;	/* get plane no. */
			}
			cnsBuf = (*mic++) << 8;
			cnsBuf |= (*mic++) & 0x00ff;
			big5buf = CNStoBIG5(cnsBuf, c1);
			if (big5buf == 0)
			{					/* cannot convert to Big5! */
				mic -= l;
				printBogusChar(&mic, &p);
			}
			else
			{
				*p++ = (big5buf >> 8) & 0x00ff;
				*p++ = big5buf & 0x00ff;
			}
		}
		else if (c1 <= 0x7f)	/* ASCII */
			*p++ = c1;
		else
		{						/* cannot convert to Big5! */
			mic--;
			printBogusChar(&mic, &p);
		}
677
	}
B
Bruce Momjian 已提交
678
	*p = '\0';
679 680
}

B
Bruce Momjian 已提交
681
/*
M
 
Marc G. Fournier 已提交
682
 * LATINn ---> MIC
B
Bruce Momjian 已提交
683
 */
684 685
static void
latin2mic(unsigned char *l, unsigned char *p, int len, int lc)
B
Bruce Momjian 已提交
686
{
687 688 689 690 691 692 693 694 695 696 697
	int			c1;

	while (len-- > 0 && (c1 = *l++))
	{
		if (c1 > 0x7f)
		{						/* Latin1? */
			*p++ = lc;
		}
		*p++ = c1;
	}
	*p = '\0';
B
Bruce Momjian 已提交
698 699 700
}

/*
M
 
Marc G. Fournier 已提交
701
 * MIC ---> LATINn
B
Bruce Momjian 已提交
702
 */
703 704
static void
mic2latin(unsigned char *mic, unsigned char *p, int len, int lc)
B
Bruce Momjian 已提交
705
{
706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724
	int			c1;

	while (len > 0 && (c1 = *mic))
	{
		len -= pg_mic_mblen(mic++);

		if (c1 == lc)
			*p++ = *mic++;
		else if (c1 > 0x7f)
		{
			mic--;
			printBogusChar(&mic, &p);
		}
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
	}
	*p = '\0';
B
Bruce Momjian 已提交
725 726
}

727 728
static void
latin12mic(unsigned char *l, unsigned char *p, int len)
B
Bruce Momjian 已提交
729
{
730
	latin2mic(l, p, len, LC_ISO8859_1);
B
Bruce Momjian 已提交
731
}
732 733
static void
mic2latin1(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
734
{
735
	mic2latin(mic, p, len, LC_ISO8859_1);
B
Bruce Momjian 已提交
736
}
737 738
static void
latin22mic(unsigned char *l, unsigned char *p, int len)
B
Bruce Momjian 已提交
739
{
740
	latin2mic(l, p, len, LC_ISO8859_2);
B
Bruce Momjian 已提交
741
}
742 743
static void
mic2latin2(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
744
{
745
	mic2latin(mic, p, len, LC_ISO8859_2);
B
Bruce Momjian 已提交
746
}
747 748
static void
latin32mic(unsigned char *l, unsigned char *p, int len)
B
Bruce Momjian 已提交
749
{
750
	latin2mic(l, p, len, LC_ISO8859_3);
B
Bruce Momjian 已提交
751
}
752 753
static void
mic2latin3(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
754
{
755
	mic2latin(mic, p, len, LC_ISO8859_3);
B
Bruce Momjian 已提交
756
}
757 758
static void
latin42mic(unsigned char *l, unsigned char *p, int len)
B
Bruce Momjian 已提交
759
{
760
	latin2mic(l, p, len, LC_ISO8859_4);
B
Bruce Momjian 已提交
761
}
762 763
static void
mic2latin4(unsigned char *mic, unsigned char *p, int len)
B
Bruce Momjian 已提交
764
{
765
	mic2latin(mic, p, len, LC_ISO8859_4);
B
Bruce Momjian 已提交
766
}
767

768
#ifdef NOT_USED
769 770
static void
latin52mic(unsigned char *l, unsigned char *p, int len)
B
Bruce Momjian 已提交
771
{
772
	latin2mic(l, p, len, LC_ISO8859_5);
B
Bruce Momjian 已提交
773
}
774 775
static void
mic2latin5(unsigned char *mic, unsigned char *p, int len)
M
 
Marc G. Fournier 已提交
776
{
777
	mic2latin(mic, p, len, LC_ISO8859_5);
M
 
Marc G. Fournier 已提交
778
}
B
Bruce Momjian 已提交
779

780
#endif
M
 
Marc G. Fournier 已提交
781

782 783 784
/*
 * ASCII ---> MIC
 */
785 786
static void
ascii2mic(unsigned char *l, unsigned char *p, int len)
787
{
788
	int			c1;
789

790 791 792
	while (len-- > 0 && (c1 = *l++))
		*p++ = (c1 & 0x7f);
	*p = '\0';
793 794 795 796 797
}

/*
 * MIC ---> ASCII
 */
798 799
static void
mic2ascii(unsigned char *mic, unsigned char *p, int len)
800
{
801 802
	int			c1;

803
	while (len-- > 0 && (c1 = *mic))
804 805 806 807 808 809 810
	{
		if (c1 > 0x7f)
			printBogusChar(&mic, &p);
		else
		{						/* should be ASCII */
			*p++ = c1;
		}
811
		mic++;
812 813
	}
	*p = '\0';
814 815
}

B
Bruce Momjian 已提交
816
/*
T
Tatsuo Ishii 已提交
817 818 819 820 821 822 823 824 825 826
 * Cyrillic support
 * currently supported Cyrillic encodings:
 *
 * KOI8-R (this is the charset for the mule internal code
 *		for Cyrillic)
 * ISO-8859-5
 * Microsoft's CP1251(windows-1251)
 * Alternativny Variant (MS-DOS CP866)
 */

B
Bruce Momjian 已提交
827
/* koi2mic: KOI8-R to Mule internal code */
T
Tatsuo Ishii 已提交
828 829 830
static void
koi2mic(unsigned char *l, unsigned char *p, int len)
{
B
Bruce Momjian 已提交
831
	latin2mic(l, p, len, LC_KOI8_R);
T
Tatsuo Ishii 已提交
832 833 834 835 836 837
}

/* mic2koi: Mule internal code to KOI8-R */
static void
mic2koi(unsigned char *mic, unsigned char *p, int len)
{
B
Bruce Momjian 已提交
838
	mic2latin(mic, p, len, LC_KOI8_R);
T
Tatsuo Ishii 已提交
839 840 841 842 843 844 845 846 847 848 849 850
}

/*
 * latin2mic_with_table: a generic single byte charset encoding
 * conversion from a local charset to the mule internal code.
 * with a encoding conversion table.
 * the table is ordered according to the local charset,
 * starting from 128 (0x80). each entry in the table
 * holds the corresponding code point for the mule internal code.
 */
static void
latin2mic_with_table(
B
Bruce Momjian 已提交
851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
					 unsigned char *l,	/* local charset string (source) */
					 unsigned char *p,	/* pointer to store mule internal
										 * code (destination) */
					 int len,	/* length of l */
					 int lc,	/* leading character of p */
					 unsigned char *tab /* code conversion table */
)
{
	unsigned char c1,
				c2;

	while (len-- > 0 && (c1 = *l++))
	{
		if (c1 < 128)
			*p++ = c1;
		else
		{
			c2 = tab[c1 - 128];
			if (c2)
			{
				*p++ = lc;
				*p++ = c2;
			}
			else
			{
				*p++ = ' ';		/* cannot convert */
			}
		}
	}
	*p = '\0';
T
Tatsuo Ishii 已提交
881 882 883 884 885 886 887
}

/*
 * mic2latin_with_table: a generic single byte charset encoding
 * conversion from the mule internal code to a local charset
 * with a encoding conversion table.
 * the table is ordered according to the second byte of the mule
B
Bruce Momjian 已提交
888
 * internal code starting from 128 (0x80).
T
Tatsuo Ishii 已提交
889 890 891 892 893
 * each entry in the table
 * holds the corresponding code point for the local code.
 */
static void
mic2latin_with_table(
B
Bruce Momjian 已提交
894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
					 unsigned char *mic,		/* mule internal code
												 * (source) */
					 unsigned char *p,	/* local code (destination) */
					 int len,	/* length of p */
					 int lc,	/* leading character */
					 unsigned char *tab /* code conversion table */
)
{

	unsigned char c1,
				c2;

	while (len-- > 0 && (c1 = *mic++))
	{
		if (c1 < 128)
			*p++ = c1;
		else if (c1 == lc)
		{
			c1 = *mic++;
			len--;
			c2 = tab[c1 - 128];
			if (c2)
				*p++ = c2;
			else
			{
				*p++ = ' ';		/* cannot convert */
			}
		}
		else
		{
			*p++ = ' ';			/* bogus character */
		}
	}
	*p = '\0';
}

/* iso2mic: ISO-8859-5 to Mule internal code */
T
Tatsuo Ishii 已提交
931 932 933
static void
iso2mic(unsigned char *l, unsigned char *p, int len)
{
934
	static unsigned char iso2koi[] = {
B
Bruce Momjian 已提交
935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0xe1, 0xe2, 0xf7, 0xe7, 0xe4, 0xe5, 0xf6, 0xfa,
		0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, 0xf0,
		0xf2, 0xf3, 0xf4, 0xf5, 0xe6, 0xe8, 0xe3, 0xfe,
		0xfb, 0xfd, 0xff, 0xf9, 0xf8, 0xfc, 0xe0, 0xf1,
		0xc1, 0xc2, 0xd7, 0xc7, 0xc4, 0xc5, 0xd6, 0xda,
		0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, 0xd0,
		0xd2, 0xd3, 0xd4, 0xd5, 0xc6, 0xc8, 0xc3, 0xde,
		0xdb, 0xdd, 0xdf, 0xd9, 0xd8, 0xdc, 0xc0, 0xd1,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
	};

	latin2mic_with_table(l, p, len, LC_KOI8_R, iso2koi);
T
Tatsuo Ishii 已提交
954 955 956 957 958 959
}

/* mic2iso: Mule internal code to ISO8859-5 */
static void
mic2iso(unsigned char *mic, unsigned char *p, int len)
{
960
	static unsigned char koi2iso[] = {
B
Bruce Momjian 已提交
961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0xee, 0xd0, 0xd1, 0xe6, 0xd4, 0xd5, 0xe4, 0xd3,
		0xe5, 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde,
		0xdf, 0xef, 0xe0, 0xe1, 0xe2, 0xe3, 0xd6, 0xd2,
		0xec, 0xeb, 0xd7, 0xe8, 0xed, 0xe9, 0xe7, 0xea,
		0xce, 0xb0, 0xb1, 0xc6, 0xb4, 0xb5, 0xc4, 0xb3,
		0xc5, 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe,
		0xbf, 0xcf, 0xc0, 0xc1, 0xc2, 0xc3, 0xb6, 0xb2,
		0xcc, 0xcb, 0xb7, 0xc8, 0xcd, 0xc9, 0xc7, 0xca
	};

	mic2latin_with_table(mic, p, len, LC_KOI8_R, koi2iso);
}

/* win2mic: CP1251 to Mule internal code */
T
Tatsuo Ishii 已提交
983 984 985
static void
win2mic(unsigned char *l, unsigned char *p, int len)
{
986
	static unsigned char win2koi[] = {
B
Bruce Momjian 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xbd, 0x00, 0x00,
		0xb3, 0x00, 0xb4, 0x00, 0x00, 0x00, 0x00, 0xb7,
		0x00, 0x00, 0xb6, 0xa6, 0xad, 0x00, 0x00, 0x00,
		0xa3, 0x00, 0xa4, 0x00, 0x00, 0x00, 0x00, 0xa7,
		0xe1, 0xe2, 0xf7, 0xe7, 0xe4, 0xe5, 0xf6, 0xfa,
		0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, 0xf0,
		0xf2, 0xf3, 0xf4, 0xf5, 0xe6, 0xe8, 0xe3, 0xfe,
		0xfb, 0xfd, 0xff, 0xf9, 0xf8, 0xfc, 0xe0, 0xf1,
		0xc1, 0xc2, 0xd7, 0xc7, 0xc4, 0xc5, 0xd6, 0xda,
		0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, 0xd0,
		0xd2, 0xd3, 0xd4, 0xd5, 0xc6, 0xc8, 0xc3, 0xde,
		0xdb, 0xdd, 0xdf, 0xd9, 0xd8, 0xdc, 0xc0, 0xd1
	};

	latin2mic_with_table(l, p, len, LC_KOI8_R, win2koi);
T
Tatsuo Ishii 已提交
1006 1007 1008 1009 1010 1011
}

/* mic2win: Mule internal code to CP1251 */
static void
mic2win(unsigned char *mic, unsigned char *p, int len)
{
1012
	static unsigned char koi2win[] = {
B
Bruce Momjian 已提交
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0xb8, 0xba, 0x00, 0xb3, 0xbf,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xb4, 0x00, 0x00,
		0x00, 0x00, 0x00, 0xa8, 0xaa, 0x00, 0xb2, 0xaf,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xa5, 0x00, 0x00,
		0xfe, 0xe0, 0xe1, 0xf6, 0xe4, 0xe5, 0xf4, 0xe3,
		0xf5, 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee,
		0xef, 0xff, 0xf0, 0xf1, 0xf2, 0xf3, 0xe6, 0xe2,
		0xfc, 0xfb, 0xe7, 0xf8, 0xfd, 0xf9, 0xf7, 0xfa,
		0xde, 0xc0, 0xc1, 0xd6, 0xc4, 0xc5, 0xd4, 0xc3,
		0xd5, 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce,
		0xcf, 0xdf, 0xd0, 0xd1, 0xd2, 0xd3, 0xc6, 0xc2,
		0xdc, 0xdb, 0xc7, 0xd8, 0xdd, 0xd9, 0xd7, 0xda
	};

	mic2latin_with_table(mic, p, len, LC_KOI8_R, koi2win);
}

/* alt2mic: CP866 to Mule internal code */
T
Tatsuo Ishii 已提交
1035 1036 1037
static void
alt2mic(unsigned char *l, unsigned char *p, int len)
{
1038
	static unsigned char alt2koi[] = {
B
Bruce Momjian 已提交
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
		0xe1, 0xe2, 0xf7, 0xe7, 0xe4, 0xe5, 0xf6, 0xfa,
		0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, 0xf0,
		0xf2, 0xf3, 0xf4, 0xf5, 0xe6, 0xe8, 0xe3, 0xfe,
		0xfb, 0xfd, 0xff, 0xf9, 0xf8, 0xfc, 0xe0, 0xf1,
		0xc1, 0xc2, 0xd7, 0xc7, 0xc4, 0xc5, 0xd6, 0xda,
		0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, 0xd0,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xbd, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0xd2, 0xd3, 0xd4, 0xd5, 0xc6, 0xc8, 0xc3, 0xde,
		0xdb, 0xdd, 0xdf, 0xd9, 0xd8, 0xdc, 0xc0, 0xd1,
		0xb3, 0xa3, 0xb4, 0xa4, 0xb7, 0xa7, 0x00, 0x00,
		0xb6, 0xa6, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
	};

	latin2mic_with_table(l, p, len, LC_KOI8_R, alt2koi);
T
Tatsuo Ishii 已提交
1058 1059 1060 1061 1062 1063
}

/* mic2alt: Mule internal code to CP866 */
static void
mic2alt(unsigned char *mic, unsigned char *p, int len)
{
1064
	static unsigned char koi2alt[] = {
B
Bruce Momjian 已提交
1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0xf1, 0xf3, 0x00, 0xf9, 0xf5,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xad, 0x00, 0x00,
		0x00, 0x00, 0x00, 0xf0, 0xf2, 0x00, 0xf8, 0xf4,
		0x00, 0x00, 0x00, 0x00, 0x00, 0xbd, 0x00, 0x00,
		0xee, 0xa0, 0xa1, 0xe6, 0xa4, 0xa5, 0xe4, 0xa3,
		0xe5, 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae,
		0xaf, 0xef, 0xe0, 0xe1, 0xe2, 0xe3, 0xa6, 0xa2,
		0xec, 0xeb, 0xa7, 0xe8, 0xed, 0xe9, 0xe7, 0xea,
		0x9e, 0x80, 0x81, 0x96, 0x84, 0x85, 0x94, 0x83,
		0x95, 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e,
		0x8f, 0x9f, 0x90, 0x91, 0x92, 0x93, 0x86, 0x82,
		0x9c, 0x9b, 0x87, 0x98, 0x9d, 0x99, 0x97, 0x9a
	};

	mic2latin_with_table(mic, p, len, LC_KOI8_R, koi2alt);
T
Tatsuo Ishii 已提交
1084 1085 1086 1087 1088 1089
}

/*
 * end of Cyrillic support
 */

1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142

/*-----------------------------------------------------------------
 * WIN1250
 * Microsoft's CP1250(windows-1250)
 *-----------------------------------------------------------------*/
static void
win12502mic(unsigned char *l, unsigned char *p, int len)
{
	static unsigned char win1250_2_iso88592[] = {
		0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
		0x88, 0x89, 0xA9, 0x8B, 0xA6, 0xAB, 0xAE, 0xAC,
		0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
		0x98, 0x99, 0xB9, 0x9B, 0xB6, 0xBB, 0xBE, 0xBC,
		0xA0, 0xB7, 0xA2, 0xA3, 0xA4, 0xA1, 0x00, 0xA7,
		0xA8, 0x00, 0xAA, 0x00, 0x00, 0xAD, 0x00, 0xAF,
		0xB0, 0x00, 0xB2, 0xB3, 0xB4, 0x00, 0x00, 0x00,
		0xB8, 0xB1, 0xBA, 0x00, 0xA5, 0xBD, 0xB5, 0xBF,
		0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
		0xC8, 0xC9, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF,
		0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7,
		0xD8, 0xD9, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF,
		0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7,
		0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF,
		0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
		0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF
	};

	latin2mic_with_table(l, p, len, LC_ISO8859_2, win1250_2_iso88592);
}
static void
mic2win1250(unsigned char *mic, unsigned char *p, int len)
{
	static unsigned char iso88592_2_win1250[] = {
		0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
		0x88, 0x89, 0x00, 0x8B, 0x00, 0x00, 0x00, 0x00,
		0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
		0x98, 0x99, 0x00, 0x9B, 0x00, 0x00, 0x00, 0x00,
		0xA0, 0xA5, 0xA2, 0xA3, 0xA4, 0xBC, 0x8C, 0xA7,
		0xA8, 0x8A, 0xAA, 0x8D, 0x8F, 0xAD, 0x8E, 0xAF,
		0xB0, 0xB9, 0xB2, 0xB3, 0xB4, 0xBE, 0x9C, 0xA1,
		0xB8, 0x9A, 0xBA, 0x9D, 0x9F, 0xBD, 0x9E, 0xBF,
		0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
		0xC8, 0xC9, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF,
		0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7,
		0xD8, 0xD9, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF,
		0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7,
		0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF,
		0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
		0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF
	};

	mic2latin_with_table(mic, p, len, LC_ISO8859_2, iso88592_2_win1250);
}
1143

1144 1145
#ifdef UNICODE_CONVERSION

1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
/*
 * UNICODE(UTF-8) support
 */

/*
 * ASCII ---> UTF-8
 */
static void
ascii2utf(unsigned char *ascii, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1156
	ascii2mic(ascii, utf, len);
1157 1158 1159 1160 1161 1162 1163 1164
}

/*
 * UTF-8 ---> ASCII
 */
static void
utf2ascii(unsigned char *utf, unsigned char *ascii, int len)
{
B
Bruce Momjian 已提交
1165
	mic2ascii(utf, ascii, len);
1166 1167 1168 1169 1170 1171 1172 1173
}

/*
 * ISO8859-1 ---> UTF-8
 */
static void
latin1_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1174
	unsigned short c;
1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194

	while (len-- > 0 && (c = *iso++))
	{
		if (c < 0x80)
			*utf++ = c;
		else
		{
			*utf++ = (c >> 6) | 0xc0;
			*utf++ = (c & 0x003f) | 0x80;
		}
	}
	*utf = '\0';
}

/*
 * UTF-8 ---> ISO8859-1
 */
static void
utf_to_latin1(unsigned char *utf, unsigned char *iso, int len)
{
B
Bruce Momjian 已提交
1195 1196 1197
	unsigned short c,
				c1,
				c2;
1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223

	while (len > 0 && (c = *utf++))
	{
		if ((c & 0xe0) == 0xc0)
		{
			c1 = c & 0x1f;
			c2 = *utf++ & 0x3f;
			*iso = c1 << 6;
			*iso++ |= c2;
			len -= 2;
		}
		else if ((c & 0xe0) == 0xe0)
			elog(ERROR, "Could not convert UTF-8 to ISO8859-1");
		else
		{
			*iso++ = c;
			len--;
		}
	}
	*iso = '\0';
}

/*
 * comparison routine for bsearch()
 * this routine is intended for UTF-8 -> local code
 */
B
Bruce Momjian 已提交
1224 1225
static int
compare1(const void *p1, const void *p2)
1226
{
B
Bruce Momjian 已提交
1227 1228
	unsigned int v1,
				v2;
1229

B
Bruce Momjian 已提交
1230 1231 1232
	v1 = *(unsigned int *) p1;
	v2 = ((pg_utf_to_local *) p2)->utf;
	return (v1 - v2);
1233 1234 1235 1236 1237 1238
}

/*
 * comparison routine for bsearch()
 * this routine is intended for local code -> UTF-8
 */
B
Bruce Momjian 已提交
1239 1240
static int
compare2(const void *p1, const void *p2)
1241
{
B
Bruce Momjian 已提交
1242 1243
	unsigned int v1,
				v2;
1244

B
Bruce Momjian 已提交
1245 1246 1247
	v1 = *(unsigned int *) p1;
	v2 = ((pg_local_to_utf *) p2)->code;
	return (v1 - v2);
1248 1249 1250 1251
}

/*
 * UTF-8 ---> local code
1252 1253
 *
 * utf: input UTF-8 string. Its length is limited by "len" parameter
B
Bruce Momjian 已提交
1254
 *		or a null terminater.
1255 1256 1257
 * iso: pointer to the output.
 * map: the conversion map.
 * size: the size of the conversion map.
1258 1259 1260
 */
static void
utf_to_local(unsigned char *utf, unsigned char *iso,
B
Bruce Momjian 已提交
1261
			 pg_utf_to_local * map, int size, int len)
1262 1263
{
	unsigned int iutf;
B
Bruce Momjian 已提交
1264
	int			l;
1265 1266
	pg_utf_to_local *p;

B
Bruce Momjian 已提交
1267
	for (; len > 0 && *utf; len -= l)
1268 1269 1270 1271 1272 1273 1274 1275
	{
		l = pg_utf_mblen(utf);
		if (l == 1)
		{
			*iso++ = *utf++;
			continue;
		}
		else if (l == 2)
B
Bruce Momjian 已提交
1276
		{
1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287
			iutf = *utf++ << 8;
			iutf |= *utf++;
		}
		else
		{
			iutf = *utf++ << 16;
			iutf |= *utf++ << 8;
			iutf |= *utf++;
		}
		p = bsearch(&iutf, map, size,
					sizeof(pg_utf_to_local), compare1);
1288
		if (p == NULL)
1289
		{
1290
			elog(NOTICE, "utf_to_latin: could not convert UTF-8 (0x%04x) Ignored", iutf);
1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
			continue;
		}
		if (p->code & 0xff000000)
			*iso++ = p->code >> 24;
		if (p->code & 0x00ff0000)
			*iso++ = (p->code & 0x00ff0000) >> 16;
		if (p->code & 0x0000ff00)
			*iso++ = (p->code & 0x0000ff00) >> 8;
		if (p->code & 0x000000ff)
			*iso++ = p->code & 0x000000ff;
	}
	*iso = '\0';
}

/*
 * UTF-8 ---> ISO8859-2
 */
static void
utf_to_latin2(unsigned char *utf, unsigned char *iso, int len)
{
B
Bruce Momjian 已提交
1311
	utf_to_local(utf, iso, ULmapISO8859_2, sizeof(ULmapISO8859_2) / sizeof(pg_utf_to_local), len);
1312 1313 1314 1315 1316 1317 1318 1319
}

/*
 * UTF-8 ---> ISO8859-3
 */
static void
utf_to_latin3(unsigned char *utf, unsigned char *iso, int len)
{
B
Bruce Momjian 已提交
1320
	utf_to_local(utf, iso, ULmapISO8859_3, sizeof(ULmapISO8859_3) / sizeof(pg_utf_to_local), len);
1321 1322 1323 1324 1325 1326 1327 1328
}

/*
 * UTF-8 ---> ISO8859-4
 */
static void
utf_to_latin4(unsigned char *utf, unsigned char *iso, int len)
{
B
Bruce Momjian 已提交
1329
	utf_to_local(utf, iso, ULmapISO8859_4, sizeof(ULmapISO8859_4) / sizeof(pg_utf_to_local), len);
1330 1331 1332 1333 1334 1335 1336 1337 1338
}

/*
 * UTF-8 ---> ISO8859-5
 */
static void
utf_to_latin5(unsigned char *utf, unsigned char *iso, int len)

{
B
Bruce Momjian 已提交
1339
	utf_to_local(utf, iso, ULmapISO8859_5, sizeof(ULmapISO8859_5) / sizeof(pg_utf_to_local), len);
1340
}
1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373
/*
 * Cyrillic charsets
 */

/*
 * UTF-8 --->KOI8 
 */
static void
utf_to_KOI8(unsigned char *utf, unsigned char *iso, int len)

{
	utf_to_local(utf, iso, ULmap_KOI8, sizeof(ULmap_KOI8) / sizeof(pg_utf_to_local), len);
}

/*
 * UTF-8 --->WIN 
 */
static void
utf_to_WIN(unsigned char *utf, unsigned char *iso, int len)

{
	utf_to_local(utf, iso, ULmap_WIN, sizeof(ULmap_WIN) / sizeof(pg_utf_to_local), len);
}

/*
 * UTF-8 --->ALT 
 */
static void
utf_to_ALT(unsigned char *utf, unsigned char *iso, int len)

{
	utf_to_local(utf, iso, ULmap_ALT, sizeof(ULmap_ALT) / sizeof(pg_utf_to_local), len);
}
1374 1375 1376 1377 1378 1379

/*
 * local code ---> UTF-8
 */
static void
local_to_utf(unsigned char *iso, unsigned char *utf,
B
Bruce Momjian 已提交
1380
			 pg_local_to_utf * map, int size, int encoding, int len)
1381 1382
{
	unsigned int iiso;
B
Bruce Momjian 已提交
1383
	int			l;
1384 1385 1386 1387 1388 1389 1390
	pg_local_to_utf *p;
	pg_encoding_conv_tbl *e;

	e = pg_get_enc_ent(encoding);
	if (e == 0)
		elog(ERROR, "Invalid encoding number %d", encoding);

B
Bruce Momjian 已提交
1391
	for (; len > 0 && *iso; len -= l)
1392 1393 1394 1395 1396 1397 1398 1399
	{
		if (*iso < 0x80)
		{
			*utf++ = *iso++;
			l = 1;
			continue;
		}

1400
		l = pg_encoding_mblen(encoding, iso);
1401 1402 1403 1404

		if (l == 1)
			iiso = *iso++;
		else if (l == 2)
B
Bruce Momjian 已提交
1405
		{
1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
			iiso = *iso++ << 8;
			iiso |= *iso++;
		}
		else if (l == 3)
		{
			iiso = *iso++ << 16;
			iiso |= *iso++ << 8;
			iiso |= *iso++;
		}
		else if (l == 4)
		{
			iiso = *iso++ << 24;
			iiso |= *iso++ << 16;
			iiso |= *iso++ << 8;
			iiso |= *iso++;
		}
		p = bsearch(&iiso, map, size,
					sizeof(pg_local_to_utf), compare2);
		if (p == NULL)
		{
B
Bruce Momjian 已提交
1426 1427
			elog(NOTICE, "local_to_utf: could not convert (0x%04x) %s to UTF-8. Ignored",
				 iiso, e->name);
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447
			continue;
		}
		if (p->utf & 0xff000000)
			*utf++ = p->utf >> 24;
		if (p->utf & 0x00ff0000)
			*utf++ = (p->utf & 0x00ff0000) >> 16;
		if (p->utf & 0x0000ff00)
			*utf++ = (p->utf & 0x0000ff00) >> 8;
		if (p->utf & 0x000000ff)
			*utf++ = p->utf & 0x000000ff;
	}
	*utf = '\0';
}

/*
 * ISO-8859-2 ---> UTF-8
 */
static void
latin2_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1448
	local_to_utf(iso, utf, LUmapISO8859_2, sizeof(LUmapISO8859_2) / sizeof(pg_local_to_utf), LATIN2, len);
1449 1450 1451 1452 1453 1454 1455 1456
}

/*
 * ISO-8859-3 ---> UTF-8
 */
static void
latin3_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1457
	local_to_utf(iso, utf, LUmapISO8859_3, sizeof(LUmapISO8859_3) / sizeof(pg_local_to_utf), LATIN3, len);
1458 1459 1460 1461 1462 1463 1464 1465
}

/*
 * ISO-8859-4 ---> UTF-8
 */
static void
latin4_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1466
	local_to_utf(iso, utf, LUmapISO8859_4, sizeof(LUmapISO8859_4) / sizeof(pg_local_to_utf), LATIN4, len);
1467 1468 1469 1470 1471 1472 1473 1474
}

/*
 * ISO-8859-5 ---> UTF-8
 */
static void
latin5_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
B
Bruce Momjian 已提交
1475
	local_to_utf(iso, utf, LUmapISO8859_5, sizeof(LUmapISO8859_5) / sizeof(pg_local_to_utf), LATIN5, len);
1476 1477
}

1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503
/*
 * KOI8 ---> UTF-8
 */
static void
KOI8_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
	local_to_utf(iso, utf, LUmapKOI8, sizeof(LUmapKOI8) / sizeof(pg_local_to_utf), KOI8, len);
}

/*
 * WIN ---> UTF-8
 */
static void
WIN_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
	local_to_utf(iso, utf, LUmapWIN, sizeof(LUmapWIN) / sizeof(pg_local_to_utf), WIN, len);
}

/*
 * ALT ---> UTF-8
 */
static void
ALT_to_utf(unsigned char *iso, unsigned char *utf, int len)
{
	local_to_utf(iso, utf, LUmapALT, sizeof(LUmapALT) / sizeof(pg_local_to_utf), ALT, len);
}
1504 1505 1506 1507 1508 1509 1510
/*
 * UTF-8 ---> EUC_JP
 */
static void
utf_to_euc_jp(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1511 1512
	utf_to_local(utf, euc, ULmapEUC_JP,
				 sizeof(ULmapEUC_JP) / sizeof(pg_utf_to_local), len);
1513 1514 1515 1516 1517 1518 1519 1520
}

/*
 * EUC_JP ---> UTF-8
 */
static void
euc_jp_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
1521
	local_to_utf(euc, utf, LUmapEUC_JP,
B
Bruce Momjian 已提交
1522
			 sizeof(LUmapEUC_JP) / sizeof(pg_local_to_utf), EUC_JP, len);
1523 1524 1525 1526 1527 1528 1529 1530 1531
}

/*
 * UTF-8 ---> EUC_CN
 */
static void
utf_to_euc_cn(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1532 1533
	utf_to_local(utf, euc, ULmapEUC_CN,
				 sizeof(ULmapEUC_CN) / sizeof(pg_utf_to_local), len);
1534 1535 1536 1537 1538 1539 1540 1541 1542
}

/*
 * EUC_CN ---> UTF-8
 */
static void
euc_cn_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
	local_to_utf(euc, utf, LUmapEUC_CN,
B
Bruce Momjian 已提交
1543
			 sizeof(LUmapEUC_CN) / sizeof(pg_local_to_utf), EUC_CN, len);
1544 1545 1546 1547 1548 1549 1550 1551 1552
}

/*
 * UTF-8 ---> EUC_KR
 */
static void
utf_to_euc_kr(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1553 1554
	utf_to_local(utf, euc, ULmapEUC_KR,
				 sizeof(ULmapEUC_KR) / sizeof(pg_utf_to_local), len);
1555 1556 1557 1558 1559 1560 1561 1562 1563
}

/*
 * EUC_KR ---> UTF-8
 */
static void
euc_kr_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
	local_to_utf(euc, utf, LUmapEUC_KR,
B
Bruce Momjian 已提交
1564
			 sizeof(LUmapEUC_KR) / sizeof(pg_local_to_utf), EUC_KR, len);
1565 1566 1567 1568 1569 1570 1571 1572 1573
}

/*
 * UTF-8 ---> EUC_TW
 */
static void
utf_to_euc_tw(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1574 1575
	utf_to_local(utf, euc, ULmapEUC_TW,
				 sizeof(ULmapEUC_TW) / sizeof(pg_utf_to_local), len);
1576 1577 1578 1579 1580 1581 1582 1583 1584
}

/*
 * EUC_TW ---> UTF-8
 */
static void
euc_tw_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
	local_to_utf(euc, utf, LUmapEUC_TW,
B
Bruce Momjian 已提交
1585
			 sizeof(LUmapEUC_TW) / sizeof(pg_local_to_utf), EUC_TW, len);
1586 1587 1588 1589 1590 1591 1592 1593 1594
}

/*
 * UTF-8 ---> SJIS
 */
static void
utf_to_sjis(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1595 1596
	utf_to_local(utf, euc, ULmapSJIS,
				 sizeof(ULmapSJIS) / sizeof(pg_utf_to_local), len);
1597 1598 1599 1600 1601 1602 1603 1604 1605
}

/*
 * SJIS ---> UTF-8
 */
static void
sjis_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
	local_to_utf(euc, utf, LUmapSJIS,
B
Bruce Momjian 已提交
1606
				 sizeof(LUmapSJIS) / sizeof(pg_local_to_utf), SJIS, len);
1607 1608 1609 1610 1611 1612 1613 1614 1615
}

/*
 * UTF-8 ---> BIG5
 */
static void
utf_to_big5(unsigned char *utf, unsigned char *euc, int len)

{
B
Bruce Momjian 已提交
1616 1617
	utf_to_local(utf, euc, ULmapBIG5,
				 sizeof(ULmapBIG5) / sizeof(pg_utf_to_local), len);
1618 1619 1620 1621 1622 1623 1624 1625 1626
}

/*
 * BIG5 ---> UTF-8
 */
static void
big5_to_utf(unsigned char *euc, unsigned char *utf, int len)
{
	local_to_utf(euc, utf, LUmapBIG5,
B
Bruce Momjian 已提交
1627
				 sizeof(LUmapBIG5) / sizeof(pg_local_to_utf), BIG5, len);
1628 1629
}

1630 1631
/*-----------------------------------------------------------------*/

M
 
Marc G. Fournier 已提交
1632
pg_encoding_conv_tbl pg_conv_tbl[] = {
1633
	{SQL_ASCII, "SQL_ASCII", 0, ascii2mic, mic2ascii,
B
Bruce Momjian 已提交
1634
	ascii2utf, utf2ascii},		/* SQL/ASCII */
1635
	{EUC_JP, "EUC_JP", 0, euc_jp2mic, mic2euc_jp,
B
Bruce Momjian 已提交
1636
	euc_jp_to_utf, utf_to_euc_jp},		/* EUC_JP */
1637
	{EUC_CN, "EUC_CN", 0, euc_cn2mic, mic2euc_cn,
B
Bruce Momjian 已提交
1638
	euc_cn_to_utf, utf_to_euc_cn},		/* EUC_CN */
1639 1640 1641
	{EUC_KR, "EUC_KR", 0, euc_kr2mic, mic2euc_kr,
	euc_kr_to_utf, utf_to_euc_kr},		/* EUC_KR */
	{EUC_TW, "EUC_TW", 0, euc_tw2mic, mic2euc_tw,
B
Bruce Momjian 已提交
1642
	euc_tw_to_utf, utf_to_euc_tw},		/* EUC_TW */
1643 1644
	{UNICODE, "UNICODE", 0, 0, 0},		/* UNICODE */
	{MULE_INTERNAL, "MULE_INTERNAL", 0, 0, 0},	/* MULE_INTERNAL */
1645
	{LATIN1, "LATIN1", 0, latin12mic, mic2latin1,
B
Bruce Momjian 已提交
1646
	latin1_to_utf, utf_to_latin1},		/* ISO 8859 Latin 1 */
1647
	{LATIN2, "LATIN2", 0, latin22mic, mic2latin2,
B
Bruce Momjian 已提交
1648
	latin2_to_utf, utf_to_latin2},		/* ISO 8859 Latin 2 */
1649
	{LATIN3, "LATIN3", 0, latin32mic, mic2latin3,
B
Bruce Momjian 已提交
1650
	latin3_to_utf, utf_to_latin3},		/* ISO 8859 Latin 3 */
1651
	{LATIN4, "LATIN4", 0, latin42mic, mic2latin4,
B
Bruce Momjian 已提交
1652
	latin4_to_utf, utf_to_latin4},		/* ISO 8859 Latin 4 */
1653
	{LATIN5, "LATIN5", 0, iso2mic, mic2iso,
B
Bruce Momjian 已提交
1654
	latin5_to_utf, utf_to_latin5},		/* ISO 8859 Latin 5 */
1655
	{KOI8, "KOI8", 0, koi2mic, mic2koi,
1656
	KOI8_to_utf, utf_to_KOI8},						/* KOI8-R */
1657
	{WIN, "WIN", 0, win2mic, mic2win,
1658
	WIN_to_utf , utf_to_WIN},						/* CP1251 */
1659
	{ALT, "ALT", 0, alt2mic, mic2alt,
1660
	ALT_to_utf, utf_to_ALT},						/* CP866 */
1661
	{SJIS, "SJIS", 1, sjis2mic, mic2sjis,
B
Bruce Momjian 已提交
1662
	sjis_to_utf, utf_to_sjis},	/* SJIS */
1663
	{BIG5, "BIG5", 1, big52mic, mic2big5,
B
Bruce Momjian 已提交
1664
	big5_to_utf, utf_to_big5},	/* Big5 */
1665
	{WIN1250, "WIN1250", 1, win12502mic, mic2win1250,
B
Bruce Momjian 已提交
1666 1667
	0, 0},						/* WIN 1250 */
	{-1, "", 0, 0, 0, 0}		/* end mark */
1668 1669 1670 1671 1672 1673
};

#else

pg_encoding_conv_tbl pg_conv_tbl[] = {
	{SQL_ASCII, "SQL_ASCII", 0, ascii2mic, mic2ascii,
B
Bruce Momjian 已提交
1674
	0, 0},						/* SQL/ASCII */
1675
	{EUC_JP, "EUC_JP", 0, euc_jp2mic, mic2euc_jp,
B
Bruce Momjian 已提交
1676
	0, 0},						/* EUC_JP */
1677
	{EUC_CN, "EUC_CN", 0, euc_cn2mic, mic2euc_cn,
B
Bruce Momjian 已提交
1678
	0, 0},						/* EUC_CN */
1679 1680 1681 1682 1683
	{EUC_KR, "EUC_KR", 0, euc_kr2mic, mic2euc_kr},		/* EUC_KR */
	{EUC_TW, "EUC_TW", 0, euc_tw2mic, mic2euc_tw},		/* EUC_TW */
	{UNICODE, "UNICODE", 0, 0, 0},		/* UNICODE */
	{MULE_INTERNAL, "MULE_INTERNAL", 0, 0, 0},	/* MULE_INTERNAL */
	{LATIN1, "LATIN1", 0, latin12mic, mic2latin1,
B
Bruce Momjian 已提交
1684
	0, 0},						/* ISO 8859 Latin 1 */
1685
	{LATIN2, "LATIN2", 0, latin22mic, mic2latin2,
B
Bruce Momjian 已提交
1686
	0, 0},						/* ISO 8859 Latin 2 */
1687
	{LATIN3, "LATIN3", 0, latin32mic, mic2latin3,
B
Bruce Momjian 已提交
1688
	0, 0},						/* ISO 8859 Latin 3 */
1689
	{LATIN4, "LATIN4", 0, latin42mic, mic2latin4,
B
Bruce Momjian 已提交
1690
	0, 0},						/* ISO 8859 Latin 4 */
1691
	{LATIN5, "LATIN5", 0, iso2mic, mic2iso,
B
Bruce Momjian 已提交
1692
	0, 0},						/* ISO 8859 Latin 5 */
1693
	{KOI8, "KOI8", 0, koi2mic, mic2koi,
B
Bruce Momjian 已提交
1694
	0, 0},						/* KOI8-R */
1695
	{WIN, "WIN", 0, win2mic, mic2win,
B
Bruce Momjian 已提交
1696
	0, 0},						/* CP1251 */
1697
	{ALT, "ALT", 0, alt2mic, mic2alt,
B
Bruce Momjian 已提交
1698
	0, 0},						/* CP866 */
1699
	{SJIS, "SJIS", 1, sjis2mic, mic2sjis,
B
Bruce Momjian 已提交
1700
	0, 0},						/* SJIS */
1701
	{BIG5, "BIG5", 1, big52mic, mic2big5,
B
Bruce Momjian 已提交
1702
	0, 0},						/* Big5 */
1703
	{WIN1250, "WIN1250", 1, win12502mic, mic2win1250,
B
Bruce Momjian 已提交
1704 1705
	0, 0},						/* WIN 1250 */
	{-1, "", 0, 0, 0, 0}		/* end mark */
M
 
Marc G. Fournier 已提交
1706
};
1707

B
Bruce Momjian 已提交
1708
#endif	 /* UNICODE_CONVERSION */