summaryrefslogtreecommitdiff
path: root/source3/lib/charcnv.c
blob: b06d869bcc3eccbd6a1cf5f8b311d57cd3ca8802 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
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
338
339
340
341
342
343
344
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
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
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
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
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
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
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
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
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
881
882
883
884
885
886
887
888
889
890
891
892
893
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
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
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
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
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
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
/* 
   Unix SMB/CIFS implementation.
   Character set conversion Extensions
   Copyright (C) Igor Vergeichik <iverg@mail.ru> 2001
   Copyright (C) Andrew Tridgell 2001
   Copyright (C) Simo Sorce 2001
   Copyright (C) Martin Pool 2003
   
   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.
   
   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.
   
   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

*/
#include "includes.h"

/**
 * @file
 *
 * @brief Character-set conversion routines built on our iconv.
 * 
 * @note Samba's internal character set (at least in the 3.0 series)
 * is always the same as the one for the Unix filesystem.  It is
 * <b>not</b> necessarily UTF-8 and may be different on machines that
 * need i18n filenames to be compatible with Unix software.  It does
 * have to be a superset of ASCII.  All multibyte sequences must start
 * with a byte with the high bit set.
 *
 * @sa lib/iconv.c
 */


static smb_iconv_t conv_handles[NUM_CHARSETS][NUM_CHARSETS];
static BOOL conv_silent; /* Should we do a debug if the conversion fails ? */
/* Unsafe unix charsets which could contain '\\' as second byte of mb character */
static const char *conv_unsafe_charsets[] = {
	    "CP932",
	    "EUC-JP",
	    NULL};
/* Global variable which is set to True in init_iconv() if unix charset is unsafe
   w.r.t. '\\' in second byte of mb character. Otherwise it is set to False.
*/
BOOL is_unix_charset_unsafe;

/**
 * Return the name of a charset to give to iconv().
 **/
static const char *charset_name(charset_t ch)
{
	const char *ret = NULL;

	if (ch == CH_UCS2) ret = "UCS-2LE";
	else if (ch == CH_UNIX) ret = lp_unix_charset();
	else if (ch == CH_DOS) ret = lp_dos_charset();
	else if (ch == CH_DISPLAY) ret = lp_display_charset();
	else if (ch == CH_UTF8) ret = "UTF8";

#if defined(HAVE_NL_LANGINFO) && defined(CODESET)
	if (ret && !strcmp(ret, "LOCALE")) {
		const char *ln = NULL;

#ifdef HAVE_SETLOCALE
		setlocale(LC_ALL, "");
#endif
		ln = nl_langinfo(CODESET);
		if (ln) {
			/* Check whether the charset name is supported
			   by iconv */
			smb_iconv_t handle = smb_iconv_open(ln,"UCS-2LE");
			if (handle == (smb_iconv_t) -1) {
				DEBUG(5,("Locale charset '%s' unsupported, using ASCII instead\n", ln));
				ln = NULL;
			} else {
				DEBUG(5,("Substituting charset '%s' for LOCALE\n", ln));
				smb_iconv_close(handle);
			}
		}
		ret = ln;
	}
#endif

	if (!ret || !*ret) ret = "ASCII";
	return ret;
}

void lazy_initialize_conv(void)
{
	static int initialized = False;

	if (!initialized) {
		initialized = True;
		load_case_tables();
		init_iconv();
	}
}

/**
 * Initialize iconv conversion descriptors.
 *
 * This is called the first time it is needed, and also called again
 * every time the configuration is reloaded, because the charset or
 * codepage might have changed.
 **/
void init_iconv(void)
{
	int c1, c2;
	BOOL did_reload = False;
	const char **unsafe_charset = conv_unsafe_charsets;

	/* so that charset_name() works we need to get the UNIX<->UCS2 going
	   first */
	if (!conv_handles[CH_UNIX][CH_UCS2])
		conv_handles[CH_UNIX][CH_UCS2] = smb_iconv_open("UCS-2LE", "ASCII");

	if (!conv_handles[CH_UCS2][CH_UNIX])
		conv_handles[CH_UCS2][CH_UNIX] = smb_iconv_open("ASCII", "UCS-2LE");

	for (c1=0;c1<NUM_CHARSETS;c1++) {
		for (c2=0;c2<NUM_CHARSETS;c2++) {
			const char *n1 = charset_name((charset_t)c1);
			const char *n2 = charset_name((charset_t)c2);
			if (conv_handles[c1][c2] &&
			    strcmp(n1, conv_handles[c1][c2]->from_name) == 0 &&
			    strcmp(n2, conv_handles[c1][c2]->to_name) == 0)
				continue;

			did_reload = True;

			if (conv_handles[c1][c2])
				smb_iconv_close(conv_handles[c1][c2]);

			conv_handles[c1][c2] = smb_iconv_open(n2,n1);
			if (conv_handles[c1][c2] == (smb_iconv_t)-1) {
				DEBUG(0,("Conversion from %s to %s not supported\n",
					 charset_name((charset_t)c1), charset_name((charset_t)c2)));
				conv_handles[c1][c2] = NULL;
			}
		}
	}

	if (did_reload) {
		/* XXX: Does this really get called every time the dos
		 * codepage changes? */
		/* XXX: Is the did_reload test too strict? */
		conv_silent = True;
		init_doschar_table();
		init_valid_table();
		conv_silent = False;
	}
	
	while(*unsafe_charset && strcmp(*unsafe_charset, conv_handles[CH_UCS2][CH_UNIX]->to_name)) {
		unsafe_charset++;
	}
	
	if (*unsafe_charset) {
		is_unix_charset_unsafe = True;
	} else {
		is_unix_charset_unsafe = False;
	}
}

/**
 * Convert string from one encoding to another, making error checking etc
 * Slow path version - uses (slow) iconv.
 *
 * @param src pointer to source string (multibyte or singlebyte)
 * @param srclen length of the source string in bytes
 * @param dest pointer to destination string (multibyte or singlebyte)
 * @param destlen maximal length allowed for string
 * @param allow_bad_conv determines if a "best effort" conversion is acceptable (never returns errors)
 * @returns the number of bytes occupied in the destination
 *
 * Ensure the srclen contains the terminating zero.
 *
 **/

static size_t convert_string_internal(charset_t from, charset_t to,
		      void const *src, size_t srclen, 
		      void *dest, size_t destlen, BOOL allow_bad_conv)
{
	size_t i_len, o_len;
	size_t retval;
	const char* inbuf = (const char*)src;
	char* outbuf = (char*)dest;
	smb_iconv_t descriptor;

	lazy_initialize_conv();

	descriptor = conv_handles[from][to];

	if (srclen == (size_t)-1) {
		if (from == CH_UCS2) {
			srclen = (strlen_w((const smb_ucs2_t *)src)+1) * 2;
		} else {
			srclen = strlen((const char *)src)+1;
		}
	}


	if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
		if (!conv_silent)
			DEBUG(0,("convert_string_internal: Conversion not supported.\n"));
		return (size_t)-1;
	}

	i_len=srclen;
	o_len=destlen;

 again:

	retval = smb_iconv(descriptor, (char **)&inbuf, &i_len, &outbuf, &o_len);
	if(retval==(size_t)-1) {
	    	const char *reason="unknown error";
		switch(errno) {
			case EINVAL:
				reason="Incomplete multibyte sequence";
				if (!conv_silent)
					DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
				if (allow_bad_conv)
					goto use_as_is;
				break;
			case E2BIG:
				reason="No more room"; 
				if (!conv_silent)
					DEBUG(3, ("convert_string_internal: Required %lu, available %lu\n",
						(unsigned long)srclen, (unsigned long)destlen));
				/* we are not sure we need srclen bytes,
			          may be more, may be less.
				  We only know we need more than destlen
				  bytes ---simo */
		               break;
			case EILSEQ:
				reason="Illegal multibyte sequence";
				if (!conv_silent)
					DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
				if (allow_bad_conv)
					goto use_as_is;
				break;
			default:
				if (!conv_silent)
					DEBUG(0,("convert_string_internal: Conversion error: %s(%s)\n",reason,inbuf));
				break;
		}
		/* smb_panic(reason); */
	}
	return destlen-o_len;

 use_as_is:

	/* 
	 * Conversion not supported. This is actually an error, but there are so
	 * many misconfigured iconv systems and smb.conf's out there we can't just
	 * fail. Do a very bad conversion instead.... JRA.
	 */

	{
		if (o_len == 0 || i_len == 0)
			return destlen - o_len;

		if (from == CH_UCS2 && to != CH_UCS2) {
			/* Can't convert from ucs2 to multibyte. Just truncate this char to ascii. */
			if (i_len < 2)
				return destlen - o_len;
			if (i_len >= 2) {
				*outbuf = inbuf[0];

				outbuf++;
				o_len--;

				inbuf += 2;
				i_len -= 2;
			}

			if (o_len == 0 || i_len == 0)
				return destlen - o_len;

			/* Keep trying with the next char... */
			goto again;

		} else if (from != CH_UCS2 && to == CH_UCS2) {
			/* Can't convert to ucs2 - just widen by adding zero. */
			if (o_len < 2)
				return destlen - o_len;

			outbuf[0] = inbuf[0];
			outbuf[1] = '\0';

			inbuf++;
			i_len--;

			outbuf += 2;
			o_len -= 2;

			if (o_len == 0 || i_len == 0)
				return destlen - o_len;

			/* Keep trying with the next char... */
			goto again;

		} else if (from != CH_UCS2 && to != CH_UCS2) {
			/* Failed multibyte to multibyte. Just copy 1 char and
				try again. */
			outbuf[0] = inbuf[0];

			inbuf++;
			i_len--;

			outbuf++;
			o_len--;

			if (o_len == 0 || i_len == 0)
				return destlen - o_len;

			/* Keep trying with the next char... */
			goto again;

		} else {
			/* Keep compiler happy.... */
			return destlen - o_len;
		}
	}
}

/**
 * Convert string from one encoding to another, making error checking etc
 * Fast path version - handles ASCII first.
 *
 * @param src pointer to source string (multibyte or singlebyte)
 * @param srclen length of the source string in bytes, or -1 for nul terminated.
 * @param dest pointer to destination string (multibyte or singlebyte)
 * @param destlen maximal length allowed for string - *NEVER* -1.
 * @param allow_bad_conv determines if a "best effort" conversion is acceptable (never returns errors)
 * @returns the number of bytes occupied in the destination
 *
 * Ensure the srclen contains the terminating zero.
 *
 * This function has been hand-tuned to provide a fast path.
 * Don't change unless you really know what you are doing. JRA.
 **/

size_t convert_string(charset_t from, charset_t to,
		      void const *src, size_t srclen, 
		      void *dest, size_t destlen, BOOL allow_bad_conv)
{
	/*
	 * NB. We deliberately don't do a strlen here if srclen == -1.
	 * This is very expensive over millions of calls and is taken
	 * care of in the slow path in convert_string_internal. JRA.
	 */

#ifdef DEVELOPER
	SMB_ASSERT(destlen != (size_t)-1);
#endif

	if (srclen == 0)
		return 0;

	if (from != CH_UCS2 && to != CH_UCS2) {
		const unsigned char *p = (const unsigned char *)src;
		unsigned char *q = (unsigned char *)dest;
		size_t slen = srclen;
		size_t dlen = destlen;
		unsigned char lastp;
		size_t retval = 0;

		/* If all characters are ascii, fast path here. */
		while (slen && dlen) {
			if ((lastp = *p) <= 0x7f) {
				*q++ = *p++;
				if (slen != (size_t)-1) {
					slen--;
				}
				dlen--;
				retval++;
				if (!lastp)
					break;
			} else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
				goto general_case;
#else
				return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
			}
		}
		return retval;
	} else if (from == CH_UCS2 && to != CH_UCS2) {
		const unsigned char *p = (const unsigned char *)src;
		unsigned char *q = (unsigned char *)dest;
		size_t retval = 0;
		size_t slen = srclen;
		size_t dlen = destlen;
		unsigned char lastp;

		/* If all characters are ascii, fast path here. */
		while (((slen == (size_t)-1) || (slen >= 2)) && dlen) {
			if (((lastp = *p) <= 0x7f) && (p[1] == 0)) {
				*q++ = *p;
				if (slen != (size_t)-1) {
					slen -= 2;
				}
				p += 2;
				dlen--;
				retval++;
				if (!lastp)
					break;
			} else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
				goto general_case;
#else
				return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
			}
		}
		return retval;
	} else if (from != CH_UCS2 && to == CH_UCS2) {
		const unsigned char *p = (const unsigned char *)src;
		unsigned char *q = (unsigned char *)dest;
		size_t retval = 0;
		size_t slen = srclen;
		size_t dlen = destlen;
		unsigned char lastp;

		/* If all characters are ascii, fast path here. */
		while (slen && (dlen >= 2)) {
			if ((lastp = *p) <= 0x7F) {
				*q++ = *p++;
				*q++ = '\0';
				if (slen != (size_t)-1) {
					slen--;
				}
				dlen -= 2;
				retval += 2;
				if (!lastp)
					break;
			} else {
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
				goto general_case;
#else
				return retval + convert_string_internal(from, to, p, slen, q, dlen, allow_bad_conv);
#endif
			}
		}
		return retval;
	}

#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
  general_case:
#endif
	return convert_string_internal(from, to, src, srclen, dest, destlen, allow_bad_conv);
}

/**
 * Convert between character sets, allocating a new buffer for the result.
 *
 * @param ctx TALLOC_CTX to use to allocate with. If NULL use malloc.
 * @param srclen length of source buffer.
 * @param dest always set at least to NULL
 * @note -1 is not accepted for srclen.
 *
 * @returns Size in bytes of the converted string; or -1 in case of error.
 *
 * Ensure the srclen contains the terminating zero.
 * 
 * I hate the goto's in this function. It's embarressing.....
 * There has to be a cleaner way to do this. JRA.
 **/

size_t convert_string_allocate(TALLOC_CTX *ctx, charset_t from, charset_t to,
			       void const *src, size_t srclen, void **dest, BOOL allow_bad_conv)
{
	size_t i_len, o_len, destlen = MAX(srclen, 512);
	size_t retval;
	const char *inbuf = (const char *)src;
	char *outbuf = NULL, *ob = NULL;
	smb_iconv_t descriptor;

	*dest = NULL;

	if (src == NULL || srclen == (size_t)-1)
		return (size_t)-1;
	if (srclen == 0)
		return 0;

	lazy_initialize_conv();

	descriptor = conv_handles[from][to];

	if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
		if (!conv_silent)
			DEBUG(0,("convert_string_allocate: Conversion not supported.\n"));
		if (allow_bad_conv)
			goto use_as_is;
		return (size_t)-1;
	}

  convert:

	if ((destlen*2) < destlen) {
		/* wrapped ! abort. */
		if (!conv_silent)
			DEBUG(0, ("convert_string_allocate: destlen wrapped !\n"));
		if (!ctx)
			SAFE_FREE(outbuf);
		return (size_t)-1;
	} else {
		destlen = destlen * 2;
	}

	if (ctx)
		ob = (char *)talloc_realloc(ctx, ob, destlen);
	else
		ob = (char *)Realloc(ob, destlen);

	if (!ob) {
		DEBUG(0, ("convert_string_allocate: realloc failed!\n"));
		if (!ctx)
			SAFE_FREE(outbuf);
		return (size_t)-1;
	} else {
		outbuf = ob;
	}
	i_len = srclen;
	o_len = destlen;

 again:

	retval = smb_iconv(descriptor,
			   (char **)&inbuf, &i_len,
			   &outbuf, &o_len);
	if(retval == (size_t)-1) 		{
	    	const char *reason="unknown error";
		switch(errno) {
			case EINVAL:
				reason="Incomplete multibyte sequence";
				if (!conv_silent)
					DEBUG(3,("convert_string_allocate: Conversion error: %s(%s)\n",reason,inbuf));
				if (allow_bad_conv)
					goto use_as_is;
				break;
			case E2BIG:
				goto convert;		
			case EILSEQ:
				reason="Illegal multibyte sequence";
				if (!conv_silent)
					DEBUG(3,("convert_string_allocate: Conversion error: %s(%s)\n",reason,inbuf));
				if (allow_bad_conv)
					goto use_as_is;
				break;
		}
		if (!conv_silent)
			DEBUG(0,("Conversion error: %s(%s)\n",reason,inbuf));
		/* smb_panic(reason); */
		return (size_t)-1;
	}

  out:

	destlen = destlen - o_len;
	if (ctx)
		*dest = (char *)talloc_realloc(ctx,ob,destlen);
	else
		*dest = (char *)Realloc(ob,destlen);
	if (destlen && !*dest) {
		DEBUG(0, ("convert_string_allocate: out of memory!\n"));
		if (!ctx)
			SAFE_FREE(ob);
		return (size_t)-1;
	}

	return destlen;

 use_as_is:

	/* 
	 * Conversion not supported. This is actually an error, but there are so
	 * many misconfigured iconv systems and smb.conf's out there we can't just
	 * fail. Do a very bad conversion instead.... JRA.
	 */

	{
		if (o_len == 0 || i_len == 0)
			goto out;

		if (from == CH_UCS2 && to != CH_UCS2) {
			/* Can't convert from ucs2 to multibyte. Just truncate this char to ascii. */
			if (i_len < 2)
				goto out;

			if (i_len >= 2) {
				*outbuf = inbuf[0];

				outbuf++;
				o_len--;

				inbuf += 2;
				i_len -= 2;
			}

			if (o_len == 0 || i_len == 0)
				goto out;

			/* Keep trying with the next char... */
			goto again;

		} else if (from != CH_UCS2 && to == CH_UCS2) {
			/* Can't convert to ucs2 - just widen by adding zero. */
			if (o_len < 2)
				goto out;

			outbuf[0] = inbuf[0];
			outbuf[1] = '\0';

			inbuf++;
			i_len--;

			outbuf += 2;
			o_len -= 2;

			if (o_len == 0 || i_len == 0)
				goto out;

			/* Keep trying with the next char... */
			goto again;

		} else if (from != CH_UCS2 && to != CH_UCS2) {
			/* Failed multibyte to multibyte. Just copy 1 char and
				try again. */
			outbuf[0] = inbuf[0];

			inbuf++;
			i_len--;

			outbuf++;
			o_len--;

			if (o_len == 0 || i_len == 0)
				goto out;

			/* Keep trying with the next char... */
			goto again;

		} else {
			/* Keep compiler happy.... */
			goto out;
		}
	}
}

/**
 * Convert between character sets, allocating a new buffer using talloc for the result.
 *
 * @param srclen length of source buffer.
 * @param dest always set at least to NULL 
 * @note -1 is not accepted for srclen.
 *
 * @returns Size in bytes of the converted string; or -1 in case of error.
 **/
static size_t convert_string_talloc(TALLOC_CTX *ctx, charset_t from, charset_t to,
		      		void const *src, size_t srclen, void **dest, BOOL allow_bad_conv)
{
	size_t dest_len;

	*dest = NULL;
	dest_len=convert_string_allocate(ctx, from, to, src, srclen, dest, allow_bad_conv);
	if (dest_len == (size_t)-1)
		return (size_t)-1;
	if (*dest == NULL)
		return (size_t)-1;
	return dest_len;
}

size_t unix_strupper(const char *src, size_t srclen, char *dest, size_t destlen)
{
	size_t size;
	smb_ucs2_t *buffer;
	
	size = push_ucs2_allocate(&buffer, src);
	if (size == (size_t)-1) {
		smb_panic("failed to create UCS2 buffer");
	}
	if (!strupper_w(buffer) && (dest == src)) {
		free(buffer);
		return srclen;
	}
	
	size = convert_string(CH_UCS2, CH_UNIX, buffer, size, dest, destlen, True);
	free(buffer);
	return size;
}

/**
 strdup() a unix string to upper case.
 Max size is pstring.
**/

char *strdup_upper(const char *s)
{
	pstring out_buffer;
	const unsigned char *p = (const unsigned char *)s;
	unsigned char *q = (unsigned char *)out_buffer;

	/* this is quite a common operation, so we want it to be
	   fast. We optimise for the ascii case, knowing that all our
	   supported multi-byte character sets are ascii-compatible
	   (ie. they match for the first 128 chars) */

	while (1) {
		if (*p & 0x80)
			break;
		*q++ = toupper(*p);
		if (!*p)
			break;
		p++;
		if (p - ( const unsigned char *)s >= sizeof(pstring))
			break;
	}

	if (*p) {
		/* MB case. */
		size_t size;
		wpstring buffer;
		size = convert_string(CH_UNIX, CH_UCS2, s, -1, buffer, sizeof(buffer), True);
		if (size == (size_t)-1) {
			return NULL;
		}

		strupper_w(buffer);
	
		size = convert_string(CH_UCS2, CH_UNIX, buffer, -1, out_buffer, sizeof(out_buffer), True);
		if (size == (size_t)-1) {
			return NULL;
		}
	}

	return strdup(out_buffer);
}

size_t unix_strlower(const char *src, size_t srclen, char *dest, size_t destlen)
{
	size_t size;
	smb_ucs2_t *buffer = NULL;
	
	size = convert_string_allocate(NULL, CH_UNIX, CH_UCS2, src, srclen,
				       (void **) &buffer, True);
	if (size == (size_t)-1 || !buffer) {
		smb_panic("failed to create UCS2 buffer");
	}
	if (!strlower_w(buffer) && (dest == src)) {
		SAFE_FREE(buffer);
		return srclen;
	}
	size = convert_string(CH_UCS2, CH_UNIX, buffer, size, dest, destlen, True);
	SAFE_FREE(buffer);
	return size;
}

/**
 strdup() a unix string to lower case.
**/

char *strdup_lower(const char *s)
{
	size_t size;
	smb_ucs2_t *buffer = NULL;
	char *out_buffer;
	
	size = push_ucs2_allocate(&buffer, s);
	if (size == -1 || !buffer) {
		return NULL;
	}

	strlower_w(buffer);
	
	size = pull_ucs2_allocate(&out_buffer, buffer);
	SAFE_FREE(buffer);

	if (size == (size_t)-1) {
		return NULL;
	}
	
	return out_buffer;
}

static size_t ucs2_align(const void *base_ptr, const void *p, int flags)
{
	if (flags & (STR_NOALIGN|STR_ASCII))
		return 0;
	return PTR_DIFF(p, base_ptr) & 1;
}


/**
 * Copy a string from a char* unix src to a dos codepage string destination.
 *
 * @return the number of bytes occupied by the string in the destination.
 *
 * @param flags can include
 * <dl>
 * <dt>STR_TERMINATE</dt> <dd>means include the null termination</dd>
 * <dt>STR_UPPER</dt> <dd>means uppercase in the destination</dd>
 * </dl>
 *
 * @param dest_len the maximum length in bytes allowed in the
 * destination.  If @p dest_len is -1 then no maximum is used.
 **/
size_t push_ascii(void *dest, const char *src, size_t dest_len, int flags)
{
	size_t src_len = strlen(src);
	pstring tmpbuf;

	/* treat a pstring as "unlimited" length */
	if (dest_len == (size_t)-1)
		dest_len = sizeof(pstring);

	if (flags & STR_UPPER) {
		pstrcpy(tmpbuf, src);
		strupper_m(tmpbuf);
		src = tmpbuf;
	}

	if (flags & (STR_TERMINATE | STR_TERMINATE_ASCII))
		src_len++;

	return convert_string(CH_UNIX, CH_DOS, src, src_len, dest, dest_len, True);
}

size_t push_ascii_fstring(void *dest, const char *src)
{
	return push_ascii(dest, src, sizeof(fstring), STR_TERMINATE);
}

size_t push_ascii_pstring(void *dest, const char *src)
{
	return push_ascii(dest, src, sizeof(pstring), STR_TERMINATE);
}

/********************************************************************
 Push an nstring - ensure null terminated. Written by
 moriyama@miraclelinux.com (MORIYAMA Masayuki).
********************************************************************/

size_t push_ascii_nstring(void *dest, const char *src)
{
	size_t i, buffer_len, dest_len;
	smb_ucs2_t *buffer;

	conv_silent = True;
	buffer_len = push_ucs2_allocate(&buffer, src);
	if (buffer_len == (size_t)-1) {
		smb_panic("failed to create UCS2 buffer");
	}

	dest_len = 0;
	for (i = 0; buffer[i] != 0 && (i < buffer_len); i++) {
		unsigned char mb[10];
		/* Convert one smb_ucs2_t character at a time. */
		size_t mb_len = convert_string(CH_UCS2, CH_DOS, buffer+i, sizeof(smb_ucs2_t), mb, sizeof(mb), False);
		if ((mb_len != (size_t)-1) && (dest_len + mb_len <= MAX_NETBIOSNAME_LEN - 1)) {
			memcpy((char *)dest + dest_len, mb, mb_len);
			dest_len += mb_len;
		} else {
			errno = E2BIG;
			break;
		}
	}
	((char *)dest)[dest_len] = '\0';

	SAFE_FREE(buffer);
	conv_silent = False;
	return dest_len;
}

/**
 * Copy a string from a dos codepage source to a unix char* destination.
 *
 * The resulting string in "dest" is always null terminated.
 *
 * @param flags can have:
 * <dl>
 * <dt>STR_TERMINATE</dt>
 * <dd>STR_TERMINATE means the string in @p src
 * is null terminated, and src_len is ignored.</dd>
 * </dl>
 *
 * @param src_len is the length of the source area in bytes.
 * @returns the number of bytes occupied by the string in @p src.
 **/
size_t pull_ascii(char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
	size_t ret;

	if (dest_len == (size_t)-1)
		dest_len = sizeof(pstring);

	if (flags & STR_TERMINATE) {
		if (src_len == (size_t)-1) {
			src_len = strlen(src) + 1;
		} else {
			size_t len = strnlen(src, src_len);
			if (len < src_len)
				len++;
			src_len = len;
		}
	}

	ret = convert_string(CH_DOS, CH_UNIX, src, src_len, dest, dest_len, True);
	if (ret == (size_t)-1) {
		dest_len = 0;
	}

	if (dest_len)
		dest[MIN(ret, dest_len-1)] = 0;
	else 
		dest[0] = 0;

	return src_len;
}

size_t pull_ascii_pstring(char *dest, const void *src)
{
	return pull_ascii(dest, src, sizeof(pstring), -1, STR_TERMINATE);
}

size_t pull_ascii_fstring(char *dest, const void *src)
{
	return pull_ascii(dest, src, sizeof(fstring), -1, STR_TERMINATE);
}

/* When pulling an nstring it can expand into a larger size (dos cp -> utf8). Cope with this. */

size_t pull_ascii_nstring(char *dest, size_t dest_len, const void *src)
{
	return pull_ascii(dest, src, dest_len, sizeof(nstring), STR_TERMINATE);
}

/**
 * Copy a string from a char* src to a unicode destination.
 *
 * @returns the number of bytes occupied by the string in the destination.
 *
 * @param flags can have:
 *
 * <dl>
 * <dt>STR_TERMINATE <dd>means include the null termination.
 * <dt>STR_UPPER     <dd>means uppercase in the destination.
 * <dt>STR_NOALIGN   <dd>means don't do alignment.
 * </dl>
 *
 * @param dest_len is the maximum length allowed in the
 * destination. If dest_len is -1 then no maxiumum is used.
 **/

size_t push_ucs2(const void *base_ptr, void *dest, const char *src, size_t dest_len, int flags)
{
	size_t len=0;
	size_t src_len;
	size_t ret;

	/* treat a pstring as "unlimited" length */
	if (dest_len == (size_t)-1)
		dest_len = sizeof(pstring);

	if (flags & STR_TERMINATE)
		src_len = (size_t)-1;
	else
		src_len = strlen(src);

	if (ucs2_align(base_ptr, dest, flags)) {
		*(char *)dest = 0;
		dest = (void *)((char *)dest + 1);
		if (dest_len)
			dest_len--;
		len++;
	}

	/* ucs2 is always a multiple of 2 bytes */
	dest_len &= ~1;

	ret =  convert_string(CH_UNIX, CH_UCS2, src, src_len, dest, dest_len, True);
	if (ret == (size_t)-1) {
		return 0;
	}

	len += ret;

	if (flags & STR_UPPER) {
		smb_ucs2_t *dest_ucs2 = dest;
		size_t i;
		for (i = 0; i < (dest_len / 2) && dest_ucs2[i]; i++) {
			smb_ucs2_t v = toupper_w(dest_ucs2[i]);
			if (v != dest_ucs2[i]) {
				dest_ucs2[i] = v;
			}
		}
	}

	return len;
}


/**
 * Copy a string from a unix char* src to a UCS2 destination,
 * allocating a buffer using talloc().
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 *         or -1 in case of error.
 **/
size_t push_ucs2_talloc(TALLOC_CTX *ctx, smb_ucs2_t **dest, const char *src)
{
	size_t src_len = strlen(src)+1;

	*dest = NULL;
	return convert_string_talloc(ctx, CH_UNIX, CH_UCS2, src, src_len, (void **)dest, True);
}


/**
 * Copy a string from a unix char* src to a UCS2 destination, allocating a buffer
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 *         or -1 in case of error.
 **/

size_t push_ucs2_allocate(smb_ucs2_t **dest, const char *src)
{
	size_t src_len = strlen(src)+1;

	*dest = NULL;
	return convert_string_allocate(NULL, CH_UNIX, CH_UCS2, src, src_len, (void **)dest, True);
}

/**
 Copy a string from a char* src to a UTF-8 destination.
 Return the number of bytes occupied by the string in the destination
 Flags can have:
  STR_TERMINATE means include the null termination
  STR_UPPER     means uppercase in the destination
 dest_len is the maximum length allowed in the destination. If dest_len
 is -1 then no maxiumum is used.
**/

static size_t push_utf8(void *dest, const char *src, size_t dest_len, int flags)
{
	size_t src_len = strlen(src);
	pstring tmpbuf;

	/* treat a pstring as "unlimited" length */
	if (dest_len == (size_t)-1)
		dest_len = sizeof(pstring);

	if (flags & STR_UPPER) {
		pstrcpy(tmpbuf, src);
		strupper_m(tmpbuf);
		src = tmpbuf;
	}

	if (flags & STR_TERMINATE)
		src_len++;

	return convert_string(CH_UNIX, CH_UTF8, src, src_len, dest, dest_len, True);
}

size_t push_utf8_fstring(void *dest, const char *src)
{
	return push_utf8(dest, src, sizeof(fstring), STR_TERMINATE);
}

/**
 * Copy a string from a unix char* src to a UTF-8 destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t push_utf8_talloc(TALLOC_CTX *ctx, char **dest, const char *src)
{
	size_t src_len = strlen(src)+1;

	*dest = NULL;
	return convert_string_talloc(ctx, CH_UNIX, CH_UTF8, src, src_len, (void**)dest, True);
}

/**
 * Copy a string from a unix char* src to a UTF-8 destination, allocating a buffer
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t push_utf8_allocate(char **dest, const char *src)
{
	size_t src_len = strlen(src)+1;

	*dest = NULL;
	return convert_string_allocate(NULL, CH_UNIX, CH_UTF8, src, src_len, (void **)dest, True);	
}

/**
 Copy a string from a ucs2 source to a unix char* destination.
 Flags can have:
  STR_TERMINATE means the string in src is null terminated.
  STR_NOALIGN   means don't try to align.
 if STR_TERMINATE is set then src_len is ignored if it is -1.
 src_len is the length of the source area in bytes
 Return the number of bytes occupied by the string in src.
 The resulting string in "dest" is always null terminated.
**/

size_t pull_ucs2(const void *base_ptr, char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
	size_t ret;

	if (dest_len == (size_t)-1)
		dest_len = sizeof(pstring);

	if (ucs2_align(base_ptr, src, flags)) {
		src = (const void *)((const char *)src + 1);
		if (src_len != (size_t)-1)
			src_len--;
	}

	if (flags & STR_TERMINATE) {
		/* src_len -1 is the default for null terminated strings. */
		if (src_len != (size_t)-1) {
			size_t len = strnlen_w(src, src_len/2);
			if (len < src_len/2)
				len++;
			src_len = len*2;
		}
	}

	/* ucs2 is always a multiple of 2 bytes */
	if (src_len != (size_t)-1)
		src_len &= ~1;
	
	ret = convert_string(CH_UCS2, CH_UNIX, src, src_len, dest, dest_len, True);
	if (ret == (size_t)-1) {
		return 0;
	}

	if (src_len == (size_t)-1)
		src_len = ret*2;
		
	if (dest_len)
		dest[MIN(ret, dest_len-1)] = 0;
	else 
		dest[0] = 0;

	return src_len;
}

size_t pull_ucs2_pstring(char *dest, const void *src)
{
	return pull_ucs2(NULL, dest, src, sizeof(pstring), -1, STR_TERMINATE);
}

size_t pull_ucs2_fstring(char *dest, const void *src)
{
	return pull_ucs2(NULL, dest, src, sizeof(fstring), -1, STR_TERMINATE);
}

/**
 * Copy a string from a UCS2 src to a unix char * destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_ucs2_talloc(TALLOC_CTX *ctx, char **dest, const smb_ucs2_t *src)
{
	size_t src_len = (strlen_w(src)+1) * sizeof(smb_ucs2_t);
	*dest = NULL;
	return convert_string_talloc(ctx, CH_UCS2, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UCS2 src to a unix char * destination, allocating a buffer
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_ucs2_allocate(char **dest, const smb_ucs2_t *src)
{
	size_t src_len = (strlen_w(src)+1) * sizeof(smb_ucs2_t);
	*dest = NULL;
	return convert_string_allocate(NULL, CH_UCS2, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UTF-8 src to a unix char * destination, allocating a buffer using talloc
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_utf8_talloc(TALLOC_CTX *ctx, char **dest, const char *src)
{
	size_t src_len = strlen(src)+1;
	*dest = NULL;
	return convert_string_talloc(ctx, CH_UTF8, CH_UNIX, src, src_len, (void **)dest, True);
}

/**
 * Copy a string from a UTF-8 src to a unix char * destination, allocating a buffer
 *
 * @param dest always set at least to NULL 
 *
 * @returns The number of bytes occupied by the string in the destination
 **/

size_t pull_utf8_allocate(char **dest, const char *src)
{
	size_t src_len = strlen(src)+1;
	*dest = NULL;
	return convert_string_allocate(NULL, CH_UTF8, CH_UNIX, src, src_len, (void **)dest, True);
}
 
/**
 Copy a string from a char* src to a unicode or ascii
 dos codepage destination choosing unicode or ascii based on the 
 flags in the SMB buffer starting at base_ptr.
 Return the number of bytes occupied by the string in the destination.
 flags can have:
  STR_TERMINATE means include the null termination.
  STR_UPPER     means uppercase in the destination.
  STR_ASCII     use ascii even with unicode packet.
  STR_NOALIGN   means don't do alignment.
 dest_len is the maximum length allowed in the destination. If dest_len
 is -1 then no maxiumum is used.
**/

size_t push_string_fn(const char *function, unsigned int line, const void *base_ptr, void *dest, const char *src, size_t dest_len, int flags)
{
#ifdef DEVELOPER
	/* We really need to zero fill here, not clobber
	 * region, as we want to ensure that valgrind thinks
	 * all of the outgoing buffer has been written to
	 * so a send() or write() won't trap an error.
	 * JRA.
	 */
#if 0
	if (dest_len != (size_t)-1)
		clobber_region(function, line, dest, dest_len);
#else
	if (dest_len != (size_t)-1)
		memset(dest, '\0', dest_len);
#endif
#endif

	if (!(flags & STR_ASCII) && \
	    ((flags & STR_UNICODE || \
	      (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
		return push_ucs2(base_ptr, dest, src, dest_len, flags);
	}
	return push_ascii(dest, src, dest_len, flags);
}


/**
 Copy a string from a unicode or ascii source (depending on
 the packet flags) to a char* destination.
 Flags can have:
  STR_TERMINATE means the string in src is null terminated.
  STR_UNICODE   means to force as unicode.
  STR_ASCII     use ascii even with unicode packet.
  STR_NOALIGN   means don't do alignment.
 if STR_TERMINATE is set then src_len is ignored is it is -1
 src_len is the length of the source area in bytes.
 Return the number of bytes occupied by the string in src.
 The resulting string in "dest" is always null terminated.
**/

size_t pull_string_fn(const char *function, unsigned int line, const void *base_ptr, char *dest, const void *src, size_t dest_len, size_t src_len, int flags)
{
#ifdef DEVELOPER
	if (dest_len != (size_t)-1)
		clobber_region(function, line, dest, dest_len);
#endif

	if (!(flags & STR_ASCII) && \
	    ((flags & STR_UNICODE || \
	      (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
		return pull_ucs2(base_ptr, dest, src, dest_len, src_len, flags);
	}
	return pull_ascii(dest, src, dest_len, src_len, flags);
}

size_t align_string(const void *base_ptr, const char *p, int flags)
{
	if (!(flags & STR_ASCII) && \
	    ((flags & STR_UNICODE || \
	      (SVAL(base_ptr, smb_flg2) & FLAGS2_UNICODE_STRINGS)))) {
		return ucs2_align(base_ptr, p, flags);
	}
	return 0;
}