summaryrefslogtreecommitdiff
path: root/lib/util/charset/convert_string.c
blob: 2e666802c9e06b6f8ced38f1fcdd6443b06a7fd0 (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
/*
   Unix SMB/CIFS implementation.
   Character set conversion Extensions
   Copyright (C) Igor Vergeichik <iverg@mail.ru> 2001
   Copyright (C) Andrew Tridgell 2001-2011
   Copyright (C) Andrew Bartlett 2011
   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 3 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, see <http://www.gnu.org/licenses/>.

*/
#include "includes.h"
#include "system/iconv.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
 */


/**
 * 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 converted size is the number of bytes occupied in the destination
 *
 * @returns false and sets errno on fail, true on success.
 *
 * Ensure the srclen contains the terminating zero.
 *
 **/

static bool convert_string_internal(struct smb_iconv_handle *ic,
				    charset_t from, charset_t to,
				    void const *src, size_t srclen,
				    void *dest, size_t destlen, size_t *converted_size)
{
	size_t i_len, o_len;
	size_t retval;
	const char* inbuf = (const char*)src;
	char* outbuf = (char*)dest;
	smb_iconv_t descriptor;

	descriptor = get_conv_handle(ic, from, to);

	if (srclen == (size_t)-1) {
		if (from == CH_UTF16LE || from == CH_UTF16BE) {
			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) {
		errno = EINVAL;
		return false;
	}

	i_len=srclen;
	o_len=destlen;

	retval = smb_iconv(descriptor, &inbuf, &i_len, &outbuf, &o_len);
	*converted_size = destlen-o_len;

	return (retval != (size_t)-1);
}

/**
 * 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 converted size is the number of bytes occupied in the destination
 *
 * @returns false and sets errno on fail, true on success.
 *
 * 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.
 **/

bool convert_string_error_handle(struct smb_iconv_handle *ic,
				 charset_t from, charset_t to,
				 void const *src, size_t srclen,
				 void *dest, size_t destlen,
				 size_t *converted_size)
{
	/*
	 * 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) {
		*converted_size = 0;
		return true;
	}

	if (from != CH_UTF16LE && from != CH_UTF16BE && to != CH_UTF16LE && to != CH_UTF16BE) {
		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 = '\0';
		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
				bool ret = convert_string_internal(ic, from, to, p, slen, q, dlen, converted_size);
				*converted_size += retval;
				return ret;
#endif
			}
		}

		*converted_size = retval;

		if (!dlen) {
			/* Even if we fast path we should note if we ran out of room. */
			if (((slen != (size_t)-1) && slen) ||
					((slen == (size_t)-1) && lastp)) {
				errno = E2BIG;
				return false;
			}
		}
		return true;
	} else if (from == CH_UTF16LE && to != CH_UTF16LE) {
		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 = '\0';
#ifndef BROKEN_UNICODE_COMPOSE_CHARACTERS
		bool ret;
#endif

		if (slen == (size_t)-1) {
			while (dlen &&
			       ((lastp = *p) <= 0x7f) && (p[1] == 0)) {
				*q++ = *p;
				p += 2;
				dlen--;
				retval++;
				if (!lastp)
					break;
			}
			if (lastp != 0) goto slow_path;
		} else {
			while (slen >= 2 && dlen &&
			       (*p <= 0x7f) && (p[1] == 0)) {
				*q++ = *p;
				slen -= 2;
				p += 2;
				dlen--;
				retval++;
			}
			if (slen != 0) goto slow_path;
		}

		*converted_size = retval;

		if (!dlen) {
			/* Even if we fast path we should note if we ran out of room. */
			if (((slen != (size_t)-1) && slen) ||
					((slen == (size_t)-1) && lastp)) {
				errno = E2BIG;
				return false;
			}
		}
		return true;

	slow_path:
		/* come here when we hit a character we can't deal
		 * with in the fast path
		 */
#ifdef BROKEN_UNICODE_COMPOSE_CHARACTERS
		goto general_case;
#else
		ret = convert_string_internal(ic, from, to, p, slen, q, dlen, converted_size);
		*converted_size += retval;
		return ret;
#endif

	} else if (from != CH_UTF16LE && from != CH_UTF16BE && to == CH_UTF16LE) {
		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 = '\0';

		/* If all characters are ascii, fast path here. */
		while (slen && (dlen >= 1)) {
			if (dlen >=2 && (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
				bool ret = convert_string_internal(ic, from, to, p, slen, q, dlen, converted_size);
				*converted_size += retval;
				return ret;
#endif
			}
		}

		*converted_size = retval;

		if (!dlen) {
			/* Even if we fast path we should note if we ran out of room. */
			if (((slen != (size_t)-1) && slen) ||
					((slen == (size_t)-1) && lastp)) {
				errno = E2BIG;
				return false;
			}
		}
		return true;
	}

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

bool convert_string_handle(struct smb_iconv_handle *ic,
			   charset_t from, charset_t to,
			   void const *src, size_t srclen,
			   void *dest, size_t destlen,
			   size_t *converted_size)
{
	bool ret = convert_string_error_handle(ic, from, to, src, srclen, dest, destlen, converted_size);

	if(ret==false) {
		const char *reason="unknown error";
		switch(errno) {
			case EINVAL:
				reason="Incomplete multibyte sequence";
				DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",
					 reason, (const char *)src));
				break;
			case E2BIG:
			{
				reason="No more room";
				if (from == CH_UNIX) {
					DEBUG(3,("E2BIG: convert_string(%s,%s): srclen=%u destlen=%u - '%s'\n",
						 charset_name(ic, from), charset_name(ic, to),
						 (unsigned int)srclen, (unsigned int)destlen, (const char *)src));
				} else {
					DEBUG(3,("E2BIG: convert_string(%s,%s): srclen=%u destlen=%u\n",
						 charset_name(ic, from), charset_name(ic, to),
						 (unsigned int)srclen, (unsigned int)destlen));
				}
				break;
			}
			case EILSEQ:
				reason="Illegal multibyte sequence";
				DEBUG(3,("convert_string_internal: Conversion error: %s(%s)\n",
					 reason, (const char *)src));
				break;
			default:
				DEBUG(0,("convert_string_internal: Conversion error: %s(%s)\n",
					 reason, (const char *)src));
				break;
		}
		/* smb_panic(reason); */
	}
	return ret;
}


/**
 * 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
 * @parm converted_size set to the number of bytes occupied by the string in
 * the destination on success.
 * @note -1 is not accepted for srclen.
 *
 * @return true if new buffer was correctly allocated, and string was
 * converted.
 *
 * Ensure the srclen contains the terminating zero.
 *
 * I hate the goto's in this function. It's emberrassing.....
 * There has to be a cleaner way to do this. JRA.
 */
bool convert_string_talloc_handle(TALLOC_CTX *ctx, struct smb_iconv_handle *ic,
				  charset_t from, charset_t to,
				  void const *src, size_t srclen, void *dst,
				  size_t *converted_size)

{
	size_t i_len, o_len, destlen = (srclen * 3) / 2;
	size_t retval;
	const char *inbuf = (const char *)src;
	char *outbuf = NULL, *ob = NULL;
	smb_iconv_t descriptor;
	void **dest = (void **)dst;

	*dest = NULL;

	if (src == NULL || srclen == (size_t)-1) {
		errno = EINVAL;
		return false;
	}

	if (srclen == 0) {
		/* We really should treat this as an error, but
		   there are too many callers that need this to
		   return a NULL terminated string in the correct
		   character set. */
		if (to == CH_UTF16LE|| to == CH_UTF16BE || to == CH_UTF16MUNGED) {
			destlen = 2;
		} else {
			destlen = 1;
		}
		ob = talloc_zero_array(ctx, char, destlen);
		if (ob == NULL) {
			errno = ENOMEM;
			return false;
		}
		if (converted_size != NULL) {
			*converted_size = destlen;
		}
		*dest = ob;
		return true;
	}

	descriptor = get_conv_handle(ic, from, to);

	if (descriptor == (smb_iconv_t)-1 || descriptor == (smb_iconv_t)0) {
		DEBUG(0,("convert_string_talloc: Conversion not supported.\n"));
		errno = EOPNOTSUPP;
		return false;
	}

  convert:

	/* +2 is for ucs2 null termination. */
	if ((destlen*2)+2 < destlen) {
		/* wrapped ! abort. */
		DEBUG(0, ("convert_string_talloc: destlen wrapped !\n"));
		TALLOC_FREE(outbuf);
		errno = EOPNOTSUPP;
		return false;
	} else {
		destlen = destlen * 2;
	}

	/* +2 is for ucs2 null termination. */
	ob = talloc_realloc(ctx, ob, char, destlen + 2);

	if (!ob) {
		DEBUG(0, ("convert_string_talloc: realloc failed!\n"));
		errno = ENOMEM;
		return false;
	}
	outbuf = ob;
	i_len = srclen;
	o_len = destlen;

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

	destlen = destlen - o_len;
	/* Don't shrink unless we're reclaiming a lot of
	 * space. This is in the hot codepath and these
	 * reallocs *cost*. JRA.
	 */
	if (o_len > 1024) {
		/* We're shrinking here so we know the +2 is safe from wrap. */
		ob = talloc_realloc(ctx,ob, char, destlen + 2);
	}

	if (destlen && !ob) {
		DEBUG(0, ("convert_string_talloc: out of memory!\n"));
		errno = ENOMEM;
		return false;
	}

	*dest = ob;

	/* Must ucs2 null terminate in the extra space we allocated. */
	ob[destlen] = '\0';
	ob[destlen+1] = '\0';

	/* Ensure we can never return a *converted_size of zero. */
	if (destlen == 0) {
		/* As we're now returning false on a bad smb_iconv call,
		   this should never happen. But be safe anyway. */
		if (to == CH_UTF16LE|| to == CH_UTF16BE || to == CH_UTF16MUNGED) {
			destlen = 2;
		} else {
			destlen = 1;
		}
	}

	if (converted_size != NULL) {
		*converted_size = destlen;
	}
	return true;
}

/**
 * Convert string from one encoding to another, making error checking etc
 *
 * @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 converted_size the number of bytes occupied in the destination
 *
 * @returns true on success, false on fail.
 **/
_PUBLIC_ bool convert_string(charset_t from, charset_t to,
			       void const *src, size_t srclen,
			       void *dest, size_t destlen,
			       size_t *converted_size)
{
	return convert_string_handle(get_iconv_handle(), from, to,
					src, srclen,
					dest, destlen, converted_size);
}

/**
 * Convert string from one encoding to another, making error checking etc
 *
 * @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 converted_size the number of bytes occupied in the destination
 *
 * @returns true on success, false on fail.
 **/
_PUBLIC_ bool convert_string_error(charset_t from, charset_t to,
				   void const *src, size_t srclen,
				   void *dest, size_t destlen,
				   size_t *converted_size)
{
	return convert_string_error_handle(get_iconv_handle(), from, to,
					   src, srclen,
					   dest, destlen, converted_size);
}

/**
 * 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
 * @param converted_size Size in bytes of the converted string
 * @note -1 is not accepted for srclen.
 *
 * @returns boolean indication whether the conversion succeeded
 **/

_PUBLIC_ bool convert_string_talloc(TALLOC_CTX *ctx,
				    charset_t from, charset_t to,
				    void const *src, size_t srclen,
				    void *dest, size_t *converted_size)
{
	return convert_string_talloc_handle(ctx, get_iconv_handle(),
						 from, to, src, srclen, dest,
						 converted_size);
}