summary refs log tree commit diff
path: root/string/strxfrm.c
blob: 9fd952600832515b866237168ebef589c1bc40ba (plain) (blame)
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
/* Copyright (C) 1995, 1996, 1997, 1998, 1999 Free Software Foundation, Inc.
   This file is part of the GNU C Library.
   Written by Ulrich Drepper <drepper@cygnus.com>, 1995.

   The GNU C Library is free software; you can redistribute it and/or
   modify it under the terms of the GNU Library General Public License as
   published by the Free Software Foundation; either version 2 of the
   License, or (at your option) any later version.

   The GNU C Library 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
   Library General Public License for more details.

   You should have received a copy of the GNU Library General Public
   License along with the GNU C Library; see the file COPYING.LIB.  If not,
   write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
   Boston, MA 02111-1307, USA.  */

#include <langinfo.h>
#include <stddef.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>

#ifndef STRING_TYPE
# define STRING_TYPE char
# define USTRING_TYPE unsigned char
# ifdef USE_IN_EXTENDED_LOCALE_MODEL
#  define STRXFRM __strxfrm_l
# else
#  define STRXFRM strxfrm
# endif
# define STRCMP strcmp
# define STRLEN strlen
# define STPNCPY __stpncpy
# define WEIGHT_H "../locale/weight.h"
# define SUFFIX	MB
# define L(arg) arg
#endif

#define CONCAT(a,b) CONCAT1(a,b)
#define CONCAT1(a,b) a##b

#include "../locale/localeinfo.h"


#ifndef WIDE_CHAR_VERSION
/* These are definitions used by some of the functions for handling
   UTF-8 encoding below.  */
static const uint32_t encoding_mask[] =
{
  ~0x7ff, ~0xffff, ~0x1fffff, ~0x3ffffff
};

static const unsigned char encoding_byte[] =
{
  0xc0, 0xe0, 0xf0, 0xf8, 0xfc
};


/* We need UTF-8 encoding of numbers.  */
static inline int
utf8_encode (char *buf, int val)
{
  char *startp = buf;
  int retval;

  if (val < 0x80)
    {
      *buf++ = (char) val;
      retval = 1;
    }
  else
    {
      int step;

      for (step = 2; step < 6; ++step)
	if ((val & encoding_mask[step - 2]) == 0)
	  break;
      retval = step;

      *buf = encoding_byte[step - 2];
      --step;
      do
	{
	  buf[step] = 0x80 | (val & 0x3f);
	  val >>= 6;
	}
      while (--step > 0);
      *buf |= val;
    }

  return buf - startp;
}
#endif


#ifndef USE_IN_EXTENDED_LOCALE_MODEL
size_t
STRXFRM (STRING_TYPE *dest, const STRING_TYPE *src, size_t n)
#else
size_t
STRXFRM (STRING_TYPE *dest, const STRING_TYPE *src, size_t n, __locale_t l)
#endif
{
#ifdef USE_IN_EXTENDED_LOCALE_MODEL
  struct locale_data *current = l->__locales[LC_COLLATE];
  uint_fast32_t nrules = *((uint32_t *) current->values[_NL_ITEM_INDEX (_NL_COLLATE_NRULES)].string);
#else
  uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
#endif
  /* We don't assign the following values right away since it might be
     unnecessary in case there are no rules.  */
  const unsigned char *rulesets;
  const int32_t *table;
  const USTRING_TYPE *weights;
  const USTRING_TYPE *extra;
  const int32_t *indirect;
  uint_fast32_t pass;
  size_t needed;
  const USTRING_TYPE *usrc;
  size_t srclen = STRLEN (src);
  int32_t *idxarr;
  unsigned char *rulearr;
  size_t idxmax;
  size_t idxcnt;
  int use_malloc;
#ifdef WIDE_CHAR_VERSION
  size_t size;
  size_t layers;
  const wint_t *names;
#endif

#include WEIGHT_H

  if (nrules == 0)
    {
      if (n != 0)
	STPNCPY (dest, src, n);

      return srclen;
    }

#ifdef USE_IN_EXTENDED_LOCALE_MODEL
  rulesets = (const unsigned char *)
    current->values[_NL_ITEM_INDEX (_NL_COLLATE_RULESETS)].string;
  table = (const int32_t *)
    current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_TABLE,SUFFIX))].string;
  weights = (const USTRING_TYPE *)
    current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_WEIGHT,SUFFIX))].string;
  extra = (const USTRING_TYPE *)
    current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_EXTRA,SUFFIX))].string;
  indirect = (const int32_t *)
    current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_INDIRECT,SUFFIX))].string;
# ifdef WIDE_CHAR_VERSION
  names = (const wint_t *)
    current->values[_NL_ITEM_INDEX (_NL_COLLATE_NAMES)].string;
  size = current->values[_NL_ITEM_INDEX (_NL_COLLATE_HASH_SIZE)].word;
  layers = current->values[_NL_ITEM_INDEX (_NL_COLLATE_HASH_LAYERS)].word;
# endif
#else
  rulesets = (const unsigned char *)
    _NL_CURRENT (LC_COLLATE, _NL_COLLATE_RULESETS);
  table = (const int32_t *)
    _NL_CURRENT (LC_COLLATE, CONCAT(_NL_COLLATE_TABLE,SUFFIX));
  weights = (const USTRING_TYPE *)
    _NL_CURRENT (LC_COLLATE, CONCAT(_NL_COLLATE_WEIGHT,SUFFIX));
  extra = (const USTRING_TYPE *)
    _NL_CURRENT (LC_COLLATE, CONCAT(_NL_COLLATE_EXTRA,SUFFIX));
  indirect = (const int32_t *)
    _NL_CURRENT (LC_COLLATE, CONCAT(_NL_COLLATE_INDIRECT,SUFFIX));
# ifdef WIDE_CHAR_VERSION
  names = (const wint_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_NAMES);
  size = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_HASH_SIZE);
  layers = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_HASH_LAYERS);
# endif
#endif
  use_malloc = 0;

  /* Handle an empty string as a special case.  */
  if (srclen == 0)
    {
      if (n != 0)
        *dest = L('\0');
      return 1;
    }

  /* We need the elements of the string as unsigned values since they
     are used as indeces.  */
  usrc = (const USTRING_TYPE *) src;

  /* Perform the first pass over the string and while doing this find
     and store the weights for each character.  Since we want this to
     be as fast as possible we are using `alloca' to store the temporary
     values.  But since there is no limit on the length of the string
     we have to use `malloc' if the string is too long.  We should be
     very conservative here.  */
  if (srclen >= 16384)
    {
      idxarr = (int32_t *) malloc (srclen * (sizeof (int32_t) + 1));
      rulearr = (unsigned char *) &idxarr[srclen];

      if (idxarr == NULL)
	/* No memory.  Well, go with the stack then.

	   XXX Once this implementation is stable we will handle this
	   differently.  Instead of precomputing the indeces we will
	   do this in time.  This means, though, that this happens for
	   every pass again.  */
	goto try_stack;
      use_malloc = 1;
    }
  else
    {
    try_stack:
      idxarr = (int32_t *) alloca (srclen * sizeof (int32_t));
      rulearr = (unsigned char *) alloca (srclen);
    }

  idxmax = 0;
  do
    {
      int32_t tmp = findidx (&usrc);
      rulearr[idxmax] = tmp >> 24;
      idxarr[idxmax] = tmp & 0xffffff;

      ++idxmax;
    }
  while (*usrc != L('\0'));

  /* Now the passes over the weights.  We now use the indeces we found
     before.  */
  needed = 0;
  for (pass = 0; pass < nrules; ++pass)
    {
      size_t backw_stop = ~0ul;
      int rule = rulesets[rulearr[0] * nrules + pass];
      /* We assume that if a rule has defined `position' in one section
	 this is true for all of them.  */
      int position = rule & sort_position;

      if (position == 0)
	{
	  for (idxcnt = 0; idxcnt < idxmax; ++idxcnt)
	    {
	      if ((rule & sort_forward) != 0)
		{
		  size_t len;

		  if (backw_stop != ~0ul)
		    {
		      /* Handle the pushed elements now.  */
		      size_t backw;

		      for (backw = idxcnt - 1; backw >= backw_stop; --backw)
			{
			  len = weights[idxarr[backw]++];

			  if (needed + len < n)
			    while (len-- > 0)
			      dest[needed++] = weights[idxarr[backw]++];
			  else
			    {
				/* No more characters fit into the buffer.  */
			      needed += len;
			      idxarr[backw] += len;
			    }
			}

		      backw_stop = ~0ul;
		    }

		  /* Now handle the forward element.  */
		  len = weights[idxarr[idxcnt]++];
		  if (needed + len < n)
		    while (len-- > 0)
		      dest[needed++] = weights[idxarr[idxcnt]++];
		  else
		    {
		      /* No more characters fit into the buffer.  */
		      needed += len;
		      idxarr[idxcnt] += len;
		    }
		}
	      else
		{
		  /* Remember where the backwards series started.  */
		  if (backw_stop == ~0ul)
		    backw_stop = idxcnt;
		}

	      rule = rulesets[rulearr[idxcnt + 1] * nrules + pass];
	    }


	  if (backw_stop != ~0ul)
	    {
	      /* Handle the pushed elements now.  */
	      size_t backw;

	      backw = idxcnt;
	      while (backw > backw_stop)
		{
		  size_t len = weights[idxarr[--backw]++];

		  if (needed + len < n)
		    while (len-- > 0)
		      dest[needed++] = weights[idxarr[backw]++];
		  else
		    {
		      /* No more characters fit into the buffer.  */
		      needed += len;
		      idxarr[backw] += len;
		    }
		}
	    }
	}
      else
	{
	  int val = 1;
#ifndef WIDE_CHAR_VERSION
	  char buf[7];
	  size_t buflen;
#endif
	  size_t i;

	  for (idxcnt = 0; idxcnt < idxmax; ++idxcnt)
	    {
	      if ((rule & sort_forward) != 0)
		{
		  size_t len;

		  if (backw_stop != ~0ul)
		    {
		     /* Handle the pushed elements now.  */
		      size_t backw;

		      for (backw = idxcnt - 1; backw >= backw_stop; --backw)
			{
			  len = weights[idxarr[backw]++];
			  if (len != 0)
			    {
#ifdef WIDE_CHAR_VERSION
			      if (needed + 1 + len < n)
				{
				  dest[needed] = val;
				  for (i = 0; i < len; ++i)
				    dest[needed + 1 + i] =
				      weights[idxarr[backw] + i];
				}
			      needed += 1 + len;
#else
			      buflen = utf8_encode (buf, val);
			      if (needed + buflen + len < n)
				{
				  for (i = 0; i < buflen; ++i)
				    dest[needed + i] = buf[i];
				  for (i = 0; i < len; ++i)
				    dest[needed + buflen + i] =
				      weights[idxarr[backw] + i];
				}
			      needed += buflen + len;
#endif
			      idxarr[backw] += len;
			      val = 1;
			    }
			  else
			    ++val;
			}

		      backw_stop = ~0ul;
		    }

		  /* Now handle the forward element.  */
		  len = weights[idxarr[idxcnt]++];
		  if (len != 0)
		    {
#ifdef WIDE_CHAR_VERSION
		      if (needed + 1+ len < n)
			{
			  dest[needed] = val;
			  for (i = 0; i < len; ++i)
			    dest[needed + 1 + i] =
			      weights[idxarr[idxcnt] + i];
			}
		      needed += 1 + len;
#else
		      buflen = utf8_encode (buf, val);
		      if (needed + buflen + len < n)
			{
			  for (i = 0; i < buflen; ++i)
			    dest[needed + i] = buf[i];
			  for (i = 0; i < len; ++i)
			    dest[needed + buflen + i] =
			      weights[idxarr[idxcnt] + i];
			}
		      needed += buflen + len;
#endif
		      idxarr[idxcnt] += len;
		      val = 1;
		    }
		  else
		    /* Note that we don't have to increment `idxarr[idxcnt]'
		       since the length is zero.  */
		    ++val;
		}
	      else
		{
		  /* Remember where the backwards series started.  */
		  if (backw_stop == ~0ul)
		    backw_stop = idxcnt;
		}

	      rule = rulesets[rulearr[idxcnt + 1] * nrules + pass];
	    }

	  if (backw_stop != ~0)
	    {
	      /* Handle the pushed elements now.  */
	      size_t backw;

	      backw = idxmax - 1;
	      while (backw > backw_stop)
		{
		  size_t len = weights[idxarr[--backw]++];
		  if (len != 0)
		    {
#ifdef WIDE_CHAR_VERSION
		      if (needed + 1 + len < n)
			{
			  dest[needed] = val;
			  for (i = 0; i < len; ++i)
			    dest[needed + 1 + i] =
			      weights[idxarr[backw] + i];
			}
		      needed += 1 + len;
#else
		      buflen = utf8_encode (buf, val);
		      if (needed + buflen + len < n)
			{
			  for (i = 0; i < buflen; ++i)
			    dest[needed + i] = buf[i];
			  for (i = 0; i < len; ++i)
			    dest[needed + buflen + i] =
			      weights[idxarr[backw] + i];
			}
		      needed += buflen + len;
#endif
		      idxarr[backw] += len;
		      val = 1;
		    }
		  else
		    ++val;
		}
	    }
	}

      /* Finally store the byte to separate the passes or terminate
	 the string.  */
      if (needed < n)
	dest[needed] = pass + 1 < nrules ? L('\1') : L('\0');
      ++needed;
    }

  /* This is a little optimization: many collation specifications have
     a `position' rule at the end and if no non-ignored character
     is found the last \1 byte is immediately followed by a \0 byte
     signalling this.  We can avoid the \1 byte(s).  */
  if (needed <= n && needed > 2 && dest[needed - 2] == L('\1'))
    {
      /* Remove the \1 byte.  */
      --needed;
      dest[needed - 1] = L('\0');
    }

  /* Free the memory if needed.  */
  if (use_malloc)
    free (idxarr);

  return needed;
}