about summary refs log tree commit diff
path: root/string/strxfrm.c
blob: 820ad9394707b1367e4e2884920d34f228d54058 (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
/* Copyright (C) 1995, 1996 Free Software Foundation, Inc.
   This file is part of the GNU C Library.
   Written by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 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 <stddef.h>
#include <stdlib.h>
#include <string.h>

#ifndef WIDE_VERSION
# define STRING_TYPE char
# define USTRING_TYPE unsigned char
# define L_(Ch) Ch
# define STRXFRM strxfrm
# define STRLEN strlen
# define STPNCPY __stpncpy
#endif

/* Include the shared helper functions.  `strxfrm'/`wcsxfrm' also use
   these functions.  */
#include "../locale/weight.h"


#ifndef WIDE_VERSION
/* Write 32 bit value UTF-8 encoded but only if enough space is left.  */
static __inline size_t
print_val (u_int32_t value, char *dest, size_t max, size_t act)
{
  char tmp[6];
  int idx = 0;

  if (value < 0x80)
    tmp[idx++] = (char) value;
  else
    {
      tmp[idx++] = '\x80' + (char) (value & 0x3f);
      value >>= 6;

      if (value < 0x20)
	tmp[idx++] = '\xc0' + (char) value;
      else
	{
	  tmp[idx++] = '\x80' + (char) (value & 0x3f);
	  value >>= 6;

	  if (value < 0x10)
	    tmp[idx++] = '\xe0' + (char) value;
	  else
	    {
	      tmp[idx++] = '\x80' + (char) (value & 0x3f);
	      value >>= 6;

	      if (value < 0x08)
		tmp[idx++] = '\xf0' + (char) value;
	      else
		{
		  tmp[idx++] = '\x80' + (char) (value & 0x3f);
		  value >>= 6;

		  if (value < 0x04)
		    tmp[idx++] = '\xf8' + (char) value;
		  else
		    {
		      tmp[idx++] = '\x80' + (char) (value & 0x3f);
		      tmp[idx++] = '\xfc' + (char) (value >> 6);
		    }
		}
	    }
	}
    }

  while (idx-- > 0)
    {
      if (act < max)
	dest[act] = tmp[idx];
      ++act;
    }

  return act;
}
#else
static __inline size_t
print_val (u_int32_t value, wchar_t *dest, size_t max, size_t act)
{
  /* We cannot really assume wchar_t is 32 bits wide.  But it is for
     GCC and so we don't do much optimization for the other case.  */
  if (sizeof (wchar_t) == 4)
    {
      if (act < max)
	dest[act] = (wchar_t) value;
      ++act;
    }
  else
    {
      wchar_t tmp[3];
      size_t idx = 0;

      if (value < 0x8000)
	tmp[idx++] = (wchar_t) act;
      else
	{
	  tmp[idx++] = (wchar_t) (0x8000 + (value & 0x3fff));
	  value >>= 14;
	  if (value < 0x2000)
	    tmp[idx++] = (wchar_t) (0xc000 + value);
	  else
	    {
	      tmp[idx++] = (wchar_t) (0x8000 + (value & 0x3fff));
	      value >>= 14;
	      tmp[idx++] = (wchar_t) (0xe000 + value);
	    }
	}
      while (idx-- > 0)
	{
	  if (act < max)
	    dest[act] = tmp[idx];
	  ++act;
	}
    }
  return act;
}
#endif


/* Transform SRC into a form such that the result of strcmp
   on two strings that have been transformed by strxfrm is
   the same as the result of strcoll on the two strings before
   their transformation.  The transformed string is put in at
   most N characters of DEST and its length is returned.  */
size_t
STRXFRM (STRING_TYPE *dest, const STRING_TYPE *src, size_t n)
{
  weight_t *forw = NULL;
  weight_t *backw = NULL;
  size_t pass;
  size_t written;

  /* If the current locale does not specify locale data we use normal
     8-bit string comparison.  */
  if (collate_nrules == 0)
    {
      if (n != 0)
	STPNCPY (dest, src, n);

      return STRLEN (src);
    }

  /* Get full information about the string.  This means we get
     information for all passes in a special data structure.  */
  get_string (src, forw, backw);

  /* Now we have all the information.  In at most the given number of
     passes we can finally decide about the order.  */
  written = 0;
  for (pass = 0; pass < collate_nrules; ++pass)
    {
      int forward = (collate_rules[pass] & sort_forward) != 0;
      const weight_t *run = forward ? forw : backw;
      int idx = forward ? 0 : run->data[pass].number - 1;

      do
	{
	  int ignore = 0;
	  u_int32_t w;

	  /* Here we have to check for IGNORE entries.  If these are
	     found we count them and go on with he next value.  */
	  while ((w = run->data[pass].value[idx]) == (u_int32_t) IGNORE_CHAR)
	    {
	      ++ignore;
	      if ((forward && ++idx >= run->data[pass].number)
		  || (!forward && --idx < 0))
		{
		  weight_t *nextp = forward ? run->next : run->prev;
		  if (nextp == NULL)
		    {
		      w = 0;
		      break;
		    }
		  run = nextp;
		  idx = forward ? 0 : run->data[pass].number - 1;
		}
	    }

	  /* Now we have information of the number of ignored weights
	     and the value of the next weight.  We have to add 2
	     because 0 means EOS and 1 is the intermediate string end.  */
	  if ((collate_rules[pass] & sort_position) != 0)
	    written = print_val (ignore + 2, dest, n, written);

	  if (w != 0)
	    written = print_val (w, dest, n, written);

	  /* We have to increment the index counters.  */
	  if ((forward && ++idx >= run->data[pass].number)
	      || (!forward && --idx < 0))
	    if (forward)
	      {
		run = run->next;
		idx = 0;
	      }
	    else
	      {
		run = run->prev;
		if (run != NULL)
		  idx = run->data[pass].number - 1;
	      }
	}
      while (run != NULL);

      /* Write marker for end of word.  */
      if (pass + 1 < collate_nrules)
	written = print_val (1, dest, n, written);
    }

  /* Terminate string.  */
  if (written < n)
    dest[written] = L_('\0');

  /* Return length without counting the terminating '\0'.  */
  return written;
}