Subversion Repositories HelenOS

Rev

Rev 4581 | Details | Compare with Previous | Last modification | View Log | RSS feed

Rev Author Line No. Line
999 palkovsky 1
/*
2071 jermar 2
 * Copyright (c) 2005 Martin Decky
4153 mejdrech 3
 * Copyright (c) 2008 Jiri Svoboda
999 palkovsky 4
 * All rights reserved.
5
 *
6
 * Redistribution and use in source and binary forms, with or without
7
 * modification, are permitted provided that the following conditions
8
 * are met:
9
 *
10
 * - Redistributions of source code must retain the above copyright
11
 *   notice, this list of conditions and the following disclaimer.
12
 * - Redistributions in binary form must reproduce the above copyright
13
 *   notice, this list of conditions and the following disclaimer in the
14
 *   documentation and/or other materials provided with the distribution.
15
 * - The name of the author may not be used to endorse or promote products
16
 *   derived from this software without specific prior written permission.
17
 *
18
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
19
 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
20
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
21
 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
22
 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
23
 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
24
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
25
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
26
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
27
 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
28
 */
29
 
1719 decky 30
/** @addtogroup libc
1653 cejka 31
 * @{
32
 */
33
/** @file
34
 */
35
 
999 palkovsky 36
#include <string.h>
4153 mejdrech 37
#include <stdlib.h>
4327 mejdrech 38
#include <assert.h>
4153 mejdrech 39
#include <limits.h>
1314 cejka 40
#include <ctype.h>
3238 jermar 41
#include <malloc.h>
4263 mejdrech 42
#include <errno.h>
43
#include <align.h>
4327 mejdrech 44
#include <mem.h>
4263 mejdrech 45
#include <string.h>
999 palkovsky 46
 
4263 mejdrech 47
/** Byte mask consisting of lowest @n bits (out of 8) */
48
#define LO_MASK_8(n)  ((uint8_t) ((1 << (n)) - 1))
49
 
50
/** Byte mask consisting of lowest @n bits (out of 32) */
51
#define LO_MASK_32(n)  ((uint32_t) ((1 << (n)) - 1))
52
 
53
/** Byte mask consisting of highest @n bits (out of 8) */
54
#define HI_MASK_8(n)  (~LO_MASK_8(8 - (n)))
55
 
56
/** Number of data bits in a UTF-8 continuation byte */
57
#define CONT_BITS  6
58
 
59
/** Decode a single character from a string.
60
 *
61
 * Decode a single character from a string of size @a size. Decoding starts
62
 * at @a offset and this offset is moved to the beginning of the next
63
 * character. In case of decoding error, offset generally advances at least
64
 * by one. However, offset is never moved beyond size.
65
 *
66
 * @param str    String (not necessarily NULL-terminated).
67
 * @param offset Byte offset in string where to start decoding.
68
 * @param size   Size of the string (in bytes).
69
 *
70
 * @return Value of decoded character, U_SPECIAL on decoding error or
71
 *         NULL if attempt to decode beyond @a size.
72
 *
73
 */
74
wchar_t str_decode(const char *str, size_t *offset, size_t size)
75
{
76
    if (*offset + 1 > size)
77
        return 0;
78
 
79
    /* First byte read from string */
80
    uint8_t b0 = (uint8_t) str[(*offset)++];
81
 
82
    /* Determine code length */
83
 
84
    unsigned int b0_bits;  /* Data bits in first byte */
85
    unsigned int cbytes;   /* Number of continuation bytes */
86
 
87
    if ((b0 & 0x80) == 0) {
88
        /* 0xxxxxxx (Plain ASCII) */
89
        b0_bits = 7;
90
        cbytes = 0;
91
    } else if ((b0 & 0xe0) == 0xc0) {
92
        /* 110xxxxx 10xxxxxx */
93
        b0_bits = 5;
94
        cbytes = 1;
95
    } else if ((b0 & 0xf0) == 0xe0) {
96
        /* 1110xxxx 10xxxxxx 10xxxxxx */
97
        b0_bits = 4;
98
        cbytes = 2;
99
    } else if ((b0 & 0xf8) == 0xf0) {
100
        /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
101
        b0_bits = 3;
102
        cbytes = 3;
103
    } else {
104
        /* 10xxxxxx -- unexpected continuation byte */
105
        return U_SPECIAL;
106
    }
107
 
108
    if (*offset + cbytes > size)
109
        return U_SPECIAL;
110
 
111
    wchar_t ch = b0 & LO_MASK_8(b0_bits);
112
 
113
    /* Decode continuation bytes */
114
    while (cbytes > 0) {
115
        uint8_t b = (uint8_t) str[(*offset)++];
116
 
117
        /* Must be 10xxxxxx */
118
        if ((b & 0xc0) != 0x80)
119
            return U_SPECIAL;
120
 
121
        /* Shift data bits to ch */
122
        ch = (ch << CONT_BITS) | (wchar_t) (b & LO_MASK_8(CONT_BITS));
123
        cbytes--;
124
    }
125
 
126
    return ch;
127
}
128
 
129
/** Encode a single character to string representation.
130
 *
131
 * Encode a single character to string representation (i.e. UTF-8) and store
132
 * it into a buffer at @a offset. Encoding starts at @a offset and this offset
133
 * is moved to the position where the next character can be written to.
134
 *
135
 * @param ch     Input character.
136
 * @param str    Output buffer.
137
 * @param offset Byte offset where to start writing.
138
 * @param size   Size of the output buffer (in bytes).
139
 *
140
 * @return EOK if the character was encoded successfully, EOVERFLOW if there
4581 mejdrech 141
 *         was not enough space in the output buffer or EINVAL if the character
142
 *         code was invalid.
4263 mejdrech 143
 */
144
int chr_encode(const wchar_t ch, char *str, size_t *offset, size_t size)
145
{
146
    if (*offset >= size)
147
        return EOVERFLOW;
148
 
149
    if (!chr_check(ch))
150
        return EINVAL;
151
 
152
    /* Unsigned version of ch (bit operations should only be done
153
       on unsigned types). */
154
    uint32_t cc = (uint32_t) ch;
155
 
156
    /* Determine how many continuation bytes are needed */
157
 
158
    unsigned int b0_bits;  /* Data bits in first byte */
159
    unsigned int cbytes;   /* Number of continuation bytes */
160
 
161
    if ((cc & ~LO_MASK_32(7)) == 0) {
162
        b0_bits = 7;
163
        cbytes = 0;
164
    } else if ((cc & ~LO_MASK_32(11)) == 0) {
165
        b0_bits = 5;
166
        cbytes = 1;
167
    } else if ((cc & ~LO_MASK_32(16)) == 0) {
168
        b0_bits = 4;
169
        cbytes = 2;
170
    } else if ((cc & ~LO_MASK_32(21)) == 0) {
171
        b0_bits = 3;
172
        cbytes = 3;
173
    } else {
174
        /* Codes longer than 21 bits are not supported */
175
        return EINVAL;
176
    }
177
 
178
    /* Check for available space in buffer */
179
    if (*offset + cbytes >= size)
180
        return EOVERFLOW;
181
 
182
    /* Encode continuation bytes */
183
    unsigned int i;
184
    for (i = cbytes; i > 0; i--) {
185
        str[*offset + i] = 0x80 | (cc & LO_MASK_32(CONT_BITS));
186
        cc = cc >> CONT_BITS;
187
    }
188
 
189
    /* Encode first byte */
190
    str[*offset] = (cc & LO_MASK_32(b0_bits)) | HI_MASK_8(8 - b0_bits - 1);
191
 
192
    /* Advance offset */
193
    *offset += cbytes + 1;
194
 
195
    return EOK;
196
}
197
 
198
/** Get size of string.
199
 *
200
 * Get the number of bytes which are used by the string @a str (excluding the
201
 * NULL-terminator).
202
 *
203
 * @param str String to consider.
204
 *
205
 * @return Number of bytes used by the string
206
 *
207
 */
208
size_t str_size(const char *str)
209
{
210
    size_t size = 0;
211
 
212
    while (*str++ != 0)
213
        size++;
214
 
215
    return size;
216
}
217
 
218
/** Get size of wide string.
219
 *
220
 * Get the number of bytes which are used by the wide string @a str (excluding the
221
 * NULL-terminator).
222
 *
223
 * @param str Wide string to consider.
224
 *
225
 * @return Number of bytes used by the wide string
226
 *
227
 */
228
size_t wstr_size(const wchar_t *str)
229
{
230
    return (wstr_length(str) * sizeof(wchar_t));
231
}
232
 
233
/** Get size of string with length limit.
234
 *
235
 * Get the number of bytes which are used by up to @a max_len first
236
 * characters in the string @a str. If @a max_len is greater than
237
 * the length of @a str, the entire string is measured (excluding the
238
 * NULL-terminator).
239
 *
240
 * @param str     String to consider.
241
 * @param max_len Maximum number of characters to measure.
242
 *
243
 * @return Number of bytes used by the characters.
244
 *
245
 */
4581 mejdrech 246
size_t str_lsize(const char *str, size_t max_len)
4263 mejdrech 247
{
4581 mejdrech 248
    size_t len = 0;
4263 mejdrech 249
    size_t offset = 0;
250
 
251
    while (len < max_len) {
252
        if (str_decode(str, &offset, STR_NO_LIMIT) == 0)
253
            break;
254
 
255
        len++;
256
    }
257
 
258
    return offset;
259
}
260
 
261
/** Get size of wide string with length limit.
262
 *
263
 * Get the number of bytes which are used by up to @a max_len first
264
 * wide characters in the wide string @a str. If @a max_len is greater than
265
 * the length of @a str, the entire wide string is measured (excluding the
266
 * NULL-terminator).
267
 *
268
 * @param str     Wide string to consider.
269
 * @param max_len Maximum number of wide characters to measure.
270
 *
271
 * @return Number of bytes used by the wide characters.
272
 *
273
 */
4581 mejdrech 274
size_t wstr_lsize(const wchar_t *str, size_t max_len)
4263 mejdrech 275
{
276
    return (wstr_nlength(str, max_len * sizeof(wchar_t)) * sizeof(wchar_t));
277
}
278
 
279
/** Get number of characters in a string.
280
 *
281
 * @param str NULL-terminated string.
282
 *
283
 * @return Number of characters in string.
284
 *
285
 */
4581 mejdrech 286
size_t str_length(const char *str)
4263 mejdrech 287
{
4581 mejdrech 288
    size_t len = 0;
4263 mejdrech 289
    size_t offset = 0;
290
 
291
    while (str_decode(str, &offset, STR_NO_LIMIT) != 0)
292
        len++;
293
 
294
    return len;
295
}
296
 
297
/** Get number of characters in a wide string.
298
 *
299
 * @param str NULL-terminated wide string.
300
 *
301
 * @return Number of characters in @a str.
302
 *
303
 */
4581 mejdrech 304
size_t wstr_length(const wchar_t *wstr)
4263 mejdrech 305
{
4581 mejdrech 306
    size_t len = 0;
4263 mejdrech 307
 
308
    while (*wstr++ != 0)
309
        len++;
310
 
311
    return len;
312
}
313
 
314
/** Get number of characters in a string with size limit.
315
 *
316
 * @param str  NULL-terminated string.
317
 * @param size Maximum number of bytes to consider.
318
 *
319
 * @return Number of characters in string.
320
 *
321
 */
4581 mejdrech 322
size_t str_nlength(const char *str, size_t size)
4263 mejdrech 323
{
4581 mejdrech 324
    size_t len = 0;
4263 mejdrech 325
    size_t offset = 0;
326
 
327
    while (str_decode(str, &offset, size) != 0)
328
        len++;
329
 
330
    return len;
331
}
332
 
333
/** Get number of characters in a string with size limit.
334
 *
335
 * @param str  NULL-terminated string.
336
 * @param size Maximum number of bytes to consider.
337
 *
338
 * @return Number of characters in string.
339
 *
340
 */
4581 mejdrech 341
size_t wstr_nlength(const wchar_t *str, size_t size)
4263 mejdrech 342
{
4581 mejdrech 343
    size_t len = 0;
344
    size_t limit = ALIGN_DOWN(size, sizeof(wchar_t));
345
    size_t offset = 0;
4263 mejdrech 346
 
347
    while ((offset < limit) && (*str++ != 0)) {
348
        len++;
349
        offset += sizeof(wchar_t);
350
    }
351
 
352
    return len;
353
}
354
 
355
/** Check whether character is plain ASCII.
356
 *
357
 * @return True if character is plain ASCII.
358
 *
359
 */
360
bool ascii_check(wchar_t ch)
361
{
362
    if ((ch >= 0) && (ch <= 127))
363
        return true;
364
 
365
    return false;
366
}
367
 
368
/** Check whether character is valid
369
 *
370
 * @return True if character is a valid Unicode code point.
371
 *
372
 */
373
bool chr_check(wchar_t ch)
374
{
375
    if ((ch >= 0) && (ch <= 1114111))
376
        return true;
377
 
378
    return false;
379
}
380
 
381
/** Compare two NULL terminated strings.
382
 *
383
 * Do a char-by-char comparison of two NULL-terminated strings.
384
 * The strings are considered equal iff they consist of the same
385
 * characters on the minimum of their lengths.
386
 *
387
 * @param s1 First string to compare.
388
 * @param s2 Second string to compare.
389
 *
390
 * @return 0 if the strings are equal, -1 if first is smaller,
391
 *         1 if second smaller.
392
 *
393
 */
394
int str_cmp(const char *s1, const char *s2)
395
{
396
    wchar_t c1 = 0;
397
    wchar_t c2 = 0;
398
 
399
    size_t off1 = 0;
400
    size_t off2 = 0;
401
 
402
    while (true) {
403
        c1 = str_decode(s1, &off1, STR_NO_LIMIT);
404
        c2 = str_decode(s2, &off2, STR_NO_LIMIT);
405
 
406
        if (c1 < c2)
407
            return -1;
408
 
409
        if (c1 > c2)
410
            return 1;
411
 
412
        if (c1 == 0 || c2 == 0)
413
            break;     
414
    }
415
 
416
    return 0;
417
}
418
 
419
/** Compare two NULL terminated strings with length limit.
420
 *
421
 * Do a char-by-char comparison of two NULL-terminated strings.
422
 * The strings are considered equal iff they consist of the same
423
 * characters on the minimum of their lengths and the length limit.
424
 *
425
 * @param s1      First string to compare.
426
 * @param s2      Second string to compare.
427
 * @param max_len Maximum number of characters to consider.
428
 *
429
 * @return 0 if the strings are equal, -1 if first is smaller,
430
 *         1 if second smaller.
431
 *
432
 */
4581 mejdrech 433
int str_lcmp(const char *s1, const char *s2, size_t max_len)
4263 mejdrech 434
{
435
    wchar_t c1 = 0;
436
    wchar_t c2 = 0;
437
 
438
    size_t off1 = 0;
439
    size_t off2 = 0;
440
 
4581 mejdrech 441
    size_t len = 0;
4263 mejdrech 442
 
443
    while (true) {
444
        if (len >= max_len)
445
            break;
446
 
447
        c1 = str_decode(s1, &off1, STR_NO_LIMIT);
448
        c2 = str_decode(s2, &off2, STR_NO_LIMIT);
449
 
450
        if (c1 < c2)
451
            return -1;
452
 
453
        if (c1 > c2)
454
            return 1;
455
 
456
        if (c1 == 0 || c2 == 0)
457
            break;
458
 
459
        ++len; 
460
    }
461
 
462
    return 0;
463
 
464
}
465
 
4327 mejdrech 466
/** Copy string.
4263 mejdrech 467
 *
4327 mejdrech 468
 * Copy source string @a src to destination buffer @a dest.
469
 * No more than @a size bytes are written. If the size of the output buffer
470
 * is at least one byte, the output string will always be well-formed, i.e.
471
 * null-terminated and containing only complete characters.
4263 mejdrech 472
 *
4327 mejdrech 473
 * @param dst   Destination buffer.
474
 * @param count Size of the destination buffer (must be > 0).
4263 mejdrech 475
 * @param src   Source string.
4327 mejdrech 476
 */
477
void str_cpy(char *dest, size_t size, const char *src)
478
{
479
    wchar_t ch;
480
    size_t src_off;
481
    size_t dest_off;
482
 
483
    /* There must be space for a null terminator in the buffer. */
484
    assert(size > 0);
485
 
486
    src_off = 0;
487
    dest_off = 0;
488
 
489
    while ((ch = str_decode(src, &src_off, STR_NO_LIMIT)) != 0) {
490
        if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
491
            break;
492
    }
493
 
494
    dest[dest_off] = '\0';
495
}
496
 
497
/** Copy size-limited substring.
498
 *
499
 * Copy prefix of string @a src of max. size @a size to destination buffer
500
 * @a dest. No more than @a size bytes are written. The output string will
501
 * always be well-formed, i.e. null-terminated and containing only complete
502
 * characters.
503
 *
504
 * No more than @a n bytes are read from the input string, so it does not
505
 * have to be null-terminated.
506
 *
4263 mejdrech 507
 * @param dst   Destination buffer.
4327 mejdrech 508
 * @param count Size of the destination buffer (must be > 0).
509
 * @param src   Source string.
510
 * @param n Maximum number of bytes to read from @a src.
4263 mejdrech 511
 */
4327 mejdrech 512
void str_ncpy(char *dest, size_t size, const char *src, size_t n)
4263 mejdrech 513
{
514
    wchar_t ch;
4327 mejdrech 515
    size_t src_off;
516
    size_t dest_off;
517
 
518
    /* There must be space for a null terminator in the buffer. */
519
    assert(size > 0);
4263 mejdrech 520
 
4327 mejdrech 521
    src_off = 0;
522
    dest_off = 0;
523
 
524
    while ((ch = str_decode(src, &src_off, n)) != 0) {
525
        if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
4263 mejdrech 526
            break;
527
    }
4327 mejdrech 528
 
529
    dest[dest_off] = '\0';
4263 mejdrech 530
}
531
 
4327 mejdrech 532
/** Append one string to another.
533
 *
534
 * Append source string @a src to string in destination buffer @a dest.
535
 * Size of the destination buffer is @a dest. If the size of the output buffer
536
 * is at least one byte, the output string will always be well-formed, i.e.
537
 * null-terminated and containing only complete characters.
538
 *
539
 * @param dst   Destination buffer.
540
 * @param count Size of the destination buffer.
541
 * @param src   Source string.
542
 */
543
void str_append(char *dest, size_t size, const char *src)
544
{
545
    size_t dstr_size;
546
 
547
    dstr_size = str_size(dest);
548
    str_cpy(dest + dstr_size, size - dstr_size, src);
549
}
550
 
4263 mejdrech 551
/** Copy NULL-terminated wide string to string
552
 *
553
 * Copy source wide string @a src to destination buffer @a dst.
554
 * No more than @a size bytes are written. NULL-terminator is always
555
 * written after the last succesfully copied character (i.e. if the
556
 * destination buffer is has at least 1 byte, it will be always
557
 * NULL-terminated).
558
 *
559
 * @param src   Source wide string.
560
 * @param dst   Destination buffer.
561
 * @param count Size of the destination buffer.
562
 *
563
 */
564
void wstr_nstr(char *dst, const wchar_t *src, size_t size)
565
{
566
    /* No space for the NULL-terminator in the buffer */
567
    if (size == 0)
568
        return;
569
 
570
    wchar_t ch;
4581 mejdrech 571
    size_t src_idx = 0;
4263 mejdrech 572
    size_t dst_off = 0;
573
 
574
    while ((ch = src[src_idx++]) != 0) {
575
        if (chr_encode(ch, dst, &dst_off, size) != EOK)
576
            break;
577
    }
578
 
579
    if (dst_off >= size)
580
        dst[size - 1] = 0;
581
    else
582
        dst[dst_off] = 0;
583
}
584
 
585
/** Find first occurence of character in string.
586
 *
587
 * @param str String to search.
588
 * @param ch  Character to look for.
589
 *
590
 * @return Pointer to character in @a str or NULL if not found.
591
 */
4718 mejdrech 592
char *str_chr(const char *str, wchar_t ch)
4263 mejdrech 593
{
594
    wchar_t acc;
595
    size_t off = 0;
4327 mejdrech 596
    size_t last = 0;
4263 mejdrech 597
 
598
    while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
599
        if (acc == ch)
4718 mejdrech 600
            return (char *) (str + last);
4327 mejdrech 601
        last = off;
4263 mejdrech 602
    }
603
 
604
    return NULL;
605
}
606
 
4327 mejdrech 607
/** Find last occurence of character in string.
608
 *
609
 * @param str String to search.
610
 * @param ch  Character to look for.
611
 *
612
 * @return Pointer to character in @a str or NULL if not found.
613
 */
4718 mejdrech 614
char *str_rchr(const char *str, wchar_t ch)
4327 mejdrech 615
{
616
    wchar_t acc;
617
    size_t off = 0;
618
    size_t last = 0;
4581 mejdrech 619
    const char *res = NULL;
4327 mejdrech 620
 
621
    while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
622
        if (acc == ch)
623
            res = (str + last);
624
        last = off;
625
    }
626
 
4718 mejdrech 627
    return (char *) res;
4327 mejdrech 628
}
629
 
4263 mejdrech 630
/** Insert a wide character into a wide string.
631
 *
632
 * Insert a wide character into a wide string at position
633
 * @a pos. The characters after the position are shifted.
634
 *
635
 * @param str     String to insert to.
636
 * @param ch      Character to insert to.
637
 * @param pos     Character index where to insert.
638
 @ @param max_pos Characters in the buffer.
639
 *
640
 * @return True if the insertion was sucessful, false if the position
641
 *         is out of bounds.
642
 *
643
 */
4581 mejdrech 644
bool wstr_linsert(wchar_t *str, wchar_t ch, size_t pos, size_t max_pos)
4263 mejdrech 645
{
4581 mejdrech 646
    size_t len = wstr_length(str);
4263 mejdrech 647
 
648
    if ((pos > len) || (pos + 1 > max_pos))
649
        return false;
650
 
4581 mejdrech 651
    size_t i;
4263 mejdrech 652
    for (i = len; i + 1 > pos; i--)
653
        str[i + 1] = str[i];
654
 
655
    str[pos] = ch;
656
 
657
    return true;
658
}
659
 
660
/** Remove a wide character from a wide string.
661
 *
662
 * Remove a wide character from a wide string at position
663
 * @a pos. The characters after the position are shifted.
664
 *
665
 * @param str String to remove from.
666
 * @param pos Character index to remove.
667
 *
668
 * @return True if the removal was sucessful, false if the position
669
 *         is out of bounds.
670
 *
671
 */
4581 mejdrech 672
bool wstr_remove(wchar_t *str, size_t pos)
4263 mejdrech 673
{
4581 mejdrech 674
    size_t len = wstr_length(str);
4263 mejdrech 675
 
676
    if (pos >= len)
677
        return false;
678
 
4581 mejdrech 679
    size_t i;
4263 mejdrech 680
    for (i = pos + 1; i <= len; i++)
681
        str[i - 1] = str[i];
682
 
683
    return true;
684
}
685
 
3271 jermar 686
int stricmp(const char *a, const char *b)
687
{
688
    int c = 0;
689
 
690
    while (a[c] && b[c] && (!(tolower(a[c]) - tolower(b[c]))))
691
        c++;
692
 
693
    return (tolower(a[c]) - tolower(b[c]));
694
}
695
 
1314 cejka 696
/** Convert string to a number.
697
 * Core of strtol and strtoul functions.
3261 jermar 698
 *
699
 * @param nptr      Pointer to string.
700
 * @param endptr    If not NULL, function stores here pointer to the first
701
 *          invalid character.
702
 * @param base      Zero or number between 2 and 36 inclusive.
703
 * @param sgn       It's set to 1 if minus found.
704
 * @return      Result of conversion.
1314 cejka 705
 */
3261 jermar 706
static unsigned long
707
_strtoul(const char *nptr, char **endptr, int base, char *sgn)
1314 cejka 708
{
709
    unsigned char c;
710
    unsigned long result = 0;
711
    unsigned long a, b;
712
    const char *str = nptr;
713
    const char *tmpptr;
714
 
715
    while (isspace(*str))
716
        str++;
717
 
718
    if (*str == '-') {
719
        *sgn = 1;
720
        ++str;
721
    } else if (*str == '+')
722
        ++str;
723
 
724
    if (base) {
725
        if ((base == 1) || (base > 36)) {
726
            /* FIXME: set errno to EINVAL */
727
            return 0;
728
        }
3261 jermar 729
        if ((base == 16) && (*str == '0') && ((str[1] == 'x') ||
730
            (str[1] == 'X'))) {
1314 cejka 731
            str += 2;
732
        }
733
    } else {
734
        base = 10;
735
 
736
        if (*str == '0') {
737
            base = 8;
738
            if ((str[1] == 'X') || (str[1] == 'x'))  {
739
                base = 16;
740
                str += 2;
741
            }
742
        }
743
    }
744
 
745
    tmpptr = str;
746
 
747
    while (*str) {
748
        c = *str;
3261 jermar 749
        c = (c >= 'a' ? c - 'a' + 10 : (c >= 'A' ? c - 'A' + 10 :
750
            (c <= '9' ? c - '0' : 0xff)));
1314 cejka 751
        if (c > base) {
752
            break;
753
        }
754
 
755
        a = (result & 0xff) * base + c;
756
        b = (result >> 8) * base + (a >> 8);
757
 
758
        if (b > (ULONG_MAX >> 8)) {
759
            /* overflow */
760
            /* FIXME: errno = ERANGE*/
761
            return ULONG_MAX;
762
        }
763
 
764
        result = (b << 8) + (a & 0xff);
765
        ++str;
766
    }
767
 
768
    if (str == tmpptr) {
3261 jermar 769
        /*
770
         * No number was found => first invalid character is the first
771
         * character of the string.
772
         */
1314 cejka 773
        /* FIXME: set errno to EINVAL */
774
        str = nptr;
775
        result = 0;
776
    }
777
 
778
    if (endptr)
1719 decky 779
        *endptr = (char *) str;
1314 cejka 780
 
781
    if (nptr == str) {
782
        /*FIXME: errno = EINVAL*/
783
        return 0;
784
    }
785
 
786
    return result;
787
}
788
 
789
/** Convert initial part of string to long int according to given base.
3261 jermar 790
 * The number may begin with an arbitrary number of whitespaces followed by
791
 * optional sign (`+' or `-'). If the base is 0 or 16, the prefix `0x' may be
792
 * inserted and the number will be taken as hexadecimal one. If the base is 0
793
 * and the number begin with a zero, number will be taken as octal one (as with
794
 * base 8). Otherwise the base 0 is taken as decimal.
795
 *
796
 * @param nptr      Pointer to string.
797
 * @param endptr    If not NULL, function stores here pointer to the first
798
 *          invalid character.
799
 * @param base      Zero or number between 2 and 36 inclusive.
800
 * @return      Result of conversion.
1314 cejka 801
 */
802
long int strtol(const char *nptr, char **endptr, int base)
803
{
804
    char sgn = 0;
805
    unsigned long number = 0;
806
 
807
    number = _strtoul(nptr, endptr, base, &sgn);
808
 
809
    if (number > LONG_MAX) {
1719 decky 810
        if ((sgn) && (number == (unsigned long) (LONG_MAX) + 1)) {
1314 cejka 811
            /* FIXME: set 0 to errno */
812
            return number;     
813
        }
814
        /* FIXME: set ERANGE to errno */
1719 decky 815
        return (sgn ? LONG_MIN : LONG_MAX);
1314 cejka 816
    }
817
 
1719 decky 818
    return (sgn ? -number : number);
1314 cejka 819
}
820
 
821
 
822
/** Convert initial part of string to unsigned long according to given base.
3261 jermar 823
 * The number may begin with an arbitrary number of whitespaces followed by
824
 * optional sign (`+' or `-'). If the base is 0 or 16, the prefix `0x' may be
825
 * inserted and the number will be taken as hexadecimal one. If the base is 0
826
 * and the number begin with a zero, number will be taken as octal one (as with
827
 * base 8). Otherwise the base 0 is taken as decimal.
828
 *
829
 * @param nptr      Pointer to string.
830
 * @param endptr    If not NULL, function stores here pointer to the first
831
 *          invalid character
832
 * @param base      Zero or number between 2 and 36 inclusive.
833
 * @return      Result of conversion.
1314 cejka 834
 */
835
unsigned long strtoul(const char *nptr, char **endptr, int base)
836
{
837
    char sgn = 0;
838
    unsigned long number = 0;
839
 
840
    number = _strtoul(nptr, endptr, base, &sgn);
841
 
1719 decky 842
    return (sgn ? -number : number);
1314 cejka 843
}
1472 palkovsky 844
 
4327 mejdrech 845
char *str_dup(const char *src)
1472 palkovsky 846
{
4327 mejdrech 847
    size_t size = str_size(src);
848
    void *dest = malloc(size + 1);
1472 palkovsky 849
 
4327 mejdrech 850
    if (dest == NULL)
3238 jermar 851
        return (char *) NULL;
852
 
4327 mejdrech 853
    return (char *) memcpy(dest, src, size + 1);
3238 jermar 854
}
855
 
4153 mejdrech 856
char *strtok(char *s, const char *delim)
857
{
858
    static char *next;
859
 
860
    return strtok_r(s, delim, &next);
861
}
862
 
863
char *strtok_r(char *s, const char *delim, char **next)
864
{
865
    char *start, *end;
866
 
867
    if (s == NULL)
868
        s = *next;
869
 
870
    /* Skip over leading delimiters. */
4327 mejdrech 871
    while (*s && (str_chr(delim, *s) != NULL)) ++s;
4153 mejdrech 872
    start = s;
873
 
874
    /* Skip over token characters. */
4327 mejdrech 875
    while (*s && (str_chr(delim, *s) == NULL)) ++s;
4153 mejdrech 876
    end = s;
877
    *next = (*s ? s + 1 : s);
878
 
879
    if (start == end) {
880
        return NULL;    /* No more tokens. */
881
    }
882
 
883
    /* Overwrite delimiter with NULL terminator. */
884
    *end = '\0';
885
    return start;
886
}
887
 
1719 decky 888
/** @}
1653 cejka 889
 */