1 |
|
---|
2 | #include "config_xor.h"
|
---|
3 |
|
---|
4 | #include <string.h>
|
---|
5 | #include <stdio.h>
|
---|
6 | #include <sys/types.h>
|
---|
7 |
|
---|
8 | #include "sh_string.h"
|
---|
9 | #include "sh_mem.h"
|
---|
10 |
|
---|
11 | #undef FIL__
|
---|
12 | #define FIL__ _("sh_string.c")
|
---|
13 |
|
---|
14 | extern int sl_ok_adds (size_t a, size_t b);
|
---|
15 | #define SL_TRUE 1
|
---|
16 | #define SL_FALSE 0
|
---|
17 |
|
---|
18 | #include <ctype.h>
|
---|
19 | #include <errno.h>
|
---|
20 |
|
---|
21 | /* Split array at delim in at most nfields fields.
|
---|
22 | * Empty fields are returned as empty (zero-length) strings.
|
---|
23 | * Leading and trailing WS are removed from token.
|
---|
24 | * The number of fields is returned in 'nfields', their
|
---|
25 | * lengths in 'lengths'.
|
---|
26 | * A single delimiter will return two empty fields.
|
---|
27 | */
|
---|
28 | char ** split_array(char *line, unsigned int * nfields,
|
---|
29 | char delim, size_t * lengths)
|
---|
30 | {
|
---|
31 | char *a, *e, *s;
|
---|
32 | unsigned int i = 0;
|
---|
33 | int flag = 0;
|
---|
34 | char **arr;
|
---|
35 | unsigned int maxfields = (*nfields);
|
---|
36 |
|
---|
37 | arr = SH_ALLOC((maxfields+1) * sizeof (char*));
|
---|
38 |
|
---|
39 | e = line;
|
---|
40 |
|
---|
41 | do
|
---|
42 | {
|
---|
43 | /* skip leading WS
|
---|
44 | */
|
---|
45 | for (s=e; *s && isspace((int)*s); ++s) /* nothing */;
|
---|
46 |
|
---|
47 | if (*s)
|
---|
48 | {
|
---|
49 | a = s;
|
---|
50 |
|
---|
51 | /* move a to next delim
|
---|
52 | */
|
---|
53 | for (a=s; *a && *a != delim; ++a) /* nothing */;
|
---|
54 |
|
---|
55 | /* set e to next after delim
|
---|
56 | */
|
---|
57 | if (*a == delim)
|
---|
58 | {
|
---|
59 | e = a+1;
|
---|
60 | flag = 1;
|
---|
61 | }
|
---|
62 | else /* (!*a) */
|
---|
63 | {
|
---|
64 | e = a;
|
---|
65 | flag = 0;
|
---|
66 | }
|
---|
67 |
|
---|
68 | if (a != line)
|
---|
69 | {
|
---|
70 | if (i < (maxfields -1))
|
---|
71 | {
|
---|
72 |
|
---|
73 | /* chop off trailing WS
|
---|
74 | */
|
---|
75 | for (a--; isspace((int)*a) && a > s; a--) /* do nothing */;
|
---|
76 |
|
---|
77 | /* terminate string
|
---|
78 | */
|
---|
79 | ++a; *a = '\0';
|
---|
80 | }
|
---|
81 | else
|
---|
82 | {
|
---|
83 | /* If nfields < actual fields, last string
|
---|
84 | * will be remainder, therefore skip to end.
|
---|
85 | */
|
---|
86 | if ( *a )
|
---|
87 | {
|
---|
88 | do {
|
---|
89 | a++;
|
---|
90 | } while ( *a );
|
---|
91 | }
|
---|
92 | }
|
---|
93 | }
|
---|
94 | else
|
---|
95 | {
|
---|
96 | *a = '\0';
|
---|
97 | }
|
---|
98 | }
|
---|
99 | else /* (!*s) */
|
---|
100 | {
|
---|
101 | a = s;
|
---|
102 | /* (i == 0) handles the special case of splitting the empty string */
|
---|
103 | if (flag || i == 0)
|
---|
104 | {
|
---|
105 | flag = 0;
|
---|
106 | goto setnext;
|
---|
107 | }
|
---|
108 | break;
|
---|
109 | }
|
---|
110 |
|
---|
111 | setnext:
|
---|
112 | lengths[i] = (size_t) (a-s); /* a >= s always */
|
---|
113 | arr[i] = s;
|
---|
114 | ++i;
|
---|
115 |
|
---|
116 | } while (i < maxfields);
|
---|
117 |
|
---|
118 | *nfields = i;
|
---|
119 | arr[i] = NULL;
|
---|
120 |
|
---|
121 | return arr;
|
---|
122 | }
|
---|
123 |
|
---|
124 | /* Split array at whitespace in at most nfields fields.
|
---|
125 | * Multiple whitespaces are collapsed.
|
---|
126 | * Empty fields are returned as empty (zero-length) strings.
|
---|
127 | * The number of fields is returned in nfields.
|
---|
128 | * An empty string will return zero fields.
|
---|
129 | * If nfields < actual fields, last string will be remainder.
|
---|
130 | */
|
---|
131 |
|
---|
132 | #define SH_SPLIT_LIST 0
|
---|
133 | #define SH_SPLIT_WS 1
|
---|
134 |
|
---|
135 | char ** split_array_ws_int (char *line,
|
---|
136 | unsigned int * nfields, size_t * lengths,
|
---|
137 | const char *delim, int isList)
|
---|
138 | {
|
---|
139 | char *a, *e, *s;
|
---|
140 | unsigned int i = 0;
|
---|
141 | char **arr;
|
---|
142 | unsigned int maxfields = (*nfields);
|
---|
143 |
|
---|
144 | arr = SH_ALLOC((maxfields+1) * sizeof (char*));
|
---|
145 |
|
---|
146 | e = line;
|
---|
147 |
|
---|
148 | do
|
---|
149 | {
|
---|
150 | s = e;
|
---|
151 |
|
---|
152 | /* skip leading WS
|
---|
153 | */
|
---|
154 | if (isList == SH_SPLIT_WS)
|
---|
155 | {
|
---|
156 | if ( *s && isspace((int)*s) )
|
---|
157 | {
|
---|
158 | do {
|
---|
159 | ++s;
|
---|
160 | } while ( *s && isspace((int)*s) );
|
---|
161 | }
|
---|
162 | }
|
---|
163 | else
|
---|
164 | {
|
---|
165 | if ( *s && strchr(delim, (int)*s))
|
---|
166 | {
|
---|
167 | do {
|
---|
168 | ++s;
|
---|
169 | } while ( *s && strchr(delim, (int)*s));
|
---|
170 | }
|
---|
171 |
|
---|
172 | }
|
---|
173 |
|
---|
174 | if (*s)
|
---|
175 | {
|
---|
176 |
|
---|
177 | /* s is at non-ws, move a to next ws
|
---|
178 | */
|
---|
179 | a = s;
|
---|
180 | if (isList == SH_SPLIT_WS)
|
---|
181 | {
|
---|
182 | do {
|
---|
183 | a++;
|
---|
184 | } while ( *a && (!isspace((int)*a)) );
|
---|
185 | }
|
---|
186 | else
|
---|
187 | {
|
---|
188 | do {
|
---|
189 | a++;
|
---|
190 | } while ( *a && NULL == strchr(delim, (int)*a));
|
---|
191 | }
|
---|
192 |
|
---|
193 | /* next token, *a is either ws or '\0'
|
---|
194 | */
|
---|
195 | e = ( (*a) ? a+1 : a);
|
---|
196 |
|
---|
197 | /* terminate and set arr[i]
|
---|
198 | */
|
---|
199 | if (i < (maxfields-1))
|
---|
200 | {
|
---|
201 | *a = '\0';
|
---|
202 | }
|
---|
203 | else
|
---|
204 | {
|
---|
205 | /* If nfields < actual fields, last
|
---|
206 | * string will be remainder. Therefore
|
---|
207 | * skip to end.
|
---|
208 | */
|
---|
209 | if ( *a )
|
---|
210 | {
|
---|
211 | do {
|
---|
212 | a++;
|
---|
213 | } while ( *a );
|
---|
214 | }
|
---|
215 | }
|
---|
216 | lengths[i] = (size_t)(a-s); /* a >= s always */
|
---|
217 | arr[i] = s;
|
---|
218 | ++i;
|
---|
219 | }
|
---|
220 | else /* if (!*s) */
|
---|
221 | {
|
---|
222 | break;
|
---|
223 | }
|
---|
224 |
|
---|
225 | } while (i < maxfields);
|
---|
226 |
|
---|
227 | *nfields = i;
|
---|
228 | arr[i] = NULL;
|
---|
229 |
|
---|
230 | return arr;
|
---|
231 | }
|
---|
232 |
|
---|
233 | char ** split_array_ws (char *line,
|
---|
234 | unsigned int * nfields, size_t * lengths)
|
---|
235 | {
|
---|
236 | return split_array_ws_int (line, nfields, lengths, NULL, SH_SPLIT_WS);
|
---|
237 | }
|
---|
238 |
|
---|
239 | char ** split_array_list (char *line,
|
---|
240 | unsigned int * nfields, size_t * lengths)
|
---|
241 | {
|
---|
242 | return split_array_ws_int (line, nfields, lengths, ", \t", SH_SPLIT_LIST);
|
---|
243 | }
|
---|
244 |
|
---|
245 | char ** split_array_token (char *line,
|
---|
246 | unsigned int * nfields, size_t * lengths,
|
---|
247 | const char * token)
|
---|
248 | {
|
---|
249 | return split_array_ws_int (line, nfields, lengths, token, SH_SPLIT_LIST);
|
---|
250 | }
|
---|
251 |
|
---|
252 | /* return a split() of a list contained in 'PREFIX\s*( list ).*'
|
---|
253 | */
|
---|
254 | char ** split_array_braced (char *line, const char * prefix,
|
---|
255 | unsigned int * nfields, size_t * lengths)
|
---|
256 | {
|
---|
257 | char * s = line;
|
---|
258 | char * p;
|
---|
259 |
|
---|
260 | while ( *s && isspace((int)*s) ) ++s;
|
---|
261 | if (prefix && 0 != strncmp(s, prefix, strlen(prefix)))
|
---|
262 | return NULL;
|
---|
263 | s = &s[strlen(prefix)];
|
---|
264 | while ( *s && isspace((int)*s) ) ++s;
|
---|
265 | if (!s || (*s != '('))
|
---|
266 | return NULL;
|
---|
267 | ++s;
|
---|
268 | p = strchr(s, ')');
|
---|
269 | if (!p || (*p == *s))
|
---|
270 | return NULL;
|
---|
271 | *p = '\0';
|
---|
272 | return split_array_list (s, nfields, lengths);
|
---|
273 | }
|
---|
274 |
|
---|
275 | #define SH_STRING_PARCEL 120
|
---|
276 |
|
---|
277 | static
|
---|
278 | size_t sh_string_read_int(sh_string * s, FILE * fp, size_t maxlen, char *start);
|
---|
279 |
|
---|
280 | size_t sh_string_read(sh_string * s, FILE * fp, size_t maxlen)
|
---|
281 | {
|
---|
282 | return sh_string_read_int(s, fp, maxlen, NULL);
|
---|
283 | }
|
---|
284 |
|
---|
285 | size_t sh_string_read_cont(sh_string * s, FILE * fp, size_t maxlen, char *cont)
|
---|
286 | {
|
---|
287 | return sh_string_read_int(s, fp, maxlen, cont);
|
---|
288 | }
|
---|
289 |
|
---|
290 | static char * sh_str_fgets (char *s, int size, FILE *fp)
|
---|
291 | {
|
---|
292 | char * ret;
|
---|
293 | do {
|
---|
294 | clearerr(fp);
|
---|
295 | ret = fgets(s, size, fp);
|
---|
296 | } while (ret == NULL && ferror(fp) && errno == EAGAIN);
|
---|
297 |
|
---|
298 | return ret;
|
---|
299 | }
|
---|
300 |
|
---|
301 | size_t sh_string_read_int(sh_string * s, FILE * fp, size_t maxlen, char *start)
|
---|
302 | {
|
---|
303 |
|
---|
304 | /* case 0) start != NULL and first char not in 'start'
|
---|
305 | */
|
---|
306 | if (start)
|
---|
307 | {
|
---|
308 | int first;
|
---|
309 |
|
---|
310 | do {
|
---|
311 | clearerr(fp);
|
---|
312 | first = fgetc(fp);
|
---|
313 | } while (first == EOF && ferror(fp) && errno == EAGAIN);
|
---|
314 |
|
---|
315 | if (first == EOF)
|
---|
316 | {
|
---|
317 | sh_string_truncate(s, 0);
|
---|
318 | if (ferror(fp))
|
---|
319 | return -1;
|
---|
320 | return 0;
|
---|
321 | }
|
---|
322 |
|
---|
323 | if (NULL == strchr(start, first))
|
---|
324 | {
|
---|
325 | ungetc(first, fp);
|
---|
326 | return 0;
|
---|
327 | }
|
---|
328 | ungetc(first, fp);
|
---|
329 | }
|
---|
330 |
|
---|
331 | /* case 1) EOF or error
|
---|
332 | */
|
---|
333 | if (sh_str_fgets(s->str, s->siz, fp) == NULL)
|
---|
334 | {
|
---|
335 | sh_string_truncate(s, 0);
|
---|
336 | if (ferror(fp))
|
---|
337 | return -1;
|
---|
338 | return 0;
|
---|
339 | }
|
---|
340 |
|
---|
341 | /* case 2) end of line reached. strlen should always be > 0
|
---|
342 | * because of the '\n', but we check.
|
---|
343 | */
|
---|
344 | s->len = strlen(s->str);
|
---|
345 | if (s->len > 0 && (s->str)[s->len-1] == '\n')
|
---|
346 | {
|
---|
347 | (s->str)[s->len-1] = '\0';
|
---|
348 | --(s->len);
|
---|
349 | return (s->len + 1);
|
---|
350 | }
|
---|
351 |
|
---|
352 | /* case 3) incomplete string
|
---|
353 | */
|
---|
354 | for (;;) {
|
---|
355 |
|
---|
356 | if (maxlen > 0 && (s->siz+SH_STRING_PARCEL) > maxlen)
|
---|
357 | {
|
---|
358 | if (s->siz < maxlen)
|
---|
359 | sh_string_grow(s, (maxlen-s->siz));
|
---|
360 | else
|
---|
361 | return -2;
|
---|
362 | }
|
---|
363 | else
|
---|
364 | {
|
---|
365 | sh_string_grow(s, 0);
|
---|
366 | }
|
---|
367 |
|
---|
368 | if (sh_str_fgets(&(s->str[s->len]), (s->siz - s->len), fp) == NULL)
|
---|
369 | {
|
---|
370 | if (ferror(fp))
|
---|
371 | {
|
---|
372 | sh_string_truncate(s, 0);
|
---|
373 | return -1;
|
---|
374 | }
|
---|
375 | return s->len;
|
---|
376 | }
|
---|
377 |
|
---|
378 | s->len += strlen( &(s->str[s->len]) );
|
---|
379 | if (s->len > 0 && s->str[s->len-1] == '\n')
|
---|
380 | {
|
---|
381 | (s->str)[s->len-1] = '\0';
|
---|
382 | --(s->len);
|
---|
383 | return (s->len + 1);
|
---|
384 | }
|
---|
385 | }
|
---|
386 |
|
---|
387 | /* notreached */
|
---|
388 | }
|
---|
389 |
|
---|
390 | sh_string * sh_string_cat_lchar(sh_string * s, const char * str, size_t len)
|
---|
391 | {
|
---|
392 | if (sl_ok_adds(len, s->siz) == SL_TRUE)
|
---|
393 | {
|
---|
394 | if ((len + 1 + s->len) > s->siz)
|
---|
395 | {
|
---|
396 | sh_string_grow(s, ((len+1+s->len) - s->siz) );
|
---|
397 | }
|
---|
398 | memcpy(&(s->str[s->len]), str, len);
|
---|
399 | s->len += len;
|
---|
400 | s->str[s->len] = '\0';
|
---|
401 | return s;
|
---|
402 | }
|
---|
403 |
|
---|
404 | return NULL;
|
---|
405 | }
|
---|
406 |
|
---|
407 | sh_string * sh_string_set_from_char(sh_string * s, const char * str)
|
---|
408 | {
|
---|
409 | size_t len = strlen(str);
|
---|
410 |
|
---|
411 | if ((len+1) > s->siz)
|
---|
412 | {
|
---|
413 | sh_string_grow(s, ((len+1) - s->siz) );
|
---|
414 | }
|
---|
415 | memcpy(s->str, str, (len+1));
|
---|
416 | s->len = len;
|
---|
417 | return s;
|
---|
418 | }
|
---|
419 |
|
---|
420 | sh_string * sh_string_add_from_char(sh_string * s, const char * str)
|
---|
421 | {
|
---|
422 | size_t len = strlen(str);
|
---|
423 | size_t avail = (s->siz - s->len);
|
---|
424 |
|
---|
425 | if ((len+1) > avail)
|
---|
426 | {
|
---|
427 | (void) sh_string_grow(s, ((len+1) - avail) );
|
---|
428 | }
|
---|
429 | memcpy(&(s->str[s->len]), str, (len+1));
|
---|
430 | s->len += len;
|
---|
431 | return s;
|
---|
432 | }
|
---|
433 |
|
---|
434 | sh_string * sh_string_new_from_lchar(const char * str, size_t len)
|
---|
435 | {
|
---|
436 | sh_string * s;
|
---|
437 | s = SH_ALLOC(sizeof(sh_string));
|
---|
438 | s->str = SH_ALLOC(len+1);
|
---|
439 | memcpy(s->str, str, len);
|
---|
440 | s->str[len] = '\0';
|
---|
441 | s->siz = len+1;
|
---|
442 | s->len = len;
|
---|
443 | return s;
|
---|
444 | }
|
---|
445 |
|
---|
446 | sh_string * sh_string_new_from_lchar3(const char * str1, size_t len1,
|
---|
447 | const char * str2, size_t len2,
|
---|
448 | const char * str3, size_t len3)
|
---|
449 | {
|
---|
450 | sh_string * s;
|
---|
451 | size_t len = 0;
|
---|
452 |
|
---|
453 | if (sl_ok_adds(len1, len2) == SL_TRUE)
|
---|
454 | len = len1 + len2;
|
---|
455 | else
|
---|
456 | return NULL;
|
---|
457 | if (sl_ok_adds( len, len3) == SL_TRUE)
|
---|
458 | len = len + len3;
|
---|
459 | else
|
---|
460 | return NULL;
|
---|
461 |
|
---|
462 | s = SH_ALLOC(sizeof(sh_string));
|
---|
463 | s->str = SH_ALLOC(len+1);
|
---|
464 |
|
---|
465 | memcpy(s->str, str1, len1);
|
---|
466 | memcpy(&s->str[len1], str2, len2);
|
---|
467 | memcpy(&s->str[len1+len2], str3, len3);
|
---|
468 |
|
---|
469 | s->str[len] = '\0';
|
---|
470 | s->siz = len+1;
|
---|
471 | s->len = len;
|
---|
472 | return s;
|
---|
473 | }
|
---|
474 |
|
---|
475 | sh_string * sh_string_grow(sh_string * s, size_t increase)
|
---|
476 | {
|
---|
477 | char * new;
|
---|
478 |
|
---|
479 | if (increase == 0)
|
---|
480 | increase = SH_STRING_PARCEL;
|
---|
481 |
|
---|
482 | if (s && sl_ok_adds(s->siz, increase) == SL_TRUE)
|
---|
483 | {
|
---|
484 | new = SH_ALLOC(s->siz + increase);
|
---|
485 |
|
---|
486 | if (s->str)
|
---|
487 | {
|
---|
488 | memcpy(new, s->str, s->len+1);
|
---|
489 | SH_FREE(s->str);
|
---|
490 | }
|
---|
491 | else
|
---|
492 | {
|
---|
493 | new[0] = '\0';
|
---|
494 | }
|
---|
495 | s->str = new;
|
---|
496 | s->siz += increase;
|
---|
497 | return s;
|
---|
498 | }
|
---|
499 | return NULL;
|
---|
500 | }
|
---|
501 |
|
---|
502 | sh_string * sh_string_truncate(sh_string * s, size_t len)
|
---|
503 | {
|
---|
504 | if (s)
|
---|
505 | {
|
---|
506 | if (s->str && (s->len > len) )
|
---|
507 | {
|
---|
508 | s->len = len;
|
---|
509 | (s->str)[len] = '\0';
|
---|
510 | }
|
---|
511 | return s;
|
---|
512 | }
|
---|
513 | return NULL;
|
---|
514 | }
|
---|
515 |
|
---|
516 | void sh_string_destroy(sh_string ** s)
|
---|
517 | {
|
---|
518 | if (s)
|
---|
519 | {
|
---|
520 | if ((*s) && (*s)->str)
|
---|
521 | SH_FREE ((*s)->str);
|
---|
522 | SH_FREE(*s);
|
---|
523 | *s = NULL;
|
---|
524 | }
|
---|
525 | return;
|
---|
526 | }
|
---|
527 |
|
---|
528 | sh_string * sh_string_new(size_t size)
|
---|
529 | {
|
---|
530 | sh_string * s;
|
---|
531 | s = SH_ALLOC (sizeof(sh_string));
|
---|
532 | if (size == 0)
|
---|
533 | size = SH_STRING_PARCEL;
|
---|
534 | s->str = SH_ALLOC (size);
|
---|
535 | s->str[0] = '\0';
|
---|
536 | s->siz = size;
|
---|
537 | s->len = 0;
|
---|
538 | return s;
|
---|
539 | }
|
---|
540 |
|
---|
541 | /* Replaces fields in s with 'replacement'. Fields are given
|
---|
542 | * in the ordered array ovector, comprising ovecnum pairs
|
---|
543 | * ovector[i], ovector[i+1] which list offset of first char
|
---|
544 | * of field, offset of first char after field (this is how
|
---|
545 | * the pcre library does it).
|
---|
546 | */
|
---|
547 | sh_string * sh_string_replace(const sh_string * s,
|
---|
548 | const int * ovector, int ovecnum,
|
---|
549 | const char * replacement, size_t rlen)
|
---|
550 | {
|
---|
551 | sh_string * r = NULL;
|
---|
552 | char * p;
|
---|
553 | long tlen;
|
---|
554 | size_t len;
|
---|
555 | int end = 0;
|
---|
556 | int start = 0;
|
---|
557 | size_t oldlen = 0;
|
---|
558 | size_t newlen = 0;
|
---|
559 | long diff;
|
---|
560 | int i, curr, last;
|
---|
561 |
|
---|
562 | for (i = 0; i < ovecnum; ++i)
|
---|
563 | {
|
---|
564 | start = ovector[2*i]; /* offset of first char of substring */
|
---|
565 | if (start >= end)
|
---|
566 | {
|
---|
567 | end = ovector[2*i+1]; /* offset of first char after substring end*/
|
---|
568 |
|
---|
569 | if (end > start && (unsigned int)end <= (s->len + 1))
|
---|
570 | {
|
---|
571 | oldlen += (end - start);
|
---|
572 | newlen += rlen;
|
---|
573 | }
|
---|
574 | else /* inconsistency detected */
|
---|
575 | {
|
---|
576 | return NULL;
|
---|
577 | }
|
---|
578 | }
|
---|
579 | else /* overlap detected */
|
---|
580 | {
|
---|
581 | return NULL;
|
---|
582 | }
|
---|
583 | }
|
---|
584 |
|
---|
585 | diff = newlen - oldlen;
|
---|
586 |
|
---|
587 | if ((diff > 0) && ((s->len + 1 + diff) > s->siz))
|
---|
588 | {
|
---|
589 | r = sh_string_new_from_lchar(sh_string_str(s),
|
---|
590 | sh_string_len(s));
|
---|
591 | r = sh_string_grow(r, diff);
|
---|
592 | }
|
---|
593 | else
|
---|
594 | {
|
---|
595 | r = sh_string_new_from_lchar(sh_string_str(s),
|
---|
596 | sh_string_len(s));
|
---|
597 | }
|
---|
598 |
|
---|
599 |
|
---|
600 | curr = -1;
|
---|
601 | last = -1;
|
---|
602 |
|
---|
603 | for (i = 0; i < ovecnum; ++i)
|
---|
604 | {
|
---|
605 | if (ovector[2*i] >= 0)
|
---|
606 | {
|
---|
607 | curr = 2*i;
|
---|
608 | break;
|
---|
609 | }
|
---|
610 | }
|
---|
611 |
|
---|
612 | if (r && ovecnum > 0 && ovector[curr] >= 0)
|
---|
613 | {
|
---|
614 | r->len = 0; r->str[0] = '\0'; p = r->str;
|
---|
615 |
|
---|
616 | /* First part, until start of first replacement
|
---|
617 | */
|
---|
618 | if (r->siz > (unsigned int)ovector[curr]) {
|
---|
619 | memcpy(p, s->str, (size_t)ovector[curr]);
|
---|
620 | p += ovector[curr];
|
---|
621 | r->len += ovector[curr];
|
---|
622 | }
|
---|
623 | if (r->siz > (r->len + rlen)) {
|
---|
624 | memcpy(p, replacement, rlen);
|
---|
625 | p += rlen;
|
---|
626 | r->len += rlen;
|
---|
627 | }
|
---|
628 | *p = '\0';
|
---|
629 |
|
---|
630 | last = curr + 1;
|
---|
631 |
|
---|
632 | for (i = 1; i < ovecnum; ++i)
|
---|
633 | {
|
---|
634 | if (ovector[2*i] < 0)
|
---|
635 | continue;
|
---|
636 |
|
---|
637 | curr = 2*i;
|
---|
638 |
|
---|
639 | /* From end of last replacement to start of this */
|
---|
640 | tlen = (long)(ovector[curr] - ovector[last]);
|
---|
641 | if (tlen >= 0)
|
---|
642 | {
|
---|
643 | len = (size_t) tlen;
|
---|
644 |
|
---|
645 | if (tlen > 0 && r->siz > (r->len + len))
|
---|
646 | {
|
---|
647 | memcpy(p, &(s->str[ovector[last]]), (size_t)len);
|
---|
648 | p += len;
|
---|
649 | r->len += len;
|
---|
650 | }
|
---|
651 |
|
---|
652 | /* The replacement */
|
---|
653 | if (r->siz > (r->len + rlen)) {
|
---|
654 | memcpy(p, replacement, rlen);
|
---|
655 | p += rlen;
|
---|
656 | r->len += rlen;
|
---|
657 | }
|
---|
658 |
|
---|
659 | /* null terminate */
|
---|
660 | *p = '\0';
|
---|
661 |
|
---|
662 | last = curr + 1;
|
---|
663 | }
|
---|
664 | }
|
---|
665 |
|
---|
666 | /* Last part, after last replacement; includes terminating null
|
---|
667 | */
|
---|
668 | if (last > 0)
|
---|
669 | {
|
---|
670 | /* If not, nothing has been replaced, and r is still a copy of s
|
---|
671 | */
|
---|
672 | tlen = (long)((s->len + 1) - ovector[last]);
|
---|
673 | if (tlen > 0)
|
---|
674 | {
|
---|
675 | len = (size_t)tlen;
|
---|
676 | if (r->siz >= (r->len + len)) {
|
---|
677 | memcpy(p, &(s->str[ovector[2*i -1]]), (size_t)len);
|
---|
678 | p += (len - 1);
|
---|
679 | r->len += (len - 1);
|
---|
680 | *p = '\0';
|
---|
681 | }
|
---|
682 | }
|
---|
683 | }
|
---|
684 |
|
---|
685 | }
|
---|
686 |
|
---|
687 | return r;
|
---|
688 | }
|
---|
689 |
|
---|
690 |
|
---|
691 | #ifdef SH_CUTEST
|
---|
692 | #include <stdlib.h>
|
---|
693 | #include "CuTest.h"
|
---|
694 |
|
---|
695 | void Test_string (CuTest *tc) {
|
---|
696 | int status, i, max = 120;
|
---|
697 | FILE * fp;
|
---|
698 | sh_string * s = NULL;
|
---|
699 | sh_string * t;
|
---|
700 | static char template[] = "/tmp/xtest.XXXXXX";
|
---|
701 | char ** array;
|
---|
702 | char test[128];
|
---|
703 | size_t lengths[16];
|
---|
704 | unsigned int iarr;
|
---|
705 | int ovector[16];
|
---|
706 | int ovecnum;
|
---|
707 |
|
---|
708 | s = sh_string_new(0);
|
---|
709 | CuAssertPtrNotNull(tc, s);
|
---|
710 | sh_string_destroy(&s);
|
---|
711 | CuAssertTrue(tc, s == NULL);
|
---|
712 |
|
---|
713 | s = sh_string_new(0);
|
---|
714 | CuAssertPtrNotNull(tc, s);
|
---|
715 |
|
---|
716 | status = mkstemp(template);
|
---|
717 | CuAssertTrue(tc, status >= 0);
|
---|
718 |
|
---|
719 | fp = fdopen(status, "r+");
|
---|
720 | CuAssertPtrNotNull(tc, fp);
|
---|
721 |
|
---|
722 | for (i = 0; i < 80; ++i) { fputc ('a', fp); } fputc ('\n', fp); /* 0 */
|
---|
723 | for (i = 0; i < 118; ++i) { fputc ('a', fp); } fputc ('\n', fp); /* 1 */
|
---|
724 | for (i = 0; i < 119; ++i) { fputc ('a', fp); } fputc ('\n', fp); /* 2 */
|
---|
725 | for (i = 0; i < 120; ++i) { fputc ('a', fp); } fputc ('\n', fp); /* 3 */
|
---|
726 | for (i = 0; i < 121; ++i) { fputc ('a', fp); } fputc ('\n', fp); /* 4 */
|
---|
727 | for (i = 0; i < 238; ++i) { fputc ('a', fp); } fputc ('\n', fp);
|
---|
728 | for (i = 0; i < 239; ++i) { fputc ('a', fp); } fputc ('\n', fp);
|
---|
729 | for (i = 0; i < 240; ++i) { fputc ('a', fp); } fputc ('\n', fp);
|
---|
730 | for (i = 0; i < 241; ++i) { fputc ('a', fp); } fputc ('\n', fp);
|
---|
731 |
|
---|
732 | rewind(fp);
|
---|
733 |
|
---|
734 | for (i = 0; i < 9; ++i)
|
---|
735 | {
|
---|
736 | status = sh_string_read(s, fp, max);
|
---|
737 |
|
---|
738 | switch (i) {
|
---|
739 | case 0:
|
---|
740 | CuAssertTrue(tc, s->len == 80);
|
---|
741 | CuAssertTrue(tc, s->siz == 120);
|
---|
742 | CuAssertTrue(tc, status == 81);
|
---|
743 | break;
|
---|
744 | case 1:
|
---|
745 | CuAssertTrue(tc, s->len == 118);
|
---|
746 | CuAssertTrue(tc, s->siz == 120);
|
---|
747 | CuAssertTrue(tc, status == 119);
|
---|
748 | break;
|
---|
749 | case 2:
|
---|
750 | CuAssertTrue(tc, s->len == 119);
|
---|
751 | CuAssertTrue(tc, s->siz == 120);
|
---|
752 | CuAssertTrue(tc, status == -2); /* no terminating '\n', truncated */
|
---|
753 | break;
|
---|
754 | case 3:
|
---|
755 | CuAssertTrue(tc, s->len == 120);
|
---|
756 | CuAssertTrue(tc, s->siz == 240);
|
---|
757 | CuAssertTrue(tc, status == 121);
|
---|
758 | break;
|
---|
759 | case 4:
|
---|
760 | CuAssertTrue(tc, s->len == 121);
|
---|
761 | CuAssertTrue(tc, s->siz == 240);
|
---|
762 | CuAssertTrue(tc, status == 122);
|
---|
763 | break;
|
---|
764 | case 5:
|
---|
765 | CuAssertTrue(tc, s->len == 238);
|
---|
766 | CuAssertTrue(tc, s->siz == 240);
|
---|
767 | CuAssertTrue(tc, status == 239);
|
---|
768 | break;
|
---|
769 | case 6:
|
---|
770 | CuAssertTrue(tc, s->len == 239);
|
---|
771 | CuAssertTrue(tc, s->siz == 240);
|
---|
772 | CuAssertTrue(tc, status == -2); /* no terminating '\n', truncated */
|
---|
773 | break;
|
---|
774 | default:
|
---|
775 | CuAssertTrue(tc, s->len == 239);
|
---|
776 | CuAssertTrue(tc, s->siz == 240);
|
---|
777 | CuAssertTrue(tc, status == -2);
|
---|
778 | }
|
---|
779 | if (status == -2) /* read in rest of string */
|
---|
780 | { max = 240; sh_string_read(s, fp, max); }
|
---|
781 | }
|
---|
782 |
|
---|
783 | rewind(fp);
|
---|
784 |
|
---|
785 | sh_string_truncate(s, 0);
|
---|
786 | CuAssertTrue(tc, s->len == 0);
|
---|
787 |
|
---|
788 | for (i = 0; i < 9; ++i)
|
---|
789 | {
|
---|
790 | status = sh_string_read(s, fp, 240);
|
---|
791 | if (status == -2)
|
---|
792 | sh_string_read(s, fp, 240);
|
---|
793 | else
|
---|
794 | {
|
---|
795 | for (status = 0; status < (int)s->len; ++status)
|
---|
796 | {
|
---|
797 | if (s->str[status] != 'a')
|
---|
798 | {
|
---|
799 | CuFail(tc, "unexpected character");
|
---|
800 | }
|
---|
801 | }
|
---|
802 | }
|
---|
803 | }
|
---|
804 |
|
---|
805 | status = fclose(fp);
|
---|
806 | CuAssertTrue(tc, status == 0);
|
---|
807 | status = remove(template);
|
---|
808 | CuAssertTrue(tc, status == 0);
|
---|
809 |
|
---|
810 | iarr = 10; strcpy(test, "|a1|| a2| |a3 |a4|a5|");
|
---|
811 | array = split_array(test, &iarr, '|', lengths);
|
---|
812 | CuAssertIntEquals(tc,9,(int)iarr);
|
---|
813 | CuAssertStrEquals(tc,"", array[0]);
|
---|
814 | CuAssertStrEquals(tc,"a1", array[1]);
|
---|
815 | CuAssertStrEquals(tc,"", array[2]);
|
---|
816 | CuAssertStrEquals(tc,"a2", array[3]);
|
---|
817 | CuAssertStrEquals(tc,"", array[4]);
|
---|
818 | CuAssertStrEquals(tc,"a3", array[5]);
|
---|
819 | CuAssertStrEquals(tc,"a4", array[6]);
|
---|
820 | CuAssertStrEquals(tc,"a5", array[7]);
|
---|
821 | CuAssertStrEquals(tc,"", array[8]);
|
---|
822 |
|
---|
823 | CuAssertIntEquals(tc, 0, (int)lengths[0]);
|
---|
824 | CuAssertIntEquals(tc, 2, (int)lengths[1]);
|
---|
825 | CuAssertIntEquals(tc, 0, (int)lengths[2]);
|
---|
826 | CuAssertIntEquals(tc, 2, (int)lengths[3]);
|
---|
827 | CuAssertIntEquals(tc, 0, (int)lengths[4]);
|
---|
828 | CuAssertIntEquals(tc, 2, (int)lengths[5]);
|
---|
829 | CuAssertIntEquals(tc, 2, (int)lengths[6]);
|
---|
830 | CuAssertIntEquals(tc, 2, (int)lengths[7]);
|
---|
831 | CuAssertIntEquals(tc, 0, (int)lengths[8]);
|
---|
832 |
|
---|
833 | iarr = 10; strcpy(test, "a1|| a2| |a3 |a4|a5|");
|
---|
834 | array = split_array(test, &iarr, '|', lengths);
|
---|
835 | CuAssertIntEquals(tc,8,(int)iarr);
|
---|
836 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
837 | CuAssertStrEquals(tc,"", array[1]);
|
---|
838 | CuAssertStrEquals(tc,"a2", array[2]);
|
---|
839 | CuAssertStrEquals(tc,"", array[3]);
|
---|
840 | CuAssertStrEquals(tc,"a3", array[4]);
|
---|
841 | CuAssertStrEquals(tc,"a4", array[5]);
|
---|
842 | CuAssertStrEquals(tc,"a5", array[6]);
|
---|
843 | CuAssertStrEquals(tc,"", array[7]);
|
---|
844 |
|
---|
845 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
846 | CuAssertIntEquals(tc, 0, (int)lengths[1]);
|
---|
847 | CuAssertIntEquals(tc, 2, (int)lengths[2]);
|
---|
848 | CuAssertIntEquals(tc, 0, (int)lengths[3]);
|
---|
849 | CuAssertIntEquals(tc, 2, (int)lengths[4]);
|
---|
850 | CuAssertIntEquals(tc, 2, (int)lengths[5]);
|
---|
851 | CuAssertIntEquals(tc, 2, (int)lengths[6]);
|
---|
852 | CuAssertIntEquals(tc, 0, (int)lengths[7]);
|
---|
853 |
|
---|
854 | iarr = 10; strcpy(test, " a1|| a2 | |a3 |a4|a5");
|
---|
855 | array = split_array(test, &iarr, '|', lengths);
|
---|
856 | CuAssertIntEquals(tc,7,(int)iarr);
|
---|
857 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
858 | CuAssertStrEquals(tc,"", array[1]);
|
---|
859 | CuAssertStrEquals(tc,"a2", array[2]);
|
---|
860 | CuAssertStrEquals(tc,"", array[3]);
|
---|
861 | CuAssertStrEquals(tc,"a3", array[4]);
|
---|
862 | CuAssertStrEquals(tc,"a4", array[5]);
|
---|
863 | CuAssertStrEquals(tc,"a5", array[6]);
|
---|
864 |
|
---|
865 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
866 | CuAssertIntEquals(tc, 0, (int)lengths[1]);
|
---|
867 | CuAssertIntEquals(tc, 2, (int)lengths[2]);
|
---|
868 | CuAssertIntEquals(tc, 0, (int)lengths[3]);
|
---|
869 | CuAssertIntEquals(tc, 2, (int)lengths[4]);
|
---|
870 | CuAssertIntEquals(tc, 2, (int)lengths[5]);
|
---|
871 | CuAssertIntEquals(tc, 2, (int)lengths[6]);
|
---|
872 |
|
---|
873 | iarr = 10; strcpy(test, "a1|| a2 | |a3 |a4|a5 ");
|
---|
874 | array = split_array(test, &iarr, '|', lengths);
|
---|
875 | CuAssertIntEquals(tc,7,(int)iarr);
|
---|
876 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
877 | CuAssertStrEquals(tc,"", array[1]);
|
---|
878 | CuAssertStrEquals(tc,"a2", array[2]);
|
---|
879 | CuAssertStrEquals(tc,"", array[3]);
|
---|
880 | CuAssertStrEquals(tc,"a3", array[4]);
|
---|
881 | CuAssertStrEquals(tc,"a4", array[5]);
|
---|
882 | CuAssertStrEquals(tc,"a5", array[6]);
|
---|
883 |
|
---|
884 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
885 | CuAssertIntEquals(tc, 0, (int)lengths[1]);
|
---|
886 | CuAssertIntEquals(tc, 2, (int)lengths[2]);
|
---|
887 | CuAssertIntEquals(tc, 0, (int)lengths[3]);
|
---|
888 | CuAssertIntEquals(tc, 2, (int)lengths[4]);
|
---|
889 | CuAssertIntEquals(tc, 2, (int)lengths[5]);
|
---|
890 | CuAssertIntEquals(tc, 2, (int)lengths[6]);
|
---|
891 |
|
---|
892 | iarr = 10; strcpy(test, "|");
|
---|
893 | array = split_array(test, &iarr, '|', lengths);
|
---|
894 | CuAssertIntEquals(tc,2,(int)iarr);
|
---|
895 | CuAssertStrEquals(tc,"", array[0]);
|
---|
896 | CuAssertStrEquals(tc,"", array[1]);
|
---|
897 |
|
---|
898 | CuAssertIntEquals(tc, 0, (int)lengths[0]);
|
---|
899 | CuAssertIntEquals(tc, 0, (int)lengths[1]);
|
---|
900 |
|
---|
901 | iarr = 10; strcpy(test, "|||");
|
---|
902 | array = split_array(test, &iarr, '|', lengths);
|
---|
903 | CuAssertIntEquals(tc,4,(int)iarr);
|
---|
904 | CuAssertStrEquals(tc,"", array[0]);
|
---|
905 | CuAssertStrEquals(tc,"", array[1]);
|
---|
906 | CuAssertStrEquals(tc,"", array[2]);
|
---|
907 | CuAssertStrEquals(tc,"", array[3]);
|
---|
908 |
|
---|
909 | CuAssertIntEquals(tc, 0, (int)lengths[0]);
|
---|
910 | CuAssertIntEquals(tc, 0, (int)lengths[1]);
|
---|
911 | CuAssertIntEquals(tc, 0, (int)lengths[2]);
|
---|
912 | CuAssertIntEquals(tc, 0, (int)lengths[3]);
|
---|
913 |
|
---|
914 | iarr = 10; strcpy(test, " a1 ");
|
---|
915 | array = split_array(test, &iarr, '|', lengths);
|
---|
916 | CuAssertIntEquals(tc,1,(int)iarr);
|
---|
917 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
918 |
|
---|
919 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
920 |
|
---|
921 | iarr = 10; strcpy(test, "");
|
---|
922 | array = split_array(test, &iarr, '|', lengths);
|
---|
923 | CuAssertIntEquals(tc,1,(int)iarr);
|
---|
924 | CuAssertStrEquals(tc,"", array[0]);
|
---|
925 |
|
---|
926 | CuAssertIntEquals(tc, 0, (int)lengths[0]);
|
---|
927 |
|
---|
928 | /* WS separated */
|
---|
929 |
|
---|
930 | iarr = 10; strcpy(test, "a1");
|
---|
931 | array = split_array_ws (test, &iarr, lengths);
|
---|
932 | CuAssertIntEquals(tc,1,(int)iarr);
|
---|
933 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
934 |
|
---|
935 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
936 |
|
---|
937 | iarr = 10; strcpy(test, " a1");
|
---|
938 | array = split_array_ws (test, &iarr, lengths);
|
---|
939 | CuAssertIntEquals(tc,1,(int)iarr);
|
---|
940 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
941 |
|
---|
942 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
943 |
|
---|
944 | iarr = 10; strcpy(test, " a1 ");
|
---|
945 | array = split_array_ws (test, &iarr, lengths);
|
---|
946 | CuAssertIntEquals(tc,1,(int)iarr);
|
---|
947 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
948 |
|
---|
949 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
950 |
|
---|
951 | iarr = 10; strcpy(test, " ");
|
---|
952 | array = split_array_ws (test, &iarr, lengths);
|
---|
953 | CuAssertIntEquals(tc,0,(int)iarr);
|
---|
954 | CuAssertTrue(tc, array[0] == NULL);
|
---|
955 |
|
---|
956 | iarr = 10; strcpy(test, " a1 a2");
|
---|
957 | array = split_array_ws (test, &iarr, lengths);
|
---|
958 | CuAssertIntEquals(tc,2,(int)iarr);
|
---|
959 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
960 | CuAssertStrEquals(tc,"a2", array[1]);
|
---|
961 |
|
---|
962 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
963 | CuAssertIntEquals(tc, 2, (int)lengths[1]);
|
---|
964 |
|
---|
965 | iarr = 10; strcpy(test, " a1 a2 ");
|
---|
966 | array = split_array_ws (test, &iarr, lengths);
|
---|
967 | CuAssertIntEquals(tc,2,(int)iarr);
|
---|
968 | CuAssertStrEquals(tc,"a1", array[0]);
|
---|
969 | CuAssertStrEquals(tc,"a2", array[1]);
|
---|
970 |
|
---|
971 | CuAssertIntEquals(tc, 2, (int)lengths[0]);
|
---|
972 | CuAssertIntEquals(tc, 2, (int)lengths[1]);
|
---|
973 |
|
---|
974 | iarr = 10; strcpy(test, "");
|
---|
975 | array = split_array_ws (test, &iarr, lengths);
|
---|
976 | CuAssertIntEquals(tc,0,(int)iarr);
|
---|
977 | CuAssertTrue(tc, array[0] == NULL);
|
---|
978 |
|
---|
979 | iarr = 3; strcpy(test, " this is a test for remainder");
|
---|
980 | array = split_array_ws (test, &iarr, lengths);
|
---|
981 | CuAssertIntEquals(tc,3,(int)iarr);
|
---|
982 | CuAssertStrEquals(tc,"this", array[0]);
|
---|
983 | CuAssertStrEquals(tc,"is", array[1]);
|
---|
984 | CuAssertStrEquals(tc,"a test for remainder", array[2]);
|
---|
985 | for (i = 0; i < 3; ++i)
|
---|
986 | {
|
---|
987 | CuAssertIntEquals(tc, (int)strlen(array[i]), lengths[i] );
|
---|
988 | }
|
---|
989 |
|
---|
990 | /* string replace */
|
---|
991 | s = sh_string_new_from_lchar3 ("abc ", 4, "def ", 4, "ghi ", 4);
|
---|
992 | ovecnum = 2;
|
---|
993 | ovector[0] = 0; ovector[1] = 2;
|
---|
994 | ovector[2] = 4; ovector[3] = 11;
|
---|
995 |
|
---|
996 | t = sh_string_replace(s, ovector, ovecnum,
|
---|
997 | "___", 3);
|
---|
998 | CuAssertPtrNotNull(tc, t);
|
---|
999 | CuAssertStrEquals(tc, "___c ___ ", t->str);
|
---|
1000 | CuAssertIntEquals(tc, 9, (int)t->len);
|
---|
1001 |
|
---|
1002 | ovector[0] = 0; ovector[1] = 2;
|
---|
1003 | ovector[2] = 4; ovector[3] = 12;
|
---|
1004 | t = sh_string_replace(s, ovector, ovecnum,
|
---|
1005 | "___", 3);
|
---|
1006 | CuAssertPtrNotNull(tc, t);
|
---|
1007 | CuAssertStrEquals(tc, "___c ___", t->str);
|
---|
1008 | CuAssertIntEquals(tc, 8, (int)t->len);
|
---|
1009 |
|
---|
1010 | ovector[0] = 0; ovector[1] = 0;
|
---|
1011 | ovector[2] = 0; ovector[3] = 0;
|
---|
1012 | t = sh_string_replace(s, ovector, ovecnum,
|
---|
1013 | "___", 3);
|
---|
1014 | CuAssertTrue(tc, t == NULL);
|
---|
1015 |
|
---|
1016 | ovector[0] = 0; ovector[1] = 3;
|
---|
1017 | ovector[2] = 3; ovector[3] = 6;
|
---|
1018 | t = sh_string_replace(s, ovector, ovecnum,
|
---|
1019 | "___", 3);
|
---|
1020 |
|
---|
1021 | CuAssertPtrNotNull(tc, t);
|
---|
1022 | CuAssertStrEquals(tc, "______f ghi ", t->str);
|
---|
1023 | CuAssertIntEquals(tc, 12, (int)t->len);
|
---|
1024 |
|
---|
1025 | ovector[0] = 4; ovector[1] = 5;
|
---|
1026 | ovector[2] = 11; ovector[3] = 12;
|
---|
1027 | t = sh_string_replace(s, ovector, ovecnum,
|
---|
1028 | "___", 3);
|
---|
1029 | CuAssertPtrNotNull(tc, t);
|
---|
1030 | CuAssertStrEquals(tc, "abc ___ef ghi___", t->str);
|
---|
1031 | CuAssertIntEquals(tc, 16, (int)t->len);
|
---|
1032 |
|
---|
1033 | t = sh_string_replace(s, ovector, 0,
|
---|
1034 | "___", 3);
|
---|
1035 | CuAssertPtrNotNull(tc, t);
|
---|
1036 | CuAssertStrEquals(tc, s->str, t->str);
|
---|
1037 | CuAssertIntEquals(tc, (int)s->len, (int)t->len);
|
---|
1038 |
|
---|
1039 | }
|
---|
1040 |
|
---|
1041 | #endif
|
---|