Fri, 11 May 2018 19:48:19 +0200
documents (and fixes!) the UcxLogger
1 /*
2 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER.
3 *
4 * Copyright 2017 Mike Becker, Olaf Wintermann 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 are met:
8 *
9 * 1. Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 *
12 * 2. 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 *
16 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
17 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
20 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
21 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
22 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
23 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
24 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
25 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
26 * POSSIBILITY OF SUCH DAMAGE.
27 */
29 #include "ucx/string.h"
31 #include "ucx/allocator.h"
33 #include <stdlib.h>
34 #include <string.h>
35 #include <stdarg.h>
36 #include <stdint.h>
37 #include <ctype.h>
39 sstr_t sstr(char *cstring) {
40 sstr_t string;
41 string.ptr = cstring;
42 string.length = strlen(cstring);
43 return string;
44 }
46 sstr_t sstrn(char *cstring, size_t length) {
47 sstr_t string;
48 string.ptr = cstring;
49 string.length = length;
50 return string;
51 }
53 size_t sstrnlen(size_t n, sstr_t s, ...) {
54 va_list ap;
55 size_t size = s.length;
56 va_start(ap, s);
58 for (size_t i = 1 ; i < n ; i++) {
59 sstr_t str = va_arg(ap, sstr_t);
60 if(((size_t)-1) - str.length < size) {
61 size = 0;
62 break;
63 }
64 size += str.length;
65 }
66 va_end(ap);
68 return size;
69 }
71 static sstr_t sstrvcat_a(
72 UcxAllocator *a,
73 size_t count,
74 sstr_t s1,
75 sstr_t s2,
76 va_list ap) {
77 sstr_t str;
78 str.ptr = NULL;
79 str.length = 0;
80 if(count < 2) {
81 return str;
82 }
84 if(((size_t)-1) - s1.length < s2.length) {
85 return str;
86 }
88 sstr_t *strings = (sstr_t*) calloc(count, sizeof(sstr_t));
89 if(!strings) {
90 return str;
91 }
93 // get all args and overall length
94 strings[0] = s1;
95 strings[1] = s2;
96 size_t slen = s1.length + s2.length;
97 int error = 0;
98 for (size_t i=2;i<count;i++) {
99 sstr_t s = va_arg (ap, sstr_t);
100 strings[i] = s;
101 if(((size_t)-1) - s.length < slen) {
102 error = 1;
103 break;
104 }
105 slen += s.length;
106 }
107 if(error) {
108 free(strings);
109 return str;
110 }
112 // create new string
113 str.ptr = (char*) almalloc(a, slen + 1);
114 str.length = slen;
115 if(!str.ptr) {
116 free(strings);
117 str.length = 0;
118 return str;
119 }
121 // concatenate strings
122 size_t pos = 0;
123 for (size_t i=0;i<count;i++) {
124 sstr_t s = strings[i];
125 memcpy(str.ptr + pos, s.ptr, s.length);
126 pos += s.length;
127 }
129 str.ptr[str.length] = '\0';
131 free(strings);
133 return str;
134 }
136 sstr_t sstrcat(size_t count, sstr_t s1, sstr_t s2, ...) {
137 va_list ap;
138 va_start(ap, s2);
139 sstr_t s = sstrvcat_a(ucx_default_allocator(), count, s1, s2, ap);
140 va_end(ap);
141 return s;
142 }
144 sstr_t sstrcat_a(UcxAllocator *a, size_t count, sstr_t s1, sstr_t s2, ...) {
145 va_list ap;
146 va_start(ap, s2);
147 sstr_t s = sstrvcat_a(a, count, s1, s2, ap);
148 va_end(ap);
149 return s;
150 }
152 sstr_t sstrsubs(sstr_t s, size_t start) {
153 return sstrsubsl (s, start, s.length-start);
154 }
156 sstr_t sstrsubsl(sstr_t s, size_t start, size_t length) {
157 sstr_t new_sstr;
158 if (start >= s.length) {
159 new_sstr.ptr = NULL;
160 new_sstr.length = 0;
161 } else {
162 if (length > s.length-start) {
163 length = s.length-start;
164 }
165 new_sstr.ptr = &s.ptr[start];
166 new_sstr.length = length;
167 }
168 return new_sstr;
169 }
171 sstr_t sstrchr(sstr_t s, int c) {
172 for(size_t i=0;i<s.length;i++) {
173 if(s.ptr[i] == c) {
174 return sstrsubs(s, i);
175 }
176 }
177 sstr_t n;
178 n.ptr = NULL;
179 n.length = 0;
180 return n;
181 }
183 sstr_t sstrrchr(sstr_t s, int c) {
184 if (s.length > 0) {
185 for(size_t i=s.length;i>0;i--) {
186 if(s.ptr[i-1] == c) {
187 return sstrsubs(s, i-1);
188 }
189 }
190 }
191 sstr_t n;
192 n.ptr = NULL;
193 n.length = 0;
194 return n;
195 }
197 #define ptable_r(dest, useheap, ptable, index) (dest = useheap ? \
198 ((size_t*)ptable)[index] : (size_t) ((uint8_t*)ptable)[index])
200 #define ptable_w(useheap, ptable, index, src) do {\
201 if (!useheap) ((uint8_t*)ptable)[index] = (uint8_t) src;\
202 else ((size_t*)ptable)[index] = src;\
203 } while (0);
205 sstr_t sstrstr(sstr_t string, sstr_t match) {
206 if (match.length == 0) {
207 return string;
208 }
210 /* prepare default return value in case of no match */
211 sstr_t result = sstrn(NULL, 0);
213 /*
214 * IMPORTANT:
215 * our prefix table contains the prefix length PLUS ONE
216 * this is our decision, because we want to use the full range of size_t
217 * the original algorithm needs a (-1) at one single place
218 * and we want to avoid that
219 */
221 /* static prefix table */
222 static uint8_t s_prefix_table[256];
224 /* check pattern length and use appropriate prefix table */
225 /* if the pattern exceeds static prefix table, allocate on the heap */
226 register int useheap = match.length > 255;
227 register void* ptable = useheap ?
228 calloc(match.length+1, sizeof(size_t)): s_prefix_table;
230 /* keep counter in registers */
231 register size_t i, j;
233 /* fill prefix table */
234 i = 0; j = 0;
235 ptable_w(useheap, ptable, i, j);
236 while (i < match.length) {
237 while (j >= 1 && match.ptr[j-1] != match.ptr[i]) {
238 ptable_r(j, useheap, ptable, j-1);
239 }
240 i++; j++;
241 ptable_w(useheap, ptable, i, j);
242 }
244 /* search */
245 i = 0; j = 1;
246 while (i < string.length) {
247 while (j >= 1 && string.ptr[i] != match.ptr[j-1]) {
248 ptable_r(j, useheap, ptable, j-1);
249 }
250 i++; j++;
251 if (j-1 == match.length) {
252 size_t start = i - match.length;
253 result.ptr = string.ptr + start;
254 result.length = string.length - start;
255 break;
256 }
257 }
259 /* if prefix table was allocated on the heap, free it */
260 if (ptable != s_prefix_table) {
261 free(ptable);
262 }
264 return result;
265 }
267 #undef ptable_r
268 #undef ptable_w
270 sstr_t* sstrsplit(sstr_t s, sstr_t d, ssize_t *n) {
271 return sstrsplit_a(ucx_default_allocator(), s, d, n);
272 }
274 sstr_t* sstrsplit_a(UcxAllocator *allocator, sstr_t s, sstr_t d, ssize_t *n) {
275 if (s.length == 0 || d.length == 0) {
276 *n = -1;
277 return NULL;
278 }
280 /* special cases: delimiter is at least as large as the string */
281 if (d.length >= s.length) {
282 /* exact match */
283 if (sstrcmp(s, d) == 0) {
284 *n = 0;
285 return NULL;
286 } else /* no match possible */ {
287 *n = 1;
288 sstr_t *result = (sstr_t*) almalloc(allocator, sizeof(sstr_t));
289 if(result) {
290 *result = sstrdup_a(allocator, s);
291 } else {
292 *n = -2;
293 }
294 return result;
295 }
296 }
298 ssize_t nmax = *n;
299 size_t arrlen = 16;
300 sstr_t* result = (sstr_t*) alcalloc(allocator, arrlen, sizeof(sstr_t));
302 if (result) {
303 sstr_t curpos = s;
304 ssize_t j = 1;
305 while (1) {
306 sstr_t match;
307 /* optimize for one byte delimiters */
308 if (d.length == 1) {
309 match = curpos;
310 for (size_t i = 0 ; i < curpos.length ; i++) {
311 if (curpos.ptr[i] == *(d.ptr)) {
312 match.ptr = curpos.ptr + i;
313 break;
314 }
315 match.length--;
316 }
317 } else {
318 match = sstrstr(curpos, d);
319 }
320 if (match.length > 0) {
321 /* is this our last try? */
322 if (nmax == 0 || j < nmax) {
323 /* copy the current string to the array */
324 sstr_t item = sstrn(curpos.ptr, match.ptr - curpos.ptr);
325 result[j-1] = sstrdup_a(allocator, item);
326 size_t processed = item.length + d.length;
327 curpos.ptr += processed;
328 curpos.length -= processed;
330 /* allocate memory for the next string */
331 j++;
332 if (j > arrlen) {
333 arrlen *= 2;
334 size_t reallocsz;
335 sstr_t* reallocated = NULL;
336 if(!ucx_szmul(arrlen, sizeof(sstr_t), &reallocsz)) {
337 reallocated = (sstr_t*) alrealloc(
338 allocator, result, reallocsz);
339 }
340 if (reallocated) {
341 result = reallocated;
342 } else {
343 for (ssize_t i = 0 ; i < j-1 ; i++) {
344 alfree(allocator, result[i].ptr);
345 }
346 alfree(allocator, result);
347 *n = -2;
348 return NULL;
349 }
350 }
351 } else {
352 /* nmax reached, copy the _full_ remaining string */
353 result[j-1] = sstrdup_a(allocator, curpos);
354 break;
355 }
356 } else {
357 /* no more matches, copy last string */
358 result[j-1] = sstrdup_a(allocator, curpos);
359 break;
360 }
361 }
362 *n = j;
363 } else {
364 *n = -2;
365 }
367 return result;
368 }
370 int sstrcmp(sstr_t s1, sstr_t s2) {
371 if (s1.length == s2.length) {
372 return memcmp(s1.ptr, s2.ptr, s1.length);
373 } else if (s1.length > s2.length) {
374 return 1;
375 } else {
376 return -1;
377 }
378 }
380 int sstrcasecmp(sstr_t s1, sstr_t s2) {
381 if (s1.length == s2.length) {
382 #ifdef _WIN32
383 return _strnicmp(s1.ptr, s2.ptr, s1.length);
384 #else
385 return strncasecmp(s1.ptr, s2.ptr, s1.length);
386 #endif
387 } else if (s1.length > s2.length) {
388 return 1;
389 } else {
390 return -1;
391 }
392 }
394 sstr_t sstrdup(sstr_t s) {
395 return sstrdup_a(ucx_default_allocator(), s);
396 }
398 sstr_t sstrdup_a(UcxAllocator *allocator, sstr_t s) {
399 sstr_t newstring;
400 newstring.ptr = (char*)almalloc(allocator, s.length + 1);
401 if (newstring.ptr) {
402 newstring.length = s.length;
403 newstring.ptr[newstring.length] = 0;
405 memcpy(newstring.ptr, s.ptr, s.length);
406 } else {
407 newstring.length = 0;
408 }
410 return newstring;
411 }
413 sstr_t sstrtrim(sstr_t string) {
414 sstr_t newstr = string;
416 while (newstr.length > 0 && isspace(*newstr.ptr)) {
417 newstr.ptr++;
418 newstr.length--;
419 }
420 while (newstr.length > 0 && isspace(newstr.ptr[newstr.length-1])) {
421 newstr.length--;
422 }
424 return newstr;
425 }
427 int sstrprefix(sstr_t string, sstr_t prefix) {
428 if (string.length == 0) {
429 return prefix.length == 0;
430 }
431 if (prefix.length == 0) {
432 return 1;
433 }
435 if (prefix.length > string.length) {
436 return 0;
437 } else {
438 return memcmp(string.ptr, prefix.ptr, prefix.length) == 0;
439 }
440 }
442 int sstrsuffix(sstr_t string, sstr_t suffix) {
443 if (string.length == 0) {
444 return suffix.length == 0;
445 }
446 if (suffix.length == 0) {
447 return 1;
448 }
450 if (suffix.length > string.length) {
451 return 0;
452 } else {
453 return memcmp(string.ptr+string.length-suffix.length,
454 suffix.ptr, suffix.length) == 0;
455 }
456 }
458 sstr_t sstrlower(sstr_t string) {
459 sstr_t ret = sstrdup(string);
460 for (size_t i = 0; i < ret.length ; i++) {
461 ret.ptr[i] = tolower(ret.ptr[i]);
462 }
463 return ret;
464 }
466 sstr_t sstrlower_a(UcxAllocator *allocator, sstr_t string) {
467 sstr_t ret = sstrdup_a(allocator, string);
468 for (size_t i = 0; i < ret.length ; i++) {
469 ret.ptr[i] = tolower(ret.ptr[i]);
470 }
471 return ret;
472 }
474 sstr_t sstrupper(sstr_t string) {
475 sstr_t ret = sstrdup(string);
476 for (size_t i = 0; i < ret.length ; i++) {
477 ret.ptr[i] = toupper(ret.ptr[i]);
478 }
479 return ret;
480 }
482 sstr_t sstrupper_a(UcxAllocator *allocator, sstr_t string) {
483 sstr_t ret = sstrdup_a(allocator, string);
484 for (size_t i = 0; i < ret.length ; i++) {
485 ret.ptr[i] = toupper(ret.ptr[i]);
486 }
487 return ret;
488 }