util.c 19.7 KB
Newer Older
Vicent Marti committed
1
/*
Edward Thomson committed
2
 * Copyright (C) the libgit2 contributors. All rights reserved.
Vicent Marti committed
3 4 5 6
 *
 * This file is part of libgit2, distributed under the GNU GPL v2 with
 * a Linking Exception. For full terms see the included COPYING file.
 */
7 8 9

#include "util.h"

10
#include "common.h"
11

12
#ifdef GIT_WIN32
13
# include "win32/utf-conv.h"
14
# include "win32/w32_buffer.h"
15 16 17 18

# ifdef HAVE_QSORT_S
#  include <search.h>
# endif
19 20
#endif

21
#ifdef _MSC_VER
22 23 24
# include <Shlwapi.h>
#endif

25 26 27
void git_strarray_free(git_strarray *array)
{
	size_t i;
28 29 30 31

	if (array == NULL)
		return;

32
	for (i = 0; i < array->count; ++i)
33
		git__free(array->strings[i]);
34

35
	git__free(array->strings);
36 37

	memset(array, 0, sizeof(*array));
38 39
}

40 41 42 43
int git_strarray_copy(git_strarray *tgt, const git_strarray *src)
{
	size_t i;

Russell Belfer committed
44
	assert(tgt && src);
45 46 47

	memset(tgt, 0, sizeof(*tgt));

Russell Belfer committed
48
	if (!src->count)
49 50
		return 0;

Russell Belfer committed
51
	tgt->strings = git__calloc(src->count, sizeof(char *));
52
	GIT_ERROR_CHECK_ALLOC(tgt->strings);
53

Russell Belfer committed
54 55
	for (i = 0; i < src->count; ++i) {
		if (!src->strings[i])
56
			continue;
57

Russell Belfer committed
58
		tgt->strings[tgt->count] = git__strdup(src->strings[i]);
59 60
		if (!tgt->strings[tgt->count]) {
			git_strarray_free(tgt);
Russell Belfer committed
61
			memset(tgt, 0, sizeof(*tgt));
62 63 64 65 66 67 68 69 70
			return -1;
		}

		tgt->count++;
	}

	return 0;
}

71 72
int git__strntol64(int64_t *result, const char *nptr, size_t nptr_len, const char **endptr, int base)
{
73
	const char *p;
74
	int64_t n, nn;
75 76 77 78 79 80 81 82 83 84 85
	int c, ovfl, v, neg, ndig;

	p = nptr;
	neg = 0;
	n = 0;
	ndig = 0;
	ovfl = 0;

	/*
	 * White space
	 */
86 87 88 89 90
	while (nptr_len && git__isspace(*p))
		p++, nptr_len--;

	if (!nptr_len)
		goto Return;
91 92 93 94

	/*
	 * Sign
	 */
95 96
	if (*p == '-' || *p == '+') {
		if (*p == '-')
97
			neg = 1;
98 99 100 101 102 103
		p++;
		nptr_len--;
	}

	if (!nptr_len)
		goto Return;
104 105

	/*
106 107 108 109
	 * Automatically detect the base if none was given to us.
	 * Right now, we assume that a number starting with '0x'
	 * is hexadecimal and a number starting with '0' is
	 * octal.
110 111 112 113
	 */
	if (base == 0) {
		if (*p != '0')
			base = 10;
114 115 116
		else if (nptr_len > 2 && (p[1] == 'x' || p[1] == 'X'))
			base = 16;
		else
117
			base = 8;
118 119 120
	}

	if (base < 0 || 36 < base)
121 122 123
		goto Return;

	/*
124 125 126 127 128 129 130 131 132 133 134 135
	 * Skip prefix of '0x'-prefixed hexadecimal numbers. There is no
	 * need to do the same for '0'-prefixed octal numbers as a
	 * leading '0' does not have any impact. Also, if we skip a
	 * leading '0' in such a string, then we may end up with no
	 * digits left and produce an error later on which isn't one.
	 */
	if (base == 16 && nptr_len > 2 && p[0] == '0' && (p[1] == 'x' || p[1] == 'X')) {
		p += 2;
		nptr_len -= 2;
	}

	/*
136 137
	 * Non-empty sequence of digits
	 */
138
	for (; nptr_len > 0; p++,ndig++,nptr_len--) {
139 140 141 142 143 144 145 146 147 148
		c = *p;
		v = base;
		if ('0'<=c && c<='9')
			v = c - '0';
		else if ('a'<=c && c<='z')
			v = c - 'a' + 10;
		else if ('A'<=c && c<='Z')
			v = c - 'A' + 10;
		if (v >= base)
			break;
149 150
		v = neg ? -v : v;
		if (n > INT64_MAX / base || n < INT64_MIN / base) {
151
			ovfl = 1;
152 153 154 155 156 157 158 159 160 161 162
			/* Keep on iterating until the end of this number */
			continue;
		}
		nn = n * base;
		if ((v > 0 && nn > INT64_MAX - v) ||
		    (v < 0 && nn < INT64_MIN - v)) {
			ovfl = 1;
			/* Keep on iterating until the end of this number */
			continue;
		}
		n = nn + v;
163 164 165
	}

Return:
166
	if (ndig == 0) {
167
		git_error_set(GIT_ERROR_INVALID, "failed to convert string to long: not a number");
168 169
		return -1;
	}
170 171 172 173

	if (endptr)
		*endptr = p;

174
	if (ovfl) {
175
		git_error_set(GIT_ERROR_INVALID, "failed to convert string to long: overflow error");
176 177
		return -1;
	}
178

179
	*result = n;
180
	return 0;
181 182
}

183 184
int git__strntol32(int32_t *result, const char *nptr, size_t nptr_len, const char **endptr, int base)
{
185
	const char *tmp_endptr;
186 187
	int32_t tmp_int;
	int64_t tmp_long;
188
	int error;
189

190
	if ((error = git__strntol64(&tmp_long, nptr, nptr_len, &tmp_endptr, base)) < 0)
191 192 193
		return error;

	tmp_int = tmp_long & 0xFFFFFFFF;
194
	if (tmp_int != tmp_long) {
195
		int len = tmp_endptr - nptr;
196
		git_error_set(GIT_ERROR_INVALID, "failed to convert: '%.*s' is too large", len, nptr);
197 198
		return -1;
	}
199 200

	*result = tmp_int;
201 202
	if (endptr)
		*endptr = tmp_endptr;
203

204 205 206
	return error;
}

207 208 209 210 211 212 213 214 215
int git__strcmp(const char *a, const char *b)
{
	while (*a && *b && *a == *b)
		++a, ++b;
	return (int)(*(const unsigned char *)a) - (int)(*(const unsigned char *)b);
}

int git__strcasecmp(const char *a, const char *b)
{
216
	while (*a && *b && git__tolower(*a) == git__tolower(*b))
217
		++a, ++b;
218
	return ((unsigned char)git__tolower(*a) - (unsigned char)git__tolower(*b));
219 220
}

221 222 223 224 225
int git__strcasesort_cmp(const char *a, const char *b)
{
	int cmp = 0;

	while (*a && *b) {
Russell Belfer committed
226
		if (*a != *b) {
227
			if (git__tolower(*a) != git__tolower(*b))
Russell Belfer committed
228 229
				break;
			/* use case in sort order even if not in equivalence */
230
			if (!cmp)
Russell Belfer committed
231 232
				cmp = (int)(*(const uint8_t *)a) - (int)(*(const uint8_t *)b);
		}
233 234 235 236 237

		++a, ++b;
	}

	if (*a || *b)
238
		return (unsigned char)git__tolower(*a) - (unsigned char)git__tolower(*b);
239 240 241 242

	return cmp;
}

243 244 245 246 247 248 249 250 251 252 253
int git__strncmp(const char *a, const char *b, size_t sz)
{
	while (sz && *a && *b && *a == *b)
		--sz, ++a, ++b;
	if (!sz)
		return 0;
	return (int)(*(const unsigned char *)a) - (int)(*(const unsigned char *)b);
}

int git__strncasecmp(const char *a, const char *b, size_t sz)
{
Philip Kelley committed
254
	int al, bl;
255

Philip Kelley committed
256
	do {
257 258
		al = (unsigned char)git__tolower(*a);
		bl = (unsigned char)git__tolower(*b);
Philip Kelley committed
259 260
		++a, ++b;
	} while (--sz && al && al == bl);
261

Philip Kelley committed
262
	return al - bl;
263 264
}

265
void git__strntolower(char *str, size_t len)
266
{
267
	size_t i;
268 269

	for (i = 0; i < len; ++i) {
270
		str[i] = (char)git__tolower(str[i]);
271 272 273 274 275 276 277 278
	}
}

void git__strtolower(char *str)
{
	git__strntolower(str, strlen(str));
}

279
GIT_INLINE(int) prefixcmp(const char *str, size_t str_n, const char *prefix, bool icase)
280
{
281 282 283 284 285 286 287 288 289 290 291
	int s, p;

	while (str_n--) {
		s = (unsigned char)*str++;
		p = (unsigned char)*prefix++;

		if (icase) {
			s = git__tolower(s);
			p = git__tolower(p);
		}

292 293
		if (!p)
			return 0;
294 295

		if (s != p)
296 297
			return s - p;
	}
298 299

	return (0 - *prefix);
300 301
}

302
int git__prefixcmp(const char *str, const char *prefix)
303
{
304
	return prefixcmp(str, SIZE_MAX, prefix, false);
305 306
}

307
int git__prefixncmp(const char *str, size_t str_n, const char *prefix)
308
{
309 310
	return prefixcmp(str, str_n, prefix, false);
}
311

312 313 314 315
int git__prefixcmp_icase(const char *str, const char *prefix)
{
	return prefixcmp(str, SIZE_MAX, prefix, true);
}
316

317 318 319
int git__prefixncmp_icase(const char *str, size_t str_n, const char *prefix)
{
	return prefixcmp(str, str_n, prefix, true);
320 321
}

322 323 324 325 326 327 328 329
int git__suffixcmp(const char *str, const char *suffix)
{
	size_t a = strlen(str);
	size_t b = strlen(suffix);
	if (a < b)
		return -1;
	return strcmp(str + (a - b), suffix);
}
330

331
char *git__strtok(char **end, const char *sep)
332
{
333
	char *ptr = *end;
334

335 336
	while (*ptr && strchr(sep, *ptr))
		++ptr;
337

338 339 340
	if (*ptr) {
		char *start = ptr;
		*end = start + 1;
341

342 343
		while (**end && !strchr(sep, **end))
			++*end;
344

345 346 347 348 349 350 351 352 353
		if (**end) {
			**end = '\0';
			++*end;
		}

		return start;
	}

	return NULL;
354 355
}

356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
/* Similar to strtok, but does not collapse repeated tokens. */
char *git__strsep(char **end, const char *sep)
{
	char *start = *end, *ptr = *end;

	while (*ptr && !strchr(sep, *ptr))
		++ptr;

	if (*ptr) {
		*end = ptr + 1;
		*ptr = '\0';

		return start;
	}

	return NULL;
}

374 375 376 377 378 379
size_t git__linenlen(const char *buffer, size_t buffer_len)
{
	char *nl = memchr(buffer, '\n', buffer_len);
	return nl ? (size_t)(nl - buffer) + 1 : buffer_len;
}

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
/*
 * Adapted Not So Naive algorithm from http://www-igm.univ-mlv.fr/~lecroq/string/
 */
const void * git__memmem(const void *haystack, size_t haystacklen,
			 const void *needle, size_t needlelen)
{
	const char *h, *n;
	size_t j, k, l;

	if (needlelen > haystacklen || !haystacklen || !needlelen)
		return NULL;

	h = (const char *) haystack,
	n = (const char *) needle;

	if (needlelen == 1)
		return memchr(haystack, *n, haystacklen);

	if (n[0] == n[1]) {
		k = 2;
		l = 1;
	} else {
		k = 1;
		l = 2;
	}

	j = 0;
	while (j <= haystacklen - needlelen) {
		if (n[1] != h[j + 1]) {
			j += k;
		} else {
			if (memcmp(n + 2, h + j + 2, needlelen - 2) == 0 &&
			    n[0] == h[j])
				return h + j;
			j += l;
		}
	}

	return NULL;
}

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
void git__hexdump(const char *buffer, size_t len)
{
	static const size_t LINE_WIDTH = 16;

	size_t line_count, last_line, i, j;
	const char *line;

	line_count = (len / LINE_WIDTH);
	last_line = (len % LINE_WIDTH);

	for (i = 0; i < line_count; ++i) {
		line = buffer + (i * LINE_WIDTH);
		for (j = 0; j < LINE_WIDTH; ++j, ++line)
			printf("%02X ", (unsigned char)*line & 0xFF);

		printf("| ");

		line = buffer + (i * LINE_WIDTH);
		for (j = 0; j < LINE_WIDTH; ++j, ++line)
			printf("%c", (*line >= 32 && *line <= 126) ? *line : '.');

		printf("\n");
	}

	if (last_line > 0) {

		line = buffer + (line_count * LINE_WIDTH);
		for (j = 0; j < last_line; ++j, ++line)
			printf("%02X ", (unsigned char)*line & 0xFF);

		for (j = 0; j < (LINE_WIDTH - last_line); ++j)
Vicent Marti committed
452
			printf("	");
453 454 455 456 457 458 459 460 461 462 463 464

		printf("| ");

		line = buffer + (line_count * LINE_WIDTH);
		for (j = 0; j < last_line; ++j, ++line)
			printf("%c", (*line >= 32 && *line <= 126) ? *line : '.');

		printf("\n");
	}

	printf("\n");
}
465 466 467 468 469 470 471 472 473 474 475 476 477

#ifdef GIT_LEGACY_HASH
uint32_t git__hash(const void *key, int len, unsigned int seed)
{
	const uint32_t m = 0x5bd1e995;
	const int r = 24;
	uint32_t h = seed ^ len;

	const unsigned char *data = (const unsigned char *)key;

	while(len >= 4) {
		uint32_t k = *(uint32_t *)data;

478 479 480 481 482
		k *= m;
		k ^= k >> r;
		k *= m;

		h *= m;
483 484 485 486 487
		h ^= k;

		data += 4;
		len -= 4;
	}
488

489 490 491 492
	switch(len) {
	case 3: h ^= data[2] << 16;
	case 2: h ^= data[1] << 8;
	case 1: h ^= data[0];
Vicent Marti committed
493
			h *= m;
494 495 496 497 498 499 500
	};

	h ^= h >> 13;
	h *= m;
	h ^= h >> 15;

	return h;
501
}
502 503 504 505 506 507 508 509 510 511 512 513
#else
/*
	Cross-platform version of Murmurhash3
	http://code.google.com/p/smhasher/wiki/MurmurHash3
	by Austin Appleby (aappleby@gmail.com)

	This code is on the public domain.
*/
uint32_t git__hash(const void *key, int len, uint32_t seed)
{

#define MURMUR_BLOCK() {\
Vicent Marti committed
514 515 516 517 518 519 520
	k1 *= c1; \
	k1 = git__rotl(k1,11);\
	k1 *= c2;\
	h1 ^= k1;\
	h1 = h1*3 + 0x52dce729;\
	c1 = c1*5 + 0x7b7d159c;\
	c2 = c2*5 + 0x6bce6396;\
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
}

	const uint8_t *data = (const uint8_t*)key;
	const int nblocks = len / 4;

	const uint32_t *blocks = (const uint32_t *)(data + nblocks * 4);
	const uint8_t *tail = (const uint8_t *)(data + nblocks * 4);

	uint32_t h1 = 0x971e137b ^ seed;
	uint32_t k1;

	uint32_t c1 = 0x95543787;
	uint32_t c2 = 0x2ad7eb25;

	int i;

	for (i = -nblocks; i; i++) {
		k1 = blocks[i];
		MURMUR_BLOCK();
	}

	k1 = 0;

	switch(len & 3) {
	case 3: k1 ^= tail[2] << 16;
546
		/* fall through */
547
	case 2: k1 ^= tail[1] << 8;
548
		/* fall through */
549
	case 1: k1 ^= tail[0];
550
		MURMUR_BLOCK();
551 552 553 554 555 556 557 558 559 560
	}

	h1 ^= len;
	h1 ^= h1 >> 16;
	h1 *= 0x85ebca6b;
	h1 ^= h1 >> 13;
	h1 *= 0xc2b2ae35;
	h1 ^= h1 >> 16;

	return h1;
561
}
562
#endif
563

564 565 566 567 568
/**
 * A modified `bsearch` from the BSD glibc.
 *
 * Copyright (c) 1990 Regents of the University of California.
 * All rights reserved.
569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 * notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 * notice, this list of conditions and the following disclaimer in the
 * documentation and/or other materials provided with the distribution.
 * 3. [rescinded 22 July 1999]
 * 4. Neither the name of the University nor the names of its contributors
 * may be used to endorse or promote products derived from this software
 * without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
593
 */
594 595 596 597 598 599
int git__bsearch(
	void **array,
	size_t array_len,
	const void *key,
	int (*compare)(const void *, const void *),
	size_t *position)
600
{
601
	size_t lim;
602
	int cmp = -1;
603 604
	void **part, **base = array;

605
	for (lim = array_len; lim != 0; lim >>= 1) {
606 607 608
		part = base + (lim >> 1);
		cmp = (*compare)(key, *part);
		if (cmp == 0) {
609 610 611 612
			base = part;
			break;
		}
		if (cmp > 0) { /* key > p; take right partition */
613 614 615 616 617
			base = part + 1;
			lim--;
		} /* else take left partition */
	}

618 619
	if (position)
		*position = (base - array);
620

621
	return (cmp == 0) ? 0 : GIT_ENOTFOUND;
622 623 624 625 626 627 628 629 630 631
}

int git__bsearch_r(
	void **array,
	size_t array_len,
	const void *key,
	int (*compare_r)(const void *, const void *, void *),
	void *payload,
	size_t *position)
{
632
	size_t lim;
633 634 635
	int cmp = -1;
	void **part, **base = array;

636
	for (lim = array_len; lim != 0; lim >>= 1) {
637 638 639 640 641 642 643 644 645 646 647 648 649 650
		part = base + (lim >> 1);
		cmp = (*compare_r)(key, *part, payload);
		if (cmp == 0) {
			base = part;
			break;
		}
		if (cmp > 0) { /* key > p; take right partition */
			base = part + 1;
			lim--;
		} /* else take left partition */
	}

	if (position)
		*position = (base - array);
651

652
	return (cmp == 0) ? 0 : GIT_ENOTFOUND;
653 654
}

655 656
/**
 * A strcmp wrapper
657
 *
658 659 660 661 662
 * We don't want direct pointers to the CRT on Windows, we may
 * get stdcall conflicts.
 */
int git__strcmp_cb(const void *a, const void *b)
{
663 664
	return strcmp((const char *)a, (const char *)b);
}
665

666 667 668
int git__strcasecmp_cb(const void *a, const void *b)
{
	return strcasecmp((const char *)a, (const char *)b);
669
}
670 671 672 673

int git__parse_bool(int *out, const char *value)
{
	/* A missing value means true */
674 675
	if (value == NULL ||
		!strcasecmp(value, "true") ||
676 677 678 679 680 681 682
		!strcasecmp(value, "yes") ||
		!strcasecmp(value, "on")) {
		*out = 1;
		return 0;
	}
	if (!strcasecmp(value, "false") ||
		!strcasecmp(value, "no") ||
683 684
		!strcasecmp(value, "off") ||
		value[0] == '\0') {
685 686 687 688 689 690
		*out = 0;
		return 0;
	}

	return -1;
}
691 692 693 694 695

size_t git__unescape(char *str)
{
	char *scan, *pos = str;

696 697 698
	if (!str)
		return 0;

699 700 701 702 703 704 705 706 707 708 709 710 711
	for (scan = str; *scan; pos++, scan++) {
		if (*scan == '\\' && *(scan + 1) != '\0')
			scan++; /* skip '\' but include next char */
		if (pos != scan)
			*pos = *scan;
	}

	if (pos != scan) {
		*pos = '\0';
	}

	return (pos - str);
}
712

713
#if defined(HAVE_QSORT_S) || defined(HAVE_QSORT_R_BSD)
714
typedef struct {
715
	git__sort_r_cmp cmp;
716 717 718 719 720 721 722 723 724 725 726 727
	void *payload;
} git__qsort_r_glue;

static int GIT_STDLIB_CALL git__qsort_r_glue_cmp(
	void *payload, const void *a, const void *b)
{
	git__qsort_r_glue *glue = payload;
	return glue->cmp(a, b, glue->payload);
}
#endif

void git__qsort_r(
728
	void *els, size_t nel, size_t elsize, git__sort_r_cmp cmp, void *payload)
729
{
730
#if defined(HAVE_QSORT_R_BSD)
731 732
	git__qsort_r_glue glue = { cmp, payload };
	qsort_r(els, nel, elsize, &glue, git__qsort_r_glue_cmp);
733
#elif defined(HAVE_QSORT_R_GNU)
734
	qsort_r(els, nel, elsize, cmp, payload);
735 736 737 738 739
#elif defined(HAVE_QSORT_S)
	git__qsort_r_glue glue = { cmp, payload };
	qsort_s(els, nel, elsize, git__qsort_r_glue_cmp, &glue);
#else
	git__insertsort_r(els, nel, elsize, NULL, cmp, payload);
740
#endif
741 742 743 744 745 746
}

void git__insertsort_r(
	void *els, size_t nel, size_t elsize, void *swapel,
	git__sort_r_cmp cmp, void *payload)
{
747 748
	uint8_t *base = els;
	uint8_t *end = base + nel * elsize;
749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
	uint8_t *i, *j;
	bool freeswap = !swapel;

	if (freeswap)
		swapel = git__malloc(elsize);

	for (i = base + elsize; i < end; i += elsize)
		for (j = i; j > base && cmp(j, j - elsize, payload) < 0; j -= elsize) {
			memcpy(swapel, j, elsize);
			memcpy(j, j - elsize, elsize);
			memcpy(j - elsize, swapel, elsize);
		}

	if (freeswap)
		git__free(swapel);
764
}
765

766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790
/*
 * git__utf8_iterate is taken from the utf8proc project,
 * http://www.public-software-group.org/utf8proc
 *
 * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the ""Software""),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
 * DEALINGS IN THE SOFTWARE.
 */

791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865
static const int8_t utf8proc_utf8class[256] = {
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
	2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
	2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
	3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
	4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0
};

int git__utf8_charlen(const uint8_t *str, int str_len)
{
	int length, i;

	length = utf8proc_utf8class[str[0]];
	if (!length)
		return -1;

	if (str_len >= 0 && length > str_len)
		return -str_len;

	for (i = 1; i < length; i++) {
		if ((str[i] & 0xC0) != 0x80)
			return -i;
	}

	return length;
}

int git__utf8_iterate(const uint8_t *str, int str_len, int32_t *dst)
{
	int length;
	int32_t uc = -1;

	*dst = -1;
	length = git__utf8_charlen(str, str_len);
	if (length < 0)
		return -1;

	switch (length) {
		case 1:
			uc = str[0];
			break;
		case 2:
			uc = ((str[0] & 0x1F) <<  6) + (str[1] & 0x3F);
			if (uc < 0x80) uc = -1;
			break;
		case 3:
			uc = ((str[0] & 0x0F) << 12) + ((str[1] & 0x3F) <<  6)
				+ (str[2] & 0x3F);
			if (uc < 0x800 || (uc >= 0xD800 && uc < 0xE000) ||
					(uc >= 0xFDD0 && uc < 0xFDF0)) uc = -1;
			break;
		case 4:
			uc = ((str[0] & 0x07) << 18) + ((str[1] & 0x3F) << 12)
				+ ((str[2] & 0x3F) <<  6) + (str[3] & 0x3F);
			if (uc < 0x10000 || uc >= 0x110000) uc = -1;
			break;
	}

	if (uc < 0 || ((uc & 0xFFFF) >= 0xFFFE))
		return -1;

	*dst = uc;
	return length;
}
866

867 868 869 870 871
double git_time_monotonic(void)
{
	return git__timer();
}

872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887
size_t git__utf8_valid_buf_length(const uint8_t *str, size_t str_len)
{
	size_t offset = 0;

	while (offset < str_len) {
		int length = git__utf8_charlen(str + offset, str_len - offset);

		if (length < 0)
			break;

		offset += length;
	}

	return offset;
}

888 889 890 891 892 893 894 895 896 897 898 899 900 901
#ifdef GIT_WIN32
int git__getenv(git_buf *out, const char *name)
{
	wchar_t *wide_name = NULL, *wide_value = NULL;
	DWORD value_len;
	int error = -1;

	git_buf_clear(out);

	if (git__utf8_to_16_alloc(&wide_name, name) < 0)
		return -1;

	if ((value_len = GetEnvironmentVariableW(wide_name, NULL, 0)) > 0) {
		wide_value = git__malloc(value_len * sizeof(wchar_t));
902
		GIT_ERROR_CHECK_ALLOC(wide_value);
903 904 905 906 907 908 909 910 911

		value_len = GetEnvironmentVariableW(wide_name, wide_value, value_len);
	}

	if (value_len)
		error = git_buf_put_w(out, wide_value, value_len);
	else if (GetLastError() == ERROR_ENVVAR_NOT_FOUND)
		error = GIT_ENOTFOUND;
	else
912
		git_error_set(GIT_ERROR_OS, "could not read environment variable '%s'", name);
913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930

	git__free(wide_name);
	git__free(wide_value);
	return error;
}
#else
int git__getenv(git_buf *out, const char *name)
{
	const char *val = getenv(name);

	git_buf_clear(out);

	if (!val)
		return GIT_ENOTFOUND;

	return git_buf_puts(out, val);
}
#endif