read-cache.c 34.6 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5
/*
 * GIT - The information manager from hell
 *
 * Copyright (C) Linus Torvalds, 2005
 */
6
#define NO_THE_INDEX_COMPATIBILITY_MACROS
7
#include "cache.h"
8
#include "cache-tree.h"
9
#include "refs.h"
10
#include "dir.h"
11 12 13 14 15 16 17 18 19 20 21 22

/* Index extensions.
 *
 * The first letter should be 'A'..'Z' for extensions that are not
 * necessary for a correct operation (i.e. optimization data).
 * When new extensions are added that _needs_ to be understood in
 * order to correctly interpret the index file, pick character that
 * is outside the range, to cause the reader to abort.
 */

#define CACHE_EXT(s) ( (s[0]<<24)|(s[1]<<16)|(s[2]<<8)|(s[3]) )
#define CACHE_EXT_TREE 0x54524545	/* "TREE" */
23

24
struct index_state the_index;
25

26 27 28 29 30 31 32 33 34 35 36
static unsigned int hash_name(const char *name, int namelen)
{
	unsigned int hash = 0x123;

	do {
		unsigned char c = *name++;
		hash = hash*101 + c;
	} while (--namelen);
	return hash;
}

Junio C Hamano's avatar
Junio C Hamano committed
37
static void hash_index_entry(struct index_state *istate, struct cache_entry *ce)
38 39
{
	void **pos;
40
	unsigned int hash;
41

42 43 44 45 46
	if (ce->ce_flags & CE_HASHED)
		return;
	ce->ce_flags |= CE_HASHED;
	ce->next = NULL;
	hash = hash_name(ce->name, ce_namelen(ce));
47 48 49 50 51 52 53
	pos = insert_hash(hash, ce, &istate->name_hash);
	if (pos) {
		ce->next = *pos;
		*pos = ce;
	}
}

Junio C Hamano's avatar
Junio C Hamano committed
54 55 56 57 58 59 60 61 62 63 64 65 66
static void lazy_init_name_hash(struct index_state *istate)
{
	int nr;

	if (istate->name_hash_initialized)
		return;
	for (nr = 0; nr < istate->cache_nr; nr++)
		hash_index_entry(istate, istate->cache[nr]);
	istate->name_hash_initialized = 1;
}

static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
67
	ce->ce_flags &= ~CE_UNHASHED;
Junio C Hamano's avatar
Junio C Hamano committed
68 69 70 71 72
	istate->cache[nr] = ce;
	if (istate->name_hash_initialized)
		hash_index_entry(istate, ce);
}

73 74 75 76
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

77
	remove_index_entry(old);
78
	set_index_entry(istate, nr, ce);
79 80 81 82 83 84
	istate->cache_changed = 1;
}

int index_name_exists(struct index_state *istate, const char *name, int namelen)
{
	unsigned int hash = hash_name(name, namelen);
Junio C Hamano's avatar
Junio C Hamano committed
85 86 87 88
	struct cache_entry *ce;

	lazy_init_name_hash(istate);
	ce = lookup_hash(hash, &istate->name_hash);
89 90 91 92 93 94 95 96 97 98 99

	while (ce) {
		if (!(ce->ce_flags & CE_UNHASHED)) {
			if (!cache_name_compare(name, namelen, ce->name, ce->ce_flags))
				return 1;
		}
		ce = ce->next;
	}
	return 0;
}

100 101 102 103 104 105 106
/*
 * This only updates the "non-critical" parts of the directory
 * cache, ie the parts that aren't tracked by GIT, and only used
 * to validate the cache.
 */
void fill_stat_cache_info(struct cache_entry *ce, struct stat *st)
{
107 108 109 110 111 112 113
	ce->ce_ctime = st->st_ctime;
	ce->ce_mtime = st->st_mtime;
	ce->ce_dev = st->st_dev;
	ce->ce_ino = st->st_ino;
	ce->ce_uid = st->st_uid;
	ce->ce_gid = st->st_gid;
	ce->ce_size = st->st_size;
Junio C Hamano's avatar
Junio C Hamano committed
114 115

	if (assume_unchanged)
116
		ce->ce_flags |= CE_VALID;
117 118 119

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
120 121
}

Junio C Hamano's avatar
Junio C Hamano committed
122 123 124 125 126 127 128
static int ce_compare_data(struct cache_entry *ce, struct stat *st)
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
129
		if (!index_fd(sha1, fd, st, 0, OBJ_BLOB, ce->name))
130
			match = hashcmp(sha1, ce->sha1);
131
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
132 133 134 135
	}
	return match;
}

136
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
137 138 139 140 141
{
	int match = -1;
	char *target;
	void *buffer;
	unsigned long size;
142
	enum object_type type;
Junio C Hamano's avatar
Junio C Hamano committed
143 144 145 146 147 148 149 150
	int len;

	target = xmalloc(expected_size);
	len = readlink(ce->name, target, expected_size);
	if (len != expected_size) {
		free(target);
		return -1;
	}
151
	buffer = read_sha1_file(ce->sha1, &type, &size);
Junio C Hamano's avatar
Junio C Hamano committed
152 153 154 155 156 157 158 159 160 161 162
	if (!buffer) {
		free(target);
		return -1;
	}
	if (size == expected_size)
		match = memcmp(buffer, target, size);
	free(buffer);
	free(target);
	return match;
}

163 164 165 166 167 168
static int ce_compare_gitlink(struct cache_entry *ce)
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
169
	 * under GITLINK directory be a valid git directory. It
170 171 172 173 174 175 176 177 178 179
	 * might even be missing (in case nobody populated that
	 * sub-project).
	 *
	 * If so, we consider it always to match.
	 */
	if (resolve_gitlink_ref(ce->name, "HEAD", sha1) < 0)
		return 0;
	return hashcmp(sha1, ce->sha1);
}

Junio C Hamano's avatar
Junio C Hamano committed
180 181 182 183 184 185 186 187
static int ce_modified_check_fs(struct cache_entry *ce, struct stat *st)
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
188
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
189 190
			return DATA_CHANGED;
		break;
191
	case S_IFDIR:
192
		if (S_ISGITLINK(ce->ce_mode))
193
			return 0;
Junio C Hamano's avatar
Junio C Hamano committed
194 195 196 197 198 199
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
200
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
201 202 203
{
	unsigned int changed = 0;

204 205 206 207
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
208 209
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
210 211 212 213
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
214
		    (0100 & (ce->ce_mode ^ st->st_mode)))
215
			changed |= MODE_CHANGED;
216 217
		break;
	case S_IFLNK:
218 219 220
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
221
		break;
Martin Waitz's avatar
Martin Waitz committed
222
	case S_IFGITLINK:
223 224 225 226
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
227
		return changed;
228
	default:
229
		die("internal error: ce_mode is %o", ce->ce_mode);
230
	}
231
	if (ce->ce_mtime != (unsigned int) st->st_mtime)
232
		changed |= MTIME_CHANGED;
233
	if (ce->ce_ctime != (unsigned int) st->st_ctime)
234
		changed |= CTIME_CHANGED;
235

236 237
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
238
		changed |= OWNER_CHANGED;
239
	if (ce->ce_ino != (unsigned int) st->st_ino)
240
		changed |= INODE_CHANGED;
241 242 243 244 245 246 247

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
248
	if (ce->ce_dev != (unsigned int) st->st_dev)
249 250 251
		changed |= INODE_CHANGED;
#endif

252
	if (ce->ce_size != (unsigned int) st->st_size)
253
		changed |= DATA_CHANGED;
254

Junio C Hamano's avatar
Junio C Hamano committed
255 256 257
	return changed;
}

258
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
259 260 261 262 263
{
	return (istate->timestamp &&
		((unsigned int)istate->timestamp) <= ce->ce_mtime);
}

264
int ie_match_stat(const struct index_state *istate,
265 266
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
267
{
Junio C Hamano's avatar
Junio C Hamano committed
268
	unsigned int changed;
269 270
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
271 272 273 274 275

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
	 */
276
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
277 278 279
		return 0;

	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
280

Junio C Hamano's avatar
Junio C Hamano committed
281 282 283 284 285 286 287 288 289 290 291 292 293 294
	/*
	 * Within 1 second of this sequence:
	 * 	echo xyzzy >file && git-update-index --add file
	 * running this command:
	 * 	echo frotz >file
	 * would give a falsely clean cache entry.  The mtime and
	 * length match the cache, and other stat fields do not change.
	 *
	 * We could detect this at update-index time (the cache entry
	 * being registered/updated records the same time as "now")
	 * and delay the return from git-update-index, but that would
	 * effectively mean we can make at most one commit per second,
	 * which is not acceptable.  Instead, we check cache entries
	 * whose mtime are the same as the index file timestamp more
Junio C Hamano's avatar
Junio C Hamano committed
295
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
296
	 */
297
	if (!changed && is_racy_timestamp(istate, ce)) {
298 299 300 301 302
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
303

Junio C Hamano's avatar
Junio C Hamano committed
304
	return changed;
305 306
}

307
int ie_modified(const struct index_state *istate,
308
		struct cache_entry *ce, struct stat *st, unsigned int options)
309
{
Junio C Hamano's avatar
Junio C Hamano committed
310
	int changed, changed_fs;
311 312

	changed = ie_match_stat(istate, ce, st, options);
313 314 315 316 317 318 319 320 321 322 323 324 325
	if (!changed)
		return 0;
	/*
	 * If the mode or type has changed, there's no point in trying
	 * to refresh the entry - it's not going to match
	 */
	if (changed & (MODE_CHANGED | TYPE_CHANGED))
		return changed;

	/* Immediately after read-tree or update-index --cacheinfo,
	 * the length field is zero.  For other cases the ce_size
	 * should match the SHA1 recorded in the index entry.
	 */
326
	if ((changed & DATA_CHANGED) && ce->ce_size != 0)
327 328
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
329 330 331
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
332 333 334
	return 0;
}

335 336 337 338 339 340 341 342 343 344 345 346
int base_name_compare(const char *name1, int len1, int mode1,
		      const char *name2, int len2, int mode2)
{
	unsigned char c1, c2;
	int len = len1 < len2 ? len1 : len2;
	int cmp;

	cmp = memcmp(name1, name2, len);
	if (cmp)
		return cmp;
	c1 = name1[len];
	c2 = name2[len];
347
	if (!c1 && S_ISDIR(mode1))
348
		c1 = '/';
349
	if (!c2 && S_ISDIR(mode2))
350 351 352 353
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
/*
 * df_name_compare() is identical to base_name_compare(), except it
 * compares conflicting directory/file entries as equal. Note that
 * while a directory name compares as equal to a regular file, they
 * then individually compare _differently_ to a filename that has
 * a dot after the basename (because '\0' < '.' < '/').
 *
 * This is used by routines that want to traverse the git namespace
 * but then handle conflicting entries together when possible.
 */
int df_name_compare(const char *name1, int len1, int mode1,
		    const char *name2, int len2, int mode2)
{
	int len = len1 < len2 ? len1 : len2, cmp;
	unsigned char c1, c2;

	cmp = memcmp(name1, name2, len);
	if (cmp)
		return cmp;
	/* Directories and files compare equal (same length, same name) */
	if (len1 == len2)
		return 0;
	c1 = name1[len];
	if (!c1 && S_ISDIR(mode1))
		c1 = '/';
	c2 = name2[len];
	if (!c2 && S_ISDIR(mode2))
		c2 = '/';
	if (c1 == '/' && !c2)
		return 0;
	if (c2 == '/' && !c1)
		return 0;
	return c1 - c2;
}

389
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
390
{
391 392
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
393 394 395 396 397 398 399 400 401 402
	int len = len1 < len2 ? len1 : len2;
	int cmp;

	cmp = memcmp(name1, name2, len);
	if (cmp)
		return cmp;
	if (len1 < len2)
		return -1;
	if (len1 > len2)
		return 1;
Junio C Hamano's avatar
Junio C Hamano committed
403

404 405 406
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
407

408 409 410 411
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
412 413 414
	return 0;
}

415
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
416 417 418 419
{
	int first, last;

	first = 0;
420
	last = istate->cache_nr;
421 422
	while (last > first) {
		int next = (last + first) >> 1;
423
		struct cache_entry *ce = istate->cache[next];
424
		int cmp = cache_name_compare(name, namelen, ce->name, ce->ce_flags);
425
		if (!cmp)
426
			return next;
427 428 429 430 431 432
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
433
	return -first-1;
434 435
}

436
/* Remove entry, return true if there are more entries to go.. */
437
int remove_index_entry_at(struct index_state *istate, int pos)
438
{
439 440
	struct cache_entry *ce = istate->cache[pos];

441
	remove_index_entry(ce);
442 443 444
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
445
		return 0;
446 447 448
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
449 450 451
	return 1;
}

452
int remove_file_from_index(struct index_state *istate, const char *path)
453
{
454
	int pos = index_name_pos(istate, path, strlen(path));
455 456
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
457
	cache_tree_invalidate_path(istate->cache_tree, path);
458 459
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
460 461 462
	return 0;
}

463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
static int compare_name(struct cache_entry *ce, const char *path, int namelen)
{
	return namelen != ce_namelen(ce) || memcmp(path, ce->name, namelen);
}

static int index_name_pos_also_unmerged(struct index_state *istate,
	const char *path, int namelen)
{
	int pos = index_name_pos(istate, path, namelen);
	struct cache_entry *ce;

	if (pos >= 0)
		return pos;

	/* maybe unmerged? */
	pos = -1 - pos;
	if (pos >= istate->cache_nr ||
			compare_name((ce = istate->cache[pos]), path, namelen))
		return -1;

	/* order of preference: stage 2, 1, 3 */
	if (ce_stage(ce) == 1 && pos + 1 < istate->cache_nr &&
			ce_stage((ce = istate->cache[pos + 1])) == 2 &&
			!compare_name(ce, path, namelen))
		pos++;
	return pos;
}

491
int add_file_to_index(struct index_state *istate, const char *path, int verbose)
Johannes Schindelin's avatar
Johannes Schindelin committed
492
{
493
	int size, namelen, pos;
Johannes Schindelin's avatar
Johannes Schindelin committed
494 495
	struct stat st;
	struct cache_entry *ce;
496
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_RACY_IS_DIRTY;
Johannes Schindelin's avatar
Johannes Schindelin committed
497 498 499 500

	if (lstat(path, &st))
		die("%s: unable to stat (%s)", path, strerror(errno));

501 502
	if (!S_ISREG(st.st_mode) && !S_ISLNK(st.st_mode) && !S_ISDIR(st.st_mode))
		die("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
503 504

	namelen = strlen(path);
505 506 507 508
	if (S_ISDIR(st.st_mode)) {
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
509 510 511
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
512
	ce->ce_flags = namelen;
Johannes Schindelin's avatar
Johannes Schindelin committed
513 514
	fill_stat_cache_info(ce, &st);

515
	if (trust_executable_bit && has_symlinks)
516 517
		ce->ce_mode = create_ce_mode(st.st_mode);
	else {
518 519
		/* If there is an existing entry, pick the mode bits and type
		 * from it, otherwise assume unexecutable regular file.
Johannes Schindelin's avatar
Johannes Schindelin committed
520
		 */
521
		struct cache_entry *ent;
522
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
523

524
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
525
		ce->ce_mode = ce_mode_from_stat(ent, st.st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
526 527
	}

528 529 530
	pos = index_name_pos(istate, ce->name, namelen);
	if (0 <= pos &&
	    !ce_stage(istate->cache[pos]) &&
531
	    !ie_match_stat(istate, istate->cache[pos], &st, ce_option)) {
532 533
		/* Nothing changed, really */
		free(ce);
534
		ce_mark_uptodate(istate->cache[pos]);
535 536 537
		return 0;
	}

Johannes Schindelin's avatar
Johannes Schindelin committed
538 539
	if (index_path(ce->sha1, path, &st, 1))
		die("unable to index file %s", path);
540
	if (add_index_entry(istate, ce, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE))
Johannes Schindelin's avatar
Johannes Schindelin committed
541 542 543 544 545 546
		die("unable to add %s to index",path);
	if (verbose)
		printf("add '%s'\n", path);
	return 0;
}

547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
		int refresh)
{
	int size, len;
	struct cache_entry *ce;

	if (!verify_path(path))
		return NULL;

	len = strlen(path);
	size = cache_entry_size(len);
	ce = xcalloc(1, size);

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
	ce->ce_flags = create_ce_flags(len, stage);
	ce->ce_mode = create_ce_mode(mode);

	if (refresh)
		return refresh_cache_entry(ce, 0);

	return ce;
}

572
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
573 574 575 576 577
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
int ce_path_match(const struct cache_entry *ce, const char **pathspec)
{
	const char *match, *name;
	int len;

	if (!pathspec)
		return 1;

	len = ce_namelen(ce);
	name = ce->name;
	while ((match = *pathspec++) != NULL) {
		int matchlen = strlen(match);
		if (matchlen > len)
			continue;
		if (memcmp(name, match, matchlen))
			continue;
		if (matchlen && name[matchlen-1] == '/')
			return 1;
		if (name[matchlen] == '/' || !name[matchlen])
			return 1;
598 599
		if (!matchlen)
			return 1;
600 601 602 603
	}
	return 0;
}

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667
/*
 * We fundamentally don't like some paths: we don't want
 * dot or dot-dot anywhere, and for obvious reasons don't
 * want to recurse into ".git" either.
 *
 * Also, we don't want double slashes or slashes at the
 * end that can make pathnames ambiguous.
 */
static int verify_dotfile(const char *rest)
{
	/*
	 * The first character was '.', but that
	 * has already been discarded, we now test
	 * the rest.
	 */
	switch (*rest) {
	/* "." is not allowed */
	case '\0': case '/':
		return 0;

	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
		if (rest[1] != 'i')
			break;
		if (rest[2] != 't')
			break;
		rest += 2;
	/* fallthrough */
	case '.':
		if (rest[1] == '\0' || rest[1] == '/')
			return 0;
	}
	return 1;
}

int verify_path(const char *path)
{
	char c;

	goto inside;
	for (;;) {
		if (!c)
			return 1;
		if (c == '/') {
inside:
			c = *path++;
			switch (c) {
			default:
				continue;
			case '/': case '\0':
				break;
			case '.':
				if (verify_dotfile(path))
					continue;
			}
			return 0;
		}
		c = *path++;
	}
}

668 669 670
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
671
 */
672 673
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
674
{
675 676
	int retval = 0;
	int len = ce_namelen(ce);
677
	int stage = ce_stage(ce);
678
	const char *name = ce->name;
679

680 681
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
682

683
		if (len >= ce_namelen(p))
684
			break;
685 686
		if (memcmp(name, p->name, len))
			break;
687 688
		if (ce_stage(p) != stage)
			continue;
689 690
		if (p->name[len] != '/')
			continue;
691
		if (p->ce_flags & CE_REMOVE)
692
			continue;
693 694 695
		retval = -1;
		if (!ok_to_replace)
			break;
696
		remove_index_entry_at(istate, --pos);
697
	}
698 699
	return retval;
}
700

701 702 703 704
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
705 706
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
707 708
{
	int retval = 0;
709
	int stage = ce_stage(ce);
710 711
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
712

713 714
	for (;;) {
		int len;
715

716 717 718 719 720 721 722
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
723

724
		pos = index_name_pos(istate, name, create_ce_flags(len, stage));
725
		if (pos >= 0) {
726 727 728 729 730 731 732 733
			/*
			 * Found one, but not so fast.  This could
			 * be a marker that says "I was here, but
			 * I am being removed".  Such an entry is
			 * not a part of the resulting tree, and
			 * it is Ok to have a directory at the same
			 * path.
			 */
734
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
735 736 737
				retval = -1;
				if (!ok_to_replace)
					break;
738
				remove_index_entry_at(istate, pos);
739 740
				continue;
			}
741
		}
742 743
		else
			pos = -pos-1;
744 745 746 747

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
748
		 * we're ok, and we can exit.
749
		 */
750 751
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
752 753 754 755
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
756 757 758
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
759 760 761 762 763 764
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
765
		}
766
	}
767 768 769 770 771 772 773
	return retval;
}

/* We may be in a situation where we already have path/file and path
 * is being added, or we already have path and path/file is being
 * added.  Either one would result in a nonsense tree that has path
 * twice when git-write-tree tries to write it out.  Prevent it.
Junio C Hamano's avatar
Junio C Hamano committed
774
 *
775 776 777 778
 * If ok-to-replace is specified, we remove the conflicting entries
 * from the cache so the caller should recompute the insert position.
 * When this happens, we return non-zero.
 */
779 780 781
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
782
{
783 784 785 786 787
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
788
	if (ce->ce_flags & CE_REMOVE)
789 790
		return 0;

791 792 793
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
794
	 * in the array.
795
	 */
796
	retval = has_file_name(istate, ce, pos, ok_to_replace);
797

798 799 800 801
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
802
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
803 804
}

805
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
806 807
{
	int pos;
808 809
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
810
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
Junio C Hamano's avatar
Junio C Hamano committed
811

Junio C Hamano's avatar
Junio C Hamano committed
812
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
813
	pos = index_name_pos(istate, ce->name, ce->ce_flags);
814

Junio C Hamano's avatar
Junio C Hamano committed
815
	/* existing match? Just replace it. */
816
	if (pos >= 0) {
817
		replace_index_entry(istate, pos, ce);
818 819
		return 0;
	}
820
	pos = -pos-1;
821

822 823 824 825
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
826 827
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
828
			ok_to_add = 1;
829
			if (!remove_index_entry_at(istate, pos))
830 831 832 833
				break;
		}
	}

834 835
	if (!ok_to_add)
		return -1;
836 837
	if (!verify_path(ce->name))
		return -1;
838

Junio C Hamano's avatar
Junio C Hamano committed
839
	if (!skip_df_check &&
840
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
841
		if (!ok_to_replace)
842 843
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
844
		pos = index_name_pos(istate, ce->name, ce->ce_flags);
845 846
		pos = -pos-1;
	}
847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862
	return pos + 1;
}

int add_index_entry(struct index_state *istate, struct cache_entry *ce, int option)
{
	int pos;

	if (option & ADD_CACHE_JUST_APPEND)
		pos = istate->cache_nr;
	else {
		int ret;
		ret = add_index_entry_with_check(istate, ce, option);
		if (ret <= 0)
			return ret;
		pos = ret - 1;
	}
863

864
	/* Make sure the array is big enough .. */
865 866 867 868
	if (istate->cache_nr == istate->cache_alloc) {
		istate->cache_alloc = alloc_nr(istate->cache_alloc);
		istate->cache = xrealloc(istate->cache,
					istate->cache_alloc * sizeof(struct cache_entry *));
869 870 871
	}

	/* Add it in.. */
872
	istate->cache_nr++;
873
	if (istate->cache_nr > pos + 1)
874 875 876
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
877
	set_index_entry(istate, pos, ce);
878
	istate->cache_changed = 1;
879 880 881
	return 0;
}

882 883 884 885 886 887 888 889 890 891 892
/*
 * "refresh" does not calculate a new sha1 file or bring the
 * cache up-to-date for mode/content changes. But what it
 * _does_ do is to "re-match" the stat information of a file
 * with the cache, so that you can refresh the cache for a
 * file that hasn't been changed but where the stat entry is
 * out of date.
 *
 * For example, you'd want to do this after doing a "git-read-tree",
 * to link up the stat cache details with the proper files.
 */
893
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
894 895
					     struct cache_entry *ce,
					     unsigned int options, int *err)
896 897 898 899
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
900
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
901

902 903 904
	if (ce_uptodate(ce))
		return ce;

905
	if (lstat(ce->name, &st) < 0) {
906 907
		if (err)
			*err = errno;
908 909
		return NULL;
	}
910

911
	changed = ie_match_stat(istate, ce, &st, options);
912
	if (!changed) {
913 914 915 916 917 918 919 920
		/*
		 * The path is unchanged.  If we were told to ignore
		 * valid bit, then we did the actual stat check and
		 * found that the entry is unmodified.  If the entry
		 * is not marked VALID, this is the place to mark it
		 * valid again, under "assume unchanged" mode.
		 */
		if (ignore_valid && assume_unchanged &&
921
		    !(ce->ce_flags & CE_VALID))
922
			; /* mark this one VALID again */
923 924 925 926 927 928 929
		else {
			/*
			 * We do not mark the index itself "modified"
			 * because CE_UPTODATE flag is in-core only;
			 * we are not going to write this change out.
			 */
			ce_mark_uptodate(ce);
930
			return ce;
931
		}
932 933
	}

934
	if (ie_modified(istate, ce, &st, options)) {
935 936
		if (err)
			*err = EINVAL;
937 938
		return NULL;
	}
939 940 941 942 943

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
944 945 946 947 948
	/*
	 * If ignore_valid is not set, we should leave CE_VALID bit
	 * alone.  Otherwise, paths marked with --no-assume-unchanged
	 * (i.e. things to be edited) will reacquire CE_VALID bit
	 * automatically, which is not really what we want.
949
	 */
950
	if (!ignore_valid && assume_unchanged &&
951 952
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
953 954 955 956

	return updated;
}

957
int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec, char *seen)
958 959 960 961 962 963 964
{
	int i;
	int has_errors = 0;
	int really = (flags & REFRESH_REALLY) != 0;
	int allow_unmerged = (flags & REFRESH_UNMERGED) != 0;
	int quiet = (flags & REFRESH_QUIET) != 0;
	int not_new = (flags & REFRESH_IGNORE_MISSING) != 0;
965
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
966

967
	for (i = 0; i < istate->cache_nr; i++) {
968
		struct cache_entry *ce, *new;
969 970
		int cache_errno = 0;

971
		ce = istate->cache[i];
972
		if (ce_stage(ce)) {
973 974
			while ((i < istate->cache_nr) &&
			       ! strcmp(istate->cache[i]->name, ce->name))
975 976 977 978 979 980 981 982 983
				i++;
			i--;
			if (allow_unmerged)
				continue;
			printf("%s: needs merge\n", ce->name);
			has_errors = 1;
			continue;
		}

984 985 986
		if (pathspec && !match_pathspec(pathspec, ce->name, strlen(ce->name), 0, seen))
			continue;

987
		new = refresh_cache_ent(istate, ce, options, &cache_errno);
988
		if (new == ce)
989
			continue;
990 991
		if (!new) {
			if (not_new && cache_errno == ENOENT)
992
				continue;
993
			if (really && cache_errno == EINVAL) {
994 995 996
				/* If we are doing --really-refresh that
				 * means the index is not valid anymore.
				 */
997
				ce->ce_flags &= ~CE_VALID;
998
				istate->cache_changed = 1;
999 1000 1001 1002 1003 1004 1005
			}
			if (quiet)
				continue;
			printf("%s: needs update\n", ce->name);
			has_errors = 1;
			continue;
		}
1006 1007

		replace_index_entry(istate, i, new);
1008 1009 1010 1011
	}
	return has_errors;
}

1012 1013
struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really)
{
1014
	return refresh_cache_ent(&the_index, ce, really, NULL);
1015 1016
}

1017
static int verify_hdr(struct cache_header *hdr, unsigned long size)
1018 1019
{
	SHA_CTX c;
1020
	unsigned char sha1[20];
1021

1022
	if (hdr->hdr_signature != htonl(CACHE_SIGNATURE))
1023
		return error("bad signature");
1024 1025
	if (hdr->hdr_version != htonl(2))
		return error("bad index version");
1026
	SHA1_Init(&c);
1027
	SHA1_Update(&c, hdr, size - 20);
1028
	SHA1_Final(sha1, &c);
1029
	if (hashcmp(sha1, (unsigned char *)hdr + size - 20))
1030
		return error("bad index file sha1 signature");
1031 1032 1033
	return 0;
}

1034 1035
static int read_index_extension(struct index_state *istate,
				const char *ext, void *data, unsigned long sz)
1036 1037 1038
{
	switch (CACHE_EXT(ext)) {
	case CACHE_EXT_TREE:
1039
		istate->cache_tree = cache_tree_read(data, sz);
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050
		break;
	default:
		if (*ext < 'A' || 'Z' < *ext)
			return error("index uses %.4s extension, which we do not understand",
				     ext);
		fprintf(stderr, "ignoring %.4s extension\n", ext);
		break;
	}
	return 0;
}

1051
int read_index(struct index_state *istate)
1052
{
1053
	return read_index_from(istate, get_index_file());
1054 1055
}

1056 1057
static void convert_from_disk(struct ondisk_cache_entry *ondisk, struct cache_entry *ce)
{
1058 1059
	size_t len;

1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070
	ce->ce_ctime = ntohl(ondisk->ctime.sec);
	ce->ce_mtime = ntohl(ondisk->mtime.sec);
	ce->ce_dev   = ntohl(ondisk->dev);
	ce->ce_ino   = ntohl(ondisk->ino);
	ce->ce_mode  = ntohl(ondisk->mode);
	ce->ce_uid   = ntohl(ondisk->uid);
	ce->ce_gid   = ntohl(ondisk->gid);
	ce->ce_size  = ntohl(ondisk->size);
	/* On-disk flags are just 16 bits */
	ce->ce_flags = ntohs(ondisk->flags);
	hashcpy(ce->sha1, ondisk->sha1);
1071 1072 1073 1074 1075 1076 1077 1078 1079

	len = ce->ce_flags & CE_NAMEMASK;
	if (len == CE_NAMEMASK)
		len = strlen(ondisk->name);
	/*
	 * NEEDSWORK: If the original index is crafted, this copy could
	 * go unchecked.
	 */
	memcpy(ce->name, ondisk->name, len + 1);
1080 1081
}

1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095
static inline size_t estimate_cache_size(size_t ondisk_size, unsigned int entries)
{
	long per_entry;

	per_entry = sizeof(struct cache_entry) - sizeof(struct ondisk_cache_entry);

	/*
	 * Alignment can cause differences. This should be "alignof", but
	 * since that's a gcc'ism, just use the size of a pointer.
	 */
	per_entry += sizeof(void *);
	return ondisk_size + entries*per_entry;
}

1096
/* remember to discard_cache() before reading a different cache! */
1097
int read_index_from(struct index_state *istate, const char *path)
1098 1099 1100
{
	int fd, i;
	struct stat st;
1101
	unsigned long src_offset, dst_offset;
1102
	struct cache_header *hdr;
1103 1104
	void *mmap;
	size_t mmap_size;
1105 1106

	errno = EBUSY;
1107
	if (istate->alloc)
1108
		return istate->cache_nr;
1109

1110
	errno = ENOENT;
1111
	istate->timestamp = 0;
1112
	fd = open(path, O_RDONLY);
1113 1114 1115 1116 1117
	if (fd < 0) {
		if (errno == ENOENT)
			return 0;
		die("index file open failed (%s)", strerror(errno));
	}
1118

1119
	if (fstat(fd, &st))
1120
		die("cannot stat the open index (%s)", strerror(errno));
1121 1122

	errno = EINVAL;
1123 1124
	mmap_size = xsize_t(st.st_size);
	if (mmap_size < sizeof(struct cache_header) + 20)
1125 1126
		die("index file smaller than expected");

1127
	mmap = xmmap(NULL, mmap_size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
1128
	close(fd);
1129 1130
	if (mmap == MAP_FAILED)
		die("unable to map index file");
1131

1132 1133
	hdr = mmap;
	if (verify_hdr(hdr, mmap_size) < 0)
1134 1135
		goto unmap;

1136 1137 1138
	istate->cache_nr = ntohl(hdr->hdr_entries);
	istate->cache_alloc = alloc_nr(istate->cache_nr);
	istate->cache = xcalloc(istate->cache_alloc, sizeof(struct cache_entry *));
1139

1140 1141 1142 1143 1144 1145
	/*
	 * The disk format is actually larger than the in-memory format,
	 * due to space for nsec etc, so even though the in-memory one
	 * has room for a few  more flags, we can allocate using the same
	 * index size
	 */
1146
	istate->alloc = xmalloc(estimate_cache_size(mmap_size, istate->cache_nr));
1147 1148 1149

	src_offset = sizeof(*hdr);
	dst_offset = 0;
1150
	for (i = 0; i < istate->cache_nr; i++) {
1151
		struct ondisk_cache_entry *disk_ce;