read-cache.c 57.2 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
#include "tree.h"
#include "commit.h"
13
#include "blob.h"
14
#include "resolve-undo.h"
15 16
#include "strbuf.h"
#include "varint.h"
Duy Nguyen's avatar
Duy Nguyen committed
17
#include "split-index.h"
18

19 20
static struct cache_entry *refresh_cache_entry(struct cache_entry *ce,
					       unsigned int options);
21

22 23 24 25
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

26 27 28 29 30 31 32 33 34 35 36
/* 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" */
37
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
Duy Nguyen's avatar
Duy Nguyen committed
38 39 40
#define CACHE_EXT_LINK 0x6c696e6b	  /* "link" */

/* changes that can be kept in $GIT_DIR/index (basically all extensions) */
41 42
#define EXTMASK (RESOLVE_UNDO_CHANGED | CACHE_TREE_CHANGED | \
		 CE_ENTRY_ADDED)
43

44
struct index_state the_index;
45
static const char *alternate_index_output;
46

Junio C Hamano's avatar
Junio C Hamano committed
47 48 49
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
50
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
51 52
}

53 54 55 56
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

57
	remove_name_hash(istate, old);
58
	free(old);
59
	set_index_entry(istate, nr, ce);
60
	istate->cache_changed |= CE_ENTRY_CHANGED;
61 62
}

63 64 65 66 67 68 69
void rename_index_entry_at(struct index_state *istate, int nr, const char *new_name)
{
	struct cache_entry *old = istate->cache[nr], *new;
	int namelen = strlen(new_name);

	new = xmalloc(cache_entry_size(namelen));
	copy_cache_entry(new, old);
70
	new->ce_flags &= ~CE_HASHED;
71
	new->ce_namelen = namelen;
Duy Nguyen's avatar
Duy Nguyen committed
72
	new->index = 0;
73 74
	memcpy(new->name, new_name, namelen + 1);

75
	cache_tree_invalidate_path(istate, old->name);
76 77 78 79
	remove_index_entry_at(istate, nr);
	add_index_entry(istate, new, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);
}

80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
void fill_stat_data(struct stat_data *sd, struct stat *st)
{
	sd->sd_ctime.sec = (unsigned int)st->st_ctime;
	sd->sd_mtime.sec = (unsigned int)st->st_mtime;
	sd->sd_ctime.nsec = ST_CTIME_NSEC(*st);
	sd->sd_mtime.nsec = ST_MTIME_NSEC(*st);
	sd->sd_dev = st->st_dev;
	sd->sd_ino = st->st_ino;
	sd->sd_uid = st->st_uid;
	sd->sd_gid = st->st_gid;
	sd->sd_size = st->st_size;
}

int match_stat_data(const struct stat_data *sd, struct stat *st)
{
	int changed = 0;

	if (sd->sd_mtime.sec != (unsigned int)st->st_mtime)
		changed |= MTIME_CHANGED;
	if (trust_ctime && check_stat &&
	    sd->sd_ctime.sec != (unsigned int)st->st_ctime)
		changed |= CTIME_CHANGED;

#ifdef USE_NSEC
	if (check_stat && sd->sd_mtime.nsec != ST_MTIME_NSEC(*st))
		changed |= MTIME_CHANGED;
	if (trust_ctime && check_stat &&
	    sd->sd_ctime.nsec != ST_CTIME_NSEC(*st))
		changed |= CTIME_CHANGED;
#endif

	if (check_stat) {
		if (sd->sd_uid != (unsigned int) st->st_uid ||
			sd->sd_gid != (unsigned int) st->st_gid)
			changed |= OWNER_CHANGED;
		if (sd->sd_ino != (unsigned int) st->st_ino)
			changed |= INODE_CHANGED;
	}

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
	if (check_stat && sd->sd_dev != (unsigned int) st->st_dev)
			changed |= INODE_CHANGED;
#endif

	if (sd->sd_size != (unsigned int) st->st_size)
		changed |= DATA_CHANGED;

	return changed;
}

135 136 137 138 139 140 141
/*
 * 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)
{
142
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
143 144

	if (assume_unchanged)
145
		ce->ce_flags |= CE_VALID;
146 147 148

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
149 150
}

151
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
152 153 154 155 156 157
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
158
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
159
			match = hashcmp(sha1, ce->sha1);
160
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
161 162 163 164
	}
	return match;
}

165
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
166 167 168 169
{
	int match = -1;
	void *buffer;
	unsigned long size;
170
	enum object_type type;
171
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
172

173
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
174
		return -1;
175

176
	buffer = read_sha1_file(ce->sha1, &type, &size);
177 178 179 180
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
181
	}
182
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
183 184 185
	return match;
}

186
static int ce_compare_gitlink(const struct cache_entry *ce)
187 188 189 190 191
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
192
	 * under GITLINK directory be a valid git directory. It
193 194 195 196 197 198 199 200 201 202
	 * 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);
}

203
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
204 205 206 207 208 209 210
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
211
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
212 213
			return DATA_CHANGED;
		break;
214
	case S_IFDIR:
215
		if (S_ISGITLINK(ce->ce_mode))
216
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
217 218 219 220 221 222
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

223
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
224 225 226
{
	unsigned int changed = 0;

227 228 229 230
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
231 232
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
233 234 235 236
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
237
		    (0100 & (ce->ce_mode ^ st->st_mode)))
238
			changed |= MODE_CHANGED;
239 240
		break;
	case S_IFLNK:
241 242 243
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
244
		break;
Martin Waitz's avatar
Martin Waitz committed
245
	case S_IFGITLINK:
246
		/* We ignore most of the st_xxx fields for gitlinks */
247 248 249 250
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
251
		return changed;
252
	default:
253
		die("internal error: ce_mode is %o", ce->ce_mode);
254
	}
255

256
	changed |= match_stat_data(&ce->ce_stat_data, st);
257

258
	/* Racily smudged entry? */
259
	if (!ce->ce_stat_data.sd_size) {
260 261 262 263
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
264 265 266
	return changed;
}

267 268
static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
269
{
270
	return (!S_ISGITLINK(ce->ce_mode) &&
271 272 273
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
274 275 276
		(istate->timestamp.sec < ce->ce_stat_data.sd_mtime.sec ||
		 (istate->timestamp.sec == ce->ce_stat_data.sd_mtime.sec &&
		  istate->timestamp.nsec <= ce->ce_stat_data.sd_mtime.nsec))
277
#else
278
		istate->timestamp.sec <= ce->ce_stat_data.sd_mtime.sec
279 280
#endif
		 );
281 282
}

283
int ie_match_stat(const struct index_state *istate,
284
		  const struct cache_entry *ce, struct stat *st,
285
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
286
{
Junio C Hamano's avatar
Junio C Hamano committed
287
	unsigned int changed;
288
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
289
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
290
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
291 292 293 294

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
295 296
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
297
	 */
298 299
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
300
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
301 302
		return 0;

303 304 305 306 307 308 309 310
	/*
	 * Intent-to-add entries have not been added, so the index entry
	 * by definition never matches what is in the work tree until it
	 * actually gets added.
	 */
	if (ce->ce_flags & CE_INTENT_TO_ADD)
		return DATA_CHANGED | TYPE_CHANGED | MODE_CHANGED;

Junio C Hamano's avatar
Junio C Hamano committed
311
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
312

Junio C Hamano's avatar
Junio C Hamano committed
313 314 315 316 317 318 319 320 321 322 323 324 325 326
	/*
	 * 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
327
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
328
	 */
329
	if (!changed && is_racy_timestamp(istate, ce)) {
330 331 332 333 334
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
335

Junio C Hamano's avatar
Junio C Hamano committed
336
	return changed;
337 338
}

339
int ie_modified(const struct index_state *istate,
340 341
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
342
{
Junio C Hamano's avatar
Junio C Hamano committed
343
	int changed, changed_fs;
344 345

	changed = ie_match_stat(istate, ce, st, options);
346 347 348 349 350 351 352 353 354
	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;

355 356 357 358 359 360 361 362 363 364 365 366 367
	/*
	 * Immediately after read-tree or update-index --cacheinfo,
	 * the length field is zero, as we have never even read the
	 * lstat(2) information once, and we cannot trust DATA_CHANGED
	 * returned by ie_match_stat() which in turn was returned by
	 * ce_match_stat_basic() to signal that the filesize of the
	 * blob changed.  We have to actually go to the filesystem to
	 * see if the contents match, and if so, should answer "unchanged".
	 *
	 * The logic does not apply to gitlinks, as ce_match_stat_basic()
	 * already has checked the actual HEAD from the filesystem in the
	 * subproject.  If ie_match_stat() already said it is different,
	 * then we know it is.
368
	 */
369
	if ((changed & DATA_CHANGED) &&
370
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
371 372
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
373 374 375
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
376 377 378
	return 0;
}

379 380 381 382 383 384 385 386 387 388 389 390
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];
391
	if (!c1 && S_ISDIR(mode1))
392
		c1 = '/';
393
	if (!c2 && S_ISDIR(mode2))
394 395 396 397
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
/*
 * 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;
}

433
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
434
{
435 436
	int len = len1 < len2 ? len1 : len2;
	int cmp;
437 438 439 440 441 442 443 444

	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
445

446
	if (stage1 < stage2)
447
		return -1;
448
	if (stage1 > stage2)
449
		return 1;
450 451 452
	return 0;
}

453 454 455 456 457
int cache_name_compare(const char *name1, int len1, const char *name2, int len2)
{
	return cache_name_stage_compare(name1, len1, 0, name2, len2, 0);
}

458
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
459 460 461 462
{
	int first, last;

	first = 0;
463
	last = istate->cache_nr;
464 465
	while (last > first) {
		int next = (last + first) >> 1;
466
		struct cache_entry *ce = istate->cache[next];
467
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
468
		if (!cmp)
469
			return next;
470 471 472 473 474 475
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
476
	return -first-1;
477 478
}

479 480 481 482 483
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

484
/* Remove entry, return true if there are more entries to go.. */
485
int remove_index_entry_at(struct index_state *istate, int pos)
486
{
487 488
	struct cache_entry *ce = istate->cache[pos];

489
	record_resolve_undo(istate, ce);
490
	remove_name_hash(istate, ce);
491
	free(ce);
492
	istate->cache_changed |= CE_ENTRY_REMOVED;
493 494
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
495
		return 0;
496 497 498
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
499 500 501
	return 1;
}

502
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
503
 * Remove all cache entries marked for removal, that is where
504 505 506 507 508 509 510 511 512
 * CE_REMOVE is set in ce_flags.  This is much more effective than
 * calling remove_index_entry_at() for each entry to be removed.
 */
void remove_marked_cache_entries(struct index_state *istate)
{
	struct cache_entry **ce_array = istate->cache;
	unsigned int i, j;

	for (i = j = 0; i < istate->cache_nr; i++) {
513
		if (ce_array[i]->ce_flags & CE_REMOVE) {
514
			remove_name_hash(istate, ce_array[i]);
515 516
			free(ce_array[i]);
		}
517 518 519
		else
			ce_array[j++] = ce_array[i];
	}
520 521
	if (j == istate->cache_nr)
		return;
522
	istate->cache_changed |= CE_ENTRY_REMOVED;
523 524 525
	istate->cache_nr = j;
}

526
int remove_file_from_index(struct index_state *istate, const char *path)
527
{
528
	int pos = index_name_pos(istate, path, strlen(path));
529 530
	if (pos < 0)
		pos = -pos-1;
531
	cache_tree_invalidate_path(istate, path);
532 533
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
534 535 536
	return 0;
}

537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
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;
}

565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596
static int different_name(struct cache_entry *ce, struct cache_entry *alias)
{
	int len = ce_namelen(ce);
	return ce_namelen(alias) != len || memcmp(ce->name, alias->name, len);
}

/*
 * If we add a filename that aliases in the cache, we will use the
 * name that we already have - but we don't want to update the same
 * alias twice, because that implies that there were actually two
 * different files with aliasing names!
 *
 * So we use the CE_ADDED flag to verify that the alias was an old
 * one before we accept it as
 */
static struct cache_entry *create_alias_ce(struct cache_entry *ce, struct cache_entry *alias)
{
	int len;
	struct cache_entry *new;

	if (alias->ce_flags & CE_ADDED)
		die("Will not add file alias '%s' ('%s' already exists in index)", ce->name, alias->name);

	/* Ok, create the new entry using the name of the existing alias */
	len = ce_namelen(alias);
	new = xcalloc(1, cache_entry_size(len));
	memcpy(new->name, alias->name, len);
	copy_cache_entry(new, ce);
	free(ce);
	return new;
}

597
void set_object_name_for_intent_to_add_entry(struct cache_entry *ce)
598 599 600 601 602 603 604
{
	unsigned char sha1[20];
	if (write_sha1_file("", 0, blob_type, sha1))
		die("cannot create an empty blob in the object database");
	hashcpy(ce->sha1, sha1);
}

605
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
606
{
607
	int size, namelen, was_same;
608
	mode_t st_mode = st->st_mode;
609
	struct cache_entry *ce, *alias;
610
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
611 612
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
613 614 615
	int intent_only = flags & ADD_CACHE_INTENT;
	int add_option = (ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE|
			  (intent_only ? ADD_CACHE_NEW_ONLY : 0));
Johannes Schindelin's avatar
Johannes Schindelin committed
616

617
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
618
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
619 620

	namelen = strlen(path);
621
	if (S_ISDIR(st_mode)) {
622 623 624
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
625 626 627
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
628
	ce->ce_namelen = namelen;
629 630
	if (!intent_only)
		fill_stat_cache_info(ce, st);
631 632
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
633

634
	if (trust_executable_bit && has_symlinks)
635
		ce->ce_mode = create_ce_mode(st_mode);
636
	else {
637 638
		/* 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
639
		 */
640
		struct cache_entry *ent;
641
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
642

643
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
644
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
645 646
	}

647 648 649 650 651 652 653 654 655 656 657 658 659 660
	/* When core.ignorecase=true, determine if a directory of the same name but differing
	 * case already exists within the Git repository.  If it does, ensure the directory
	 * case of the file being added to the repository matches (is folded into) the existing
	 * entry's directory case.
	 */
	if (ignore_case) {
		const char *startPtr = ce->name;
		const char *ptr = startPtr;
		while (*ptr) {
			while (*ptr && *ptr != '/')
				++ptr;
			if (*ptr == '/') {
				struct cache_entry *foundce;
				++ptr;
661
				foundce = index_dir_exists(istate, ce->name, ptr - ce->name - 1);
662 663 664 665 666 667 668 669
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

670
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
671
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
672 673
		/* Nothing changed, really */
		free(ce);
674 675
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
676
		alias->ce_flags |= CE_ADDED;
677 678
		return 0;
	}
679
	if (!intent_only) {
680
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
681 682
			return error("unable to index file %s", path);
	} else
683
		set_object_name_for_intent_to_add_entry(ce);
684

685 686 687
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
688

Junio C Hamano's avatar
Junio C Hamano committed
689
	/* It was suspected to be racily clean, but it turns out to be Ok */
690 691 692 693 694 695 696
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
697
	else if (add_index_entry(istate, ce, add_option))
698
		return error("unable to add %s to index",path);
699
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
700 701 702 703
		printf("add '%s'\n", path);
	return 0;
}

704
int add_file_to_index(struct index_state *istate, const char *path, int flags)
705 706 707
{
	struct stat st;
	if (lstat(path, &st))
708
		die_errno("unable to stat '%s'", path);
709
	return add_to_index(istate, path, &st, flags);
710 711
}

712 713
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
714
		unsigned int refresh_options)
715 716 717 718
{
	int size, len;
	struct cache_entry *ce;

719 720
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
721
		return NULL;
722
	}
723 724 725 726 727 728 729

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

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
730 731
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
732 733
	ce->ce_mode = create_ce_mode(mode);

734
	return refresh_cache_entry(ce, refresh_options);
735 736
}

737
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
738 739 740 741 742
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
/*
 * 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.
	 */
758

759
	/* "." is not allowed */
760
	if (*rest == '\0' || is_dir_sep(*rest))
761 762
		return 0;

763
	switch (*rest) {
764 765 766 767 768 769 770 771 772 773 774 775
	/*
	 * ".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 '.':
776
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
777 778 779 780 781 782 783 784 785
			return 0;
	}
	return 1;
}

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

786 787 788
	if (has_dos_drive_prefix(path))
		return 0;

789 790 791 792
	goto inside;
	for (;;) {
		if (!c)
			return 1;
793
		if (is_dir_sep(c)) {
794 795
inside:
			c = *path++;
796 797 798
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
799 800 801 802 803
		}
		c = *path++;
	}
}

804 805 806
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
807
 */
808 809
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
810
{
811 812
	int retval = 0;
	int len = ce_namelen(ce);
813
	int stage = ce_stage(ce);
814
	const char *name = ce->name;
815

816 817
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
818

819
		if (len >= ce_namelen(p))
820
			break;
821 822
		if (memcmp(name, p->name, len))
			break;
823 824
		if (ce_stage(p) != stage)
			continue;
825 826
		if (p->name[len] != '/')
			continue;
827
		if (p->ce_flags & CE_REMOVE)
828
			continue;
829 830 831
		retval = -1;
		if (!ok_to_replace)
			break;
832
		remove_index_entry_at(istate, --pos);
833
	}
834 835
	return retval;
}
836

837 838 839 840
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
841 842
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
843 844
{
	int retval = 0;
845
	int stage = ce_stage(ce);
846 847
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
848

849 850
	for (;;) {
		int len;
851

852 853 854 855 856 857 858
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
859

860
		pos = index_name_stage_pos(istate, name, len, stage);
861
		if (pos >= 0) {
862 863 864 865 866 867 868 869
			/*
			 * 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.
			 */
870
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
871 872 873
				retval = -1;
				if (!ok_to_replace)
					break;
874
				remove_index_entry_at(istate, pos);
875 876
				continue;
			}
877
		}
878 879
		else
			pos = -pos-1;
880 881 882 883

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
884
		 * we're ok, and we can exit.
885
		 */
886 887
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
888 889 890 891
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
892 893 894
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
895 896 897 898 899 900
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
901
		}
902
	}
903 904 905 906 907 908 909
	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
910
 *
911 912 913 914
 * 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.
 */
915 916 917
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
918
{
919 920 921 922 923
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
924
	if (ce->ce_flags & CE_REMOVE)
925 926
		return 0;

927 928 929
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
930
	 * in the array.
931
	 */
932
	retval = has_file_name(istate, ce, pos, ok_to_replace);
933

934 935 936 937
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
938
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
939 940
}

941
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
942 943
{
	int pos;
944 945
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
946
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
947
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
948

949
	cache_tree_invalidate_path(istate, ce->name);
950
	pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
951

Junio C Hamano's avatar
Junio C Hamano committed
952
	/* existing match? Just replace it. */
953
	if (pos >= 0) {
954 955
		if (!new_only)
			replace_index_entry(istate, pos, ce);
956 957
		return 0;
	}
958
	pos = -pos-1;
959

960 961 962 963
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
964 965
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
966
			ok_to_add = 1;
967
			if (!remove_index_entry_at(istate, pos))
968 969 970 971
				break;
		}
	}

972 973
	if (!ok_to_add)
		return -1;
974
	if (!verify_path(ce->name))
975
		return error("Invalid path '%s'", ce->name);
976

Junio C Hamano's avatar
Junio C Hamano committed
977
	if (!skip_df_check &&
978
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
979
		if (!ok_to_replace)
980 981
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
982
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
983 984
		pos = -pos-1;
	}
985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000
	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;
	}
1001

1002
	/* Make sure the array is big enough .. */
1003
	ALLOC_GROW(istate->cache, istate->cache_nr + 1, istate->cache_alloc);
1004 1005

	/* Add it in.. */
1006
	istate->cache_nr++;
1007
	if (istate->cache_nr > pos + 1)
1008 1009 1010
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1011
	set_index_entry(istate, pos, ce);
1012
	istate->cache_changed |= CE_ENTRY_ADDED;
1013 1014 1015
	return 0;
}

1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026
/*
 * "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.
 */
1027
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1028
					     struct cache_entry *ce,
1029 1030
					     unsigned int options, int *err,
					     int *changed_ret)
1031 1032 1033 1034
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1035
	int refresh = options & CE_MATCH_REFRESH;
1036
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1037
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1038
	int ignore_missing = options & CE_MATCH_IGNORE_MISSING;
1039

1040
	if (!refresh || ce_uptodate(ce))
1041 1042
		return ce;

1043
	/*
1044 1045 1046
	 * CE_VALID or CE_SKIP_WORKTREE means the user promised us
	 * that the change to the work tree does not matter and told
	 * us not to worry.
1047
	 */
1048 1049 1050 1051
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1052 1053 1054 1055 1056
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1057
	if (lstat(ce->name, &st) < 0) {
1058 1059
		if (ignore_missing && errno == ENOENT)
			return ce;
1060 1061
		if (err)
			*err = errno;
1062 1063
		return NULL;
	}
1064

1065
	changed = ie_match_stat(istate, ce, &st, options);
1066 1067
	if (changed_ret)
		*changed_ret = changed;
1068
	if (!changed) {
1069 1070 1071 1072 1073 1074 1075 1076
		/*
		 * 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 &&