read-cache.c 69.2 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 "tempfile.h"
9
#include "lockfile.h"
10
#include "cache-tree.h"
11
#include "refs.h"
12
#include "dir.h"
13 14
#include "tree.h"
#include "commit.h"
15
#include "blob.h"
16
#include "resolve-undo.h"
17 18
#include "strbuf.h"
#include "varint.h"
Duy Nguyen's avatar
Duy Nguyen committed
19
#include "split-index.h"
20
#include "utf8.h"
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
#define CACHE_EXT_LINK 0x6c696e6b	  /* "link" */
39
#define CACHE_EXT_UNTRACKED 0x554E5452	  /* "UNTR" */
Duy Nguyen's avatar
Duy Nguyen committed
40 41

/* changes that can be kept in $GIT_DIR/index (basically all extensions) */
42
#define EXTMASK (RESOLVE_UNDO_CHANGED | CACHE_TREE_CHANGED | \
43
		 CE_ENTRY_ADDED | CE_ENTRY_REMOVED | CE_ENTRY_CHANGED | \
44
		 SPLIT_INDEX_ORDERED | UNTRACKED_CHANGED)
45

46
struct index_state the_index;
47
static const char *alternate_index_output;
48

Junio C Hamano's avatar
Junio C Hamano committed
49 50 51
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
52
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
53 54
}

55 56 57 58
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

59
	replace_index_entry_in_base(istate, old, ce);
60
	remove_name_hash(istate, old);
61
	free(old);
62
	set_index_entry(istate, nr, ce);
63
	ce->ce_flags |= CE_UPDATE_IN_BASE;
64
	istate->cache_changed |= CE_ENTRY_CHANGED;
65 66
}

67 68 69 70 71 72 73
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);
74
	new->ce_flags &= ~CE_HASHED;
75
	new->ce_namelen = namelen;
Duy Nguyen's avatar
Duy Nguyen committed
76
	new->index = 0;
77 78
	memcpy(new->name, new_name, namelen + 1);

79
	cache_tree_invalidate_path(istate, old->name);
80
	untracked_cache_remove_from_index(istate, old->name);
81 82 83 84
	remove_index_entry_at(istate, nr);
	add_index_entry(istate, new, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);
}

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139
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;
}

140 141 142 143 144 145 146
/*
 * 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)
{
147
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
148 149

	if (assume_unchanged)
150
		ce->ce_flags |= CE_VALID;
151 152 153

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
154 155
}

156
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
157 158
{
	int match = -1;
159
	int fd = git_open_cloexec(ce->name, O_RDONLY);
Junio C Hamano's avatar
Junio C Hamano committed
160 161 162

	if (fd >= 0) {
		unsigned char sha1[20];
163
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
164
			match = hashcmp(sha1, ce->oid.hash);
165
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
166 167 168 169
	}
	return match;
}

170
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
171 172 173 174
{
	int match = -1;
	void *buffer;
	unsigned long size;
175
	enum object_type type;
176
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
177

178
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
179
		return -1;
180

181
	buffer = read_sha1_file(ce->oid.hash, &type, &size);
182 183 184 185
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
186
	}
187
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
188 189 190
	return match;
}

191
static int ce_compare_gitlink(const struct cache_entry *ce)
192 193 194 195 196
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
197
	 * under GITLINK directory be a valid git directory. It
198 199 200 201 202 203 204
	 * 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;
205
	return hashcmp(sha1, ce->oid.hash);
206 207
}

208
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
209 210 211 212 213 214 215
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
216
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
217 218
			return DATA_CHANGED;
		break;
219
	case S_IFDIR:
220
		if (S_ISGITLINK(ce->ce_mode))
221
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
222 223 224 225 226 227
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

228
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
229 230 231
{
	unsigned int changed = 0;

232 233 234 235
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
236 237
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
238 239 240 241
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
242
		    (0100 & (ce->ce_mode ^ st->st_mode)))
243
			changed |= MODE_CHANGED;
244 245
		break;
	case S_IFLNK:
246 247 248
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
249
		break;
Martin Waitz's avatar
Martin Waitz committed
250
	case S_IFGITLINK:
251
		/* We ignore most of the st_xxx fields for gitlinks */
252 253 254 255
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
256
		return changed;
257
	default:
258
		die("internal error: ce_mode is %o", ce->ce_mode);
259
	}
260

261
	changed |= match_stat_data(&ce->ce_stat_data, st);
262

263
	/* Racily smudged entry? */
264
	if (!ce->ce_stat_data.sd_size) {
265
		if (!is_empty_blob_sha1(ce->oid.hash))
266 267 268
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
269 270 271
	return changed;
}

272 273
static int is_racy_stat(const struct index_state *istate,
			const struct stat_data *sd)
274
{
275
	return (istate->timestamp.sec &&
276 277
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
278 279 280
		(istate->timestamp.sec < sd->sd_mtime.sec ||
		 (istate->timestamp.sec == sd->sd_mtime.sec &&
		  istate->timestamp.nsec <= sd->sd_mtime.nsec))
281
#else
282
		istate->timestamp.sec <= sd->sd_mtime.sec
283
#endif
284 285 286 287 288 289 290 291
		);
}

static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
{
	return (!S_ISGITLINK(ce->ce_mode) &&
		is_racy_stat(istate, &ce->ce_stat_data));
292 293
}

294 295 296 297 298 299
int match_stat_data_racy(const struct index_state *istate,
			 const struct stat_data *sd, struct stat *st)
{
	if (is_racy_stat(istate, sd))
		return MTIME_CHANGED;
	return match_stat_data(sd, st);
300 301
}

302
int ie_match_stat(const struct index_state *istate,
303
		  const struct cache_entry *ce, struct stat *st,
304
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
305
{
Junio C Hamano's avatar
Junio C Hamano committed
306
	unsigned int changed;
307
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
308
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
309
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
310 311 312 313

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
314 315
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
316
	 */
317 318
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
319
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
320 321
		return 0;

322 323 324 325 326
	/*
	 * 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.
	 */
327
	if (ce_intent_to_add(ce))
328 329
		return DATA_CHANGED | TYPE_CHANGED | MODE_CHANGED;

Junio C Hamano's avatar
Junio C Hamano committed
330
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
331

Junio C Hamano's avatar
Junio C Hamano committed
332 333 334 335 336 337 338 339 340 341 342 343 344 345
	/*
	 * 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
346
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
347
	 */
348
	if (!changed && is_racy_timestamp(istate, ce)) {
349 350 351 352 353
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
354

Junio C Hamano's avatar
Junio C Hamano committed
355
	return changed;
356 357
}

358
int ie_modified(const struct index_state *istate,
359 360
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
361
{
Junio C Hamano's avatar
Junio C Hamano committed
362
	int changed, changed_fs;
363 364

	changed = ie_match_stat(istate, ce, st, options);
365 366 367 368 369 370 371 372 373
	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;

374 375 376 377 378 379 380 381 382 383 384 385 386
	/*
	 * 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.
387
	 */
388
	if ((changed & DATA_CHANGED) &&
389
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
390 391
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
392 393 394
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
395 396 397
	return 0;
}

398 399 400 401 402 403 404 405 406 407 408 409
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];
410
	if (!c1 && S_ISDIR(mode1))
411
		c1 = '/';
412
	if (!c2 && S_ISDIR(mode2))
413 414 415 416
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
/*
 * 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;
}

452
int name_compare(const char *name1, size_t len1, const char *name2, size_t len2)
453
{
454 455
	size_t min_len = (len1 < len2) ? len1 : len2;
	int cmp = memcmp(name1, name2, min_len);
456 457 458 459 460 461
	if (cmp)
		return cmp;
	if (len1 < len2)
		return -1;
	if (len1 > len2)
		return 1;
462 463 464 465 466 467 468 469 470 471
	return 0;
}

int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
{
	int cmp;

	cmp = name_compare(name1, len1, name2, len2);
	if (cmp)
		return cmp;
Junio C Hamano's avatar
Junio C Hamano committed
472

473
	if (stage1 < stage2)
474
		return -1;
475
	if (stage1 > stage2)
476
		return 1;
477 478 479
	return 0;
}

480
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
481 482 483 484
{
	int first, last;

	first = 0;
485
	last = istate->cache_nr;
486 487
	while (last > first) {
		int next = (last + first) >> 1;
488
		struct cache_entry *ce = istate->cache[next];
489
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
490
		if (!cmp)
491
			return next;
492 493 494 495 496 497
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
498
	return -first-1;
499 500
}

501 502 503 504 505
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

506
int remove_index_entry_at(struct index_state *istate, int pos)
507
{
508 509
	struct cache_entry *ce = istate->cache[pos];

510
	record_resolve_undo(istate, ce);
511
	remove_name_hash(istate, ce);
512
	save_or_free_index_entry(istate, ce);
513
	istate->cache_changed |= CE_ENTRY_REMOVED;
514 515
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
516
		return 0;
517 518 519
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
520 521 522
	return 1;
}

523
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
524
 * Remove all cache entries marked for removal, that is where
525 526 527 528 529 530 531 532 533
 * 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++) {
534
		if (ce_array[i]->ce_flags & CE_REMOVE) {
535
			remove_name_hash(istate, ce_array[i]);
536
			save_or_free_index_entry(istate, ce_array[i]);
537
		}
538 539 540
		else
			ce_array[j++] = ce_array[i];
	}
541 542
	if (j == istate->cache_nr)
		return;
543
	istate->cache_changed |= CE_ENTRY_REMOVED;
544 545 546
	istate->cache_nr = j;
}

547
int remove_file_from_index(struct index_state *istate, const char *path)
548
{
549
	int pos = index_name_pos(istate, path, strlen(path));
550 551
	if (pos < 0)
		pos = -pos-1;
552
	cache_tree_invalidate_path(istate, path);
553
	untracked_cache_remove_from_index(istate, path);
554 555
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
556 557 558
	return 0;
}

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586
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;
}

587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
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
 */
602 603 604
static struct cache_entry *create_alias_ce(struct index_state *istate,
					   struct cache_entry *ce,
					   struct cache_entry *alias)
605 606 607 608 609 610 611 612 613 614 615 616
{
	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);
617
	save_or_free_index_entry(istate, ce);
618 619 620
	return new;
}

621
void set_object_name_for_intent_to_add_entry(struct cache_entry *ce)
622 623 624 625
{
	unsigned char sha1[20];
	if (write_sha1_file("", 0, blob_type, sha1))
		die("cannot create an empty blob in the object database");
626
	hashcpy(ce->oid.hash, sha1);
627 628
}

629
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
630
{
631
	int size, namelen, was_same;
632
	mode_t st_mode = st->st_mode;
633
	struct cache_entry *ce, *alias;
634
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
635 636
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
637 638 639
	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
640

641
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
642
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
643 644

	namelen = strlen(path);
645
	if (S_ISDIR(st_mode)) {
646 647 648
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
649 650 651
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
652
	ce->ce_namelen = namelen;
653 654
	if (!intent_only)
		fill_stat_cache_info(ce, st);
655 656
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
657

658 659

	if (trust_executable_bit && has_symlinks) {
660
		ce->ce_mode = create_ce_mode(st_mode);
661
	} else {
662 663
		/* 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
664
		 */
665
		struct cache_entry *ent;
666
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
667

668
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
669
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
670 671
	}

672 673 674 675 676 677
	/* 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) {
678
		adjust_dirname_case(istate, ce->name);
679 680
	}

681
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
682
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
683
		/* Nothing changed, really */
684 685
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
686
		alias->ce_flags |= CE_ADDED;
687 688

		free(ce);
689 690
		return 0;
	}
691
	if (!intent_only) {
692
		if (index_path(ce->oid.hash, path, st, HASH_WRITE_OBJECT)) {
693
			free(ce);
694
			return error("unable to index file %s", path);
695
		}
696
	} else
697
		set_object_name_for_intent_to_add_entry(ce);
698

699
	if (ignore_case && alias && different_name(ce, alias))
700
		ce = create_alias_ce(istate, ce, alias);
701
	ce->ce_flags |= CE_ADDED;
702

Junio C Hamano's avatar
Junio C Hamano committed
703
	/* It was suspected to be racily clean, but it turns out to be Ok */
704 705
	was_same = (alias &&
		    !ce_stage(alias) &&
706
		    !oidcmp(&alias->oid, &ce->oid) &&
707 708 709
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
710 711 712 713 714
		free(ce);
	else if (add_index_entry(istate, ce, add_option)) {
		free(ce);
		return error("unable to add %s to index", path);
	}
715
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
716 717 718 719
		printf("add '%s'\n", path);
	return 0;
}

720
int add_file_to_index(struct index_state *istate, const char *path, int flags)
721 722 723
{
	struct stat st;
	if (lstat(path, &st))
724
		die_errno("unable to stat '%s'", path);
725
	return add_to_index(istate, path, &st, flags);
726 727
}

728 729
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
730
		unsigned int refresh_options)
731 732
{
	int size, len;
733
	struct cache_entry *ce, *ret;
734

735 736
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
737
		return NULL;
738
	}
739 740 741 742 743

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

744
	hashcpy(ce->oid.hash, sha1);
745
	memcpy(ce->name, path, len);
746 747
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
748 749
	ce->ce_mode = create_ce_mode(mode);

750
	ret = refresh_cache_entry(ce, refresh_options);
Stefan Beller's avatar
Stefan Beller committed
751
	if (ret != ce)
752
		free(ce);
Stefan Beller's avatar
Stefan Beller committed
753
	return ret;
754 755
}

756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
/*
 * Chmod an index entry with either +x or -x.
 *
 * Returns -1 if the chmod for the particular cache entry failed (if it's
 * not a regular file), -2 if an invalid flip argument is passed in, 0
 * otherwise.
 */
int chmod_index_entry(struct index_state *istate, struct cache_entry *ce,
		      char flip)
{
	if (!S_ISREG(ce->ce_mode))
		return -1;
	switch (flip) {
	case '+':
		ce->ce_mode |= 0111;
		break;
	case '-':
		ce->ce_mode &= ~0111;
		break;
	default:
		return -2;
	}
	cache_tree_invalidate_path(istate, ce->name);
	ce->ce_flags |= CE_UPDATE_IN_BASE;
	istate->cache_changed |= CE_ENTRY_CHANGED;

	return 0;
}

785
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
786 787 788 789 790
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

791 792 793 794 795 796 797 798 799 800 801 802 803 804 805
/*
 * 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.
	 */
806

807
	/* "." is not allowed */
808
	if (*rest == '\0' || is_dir_sep(*rest))
809 810
		return 0;

811
	switch (*rest) {
812 813 814 815 816
	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
817 818
	case 'G':
		if (rest[1] != 'i' && rest[1] != 'I')
819
			break;
820
		if (rest[2] != 't' && rest[2] != 'T')
821 822 823 824
			break;
		rest += 2;
	/* fallthrough */
	case '.':
825
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
826 827 828 829 830 831 832 833 834
			return 0;
	}
	return 1;
}

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

835 836 837
	if (has_dos_drive_prefix(path))
		return 0;

838 839 840 841
	goto inside;
	for (;;) {
		if (!c)
			return 1;
842
		if (is_dir_sep(c)) {
843
inside:
844 845
			if (protect_hfs && is_hfs_dotgit(path))
				return 0;
846 847
			if (protect_ntfs && is_ntfs_dotgit(path))
				return 0;
848
			c = *path++;
849 850 851
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
852 853 854 855 856
		}
		c = *path++;
	}
}

857 858 859
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
860
 */
861 862
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
863
{
864 865
	int retval = 0;
	int len = ce_namelen(ce);
866
	int stage = ce_stage(ce);
867
	const char *name = ce->name;
868

869 870
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
871

872
		if (len >= ce_namelen(p))
873
			break;
874 875
		if (memcmp(name, p->name, len))
			break;
876 877
		if (ce_stage(p) != stage)
			continue;
878 879
		if (p->name[len] != '/')
			continue;
880
		if (p->ce_flags & CE_REMOVE)
881
			continue;
882 883 884
		retval = -1;
		if (!ok_to_replace)
			break;
885
		remove_index_entry_at(istate, --pos);
886
	}
887 888
	return retval;
}
889

890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909

/*
 * Like strcmp(), but also return the offset of the first change.
 * If strings are equal, return the length.
 */
int strcmp_offset(const char *s1, const char *s2, size_t *first_change)
{
	size_t k;

	if (!first_change)
		return strcmp(s1, s2);

	for (k = 0; s1[k] == s2[k]; k++)
		if (s1[k] == '\0')
			break;

	*first_change = k;
	return (unsigned char)s1[k] - (unsigned char)s2[k];
}

910 911 912
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
913 914 915
 *
 * That is, is there another file in the index with a path
 * that matches a sub-directory in the given entry?
916
 */
917 918
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
919 920
{
	int retval = 0;
921
	int stage = ce_stage(ce);
922 923
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965
	size_t len_eq_last;
	int cmp_last = 0;

	/*
	 * We are frequently called during an iteration on a sorted
	 * list of pathnames and while building a new index.  Therefore,
	 * there is a high probability that this entry will eventually
	 * be appended to the index, rather than inserted in the middle.
	 * If we can confirm that, we can avoid binary searches on the
	 * components of the pathname.
	 *
	 * Compare the entry's full path with the last path in the index.
	 */
	if (istate->cache_nr > 0) {
		cmp_last = strcmp_offset(name,
			istate->cache[istate->cache_nr - 1]->name,
			&len_eq_last);
		if (cmp_last > 0) {
			if (len_eq_last == 0) {
				/*
				 * The entry sorts AFTER the last one in the
				 * index and their paths have no common prefix,
				 * so there cannot be a F/D conflict.
				 */
				return retval;
			} else {
				/*
				 * The entry sorts AFTER the last one in the
				 * index, but has a common prefix.  Fall through
				 * to the loop below to disect the entry's path
				 * and see where the difference is.
				 */
			}
		} else if (cmp_last == 0) {
			/*
			 * The entry exactly matches the last one in the
			 * index, but because of multiple stage and CE_REMOVE
			 * items, we fall through and let the regular search
			 * code handle it.
			 */
		}
	}
966

967
	for (;;) {
968
		size_t len;
969

970 971 972 973 974 975 976
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
977

978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038
		if (cmp_last > 0) {
			/*
			 * (len + 1) is a directory boundary (including
			 * the trailing slash).  And since the loop is
			 * decrementing "slash", the first iteration is
			 * the longest directory prefix; subsequent
			 * iterations consider parent directories.
			 */

			if (len + 1 <= len_eq_last) {
				/*
				 * The directory prefix (including the trailing
				 * slash) also appears as a prefix in the last
				 * entry, so the remainder cannot collide (because
				 * strcmp said the whole path was greater).
				 *
				 * EQ: last: xxx/A
				 *     this: xxx/B
				 *
				 * LT: last: xxx/file_A
				 *     this: xxx/file_B
				 */
				return retval;
			}

			if (len > len_eq_last) {
				/*
				 * This part of the directory prefix (excluding
				 * the trailing slash) is longer than the known
				 * equal portions, so this sub-directory cannot
				 * collide with a file.
				 *
				 * GT: last: xxxA
				 *     this: xxxB/file
				 */
				return retval;
			}

			if (istate->cache_nr > 0 &&
				ce_namelen(istate->cache[istate->cache_nr - 1]) > len) {
				/*
				 * The directory prefix lines up with part of
				 * a longer file or directory name, but sorts
				 * after it, so this sub-directory cannot
				 * collide with a file.
				 *
				 * last: xxx/yy-file (because '-' sorts before '/')
				 * this: xxx/yy/abc
				 */
				return retval;
			}

			/*
			 * This is a possible collision. Fall through and
			 * let the regular search code handle it.
			 *
			 * last: xxx
			 * this: xxx/file
			 */
		}

1039
		pos = index_name_stage_pos(istate, name, len, stage);
1040
		if (pos >= 0) {
1041 1042 1043 1044 1045 1046 1047 1048
			/*
			 * 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.
			 */
1049
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
1050 1051 1052
				retval = -1;
				if (!ok_to_replace)
					break;
1053
				remove_index_entry_at(istate, pos);
1054 1055
				continue;
			}
1056
		}
1057 1058
		else
			pos = -pos-1;
1059 1060 1061 1062

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
1063
		 * we're ok, and we can exit.
1064
		 */
1065 1066
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
1067 1068 1069 1070
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
1071 1072 1073
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
1074 1075 1076 1077 1078 1079
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
1080
		}
1081
	}
1082 1083 1084 1085 1086 1087 1088
	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
1089
 *
1090 1091 1092 1093
 * 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.
 */
1094 1095 1096
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
1097
{
1098 1099 1100 1101 1102
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
1103
	if (ce->ce_flags & CE_REMOVE)
1104 1105
		return 0;

1106 1107 1108
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
1109
	 * in the array.
1110
	 */
1111
	retval = has_file_name(istate, ce, pos, ok_to_replace);
1112

1113 1114 1115 1116
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
1117
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
1118 1119
}

1120
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
1121 1122
{
	int pos;
1123 1124
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
1125
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
1126
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
1127

1128 1129
	if (!(option & ADD_CACHE_KEEP_CACHE_TREE))
		cache_tree_invalidate_path(istate, ce->name);
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139

	/*
	 * If this entry's path sorts after the last entry in the index,
	 * we can avoid searching for it.
	 */
	if (istate->cache_nr > 0 &&
		strcmp(ce->name, istate->cache[istate->cache_nr - 1]->name) > 0)
		pos = -istate->cache_nr - 1;
	else
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
1140

Junio C Hamano's avatar
Junio C Hamano committed
1141
	/* existing match? Just replace it. */
1142
	if (pos >= 0) {
1143 1144
		if (!new_only)
			replace_index_entry(istate, pos, ce);
1145 1146
		return 0;
	}
1147
	pos = -pos-1;
1148

1149 1150
	if (!(option & ADD_CACHE_KEEP_CACHE_TREE))
		untracked_cache_add_to_index(istate, ce->name);
1151

1152 1153 1154 1155
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
1156 1157
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
1158
			ok_to_add = 1;
1159
			if (!remove_index_entry_at(istate, pos))
1160 1161 1162 1163