read-cache.c 70 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 "config.h"
9
#include "tempfile.h"
10
#include "lockfile.h"
11
#include "cache-tree.h"
12
#include "refs.h"
13
#include "dir.h"
14 15
#include "tree.h"
#include "commit.h"
16
#include "blob.h"
17
#include "resolve-undo.h"
18 19
#include "strbuf.h"
#include "varint.h"
Duy Nguyen's avatar
Duy Nguyen committed
20
#include "split-index.h"
21
#include "utf8.h"
22

23 24 25 26
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

27 28 29 30 31 32 33 34 35 36 37
/* 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" */
38
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
Duy Nguyen's avatar
Duy Nguyen committed
39
#define CACHE_EXT_LINK 0x6c696e6b	  /* "link" */
40
#define CACHE_EXT_UNTRACKED 0x554E5452	  /* "UNTR" */
Duy Nguyen's avatar
Duy Nguyen committed
41 42

/* changes that can be kept in $GIT_DIR/index (basically all extensions) */
43
#define EXTMASK (RESOLVE_UNDO_CHANGED | CACHE_TREE_CHANGED | \
44
		 CE_ENTRY_ADDED | CE_ENTRY_REMOVED | CE_ENTRY_CHANGED | \
45
		 SPLIT_INDEX_ORDERED | UNTRACKED_CHANGED)
46

47
struct index_state the_index;
48
static const char *alternate_index_output;
49

Junio C Hamano's avatar
Junio C Hamano committed
50 51 52
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
53
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
54 55
}

56 57 58 59
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

60
	replace_index_entry_in_base(istate, old, ce);
61
	remove_name_hash(istate, old);
62
	free(old);
63
	set_index_entry(istate, nr, ce);
64
	ce->ce_flags |= CE_UPDATE_IN_BASE;
65
	istate->cache_changed |= CE_ENTRY_CHANGED;
66 67
}

68 69 70 71 72 73 74
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);
75
	new->ce_flags &= ~CE_HASHED;
76
	new->ce_namelen = namelen;
Duy Nguyen's avatar
Duy Nguyen committed
77
	new->index = 0;
78 79
	memcpy(new->name, new_name, namelen + 1);

80
	cache_tree_invalidate_path(istate, old->name);
81
	untracked_cache_remove_from_index(istate, old->name);
82 83 84 85
	remove_index_entry_at(istate, nr);
	add_index_entry(istate, new, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);
}

86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
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;
}

141 142 143 144 145 146 147
/*
 * 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)
{
148
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
149 150

	if (assume_unchanged)
151
		ce->ce_flags |= CE_VALID;
152 153 154

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
155 156
}

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

	if (fd >= 0) {
163
		struct object_id oid;
164
		if (!index_fd(&oid, fd, st, OBJ_BLOB, ce->name, 0))
165
			match = oidcmp(&oid, &ce->oid);
166
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
167 168 169 170
	}
	return match;
}

171
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
172 173 174 175
{
	int match = -1;
	void *buffer;
	unsigned long size;
176
	enum object_type type;
177
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
178

179
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
180
		return -1;
181

182
	buffer = read_sha1_file(ce->oid.hash, &type, &size);
183 184 185 186
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
187
	}
188
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
189 190 191
	return match;
}

192
static int ce_compare_gitlink(const struct cache_entry *ce)
193
{
194
	struct object_id oid;
195 196 197

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
198
	 * under GITLINK directory be a valid git directory. It
199 200 201 202 203
	 * might even be missing (in case nobody populated that
	 * sub-project).
	 *
	 * If so, we consider it always to match.
	 */
204
	if (resolve_gitlink_ref(ce->name, "HEAD", oid.hash) < 0)
205
		return 0;
206
	return oidcmp(&oid, &ce->oid);
207 208
}

209
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
210 211 212 213 214 215 216
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
217
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
218 219
			return DATA_CHANGED;
		break;
220
	case S_IFDIR:
221
		if (S_ISGITLINK(ce->ce_mode))
222
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
223
		/* else fallthrough */
Junio C Hamano's avatar
Junio C Hamano committed
224 225 226 227 228 229
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

230
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
231 232 233
{
	unsigned int changed = 0;

234 235 236 237
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
238 239
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
240 241 242 243
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
244
		    (0100 & (ce->ce_mode ^ st->st_mode)))
245
			changed |= MODE_CHANGED;
246 247
		break;
	case S_IFLNK:
248 249 250
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
251
		break;
Martin Waitz's avatar
Martin Waitz committed
252
	case S_IFGITLINK:
253
		/* We ignore most of the st_xxx fields for gitlinks */
254 255 256 257
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
258
		return changed;
259
	default:
260
		die("internal error: ce_mode is %o", ce->ce_mode);
261
	}
262

263
	changed |= match_stat_data(&ce->ce_stat_data, st);
264

265
	/* Racily smudged entry? */
266
	if (!ce->ce_stat_data.sd_size) {
267
		if (!is_empty_blob_sha1(ce->oid.hash))
268 269 270
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
271 272 273
	return changed;
}

274 275
static int is_racy_stat(const struct index_state *istate,
			const struct stat_data *sd)
276
{
277
	return (istate->timestamp.sec &&
278 279
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
280 281 282
		(istate->timestamp.sec < sd->sd_mtime.sec ||
		 (istate->timestamp.sec == sd->sd_mtime.sec &&
		  istate->timestamp.nsec <= sd->sd_mtime.nsec))
283
#else
284
		istate->timestamp.sec <= sd->sd_mtime.sec
285
#endif
286 287 288 289 290 291 292 293
		);
}

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));
294 295
}

296 297 298 299 300 301
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);
302 303
}

304
int ie_match_stat(const struct index_state *istate,
305
		  const struct cache_entry *ce, struct stat *st,
306
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
307
{
Junio C Hamano's avatar
Junio C Hamano committed
308
	unsigned int changed;
309
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
310
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
311
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
312 313 314 315

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
316 317
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
318
	 */
319 320
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
321
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
322 323
		return 0;

324 325 326 327 328
	/*
	 * 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.
	 */
329
	if (ce_intent_to_add(ce))
330 331
		return DATA_CHANGED | TYPE_CHANGED | MODE_CHANGED;

Junio C Hamano's avatar
Junio C Hamano committed
332
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
333

Junio C Hamano's avatar
Junio C Hamano committed
334 335 336 337 338 339 340 341 342 343 344 345 346 347
	/*
	 * 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
348
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
349
	 */
350
	if (!changed && is_racy_timestamp(istate, ce)) {
351 352 353 354 355
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
356

Junio C Hamano's avatar
Junio C Hamano committed
357
	return changed;
358 359
}

360
int ie_modified(const struct index_state *istate,
361 362
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
363
{
Junio C Hamano's avatar
Junio C Hamano committed
364
	int changed, changed_fs;
365 366

	changed = ie_match_stat(istate, ce, st, options);
367 368 369 370 371 372 373 374 375
	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;

376 377 378 379 380 381 382 383 384 385 386 387 388
	/*
	 * 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.
389
	 */
390
	if ((changed & DATA_CHANGED) &&
391
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
392 393
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
394 395 396
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
397 398 399
	return 0;
}

400 401 402 403 404 405 406 407 408 409 410 411
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];
412
	if (!c1 && S_ISDIR(mode1))
413
		c1 = '/';
414
	if (!c2 && S_ISDIR(mode2))
415 416 417 418
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453
/*
 * 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;
}

454
int name_compare(const char *name1, size_t len1, const char *name2, size_t len2)
455
{
456 457
	size_t min_len = (len1 < len2) ? len1 : len2;
	int cmp = memcmp(name1, name2, min_len);
458 459 460 461 462 463
	if (cmp)
		return cmp;
	if (len1 < len2)
		return -1;
	if (len1 > len2)
		return 1;
464 465 466 467 468 469 470 471 472 473
	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
474

475
	if (stage1 < stage2)
476
		return -1;
477
	if (stage1 > stage2)
478
		return 1;
479 480 481
	return 0;
}

482
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
483 484 485 486
{
	int first, last;

	first = 0;
487
	last = istate->cache_nr;
488 489
	while (last > first) {
		int next = (last + first) >> 1;
490
		struct cache_entry *ce = istate->cache[next];
491
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
492
		if (!cmp)
493
			return next;
494 495 496 497 498 499
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
500
	return -first-1;
501 502
}

503 504 505 506 507
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

508
int remove_index_entry_at(struct index_state *istate, int pos)
509
{
510 511
	struct cache_entry *ce = istate->cache[pos];

512
	record_resolve_undo(istate, ce);
513
	remove_name_hash(istate, ce);
514
	save_or_free_index_entry(istate, ce);
515
	istate->cache_changed |= CE_ENTRY_REMOVED;
516 517
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
518
		return 0;
René Scharfe's avatar
René Scharfe committed
519 520
	MOVE_ARRAY(istate->cache + pos, istate->cache + pos + 1,
		   istate->cache_nr - pos);
521 522 523
	return 1;
}

524
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
525
 * Remove all cache entries marked for removal, that is where
526 527 528 529 530 531 532 533 534
 * 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++) {
535
		if (ce_array[i]->ce_flags & CE_REMOVE) {
536
			remove_name_hash(istate, ce_array[i]);
537
			save_or_free_index_entry(istate, ce_array[i]);
538
		}
539 540 541
		else
			ce_array[j++] = ce_array[i];
	}
542 543
	if (j == istate->cache_nr)
		return;
544
	istate->cache_changed |= CE_ENTRY_REMOVED;
545 546 547
	istate->cache_nr = j;
}

548
int remove_file_from_index(struct index_state *istate, const char *path)
549
{
550
	int pos = index_name_pos(istate, path, strlen(path));
551 552
	if (pos < 0)
		pos = -pos-1;
553
	cache_tree_invalidate_path(istate, path);
554
	untracked_cache_remove_from_index(istate, path);
555 556
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
557 558 559
	return 0;
}

560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
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;
}

588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
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
 */
603 604 605
static struct cache_entry *create_alias_ce(struct index_state *istate,
					   struct cache_entry *ce,
					   struct cache_entry *alias)
606 607 608 609 610 611 612 613 614 615 616 617
{
	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);
618
	save_or_free_index_entry(istate, ce);
619 620 621
	return new;
}

622
void set_object_name_for_intent_to_add_entry(struct cache_entry *ce)
623 624 625 626
{
	unsigned char sha1[20];
	if (write_sha1_file("", 0, blob_type, sha1))
		die("cannot create an empty blob in the object database");
627
	hashcpy(ce->oid.hash, sha1);
628 629
}

630
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
631
{
632
	int size, namelen, was_same;
633
	mode_t st_mode = st->st_mode;
634
	struct cache_entry *ce, *alias;
635
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
636 637
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
638 639 640
	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
641

642
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
643
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
644 645

	namelen = strlen(path);
646
	if (S_ISDIR(st_mode)) {
647 648 649
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
650 651 652
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
653
	ce->ce_namelen = namelen;
654 655
	if (!intent_only)
		fill_stat_cache_info(ce, st);
656 657
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
658

659 660

	if (trust_executable_bit && has_symlinks) {
661
		ce->ce_mode = create_ce_mode(st_mode);
662
	} else {
663 664
		/* 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
665
		 */
666
		struct cache_entry *ent;
667
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
668

669
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
670
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
671 672
	}

673 674 675 676 677 678
	/* 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) {
679
		adjust_dirname_case(istate, ce->name);
680 681
	}

682
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
683
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
684
		/* Nothing changed, really */
685 686
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
687
		alias->ce_flags |= CE_ADDED;
688 689

		free(ce);
690 691
		return 0;
	}
692
	if (!intent_only) {
693
		if (index_path(&ce->oid, path, st, HASH_WRITE_OBJECT)) {
694
			free(ce);
695
			return error("unable to index file %s", path);
696
		}
697
	} else
698
		set_object_name_for_intent_to_add_entry(ce);
699

700
	if (ignore_case && alias && different_name(ce, alias))
701
		ce = create_alias_ce(istate, ce, alias);
702
	ce->ce_flags |= CE_ADDED;
703

Junio C Hamano's avatar
Junio C Hamano committed
704
	/* It was suspected to be racily clean, but it turns out to be Ok */
705 706
	was_same = (alias &&
		    !ce_stage(alias) &&
707
		    !oidcmp(&alias->oid, &ce->oid) &&
708 709 710
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
711 712 713 714 715
		free(ce);
	else if (add_index_entry(istate, ce, add_option)) {
		free(ce);
		return error("unable to add %s to index", path);
	}
716
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
717 718 719 720
		printf("add '%s'\n", path);
	return 0;
}

721
int add_file_to_index(struct index_state *istate, const char *path, int flags)
722 723 724
{
	struct stat st;
	if (lstat(path, &st))
725
		die_errno("unable to stat '%s'", path);
726
	return add_to_index(istate, path, &st, flags);
727 728
}

729 730
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
731
		unsigned int refresh_options)
732 733
{
	int size, len;
734
	struct cache_entry *ce, *ret;
735

736 737
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
738
		return NULL;
739
	}
740 741 742 743 744

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

745
	hashcpy(ce->oid.hash, sha1);
746
	memcpy(ce->name, path, len);
747 748
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
749 750
	ce->ce_mode = create_ce_mode(mode);

751
	ret = refresh_cache_entry(ce, refresh_options);
Stefan Beller's avatar
Stefan Beller committed
752
	if (ret != ce)
753
		free(ce);
Stefan Beller's avatar
Stefan Beller committed
754
	return ret;
755 756
}

757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785
/*
 * 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;
}

786
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
787 788 789 790 791
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
/*
 * 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.
	 */
807

808
	/* "." is not allowed */
809
	if (*rest == '\0' || is_dir_sep(*rest))
810 811
		return 0;

812
	switch (*rest) {
813 814 815 816 817
	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
818 819
	case 'G':
		if (rest[1] != 'i' && rest[1] != 'I')
820
			break;
821
		if (rest[2] != 't' && rest[2] != 'T')
822 823 824 825
			break;
		rest += 2;
	/* fallthrough */
	case '.':
826
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
827 828 829 830 831 832 833 834 835
			return 0;
	}
	return 1;
}

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

836 837 838
	if (has_dos_drive_prefix(path))
		return 0;

839 840 841 842
	goto inside;
	for (;;) {
		if (!c)
			return 1;
843
		if (is_dir_sep(c)) {
844
inside:
845 846
			if (protect_hfs && is_hfs_dotgit(path))
				return 0;
847 848
			if (protect_ntfs && is_ntfs_dotgit(path))
				return 0;
849
			c = *path++;
850 851 852
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
853 854 855 856 857
		}
		c = *path++;
	}
}

858 859 860
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
861
 */
862 863
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
864
{
865 866
	int retval = 0;
	int len = ce_namelen(ce);
867
	int stage = ce_stage(ce);
868
	const char *name = ce->name;
869

870 871
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
872

873
		if (len >= ce_namelen(p))
874
			break;
875 876
		if (memcmp(name, p->name, len))
			break;
877 878
		if (ce_stage(p) != stage)
			continue;
879 880
		if (p->name[len] != '/')
			continue;
881
		if (p->ce_flags & CE_REMOVE)
882
			continue;
883 884 885
		retval = -1;
		if (!ok_to_replace)
			break;
886
		remove_index_entry_at(istate, --pos);
887
	}
888 889
	return retval;
}
890

891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910

/*
 * 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];
}

911 912 913
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
914 915 916
 *
 * That is, is there another file in the index with a path
 * that matches a sub-directory in the given entry?
917
 */
918 919
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
920 921
{
	int retval = 0;
922
	int stage = ce_stage(ce);
923 924
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966
	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.
			 */
		}
	}
967

968
	for (;;) {
969
		size_t len;
970

971 972 973 974 975 976 977
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
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 1039
		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
			 */
		}

1040
		pos = index_name_stage_pos(istate, name, len, stage);
1041
		if (pos >= 0) {
1042 1043 1044 1045 1046 1047 1048 1049
			/*
			 * 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.
			 */
1050
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
1051 1052 1053
				retval = -1;
				if (!ok_to_replace)
					break;
1054
				remove_index_entry_at(istate, pos);
1055 1056
				continue;
			}
1057
		}
1058 1059
		else
			pos = -pos-1;
1060 1061 1062 1063

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
1064
		 * we're ok, and we can exit.
1065
		 */
1066 1067
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
1068 1069 1070 1071
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
1072 1073 1074
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
1075 1076 1077 1078 1079 1080
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
1081
		}
1082
	}
1083 1084 1085 1086 1087 1088 1089
	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
1090
 *
1091 1092 1093 1094
 * 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.
 */
1095 1096 1097
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
1098
{
1099 1100 1101 1102 1103
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
1104
	if (ce->ce_flags & CE_REMOVE)
1105 1106
		return 0;

1107 1108 1109
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
1110
	 * in the array.
1111
	 */
1112
	retval = has_file_name(istate, ce, pos, ok_to_replace);
1113

1114 1115 1116 1117
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
1118
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
1119 1120
}

1121
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
1122 1123
{
	int pos;
1124 1125
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
1126
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
1127
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
1128

1129 1130
	if (!(option & ADD_CACHE_KEEP_CACHE_TREE))
		cache_tree_invalidate_path(istate, ce->name);
1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

	/*
	 * 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));
1141

Junio C Hamano's avatar
Junio C Hamano committed
1142
	/* existing match? Just replace it. */
1143
	if (pos >= 0) {
1144 1145
		if (!new_only)
			replace_index_entry(istate, pos, ce);
1146 1147
		return 0;
	}
1148
	pos = -pos-1;
1149

1150 1151
	if (!(option & ADD_CACHE_KEEP_CACHE_TREE))
		untracked_cache_add_to_index(istate, ce->name);
1152