read-cache.c 53 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"
17
#include "utf8.h"
18

19 20
static struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really);

21 22 23 24
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

25 26 27 28 29 30 31 32 33 34 35
/* 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" */
36
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
37

38
struct index_state the_index;
39

Junio C Hamano's avatar
Junio C Hamano committed
40 41 42
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
43
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
44 45
}

46 47 48 49
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

50
	remove_name_hash(istate, old);
51
	set_index_entry(istate, nr, ce);
52 53 54
	istate->cache_changed = 1;
}

55 56 57 58 59 60 61
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);
62 63
	new->ce_flags &= ~CE_STATE_MASK;
	new->ce_namelen = namelen;
64 65 66 67 68 69 70
	memcpy(new->name, new_name, namelen + 1);

	cache_tree_invalidate_path(istate->cache_tree, old->name);
	remove_index_entry_at(istate, nr);
	add_index_entry(istate, new, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);
}

71 72 73 74 75 76 77 78 79 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
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;
}

126 127 128 129 130 131 132
/*
 * 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)
{
133
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
134 135

	if (assume_unchanged)
136
		ce->ce_flags |= CE_VALID;
137 138 139

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
140 141
}

142
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
143 144 145 146 147 148
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
149
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
150
			match = hashcmp(sha1, ce->sha1);
151
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
152 153 154 155
	}
	return match;
}

156
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
157 158 159 160
{
	int match = -1;
	void *buffer;
	unsigned long size;
161
	enum object_type type;
162
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
163

164
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
165
		return -1;
166

167
	buffer = read_sha1_file(ce->sha1, &type, &size);
168 169 170 171
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
172
	}
173
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
174 175 176
	return match;
}

177
static int ce_compare_gitlink(const struct cache_entry *ce)
178 179 180 181 182
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
183
	 * under GITLINK directory be a valid git directory. It
184 185 186 187 188 189 190 191 192 193
	 * 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);
}

194
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
195 196 197 198 199 200 201
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
202
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
203 204
			return DATA_CHANGED;
		break;
205
	case S_IFDIR:
206
		if (S_ISGITLINK(ce->ce_mode))
207
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
208 209 210 211 212 213
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

214
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
215 216 217
{
	unsigned int changed = 0;

218 219 220 221
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
222 223
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
224 225 226 227
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
228
		    (0100 & (ce->ce_mode ^ st->st_mode)))
229
			changed |= MODE_CHANGED;
230 231
		break;
	case S_IFLNK:
232 233 234
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
235
		break;
Martin Waitz's avatar
Martin Waitz committed
236
	case S_IFGITLINK:
237
		/* We ignore most of the st_xxx fields for gitlinks */
238 239 240 241
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
242
		return changed;
243
	default:
244
		die("internal error: ce_mode is %o", ce->ce_mode);
245
	}
246

247
	changed |= match_stat_data(&ce->ce_stat_data, st);
248

249
	/* Racily smudged entry? */
250
	if (!ce->ce_stat_data.sd_size) {
251 252 253 254
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

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

258 259
static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
260
{
261
	return (!S_ISGITLINK(ce->ce_mode) &&
262 263 264
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
265 266 267
		(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))
268
#else
269
		istate->timestamp.sec <= ce->ce_stat_data.sd_mtime.sec
270 271
#endif
		 );
272 273
}

274
int ie_match_stat(const struct index_state *istate,
275
		  const struct cache_entry *ce, struct stat *st,
276
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
277
{
Junio C Hamano's avatar
Junio C Hamano committed
278
	unsigned int changed;
279
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
280
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
281
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
282 283 284 285

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
286 287
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
288
	 */
289 290
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
291
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
292 293
		return 0;

294 295 296 297 298 299 300 301
	/*
	 * 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
302
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
303

Junio C Hamano's avatar
Junio C Hamano committed
304 305 306 307 308 309 310 311 312 313 314 315 316 317
	/*
	 * 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
318
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
319
	 */
320
	if (!changed && is_racy_timestamp(istate, ce)) {
321 322 323 324 325
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
326

Junio C Hamano's avatar
Junio C Hamano committed
327
	return changed;
328 329
}

330
int ie_modified(const struct index_state *istate,
331 332
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
333
{
Junio C Hamano's avatar
Junio C Hamano committed
334
	int changed, changed_fs;
335 336

	changed = ie_match_stat(istate, ce, st, options);
337 338 339 340 341 342 343 344 345
	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;

346 347 348 349 350 351 352 353 354 355 356 357 358
	/*
	 * 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.
359
	 */
360
	if ((changed & DATA_CHANGED) &&
361
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
362 363
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
364 365 366
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
367 368 369
	return 0;
}

370 371 372 373 374 375 376 377 378 379 380 381
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];
382
	if (!c1 && S_ISDIR(mode1))
383
		c1 = '/';
384
	if (!c2 && S_ISDIR(mode2))
385 386 387 388
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423
/*
 * 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;
}

424
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
425
{
426 427
	int len = len1 < len2 ? len1 : len2;
	int cmp;
428 429 430 431 432 433 434 435

	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
436

437
	if (stage1 < stage2)
438
		return -1;
439
	if (stage1 > stage2)
440
		return 1;
441 442 443
	return 0;
}

444 445 446 447 448
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);
}

449
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
450 451 452 453
{
	int first, last;

	first = 0;
454
	last = istate->cache_nr;
455 456
	while (last > first) {
		int next = (last + first) >> 1;
457
		struct cache_entry *ce = istate->cache[next];
458
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
459
		if (!cmp)
460
			return next;
461 462 463 464 465 466
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
467
	return -first-1;
468 469
}

470 471 472 473 474
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

475
/* Remove entry, return true if there are more entries to go.. */
476
int remove_index_entry_at(struct index_state *istate, int pos)
477
{
478 479
	struct cache_entry *ce = istate->cache[pos];

480
	record_resolve_undo(istate, ce);
481
	remove_name_hash(istate, ce);
482 483 484
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
485
		return 0;
486 487 488
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
489 490 491
	return 1;
}

492
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
493
 * Remove all cache entries marked for removal, that is where
494 495 496 497 498 499 500 501 502 503
 * 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++) {
		if (ce_array[i]->ce_flags & CE_REMOVE)
504
			remove_name_hash(istate, ce_array[i]);
505 506 507 508 509 510 511
		else
			ce_array[j++] = ce_array[i];
	}
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

512
int remove_file_from_index(struct index_state *istate, const char *path)
513
{
514
	int pos = index_name_pos(istate, path, strlen(path));
515 516
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
517
	cache_tree_invalidate_path(istate->cache_tree, path);
518 519
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
520 521 522
	return 0;
}

523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550
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;
}

551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582
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;
}

583 584 585 586 587 588 589 590
static void record_intent_to_add(struct cache_entry *ce)
{
	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);
}

591
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
592
{
593
	int size, namelen, was_same;
594
	mode_t st_mode = st->st_mode;
595
	struct cache_entry *ce, *alias;
596
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
597 598
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
599 600 601
	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
602

603
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
604
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
605 606

	namelen = strlen(path);
607
	if (S_ISDIR(st_mode)) {
608 609 610
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
611 612 613
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
614
	ce->ce_namelen = namelen;
615 616
	if (!intent_only)
		fill_stat_cache_info(ce, st);
617 618
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
619

620
	if (trust_executable_bit && has_symlinks)
621
		ce->ce_mode = create_ce_mode(st_mode);
622
	else {
623 624
		/* 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
625
		 */
626
		struct cache_entry *ent;
627
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
628

629
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
630
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
631 632
	}

633 634 635 636 637 638 639 640 641 642 643 644 645 646
	/* 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;
647
				foundce = index_dir_exists(istate, ce->name, ptr - ce->name - 1);
648 649 650 651 652 653 654 655
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

656
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
657
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
658 659
		/* Nothing changed, really */
		free(ce);
660 661
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
662
		alias->ce_flags |= CE_ADDED;
663 664
		return 0;
	}
665
	if (!intent_only) {
666
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
667 668 669 670
			return error("unable to index file %s", path);
	} else
		record_intent_to_add(ce);

671 672 673
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
674

Junio C Hamano's avatar
Junio C Hamano committed
675
	/* It was suspected to be racily clean, but it turns out to be Ok */
676 677 678 679 680 681 682
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
683
	else if (add_index_entry(istate, ce, add_option))
684
		return error("unable to add %s to index",path);
685
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
686 687 688 689
		printf("add '%s'\n", path);
	return 0;
}

690
int add_file_to_index(struct index_state *istate, const char *path, int flags)
691 692 693
{
	struct stat st;
	if (lstat(path, &st))
694
		die_errno("unable to stat '%s'", path);
695
	return add_to_index(istate, path, &st, flags);
696 697
}

698 699 700 701 702 703 704
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
		int refresh)
{
	int size, len;
	struct cache_entry *ce;

705 706
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
707
		return NULL;
708
	}
709 710 711 712 713 714 715

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

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
716 717
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
718 719 720 721 722 723 724 725
	ce->ce_mode = create_ce_mode(mode);

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

	return ce;
}

726
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
727 728 729 730 731
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

732
int ce_path_match(const struct cache_entry *ce, const struct pathspec *pathspec)
733
{
734
	return match_pathspec_depth(pathspec, ce->name, ce_namelen(ce), 0, NULL);
735 736
}

737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
/*
 * 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.
	 */
752

753
	/* "." is not allowed */
754
	if (*rest == '\0' || is_dir_sep(*rest))
755 756
		return 0;

757
	switch (*rest) {
758 759 760 761 762
	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
763 764
	case 'G':
		if (rest[1] != 'i' && rest[1] != 'I')
765
			break;
766
		if (rest[2] != 't' && rest[2] != 'T')
767 768 769 770
			break;
		rest += 2;
	/* fallthrough */
	case '.':
771
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
772 773 774 775 776 777 778 779 780
			return 0;
	}
	return 1;
}

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

781 782 783
	if (has_dos_drive_prefix(path))
		return 0;

784 785 786 787
	goto inside;
	for (;;) {
		if (!c)
			return 1;
788
		if (is_dir_sep(c)) {
789
inside:
790 791
			if (protect_hfs && is_hfs_dotgit(path))
				return 0;
792
			c = *path++;
793 794 795
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
796 797 798 799 800
		}
		c = *path++;
	}
}

801 802 803
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
804
 */
805 806
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
807
{
808 809
	int retval = 0;
	int len = ce_namelen(ce);
810
	int stage = ce_stage(ce);
811
	const char *name = ce->name;
812

813 814
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
815

816
		if (len >= ce_namelen(p))
817
			break;
818 819
		if (memcmp(name, p->name, len))
			break;
820 821
		if (ce_stage(p) != stage)
			continue;
822 823
		if (p->name[len] != '/')
			continue;
824
		if (p->ce_flags & CE_REMOVE)
825
			continue;
826 827 828
		retval = -1;
		if (!ok_to_replace)
			break;
829
		remove_index_entry_at(istate, --pos);
830
	}
831 832
	return retval;
}
833

834 835 836 837
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
838 839
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
840 841
{
	int retval = 0;
842
	int stage = ce_stage(ce);
843 844
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
845

846 847
	for (;;) {
		int len;
848

849 850 851 852 853 854 855
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
856

857
		pos = index_name_stage_pos(istate, name, len, stage);
858
		if (pos >= 0) {
859 860 861 862 863 864 865 866
			/*
			 * 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.
			 */
867
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
868 869 870
				retval = -1;
				if (!ok_to_replace)
					break;
871
				remove_index_entry_at(istate, pos);
872 873
				continue;
			}
874
		}
875 876
		else
			pos = -pos-1;
877 878 879 880

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
881
		 * we're ok, and we can exit.
882
		 */
883 884
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
885 886 887 888
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
889 890 891
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
892 893 894 895 896 897
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
898
		}
899
	}
900 901 902 903 904 905 906
	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
907
 *
908 909 910 911
 * 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.
 */
912 913 914
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
915
{
916 917 918 919 920
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
921
	if (ce->ce_flags & CE_REMOVE)
922 923
		return 0;

924 925 926
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
927
	 * in the array.
928
	 */
929
	retval = has_file_name(istate, ce, pos, ok_to_replace);
930

931 932 933 934
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
935
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
936 937
}

938
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
939 940
{
	int pos;
941 942
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
943
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
944
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
945

Junio C Hamano's avatar
Junio C Hamano committed
946
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
947
	pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
948

Junio C Hamano's avatar
Junio C Hamano committed
949
	/* existing match? Just replace it. */
950
	if (pos >= 0) {
951 952
		if (!new_only)
			replace_index_entry(istate, pos, ce);
953 954
		return 0;
	}
955
	pos = -pos-1;
956

957 958 959 960
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
961 962
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
963
			ok_to_add = 1;
964
			if (!remove_index_entry_at(istate, pos))
965 966 967 968
				break;
		}
	}

969 970
	if (!ok_to_add)
		return -1;
971
	if (!verify_path(ce->name))
972
		return error("Invalid path '%s'", ce->name);
973

Junio C Hamano's avatar
Junio C Hamano committed
974
	if (!skip_df_check &&
975
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
976
		if (!ok_to_replace)
977 978
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
979
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
980 981
		pos = -pos-1;
	}
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997
	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;
	}
998

999
	/* Make sure the array is big enough .. */
1000 1001 1002
	if (istate->cache_nr == istate->cache_alloc) {
		istate->cache_alloc = alloc_nr(istate->cache_alloc);
		istate->cache = xrealloc(istate->cache,
1003
					istate->cache_alloc * sizeof(*istate->cache));
1004 1005 1006
	}

	/* Add it in.. */
1007
	istate->cache_nr++;
1008
	if (istate->cache_nr > pos + 1)
1009 1010 1011
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1012
	set_index_entry(istate, pos, ce);
1013
	istate->cache_changed = 1;
1014 1015 1016
	return 0;
}

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

1039 1040 1041
	if (ce_uptodate(ce))
		return ce;

1042
	/*
1043 1044 1045
	 * 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.
1046
	 */
1047 1048 1049 1050
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1051 1052 1053 1054 1055
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1056
	if (lstat(ce->name, &st) < 0) {
1057 1058
		if (err)
			*err = errno;
1059 1060
		return NULL;
	}
1061

1062
	changed = ie_match_stat(istate, ce, &st, options);
1063 1064
	if (changed_ret)
		*changed_ret = changed;
1065
	if (!changed) {
1066 1067 1068 1069 1070 1071 1072 1073
		/*
		 * 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 &&
1074
		    !(ce->ce_flags & CE_VALID))
1075
			; /* mark this one VALID again */
1076 1077 1078 1079 1080 1081
		else {
			/*
			 * We do not mark the index itself "modified"
			 * because CE_UPTODATE flag is in-core only;
			 * we are not going to write this change out.
			 */
1082 1083
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1084
			return ce;
1085
		}
1086 1087
	}

1088
	if (ie_modified(istate, ce, &st, options)) {
1089 1090
		if (err)
			*err = EINVAL;
1091 1092
		return NULL;
	}
1093 1094 1095 1096 1097

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1098 1099 1100 1101 1102
	/*
	 * If ignore_valid is not set, we should leave CE_VALID bit
	 * alone.  Otherwise, paths marked with --no-assume-unchanged
	 * (i.e. things to be edited) will reacquire CE_VALID bit
	 * automatically, which is not really what we want.
1103
	 */
1104
	if (!ignore_valid && assume_unchanged &&
1105 1106
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1107 1108 1109 1110

	return updated;
}

1111
static void show_file(const char * fmt, const char * name, int in_porcelain,
1112
		      int * first, const char *header_msg)
1113 1114 1115
{
	if (in_porcelain && *first && header_msg) {
		printf("%s\n", header_msg);
1116
		*first = 0;
1117 1118 1119 1120
	}
	printf(fmt, name);
}

1121 1122
int refresh_index(struct index_state *istate, unsigned int flags,
		  const struct pathspec *pathspec,
1123
		  char *seen, const char *header_msg)
1124 1125 1126 1127 1128 1129 1130
{
	int i;
	int has_errors = 0;
	int really = (flags & REFRESH_REALLY) != 0;
	int allow_unmerged = (flags & REFRESH_UNMERGED) != 0;
	int quiet = (flags & REFRESH_QUIET) != 0;
	int not_new = (flags & REFRESH_IGNORE_MISSING) != 0;
1131
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1132 1133
	int first