read-cache.c 53.1 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,
					       unsigned int options);
21

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

#define CE_NAMEMASK  (0x0fff)

26 27 28 29 30 31 32 33 34 35 36
/* Index extensions.
 *
 * The first letter should be 'A'..'Z' for extensions that are not
 * necessary for a correct operation (i.e. optimization data).
 * When new extensions are added that _needs_ to be understood in
 * order to correctly interpret the index file, pick character that
 * is outside the range, to cause the reader to abort.
 */

#define CACHE_EXT(s) ( (s[0]<<24)|(s[1]<<16)|(s[2]<<8)|(s[3]) )
#define CACHE_EXT_TREE 0x54524545	/* "TREE" */
37
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
38

39
struct index_state the_index;
40

Junio C Hamano's avatar
Junio C Hamano committed
41 42 43
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
44
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
45 46
}

47 48 49 50
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

51
	remove_name_hash(istate, old);
52
	set_index_entry(istate, nr, ce);
53 54 55
	istate->cache_changed = 1;
}

56 57 58 59 60 61 62
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);
63 64
	new->ce_flags &= ~CE_STATE_MASK;
	new->ce_namelen = namelen;
65 66 67 68 69 70 71
	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);
}

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
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;
}

127 128 129 130 131 132 133
/*
 * 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)
{
134
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
135 136

	if (assume_unchanged)
137
		ce->ce_flags |= CE_VALID;
138 139 140

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
141 142
}

143
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
144 145 146 147 148 149
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
150
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
151
			match = hashcmp(sha1, ce->sha1);
152
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
153 154 155 156
	}
	return match;
}

157
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
158 159 160 161
{
	int match = -1;
	void *buffer;
	unsigned long size;
162
	enum object_type type;
163
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
164

165
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
166
		return -1;
167

168
	buffer = read_sha1_file(ce->sha1, &type, &size);
169 170 171 172
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
173
	}
174
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
175 176 177
	return match;
}

178
static int ce_compare_gitlink(const struct cache_entry *ce)
179 180 181 182 183
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
184
	 * under GITLINK directory be a valid git directory. It
185 186 187 188 189 190 191 192 193 194
	 * 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);
}

195
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
196 197 198 199 200 201 202
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
203
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
204 205
			return DATA_CHANGED;
		break;
206
	case S_IFDIR:
207
		if (S_ISGITLINK(ce->ce_mode))
208
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
209 210 211 212 213 214
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

215
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
216 217 218
{
	unsigned int changed = 0;

219 220 221 222
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
223 224
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
225 226 227 228
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
229
		    (0100 & (ce->ce_mode ^ st->st_mode)))
230
			changed |= MODE_CHANGED;
231 232
		break;
	case S_IFLNK:
233 234 235
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
236
		break;
Martin Waitz's avatar
Martin Waitz committed
237
	case S_IFGITLINK:
238
		/* We ignore most of the st_xxx fields for gitlinks */
239 240 241 242
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
243
		return changed;
244
	default:
245
		die("internal error: ce_mode is %o", ce->ce_mode);
246
	}
247

248
	changed |= match_stat_data(&ce->ce_stat_data, st);
249

250
	/* Racily smudged entry? */
251
	if (!ce->ce_stat_data.sd_size) {
252 253 254 255
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
256 257 258
	return changed;
}

259 260
static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
261
{
262
	return (!S_ISGITLINK(ce->ce_mode) &&
263 264 265
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
266 267 268
		(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))
269
#else
270
		istate->timestamp.sec <= ce->ce_stat_data.sd_mtime.sec
271 272
#endif
		 );
273 274
}

275
int ie_match_stat(const struct index_state *istate,
276
		  const struct cache_entry *ce, struct stat *st,
277
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
278
{
Junio C Hamano's avatar
Junio C Hamano committed
279
	unsigned int changed;
280
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
281
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
282
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
283 284 285 286

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
287 288
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
289
	 */
290 291
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
292
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
293 294
		return 0;

295 296 297 298 299 300 301 302
	/*
	 * 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
303
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
304

Junio C Hamano's avatar
Junio C Hamano committed
305 306 307 308 309 310 311 312 313 314 315 316 317 318
	/*
	 * 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
319
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
320
	 */
321
	if (!changed && is_racy_timestamp(istate, ce)) {
322 323 324 325 326
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
327

Junio C Hamano's avatar
Junio C Hamano committed
328
	return changed;
329 330
}

331
int ie_modified(const struct index_state *istate,
332 333
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
334
{
Junio C Hamano's avatar
Junio C Hamano committed
335
	int changed, changed_fs;
336 337

	changed = ie_match_stat(istate, ce, st, options);
338 339 340 341 342 343 344 345 346
	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;

347 348 349 350 351 352 353 354 355 356 357 358 359
	/*
	 * 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.
360
	 */
361
	if ((changed & DATA_CHANGED) &&
362
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
363 364
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
365 366 367
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
368 369 370
	return 0;
}

371 372 373 374 375 376 377 378 379 380 381 382
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];
383
	if (!c1 && S_ISDIR(mode1))
384
		c1 = '/';
385
	if (!c2 && S_ISDIR(mode2))
386 387 388 389
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
/*
 * 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;
}

425
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
426
{
427 428
	int len = len1 < len2 ? len1 : len2;
	int cmp;
429 430 431 432 433 434 435 436

	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
437

438
	if (stage1 < stage2)
439
		return -1;
440
	if (stage1 > stage2)
441
		return 1;
442 443 444
	return 0;
}

445 446 447 448 449
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);
}

450
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
451 452 453 454
{
	int first, last;

	first = 0;
455
	last = istate->cache_nr;
456 457
	while (last > first) {
		int next = (last + first) >> 1;
458
		struct cache_entry *ce = istate->cache[next];
459
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
460
		if (!cmp)
461
			return next;
462 463 464 465 466 467
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
468
	return -first-1;
469 470
}

471 472 473 474 475
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

476
/* Remove entry, return true if there are more entries to go.. */
477
int remove_index_entry_at(struct index_state *istate, int pos)
478
{
479 480
	struct cache_entry *ce = istate->cache[pos];

481
	record_resolve_undo(istate, ce);
482
	remove_name_hash(istate, ce);
483 484 485
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
486
		return 0;
487 488 489
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
490 491 492
	return 1;
}

493
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
494
 * Remove all cache entries marked for removal, that is where
495 496 497 498 499 500 501 502 503 504
 * 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)
505
			remove_name_hash(istate, ce_array[i]);
506 507 508 509 510 511 512
		else
			ce_array[j++] = ce_array[i];
	}
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

513
int remove_file_from_index(struct index_state *istate, const char *path)
514
{
515
	int pos = index_name_pos(istate, path, strlen(path));
516 517
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
518
	cache_tree_invalidate_path(istate->cache_tree, path);
519 520
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
521 522 523
	return 0;
}

524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551
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;
}

552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
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;
}

584 585 586 587 588 589 590 591
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);
}

592
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
593
{
594
	int size, namelen, was_same;
595
	mode_t st_mode = st->st_mode;
596
	struct cache_entry *ce, *alias;
597
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
598 599
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
600 601 602
	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
603

604
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
605
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
606 607

	namelen = strlen(path);
608
	if (S_ISDIR(st_mode)) {
609 610 611
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
612 613 614
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
615
	ce->ce_namelen = namelen;
616 617
	if (!intent_only)
		fill_stat_cache_info(ce, st);
618 619
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
620

621
	if (trust_executable_bit && has_symlinks)
622
		ce->ce_mode = create_ce_mode(st_mode);
623
	else {
624 625
		/* 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
626
		 */
627
		struct cache_entry *ent;
628
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
629

630
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
631
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
632 633
	}

634 635 636 637 638 639 640 641 642 643 644 645 646 647
	/* 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;
648
				foundce = index_dir_exists(istate, ce->name, ptr - ce->name - 1);
649 650 651 652 653 654 655 656
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

657
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
658
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
659 660
		/* Nothing changed, really */
		free(ce);
661 662
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
663
		alias->ce_flags |= CE_ADDED;
664 665
		return 0;
	}
666
	if (!intent_only) {
667
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
668 669 670 671
			return error("unable to index file %s", path);
	} else
		record_intent_to_add(ce);

672 673 674
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
675

Junio C Hamano's avatar
Junio C Hamano committed
676
	/* It was suspected to be racily clean, but it turns out to be Ok */
677 678 679 680 681 682 683
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
684
	else if (add_index_entry(istate, ce, add_option))
685
		return error("unable to add %s to index",path);
686
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
687 688 689 690
		printf("add '%s'\n", path);
	return 0;
}

691
int add_file_to_index(struct index_state *istate, const char *path, int flags)
692 693 694
{
	struct stat st;
	if (lstat(path, &st))
695
		die_errno("unable to stat '%s'", path);
696
	return add_to_index(istate, path, &st, flags);
697 698
}

699 700
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
701
		unsigned int refresh_options)
702 703 704 705
{
	int size, len;
	struct cache_entry *ce;

706 707
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
708
		return NULL;
709
	}
710 711 712 713 714 715 716

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

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

721
	return refresh_cache_entry(ce, refresh_options);
722 723
}

724
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
725 726 727 728 729
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
/*
 * 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.
	 */
745

746
	/* "." is not allowed */
747
	if (*rest == '\0' || is_dir_sep(*rest))
748 749
		return 0;

750
	switch (*rest) {
751 752 753 754 755
	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
756 757
	case 'G':
		if (rest[1] != 'i' && rest[1] != 'I')
758
			break;
759
		if (rest[2] != 't' && rest[2] != 'T')
760 761 762 763
			break;
		rest += 2;
	/* fallthrough */
	case '.':
764
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
765 766 767 768 769 770 771 772 773
			return 0;
	}
	return 1;
}

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

774 775 776
	if (has_dos_drive_prefix(path))
		return 0;

777 778 779 780
	goto inside;
	for (;;) {
		if (!c)
			return 1;
781
		if (is_dir_sep(c)) {
782
inside:
783 784
			if (protect_hfs && is_hfs_dotgit(path))
				return 0;
785 786
			if (protect_ntfs && is_ntfs_dotgit(path))
				return 0;
787
			c = *path++;
788 789 790
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
791 792 793 794 795
		}
		c = *path++;
	}
}

796 797 798
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
799
 */
800 801
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
802
{
803 804
	int retval = 0;
	int len = ce_namelen(ce);
805
	int stage = ce_stage(ce);
806
	const char *name = ce->name;
807

808 809
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
810

811
		if (len >= ce_namelen(p))
812
			break;
813 814
		if (memcmp(name, p->name, len))
			break;
815 816
		if (ce_stage(p) != stage)
			continue;
817 818
		if (p->name[len] != '/')
			continue;
819
		if (p->ce_flags & CE_REMOVE)
820
			continue;
821 822 823
		retval = -1;
		if (!ok_to_replace)
			break;
824
		remove_index_entry_at(istate, --pos);
825
	}
826 827
	return retval;
}
828

829 830 831 832
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
833 834
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
835 836
{
	int retval = 0;
837
	int stage = ce_stage(ce);
838 839
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
840

841 842
	for (;;) {
		int len;
843

844 845 846 847 848 849 850
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
851

852
		pos = index_name_stage_pos(istate, name, len, stage);
853
		if (pos >= 0) {
854 855 856 857 858 859 860 861
			/*
			 * 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.
			 */
862
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
863 864 865
				retval = -1;
				if (!ok_to_replace)
					break;
866
				remove_index_entry_at(istate, pos);
867 868
				continue;
			}
869
		}
870 871
		else
			pos = -pos-1;
872 873 874 875

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
876
		 * we're ok, and we can exit.
877
		 */
878 879
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
880 881 882 883
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
884 885 886
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
887 888 889 890 891 892
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
893
		}
894
	}
895 896 897 898 899 900 901
	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
902
 *
903 904 905 906
 * 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.
 */
907 908 909
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
910
{
911 912 913 914 915
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
916
	if (ce->ce_flags & CE_REMOVE)
917 918
		return 0;

919 920 921
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
922
	 * in the array.
923
	 */
924
	retval = has_file_name(istate, ce, pos, ok_to_replace);
925

926 927 928 929
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
930
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
931 932
}

933
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
934 935
{
	int pos;
936 937
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
938
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
939
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
940

Junio C Hamano's avatar
Junio C Hamano committed
941
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
942
	pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
943

Junio C Hamano's avatar
Junio C Hamano committed
944
	/* existing match? Just replace it. */
945
	if (pos >= 0) {
946 947
		if (!new_only)
			replace_index_entry(istate, pos, ce);
948 949
		return 0;
	}
950
	pos = -pos-1;
951

952 953 954 955
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
956 957
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
958
			ok_to_add = 1;
959
			if (!remove_index_entry_at(istate, pos))
960 961 962 963
				break;
		}
	}

964 965
	if (!ok_to_add)
		return -1;
966
	if (!verify_path(ce->name))
967
		return error("Invalid path '%s'", ce->name);
968

Junio C Hamano's avatar
Junio C Hamano committed
969
	if (!skip_df_check &&
970
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
971
		if (!ok_to_replace)
972 973
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
974
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
975 976
		pos = -pos-1;
	}
977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
	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;
	}
993

994
	/* Make sure the array is big enough .. */
995 996 997
	if (istate->cache_nr == istate->cache_alloc) {
		istate->cache_alloc = alloc_nr(istate->cache_alloc);
		istate->cache = xrealloc(istate->cache,
998
					istate->cache_alloc * sizeof(*istate->cache));
999 1000 1001
	}

	/* Add it in.. */
1002
	istate->cache_nr++;
1003
	if (istate->cache_nr > pos + 1)
1004 1005 1006
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1007
	set_index_entry(istate, pos, ce);
1008
	istate->cache_changed = 1;
1009 1010 1011
	return 0;
}

1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
/*
 * "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.
 */
1023
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1024
					     struct cache_entry *ce,
1025 1026
					     unsigned int options, int *err,
					     int *changed_ret)
1027 1028 1029 1030
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1031
	int refresh = options & CE_MATCH_REFRESH;
1032
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1033
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1034
	int ignore_missing = options & CE_MATCH_IGNORE_MISSING;
1035

1036
	if (!refresh || ce_uptodate(ce))
1037 1038
		return ce;

1039
	/*
1040 1041 1042
	 * 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.
1043
	 */
1044 1045 1046 1047
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1048 1049 1050 1051 1052
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1053
	if (lstat(ce->name, &st) < 0) {
1054 1055
		if (ignore_missing && errno == ENOENT)
			return ce;
1056 1057
		if (err)
			*err = errno;
1058 1059
		return NULL;
	}
1060

1061
	changed = ie_match_stat(istate, ce, &st, options);
1062 1063
	if (changed_ret)
		*changed_ret = changed;
1064
	if (!changed) {
1065 1066 1067 1068 1069 1070 1071 1072
		/*
		 * 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 &&
1073
		    !(ce->ce_flags & CE_VALID))
1074
			; /* mark this one VALID again */
1075 1076 1077 1078 1079 1080
		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.
			 */
1081 1082
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1083
			return ce;
1084
		}
1085 1086
	}

1087
	if (ie_modified(istate, ce, &st, options)) {