read-cache.c 52.8 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

18 19
static struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really);

20 21 22 23
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

24 25 26 27 28 29 30 31 32 33 34
/* 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" */
35
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
36

37
struct index_state the_index;
38

Junio C Hamano's avatar
Junio C Hamano committed
39 40 41
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
42
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
43 44
}

45 46 47 48
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

49
	remove_name_hash(istate, old);
50
	set_index_entry(istate, nr, ce);
51 52 53
	istate->cache_changed = 1;
}

54 55 56 57 58 59 60
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);
61 62
	new->ce_flags &= ~CE_STATE_MASK;
	new->ce_namelen = namelen;
63 64 65 66 67 68 69
	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);
}

70 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
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;
}

125 126 127 128 129 130 131
/*
 * 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)
{
132
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
133 134

	if (assume_unchanged)
135
		ce->ce_flags |= CE_VALID;
136 137 138

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
139 140
}

141
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
142 143 144 145 146 147
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
148
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
149
			match = hashcmp(sha1, ce->sha1);
150
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
151 152 153 154
	}
	return match;
}

155
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
156 157 158 159
{
	int match = -1;
	void *buffer;
	unsigned long size;
160
	enum object_type type;
161
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
162

163
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
164
		return -1;
165

166
	buffer = read_sha1_file(ce->sha1, &type, &size);
167 168 169 170
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
171
	}
172
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
173 174 175
	return match;
}

176
static int ce_compare_gitlink(const struct cache_entry *ce)
177 178 179 180 181
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
182
	 * under GITLINK directory be a valid git directory. It
183 184 185 186 187 188 189 190 191 192
	 * 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);
}

193
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
194 195 196 197 198 199 200
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
201
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
202 203
			return DATA_CHANGED;
		break;
204
	case S_IFDIR:
205
		if (S_ISGITLINK(ce->ce_mode))
206
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
207 208 209 210 211 212
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

213
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
214 215 216
{
	unsigned int changed = 0;

217 218 219 220
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
221 222
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
223 224 225 226
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
227
		    (0100 & (ce->ce_mode ^ st->st_mode)))
228
			changed |= MODE_CHANGED;
229 230
		break;
	case S_IFLNK:
231 232 233
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
234
		break;
Martin Waitz's avatar
Martin Waitz committed
235
	case S_IFGITLINK:
236
		/* We ignore most of the st_xxx fields for gitlinks */
237 238 239 240
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
241
		return changed;
242
	default:
243
		die("internal error: ce_mode is %o", ce->ce_mode);
244
	}
245

246
	changed |= match_stat_data(&ce->ce_stat_data, st);
247

248
	/* Racily smudged entry? */
249
	if (!ce->ce_stat_data.sd_size) {
250 251 252 253
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
254 255 256
	return changed;
}

257 258
static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
259
{
260
	return (!S_ISGITLINK(ce->ce_mode) &&
261 262 263
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
264 265 266
		(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))
267
#else
268
		istate->timestamp.sec <= ce->ce_stat_data.sd_mtime.sec
269 270
#endif
		 );
271 272
}

273
int ie_match_stat(const struct index_state *istate,
274
		  const struct cache_entry *ce, struct stat *st,
275
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
276
{
Junio C Hamano's avatar
Junio C Hamano committed
277
	unsigned int changed;
278
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
279
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
280
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
281 282 283 284

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
285 286
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
287
	 */
288 289
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
290
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
291 292
		return 0;

293 294 295 296 297 298 299 300
	/*
	 * 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
301
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
302

Junio C Hamano's avatar
Junio C Hamano committed
303 304 305 306 307 308 309 310 311 312 313 314 315 316
	/*
	 * 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
317
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
318
	 */
319
	if (!changed && is_racy_timestamp(istate, ce)) {
320 321 322 323 324
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
325

Junio C Hamano's avatar
Junio C Hamano committed
326
	return changed;
327 328
}

329
int ie_modified(const struct index_state *istate,
330 331
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
332
{
Junio C Hamano's avatar
Junio C Hamano committed
333
	int changed, changed_fs;
334 335

	changed = ie_match_stat(istate, ce, st, options);
336 337 338 339 340 341 342 343 344
	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;

345 346 347 348 349 350 351 352 353 354 355 356 357
	/*
	 * 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.
358
	 */
359
	if ((changed & DATA_CHANGED) &&
360
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
361 362
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
363 364 365
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
366 367 368
	return 0;
}

369 370 371 372 373 374 375 376 377 378 379 380
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];
381
	if (!c1 && S_ISDIR(mode1))
382
		c1 = '/';
383
	if (!c2 && S_ISDIR(mode2))
384 385 386 387
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
/*
 * 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;
}

423
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
424
{
425 426
	int len = len1 < len2 ? len1 : len2;
	int cmp;
427 428 429 430 431 432 433 434

	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
435

436
	if (stage1 < stage2)
437
		return -1;
438
	if (stage1 > stage2)
439
		return 1;
440 441 442
	return 0;
}

443 444 445 446 447
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);
}

448
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
449 450 451 452
{
	int first, last;

	first = 0;
453
	last = istate->cache_nr;
454 455
	while (last > first) {
		int next = (last + first) >> 1;
456
		struct cache_entry *ce = istate->cache[next];
457
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
458
		if (!cmp)
459
			return next;
460 461 462 463 464 465
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
466
	return -first-1;
467 468
}

469 470 471 472 473
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

474
/* Remove entry, return true if there are more entries to go.. */
475
int remove_index_entry_at(struct index_state *istate, int pos)
476
{
477 478
	struct cache_entry *ce = istate->cache[pos];

479
	record_resolve_undo(istate, ce);
480
	remove_name_hash(istate, ce);
481 482 483
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
484
		return 0;
485 486 487
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
488 489 490
	return 1;
}

491 492 493 494 495 496 497 498 499 500 501 502
/*
 * Remove all cache ententries marked for removal, that is where
 * 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)
503
			remove_name_hash(istate, ce_array[i]);
504 505 506 507 508 509 510
		else
			ce_array[j++] = ce_array[i];
	}
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

511
int remove_file_from_index(struct index_state *istate, const char *path)
512
{
513
	int pos = index_name_pos(istate, path, strlen(path));
514 515
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
516
	cache_tree_invalidate_path(istate->cache_tree, path);
517 518
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
519 520 521
	return 0;
}

522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549
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;
}

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581
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;
}

582 583 584 585 586 587 588 589
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);
}

590
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
591
{
592
	int size, namelen, was_same;
593
	mode_t st_mode = st->st_mode;
594
	struct cache_entry *ce, *alias;
595
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
596 597
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
598 599 600
	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
601

602
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
603
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
604 605

	namelen = strlen(path);
606
	if (S_ISDIR(st_mode)) {
607 608 609
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
610 611 612
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
613
	ce->ce_namelen = namelen;
614 615
	if (!intent_only)
		fill_stat_cache_info(ce, st);
616 617
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
618

619
	if (trust_executable_bit && has_symlinks)
620
		ce->ce_mode = create_ce_mode(st_mode);
621
	else {
622 623
		/* 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
624
		 */
625
		struct cache_entry *ent;
626
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
627

628
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
629
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
630 631
	}

632 633 634 635 636 637 638 639 640 641 642 643 644 645
	/* 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;
646
				foundce = index_name_exists(istate, ce->name, ptr - ce->name, ignore_case);
647 648 649 650 651 652 653 654
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

655
	alias = index_name_exists(istate, ce->name, ce_namelen(ce), ignore_case);
656
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
657 658
		/* Nothing changed, really */
		free(ce);
659 660
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
661
		alias->ce_flags |= CE_ADDED;
662 663
		return 0;
	}
664
	if (!intent_only) {
665
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
666 667 668 669
			return error("unable to index file %s", path);
	} else
		record_intent_to_add(ce);

670 671 672
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
673

Junio C Hamano's avatar
Junio C Hamano committed
674
	/* It was suspected to be racily clean, but it turns out to be Ok */
675 676 677 678 679 680 681
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
682
	else if (add_index_entry(istate, ce, add_option))
683
		return error("unable to add %s to index",path);
684
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
685 686 687 688
		printf("add '%s'\n", path);
	return 0;
}

689
int add_file_to_index(struct index_state *istate, const char *path, int flags)
690 691 692
{
	struct stat st;
	if (lstat(path, &st))
693
		die_errno("unable to stat '%s'", path);
694
	return add_to_index(istate, path, &st, flags);
695 696
}

697 698 699 700 701 702 703
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;

704 705
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
706
		return NULL;
707
	}
708 709 710 711 712 713 714

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

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
715 716
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
717 718 719 720 721 722 723 724
	ce->ce_mode = create_ce_mode(mode);

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

	return ce;
}

725
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
726 727 728 729 730
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

731
int ce_path_match(const struct cache_entry *ce, const struct pathspec *pathspec)
732
{
733
	return match_pathspec_depth(pathspec, ce->name, ce_namelen(ce), 0, NULL);
734 735
}

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750
/*
 * 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.
	 */
751

752
	/* "." is not allowed */
753
	if (*rest == '\0' || is_dir_sep(*rest))
754 755
		return 0;

756
	switch (*rest) {
757 758 759 760 761 762 763 764 765 766 767 768
	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
		if (rest[1] != 'i')
			break;
		if (rest[2] != 't')
			break;
		rest += 2;
	/* fallthrough */
	case '.':
769
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
770 771 772 773 774 775 776 777 778
			return 0;
	}
	return 1;
}

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

779 780 781
	if (has_dos_drive_prefix(path))
		return 0;

782 783 784 785
	goto inside;
	for (;;) {
		if (!c)
			return 1;
786
		if (is_dir_sep(c)) {
787 788
inside:
			c = *path++;
789 790 791
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
792 793 794 795 796
		}
		c = *path++;
	}
}

797 798 799
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
800
 */
801 802
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
803
{
804 805
	int retval = 0;
	int len = ce_namelen(ce);
806
	int stage = ce_stage(ce);
807
	const char *name = ce->name;
808

809 810
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
811

812
		if (len >= ce_namelen(p))
813
			break;
814 815
		if (memcmp(name, p->name, len))
			break;
816 817
		if (ce_stage(p) != stage)
			continue;
818 819
		if (p->name[len] != '/')
			continue;
820
		if (p->ce_flags & CE_REMOVE)
821
			continue;
822 823 824
		retval = -1;
		if (!ok_to_replace)
			break;
825
		remove_index_entry_at(istate, --pos);
826
	}
827 828
	return retval;
}
829

830 831 832 833
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
834 835
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
836 837
{
	int retval = 0;
838
	int stage = ce_stage(ce);
839 840
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
841

842 843
	for (;;) {
		int len;
844

845 846 847 848 849 850 851
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
852

853
		pos = index_name_stage_pos(istate, name, len, stage);
854
		if (pos >= 0) {
855 856 857 858 859 860 861 862
			/*
			 * 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.
			 */
863
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
864 865 866
				retval = -1;
				if (!ok_to_replace)
					break;
867
				remove_index_entry_at(istate, pos);
868 869
				continue;
			}
870
		}
871 872
		else
			pos = -pos-1;
873 874 875 876

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
877
		 * we're ok, and we can exit.
878
		 */
879 880
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
881 882 883 884
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
885 886 887
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
888 889 890 891 892 893
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
894
		}
895
	}
896 897 898 899 900 901 902
	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
903
 *
904 905 906 907
 * 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.
 */
908 909 910
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
911
{
912 913 914 915 916
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
917
	if (ce->ce_flags & CE_REMOVE)
918 919
		return 0;

920 921 922
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
923
	 * in the array.
924
	 */
925
	retval = has_file_name(istate, ce, pos, ok_to_replace);
926

927 928 929 930
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
931
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
932 933
}

934
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
935 936
{
	int pos;
937 938
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
939
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
940
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
941

Junio C Hamano's avatar
Junio C Hamano committed
942
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
943
	pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
944

Junio C Hamano's avatar
Junio C Hamano committed
945
	/* existing match? Just replace it. */
946
	if (pos >= 0) {
947 948
		if (!new_only)
			replace_index_entry(istate, pos, ce);
949 950
		return 0;
	}
951
	pos = -pos-1;
952

953 954 955 956
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
957 958
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
959
			ok_to_add = 1;
960
			if (!remove_index_entry_at(istate, pos))
961 962 963 964
				break;
		}
	}

965 966
	if (!ok_to_add)
		return -1;
967
	if (!verify_path(ce->name))
968
		return error("Invalid path '%s'", ce->name);
969

Junio C Hamano's avatar
Junio C Hamano committed
970
	if (!skip_df_check &&
971
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
972
		if (!ok_to_replace)
973 974
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
975
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
976 977
		pos = -pos-1;
	}
978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993
	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;
	}
994

995
	/* Make sure the array is big enough .. */
996 997 998
	if (istate->cache_nr == istate->cache_alloc) {
		istate->cache_alloc = alloc_nr(istate->cache_alloc);
		istate->cache = xrealloc(istate->cache,
999
					istate->cache_alloc * sizeof(*istate->cache));
1000 1001 1002
	}

	/* Add it in.. */
1003
	istate->cache_nr++;
1004
	if (istate->cache_nr > pos + 1)
1005 1006 1007
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1008
	set_index_entry(istate, pos, ce);
1009
	istate->cache_changed = 1;
1010 1011 1012
	return 0;
}

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

1035 1036 1037
	if (ce_uptodate(ce))
		return ce;

1038
	/*
1039 1040 1041
	 * 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.
1042
	 */
1043 1044 1045 1046
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1047 1048 1049 1050 1051
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1052
	if (lstat(ce->name, &st) < 0) {
1053 1054
		if (err)
			*err = errno;
1055 1056
		return NULL;
	}
1057

1058
	changed = ie_match_stat(istate, ce, &st, options);
1059 1060
	if (changed_ret)
		*changed_ret = changed;
1061
	if (!changed) {
1062 1063 1064 1065 1066 1067 1068 1069
		/*
		 * 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 &&
1070
		    !(ce->ce_flags & CE_VALID))
1071
			; /* mark this one VALID again */
1072 1073 1074 1075 1076 1077
		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.
			 */
1078 1079
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1080
			return ce;
1081
		}
1082 1083
	}

1084
	if (ie_modified(istate, ce, &st, options)) {
1085 1086
		if (err)
			*err = EINVAL;
1087 1088
		return NULL;
	}
1089 1090 1091 1092 1093

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1094 1095 1096 1097 1098
	/*
	 * 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.
1099
	 */
1100
	if (!ignore_valid && assume_unchanged &&
1101 1102
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1103 1104 1105 1106

	return updated;
}

1107
static void show_file(const char * fmt, const char * name, int in_porcelain,
1108
		      int * first, const char *header_msg)
1109 1110 1111
{
	if (in_porcelain && *first && header_msg) {
		printf("%s\n", header_msg);
1112
		*first = 0;
1113 1114 1115 1116 1117
	}
	printf(fmt, name);
}

int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec,
1118
		  char *seen, const char *header_msg)
1119 1120 1121 1122 1123 1124 1125
{
	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;
1126
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1127 1128
	int first = 1;
	int in_porcelain = (flags & REFRESH_IN_PORCELAIN);
1129
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
1130
	const char *modified_fmt;
1131 1132 1133
	const char *deleted_fmt;
	const char *typechange_fmt;
	const char *added_fmt;
1134
	const char *unmerged_fmt;
1135

1136
	modified_fmt = (in_porcelain ? "M\t%s\n" : "%s: needs update\n");
1137 1138 1139
	deleted_fmt = (in_porcelain ? "D\t%s\n" : "%s: needs update\n");
	typechange_fmt = (in_porcelain ? "T\t%s\n" : "%s needs update\n");
	added_fmt = (in_porcelain ? "A\t%s\n" : "%s needs update\n");