read-cache.c 45.6 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 17
static struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really);

18 19 20 21 22 23 24 25 26 27 28
/* 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" */
29
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
30

31
struct index_state the_index;
32

Junio C Hamano's avatar
Junio C Hamano committed
33 34 35
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
36
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
37 38
}

39 40 41 42
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

43
	remove_name_hash(old);
44
	set_index_entry(istate, nr, ce);
45 46 47
	istate->cache_changed = 1;
}

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
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);
	new->ce_flags &= ~(CE_STATE_MASK | CE_NAMEMASK);
	new->ce_flags |= (namelen >= CE_NAMEMASK ? CE_NAMEMASK : namelen);
	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);
}

64 65 66 67 68 69 70
/*
 * 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)
{
71 72
	ce->ce_ctime.sec = (unsigned int)st->st_ctime;
	ce->ce_mtime.sec = (unsigned int)st->st_mtime;
73 74
	ce->ce_ctime.nsec = ST_CTIME_NSEC(*st);
	ce->ce_mtime.nsec = ST_MTIME_NSEC(*st);
75 76 77 78 79
	ce->ce_dev = st->st_dev;
	ce->ce_ino = st->st_ino;
	ce->ce_uid = st->st_uid;
	ce->ce_gid = st->st_gid;
	ce->ce_size = st->st_size;
Junio C Hamano's avatar
Junio C Hamano committed
80 81

	if (assume_unchanged)
82
		ce->ce_flags |= CE_VALID;
83 84 85

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
86 87
}

Junio C Hamano's avatar
Junio C Hamano committed
88 89 90 91 92 93 94
static int ce_compare_data(struct cache_entry *ce, struct stat *st)
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
95
		if (!index_fd(sha1, fd, st, 0, OBJ_BLOB, ce->name))
96
			match = hashcmp(sha1, ce->sha1);
97
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
98 99 100 101
	}
	return match;
}

102
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
103 104 105 106
{
	int match = -1;
	void *buffer;
	unsigned long size;
107
	enum object_type type;
108
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
109

110
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
111
		return -1;
112

113
	buffer = read_sha1_file(ce->sha1, &type, &size);
114 115 116 117
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
118
	}
119
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
120 121 122
	return match;
}

123 124 125 126 127 128
static int ce_compare_gitlink(struct cache_entry *ce)
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
129
	 * under GITLINK directory be a valid git directory. It
130 131 132 133 134 135 136 137 138 139
	 * 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);
}

Junio C Hamano's avatar
Junio C Hamano committed
140 141 142 143 144 145 146 147
static int ce_modified_check_fs(struct cache_entry *ce, struct stat *st)
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
148
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
149 150
			return DATA_CHANGED;
		break;
151
	case S_IFDIR:
152
		if (S_ISGITLINK(ce->ce_mode))
153
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
154 155 156 157 158 159
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

160
static int is_empty_blob_sha1(const unsigned char *sha1)
161 162 163 164 165 166 167 168 169
{
	static const unsigned char empty_blob_sha1[20] = {
		0xe6,0x9d,0xe2,0x9b,0xb2,0xd1,0xd6,0x43,0x4b,0x8b,
		0x29,0xae,0x77,0x5a,0xd8,0xc2,0xe4,0x8c,0x53,0x91
	};

	return !hashcmp(sha1, empty_blob_sha1);
}

Junio C Hamano's avatar
Junio C Hamano committed
170
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
171 172 173
{
	unsigned int changed = 0;

174 175 176 177
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
178 179
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
180 181 182 183
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
184
		    (0100 & (ce->ce_mode ^ st->st_mode)))
185
			changed |= MODE_CHANGED;
186 187
		break;
	case S_IFLNK:
188 189 190
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
191
		break;
Martin Waitz's avatar
Martin Waitz committed
192
	case S_IFGITLINK:
193
		/* We ignore most of the st_xxx fields for gitlinks */
194 195 196 197
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
198
		return changed;
199
	default:
200
		die("internal error: ce_mode is %o", ce->ce_mode);
201
	}
202
	if (ce->ce_mtime.sec != (unsigned int)st->st_mtime)
203
		changed |= MTIME_CHANGED;
204
	if (trust_ctime && ce->ce_ctime.sec != (unsigned int)st->st_ctime)
205
		changed |= CTIME_CHANGED;
206

207
#ifdef USE_NSEC
208
	if (ce->ce_mtime.nsec != ST_MTIME_NSEC(*st))
209
		changed |= MTIME_CHANGED;
210
	if (trust_ctime && ce->ce_ctime.nsec != ST_CTIME_NSEC(*st))
211 212 213
		changed |= CTIME_CHANGED;
#endif

214 215
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
216
		changed |= OWNER_CHANGED;
217
	if (ce->ce_ino != (unsigned int) st->st_ino)
218
		changed |= INODE_CHANGED;
219 220 221 222 223 224 225

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
226
	if (ce->ce_dev != (unsigned int) st->st_dev)
227 228 229
		changed |= INODE_CHANGED;
#endif

230
	if (ce->ce_size != (unsigned int) st->st_size)
231
		changed |= DATA_CHANGED;
232

233 234 235 236 237 238
	/* Racily smudged entry? */
	if (!ce->ce_size) {
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
239 240 241
	return changed;
}

242
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
243
{
244
	return (!S_ISGITLINK(ce->ce_mode) &&
245 246 247 248 249 250 251 252 253 254
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
		(istate->timestamp.sec < ce->ce_mtime.sec ||
		 (istate->timestamp.sec == ce->ce_mtime.sec &&
		  istate->timestamp.nsec <= ce->ce_mtime.nsec))
#else
		istate->timestamp.sec <= ce->ce_mtime.sec
#endif
		 );
255 256
}

257
int ie_match_stat(const struct index_state *istate,
258 259
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
260
{
Junio C Hamano's avatar
Junio C Hamano committed
261
	unsigned int changed;
262
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
263
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
264
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
265 266 267 268

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
269 270
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
271
	 */
272 273
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
274
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
275 276
		return 0;

277 278 279 280 281 282 283 284
	/*
	 * 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
285
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
286

Junio C Hamano's avatar
Junio C Hamano committed
287 288 289 290 291 292 293 294 295 296 297 298 299 300
	/*
	 * 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
301
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
302
	 */
303
	if (!changed && is_racy_timestamp(istate, ce)) {
304 305 306 307 308
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
309

Junio C Hamano's avatar
Junio C Hamano committed
310
	return changed;
311 312
}

313
int ie_modified(const struct index_state *istate,
314
		struct cache_entry *ce, struct stat *st, unsigned int options)
315
{
Junio C Hamano's avatar
Junio C Hamano committed
316
	int changed, changed_fs;
317 318

	changed = ie_match_stat(istate, ce, st, options);
319 320 321 322 323 324 325 326 327
	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;

328 329 330 331 332 333 334 335 336 337 338 339 340
	/*
	 * 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.
341
	 */
342 343
	if ((changed & DATA_CHANGED) &&
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_size != 0))
344 345
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
346 347 348
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
349 350 351
	return 0;
}

352 353 354 355 356 357 358 359 360 361 362 363
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];
364
	if (!c1 && S_ISDIR(mode1))
365
		c1 = '/';
366
	if (!c2 && S_ISDIR(mode2))
367 368 369 370
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405
/*
 * 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;
}

406
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
407
{
408 409
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
410 411 412 413 414 415 416 417 418 419
	int len = len1 < len2 ? len1 : len2;
	int cmp;

	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
420

421 422 423
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
424

425 426 427 428
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
429 430 431
	return 0;
}

432
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
433 434 435 436
{
	int first, last;

	first = 0;
437
	last = istate->cache_nr;
438 439
	while (last > first) {
		int next = (last + first) >> 1;
440
		struct cache_entry *ce = istate->cache[next];
441
		int cmp = cache_name_compare(name, namelen, ce->name, ce->ce_flags);
442
		if (!cmp)
443
			return next;
444 445 446 447 448 449
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
450
	return -first-1;
451 452
}

453
/* Remove entry, return true if there are more entries to go.. */
454
int remove_index_entry_at(struct index_state *istate, int pos)
455
{
456 457
	struct cache_entry *ce = istate->cache[pos];

458
	record_resolve_undo(istate, ce);
459
	remove_name_hash(ce);
460 461 462
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
463
		return 0;
464 465 466
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
467 468 469
	return 1;
}

470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489
/*
 * 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)
			remove_name_hash(ce_array[i]);
		else
			ce_array[j++] = ce_array[i];
	}
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

490
int remove_file_from_index(struct index_state *istate, const char *path)
491
{
492
	int pos = index_name_pos(istate, path, strlen(path));
493 494
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
495
	cache_tree_invalidate_path(istate->cache_tree, path);
496 497
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
498 499 500
	return 0;
}

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528
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;
}

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
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;
}

561 562 563 564 565 566 567 568
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);
}

569
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
570
{
571
	int size, namelen, was_same;
572
	mode_t st_mode = st->st_mode;
573
	struct cache_entry *ce, *alias;
574
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
575 576
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
577 578 579
	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
580

581
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
582
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
583 584

	namelen = strlen(path);
585
	if (S_ISDIR(st_mode)) {
586 587 588
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
589 590 591
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
592
	ce->ce_flags = namelen;
593 594
	if (!intent_only)
		fill_stat_cache_info(ce, st);
595 596
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
597

598
	if (trust_executable_bit && has_symlinks)
599
		ce->ce_mode = create_ce_mode(st_mode);
600
	else {
601 602
		/* 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
603
		 */
604
		struct cache_entry *ent;
605
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
606

607
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
608
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
609 610
	}

611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633
	/* 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;
				foundce = index_name_exists(&the_index, ce->name, ptr - ce->name, ignore_case);
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

634
	alias = index_name_exists(istate, ce->name, ce_namelen(ce), ignore_case);
635
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
636 637
		/* Nothing changed, really */
		free(ce);
638 639
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
640
		alias->ce_flags |= CE_ADDED;
641 642
		return 0;
	}
643 644 645 646 647 648
	if (!intent_only) {
		if (index_path(ce->sha1, path, st, 1))
			return error("unable to index file %s", path);
	} else
		record_intent_to_add(ce);

649 650 651
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
652

Junio C Hamano's avatar
Junio C Hamano committed
653
	/* It was suspected to be racily clean, but it turns out to be Ok */
654 655 656 657 658 659 660
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
661
	else if (add_index_entry(istate, ce, add_option))
662
		return error("unable to add %s to index",path);
663
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
664 665 666 667
		printf("add '%s'\n", path);
	return 0;
}

668
int add_file_to_index(struct index_state *istate, const char *path, int flags)
669 670 671
{
	struct stat st;
	if (lstat(path, &st))
672
		die_errno("unable to stat '%s'", path);
673
	return add_to_index(istate, path, &st, flags);
674 675
}

676 677 678 679 680 681 682
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;

683 684
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
685
		return NULL;
686
	}
687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702

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

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

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

	return ce;
}

703
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
704 705 706 707 708
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
int ce_path_match(const struct cache_entry *ce, const char **pathspec)
{
	const char *match, *name;
	int len;

	if (!pathspec)
		return 1;

	len = ce_namelen(ce);
	name = ce->name;
	while ((match = *pathspec++) != NULL) {
		int matchlen = strlen(match);
		if (matchlen > len)
			continue;
		if (memcmp(name, match, matchlen))
			continue;
		if (matchlen && name[matchlen-1] == '/')
			return 1;
		if (name[matchlen] == '/' || !name[matchlen])
			return 1;
729 730
		if (!matchlen)
			return 1;
731 732 733 734
	}
	return 0;
}

735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776
/*
 * 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.
	 */
	switch (*rest) {
	/* "." is not allowed */
	case '\0': case '/':
		return 0;

	/*
	 * ".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 '.':
		if (rest[1] == '\0' || rest[1] == '/')
			return 0;
	}
	return 1;
}

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

777 778 779
	if (has_dos_drive_prefix(path))
		return 0;

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

795 796 797
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
798
 */
799 800
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
801
{
802 803
	int retval = 0;
	int len = ce_namelen(ce);
804
	int stage = ce_stage(ce);
805
	const char *name = ce->name;
806

807 808
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
809

810
		if (len >= ce_namelen(p))
811
			break;
812 813
		if (memcmp(name, p->name, len))
			break;
814 815
		if (ce_stage(p) != stage)
			continue;
816 817
		if (p->name[len] != '/')
			continue;
818
		if (p->ce_flags & CE_REMOVE)
819
			continue;
820 821 822
		retval = -1;
		if (!ok_to_replace)
			break;
823
		remove_index_entry_at(istate, --pos);
824
	}
825 826
	return retval;
}
827

828 829 830 831
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
832 833
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
834 835
{
	int retval = 0;
836
	int stage = ce_stage(ce);
837 838
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
839

840 841
	for (;;) {
		int len;
842

843 844 845 846 847 848 849
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
850

851
		pos = index_name_pos(istate, name, create_ce_flags(len, stage));
852
		if (pos >= 0) {
853 854 855 856 857 858 859 860
			/*
			 * 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.
			 */
861
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
862 863 864
				retval = -1;
				if (!ok_to_replace)
					break;
865
				remove_index_entry_at(istate, pos);
866 867
				continue;
			}
868
		}
869 870
		else
			pos = -pos-1;
871 872 873 874

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
875
		 * we're ok, and we can exit.
876
		 */
877 878
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
879 880 881 882
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
883 884 885
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
886 887 888 889 890 891
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
892
		}
893
	}
894 895 896 897 898 899 900
	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
901
 *
902 903 904 905
 * 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.
 */
906 907 908
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
909
{
910 911 912 913 914
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
915
	if (ce->ce_flags & CE_REMOVE)
916 917
		return 0;

918 919 920
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
921
	 * in the array.
922
	 */
923
	retval = has_file_name(istate, ce, pos, ok_to_replace);
924

925 926 927 928
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
929
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
930 931
}

932
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
933 934
{
	int pos;
935 936
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
937
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
938
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
939

Junio C Hamano's avatar
Junio C Hamano committed
940
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
941
	pos = index_name_pos(istate, ce->name, ce->ce_flags);
942

Junio C Hamano's avatar
Junio C Hamano committed
943
	/* existing match? Just replace it. */
944
	if (pos >= 0) {
945 946
		if (!new_only)
			replace_index_entry(istate, pos, ce);
947 948
		return 0;
	}
949
	pos = -pos-1;
950

951 952 953 954
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
955 956
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
957
			ok_to_add = 1;
958
			if (!remove_index_entry_at(istate, pos))
959 960 961 962
				break;
		}
	}

963 964
	if (!ok_to_add)
		return -1;
965
	if (!verify_path(ce->name))
966
		return error("Invalid path '%s'", ce->name);
967

Junio C Hamano's avatar
Junio C Hamano committed
968
	if (!skip_df_check &&
969
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
970
		if (!ok_to_replace)
971 972
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
973
		pos = index_name_pos(istate, ce->name, ce->ce_flags);
974 975
		pos = -pos-1;
	}
976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991
	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;
	}
992

993
	/* Make sure the array is big enough .. */
994 995 996 997
	if (istate->cache_nr == istate->cache_alloc) {
		istate->cache_alloc = alloc_nr(istate->cache_alloc);
		istate->cache = xrealloc(istate->cache,
					istate->cache_alloc * sizeof(struct cache_entry *));
998 999 1000
	}

	/* Add it in.. */
1001
	istate->cache_nr++;
1002
	if (istate->cache_nr > pos + 1)
1003 1004 1005
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1006
	set_index_entry(istate, pos, ce);
1007
	istate->cache_changed = 1;
1008 1009 1010
	return 0;
}

1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
/*
 * "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.
 */
1022
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1023 1024
					     struct cache_entry *ce,
					     unsigned int options, int *err)
1025 1026 1027 1028
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1029
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1030
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1031

1032 1033 1034
	if (ce_uptodate(ce))
		return ce;

1035
	/*
1036 1037 1038
	 * 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.
1039
	 */
1040 1041 1042 1043
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1044 1045 1046 1047 1048
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1049
	if (lstat(ce->name, &st) < 0) {
1050 1051
		if (err)
			*err = errno;
1052 1053
		return NULL;
	}
1054

1055
	changed = ie_match_stat(istate, ce, &st, options);
1056
	if (!changed) {
1057 1058 1059 1060 1061 1062 1063 1064
		/*
		 * 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 &&
1065
		    !(ce->ce_flags & CE_VALID))
1066
			; /* mark this one VALID again */
1067 1068 1069 1070 1071 1072
		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.
			 */
1073 1074
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1075
			return ce;
1076
		}
1077 1078
	}

1079
	if (ie_modified(istate, ce, &st, options)) {
1080 1081
		if (err)
			*err = EINVAL;
1082 1083
		return NULL;
	}
1084 1085 1086 1087 1088

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1089 1090 1091 1092 1093
	/*
	 * 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.
1094
	 */
1095
	if (!ignore_valid && assume_unchanged &&
1096 1097
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1098 1099 1100 1101

	return updated;
}

1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113
static void show_file(const char * fmt, const char * name, int in_porcelain,
		      int * first, char *header_msg)
{
	if (in_porcelain && *first && header_msg) {
		printf("%s\n", header_msg);
		*first=0;
	}
	printf(fmt, name);
}

int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec,
		  char *seen, char *header_msg)
1114 1115 1116 1117 1118 1119 1120
{
	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;
1121
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1122 1123
	int first = 1;
	int in_porcelain = (flags & REFRESH_IN_PORCELAIN);
1124
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
1125 1126
	const char *needs_update_fmt;
	const char *needs_merge_fmt;
1127

1128 1129
	needs_update_fmt = (in_porcelain ? "M\t%s\n" : "%s: needs update\n");
	needs_merge_fmt = (in_porcelain ? "U\t%s\n" : "%s: needs merge\n");
1130
	for (i = 0; i < istate->cache_nr; i++) {
1131
		struct cache_entry *ce, *new;
1132 1133
		int cache_errno = 0;

1134
		ce = istate->cache[i];
1135 1136 1137
		if (ignore_submodules && S_ISGITLINK(ce->ce_mode))
			conti