read-cache.c 50.4 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 24 25 26 27 28 29 30
/* 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" */
31
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
32

33
struct index_state the_index;
34

Junio C Hamano's avatar
Junio C Hamano committed
35 36 37
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
38
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
39 40
}

41 42 43 44
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

45
	remove_name_hash(old);
46
	set_index_entry(istate, nr, ce);
47 48 49
	istate->cache_changed = 1;
}

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
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);
}

66 67 68 69 70 71 72
/*
 * 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)
{
73 74
	ce->ce_ctime.sec = (unsigned int)st->st_ctime;
	ce->ce_mtime.sec = (unsigned int)st->st_mtime;
75 76
	ce->ce_ctime.nsec = ST_CTIME_NSEC(*st);
	ce->ce_mtime.nsec = ST_MTIME_NSEC(*st);
77 78 79 80 81
	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
82 83

	if (assume_unchanged)
84
		ce->ce_flags |= CE_VALID;
85 86 87

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
88 89
}

Junio C Hamano's avatar
Junio C Hamano committed
90 91 92 93 94 95 96
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];
97
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
98
			match = hashcmp(sha1, ce->sha1);
99
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
100 101 102 103
	}
	return match;
}

104
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
105 106 107 108
{
	int match = -1;
	void *buffer;
	unsigned long size;
109
	enum object_type type;
110
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
111

112
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
113
		return -1;
114

115
	buffer = read_sha1_file(ce->sha1, &type, &size);
116 117 118 119
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
120
	}
121
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
122 123 124
	return match;
}

125 126 127 128 129 130
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
131
	 * under GITLINK directory be a valid git directory. It
132 133 134 135 136 137 138 139 140 141
	 * 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
142 143 144 145 146 147 148 149
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:
150
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
151 152
			return DATA_CHANGED;
		break;
153
	case S_IFDIR:
154
		if (S_ISGITLINK(ce->ce_mode))
155
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
156 157 158 159 160 161
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
162
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
163 164 165
{
	unsigned int changed = 0;

166 167 168 169
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
170 171
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
172 173 174 175
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
176
		    (0100 & (ce->ce_mode ^ st->st_mode)))
177
			changed |= MODE_CHANGED;
178 179
		break;
	case S_IFLNK:
180 181 182
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
183
		break;
Martin Waitz's avatar
Martin Waitz committed
184
	case S_IFGITLINK:
185
		/* We ignore most of the st_xxx fields for gitlinks */
186 187 188 189
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
190
		return changed;
191
	default:
192
		die("internal error: ce_mode is %o", ce->ce_mode);
193
	}
194
	if (ce->ce_mtime.sec != (unsigned int)st->st_mtime)
195
		changed |= MTIME_CHANGED;
196
	if (trust_ctime && ce->ce_ctime.sec != (unsigned int)st->st_ctime)
197
		changed |= CTIME_CHANGED;
198

199
#ifdef USE_NSEC
200
	if (ce->ce_mtime.nsec != ST_MTIME_NSEC(*st))
201
		changed |= MTIME_CHANGED;
202
	if (trust_ctime && ce->ce_ctime.nsec != ST_CTIME_NSEC(*st))
203 204 205
		changed |= CTIME_CHANGED;
#endif

206 207
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
208
		changed |= OWNER_CHANGED;
209
	if (ce->ce_ino != (unsigned int) st->st_ino)
210
		changed |= INODE_CHANGED;
211 212 213 214 215 216 217

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
218
	if (ce->ce_dev != (unsigned int) st->st_dev)
219 220 221
		changed |= INODE_CHANGED;
#endif

222
	if (ce->ce_size != (unsigned int) st->st_size)
223
		changed |= DATA_CHANGED;
224

225 226 227 228 229 230
	/* 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
231 232 233
	return changed;
}

234
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
235
{
236
	return (!S_ISGITLINK(ce->ce_mode) &&
237 238 239 240 241 242 243 244 245 246
		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
		 );
247 248
}

249
int ie_match_stat(const struct index_state *istate,
250 251
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
252
{
Junio C Hamano's avatar
Junio C Hamano committed
253
	unsigned int changed;
254
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
255
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
256
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
257 258 259 260

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
261 262
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
263
	 */
264 265
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
266
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
267 268
		return 0;

269 270 271 272 273 274 275 276
	/*
	 * 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
277
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
278

Junio C Hamano's avatar
Junio C Hamano committed
279 280 281 282 283 284 285 286 287 288 289 290 291 292
	/*
	 * 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
293
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
294
	 */
295
	if (!changed && is_racy_timestamp(istate, ce)) {
296 297 298 299 300
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
301

Junio C Hamano's avatar
Junio C Hamano committed
302
	return changed;
303 304
}

305
int ie_modified(const struct index_state *istate,
306
		struct cache_entry *ce, struct stat *st, unsigned int options)
307
{
Junio C Hamano's avatar
Junio C Hamano committed
308
	int changed, changed_fs;
309 310

	changed = ie_match_stat(istate, ce, st, options);
311 312 313 314 315 316 317 318 319
	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;

320 321 322 323 324 325 326 327 328 329 330 331 332
	/*
	 * 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.
333
	 */
334 335
	if ((changed & DATA_CHANGED) &&
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_size != 0))
336 337
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
338 339 340
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
341 342 343
	return 0;
}

344 345 346 347 348 349 350 351 352 353 354 355
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];
356
	if (!c1 && S_ISDIR(mode1))
357
		c1 = '/';
358
	if (!c2 && S_ISDIR(mode2))
359 360 361 362
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397
/*
 * 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;
}

398
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
399
{
400 401 402 403 404 405 406 407 408
	int len1, len2, len, cmp;

	len1 = flags1 & CE_NAMEMASK;
	if (CE_NAMEMASK <= len1)
		len1 = strlen(name1 + CE_NAMEMASK) + CE_NAMEMASK;
	len2 = flags2 & CE_NAMEMASK;
	if (CE_NAMEMASK <= len2)
		len2 = strlen(name2 + CE_NAMEMASK) + CE_NAMEMASK;
	len = len1 < len2 ? len1 : len2;
409 410 411 412 413 414 415 416

	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
417

418 419 420
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
421

422 423 424 425
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
426 427 428
	return 0;
}

429
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
430 431 432 433
{
	int first, last;

	first = 0;
434
	last = istate->cache_nr;
435 436
	while (last > first) {
		int next = (last + first) >> 1;
437
		struct cache_entry *ce = istate->cache[next];
438
		int cmp = cache_name_compare(name, namelen, ce->name, ce->ce_flags);
439
		if (!cmp)
440
			return next;
441 442 443 444 445 446
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
447
	return -first-1;
448 449
}

450
/* Remove entry, return true if there are more entries to go.. */
451
int remove_index_entry_at(struct index_state *istate, int pos)
452
{
453 454
	struct cache_entry *ce = istate->cache[pos];

455
	record_resolve_undo(istate, ce);
456
	remove_name_hash(ce);
457 458 459
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
460
		return 0;
461 462 463
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
464 465 466
	return 1;
}

467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486
/*
 * 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;
}

487
int remove_file_from_index(struct index_state *istate, const char *path)
488
{
489
	int pos = index_name_pos(istate, path, strlen(path));
490 491
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
492
	cache_tree_invalidate_path(istate->cache_tree, path);
493 494
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
495 496 497
	return 0;
}

498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
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;
}

526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557
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;
}

558 559 560 561 562 563 564 565
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);
}

566
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
567
{
568
	int size, namelen, was_same;
569
	mode_t st_mode = st->st_mode;
570
	struct cache_entry *ce, *alias;
571
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
572 573
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
574 575 576
	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
577

578
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
579
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
580 581

	namelen = strlen(path);
582
	if (S_ISDIR(st_mode)) {
583 584 585
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
586 587 588
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
589
	ce->ce_flags = namelen;
590 591
	if (!intent_only)
		fill_stat_cache_info(ce, st);
592 593
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
594

595
	if (trust_executable_bit && has_symlinks)
596
		ce->ce_mode = create_ce_mode(st_mode);
597
	else {
598 599
		/* 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
600
		 */
601
		struct cache_entry *ent;
602
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
603

604
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
605
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
606 607
	}

608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630
	/* 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;
				}
			}
		}
	}

631
	alias = index_name_exists(istate, ce->name, ce_namelen(ce), ignore_case);
632
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
633 634
		/* Nothing changed, really */
		free(ce);
635 636
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
637
		alias->ce_flags |= CE_ADDED;
638 639
		return 0;
	}
640
	if (!intent_only) {
641
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
642 643 644 645
			return error("unable to index file %s", path);
	} else
		record_intent_to_add(ce);

646 647 648
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
649

Junio C Hamano's avatar
Junio C Hamano committed
650
	/* It was suspected to be racily clean, but it turns out to be Ok */
651 652 653 654 655 656 657
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
658
	else if (add_index_entry(istate, ce, add_option))
659
		return error("unable to add %s to index",path);
660
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
661 662 663 664
		printf("add '%s'\n", path);
	return 0;
}

665
int add_file_to_index(struct index_state *istate, const char *path, int flags)
666 667 668
{
	struct stat st;
	if (lstat(path, &st))
669
		die_errno("unable to stat '%s'", path);
670
	return add_to_index(istate, path, &st, flags);
671 672
}

673 674 675 676 677 678 679
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;

680 681
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
682
		return NULL;
683
	}
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699

	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;
}

700
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
701 702 703 704 705
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

706
int ce_path_match(const struct cache_entry *ce, const struct pathspec *pathspec)
707
{
708
	return match_pathspec_depth(pathspec, ce->name, ce_namelen(ce), 0, NULL);
709 710
}

711 712 713 714 715 716 717 718 719 720 721 722 723 724 725
/*
 * 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.
	 */
726

727
	/* "." is not allowed */
728
	if (*rest == '\0' || is_dir_sep(*rest))
729 730
		return 0;

731
	switch (*rest) {
732 733 734 735 736 737 738 739 740 741 742 743
	/*
	 * ".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 '.':
744
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
745 746 747 748 749 750 751 752 753
			return 0;
	}
	return 1;
}

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

754 755 756
	if (has_dos_drive_prefix(path))
		return 0;

757 758 759 760
	goto inside;
	for (;;) {
		if (!c)
			return 1;
761
		if (is_dir_sep(c)) {
762 763
inside:
			c = *path++;
764 765 766
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
767 768 769 770 771
		}
		c = *path++;
	}
}

772 773 774
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
775
 */
776 777
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
778
{
779 780
	int retval = 0;
	int len = ce_namelen(ce);
781
	int stage = ce_stage(ce);
782
	const char *name = ce->name;
783

784 785
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
786

787
		if (len >= ce_namelen(p))
788
			break;
789 790
		if (memcmp(name, p->name, len))
			break;
791 792
		if (ce_stage(p) != stage)
			continue;
793 794
		if (p->name[len] != '/')
			continue;
795
		if (p->ce_flags & CE_REMOVE)
796
			continue;
797 798 799
		retval = -1;
		if (!ok_to_replace)
			break;
800
		remove_index_entry_at(istate, --pos);
801
	}
802 803
	return retval;
}
804

805 806 807 808
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
809 810
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
811 812
{
	int retval = 0;
813
	int stage = ce_stage(ce);
814 815
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
816

817 818
	for (;;) {
		int len;
819

820 821 822 823 824 825 826
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
827

828
		pos = index_name_pos(istate, name, create_ce_flags(len, stage));
829
		if (pos >= 0) {
830 831 832 833 834 835 836 837
			/*
			 * 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.
			 */
838
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
839 840 841
				retval = -1;
				if (!ok_to_replace)
					break;
842
				remove_index_entry_at(istate, pos);
843 844
				continue;
			}
845
		}
846 847
		else
			pos = -pos-1;
848 849 850 851

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
852
		 * we're ok, and we can exit.
853
		 */
854 855
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
856 857 858 859
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
860 861 862
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
863 864 865 866 867 868
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
869
		}
870
	}
871 872 873 874 875 876 877
	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
878
 *
879 880 881 882
 * 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.
 */
883 884 885
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
886
{
887 888 889 890 891
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
892
	if (ce->ce_flags & CE_REMOVE)
893 894
		return 0;

895 896 897
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
898
	 * in the array.
899
	 */
900
	retval = has_file_name(istate, ce, pos, ok_to_replace);
901

902 903 904 905
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
906
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
907 908
}

909
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
910 911
{
	int pos;
912 913
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
914
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
915
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
916

Junio C Hamano's avatar
Junio C Hamano committed
917
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
918
	pos = index_name_pos(istate, ce->name, ce->ce_flags);
919

Junio C Hamano's avatar
Junio C Hamano committed
920
	/* existing match? Just replace it. */
921
	if (pos >= 0) {
922 923
		if (!new_only)
			replace_index_entry(istate, pos, ce);
924 925
		return 0;
	}
926
	pos = -pos-1;
927

928 929 930 931
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
932 933
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
934
			ok_to_add = 1;
935
			if (!remove_index_entry_at(istate, pos))
936 937 938 939
				break;
		}
	}

940 941
	if (!ok_to_add)
		return -1;
942
	if (!verify_path(ce->name))
943
		return error("Invalid path '%s'", ce->name);
944

Junio C Hamano's avatar
Junio C Hamano committed
945
	if (!skip_df_check &&
946
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
947
		if (!ok_to_replace)
948 949
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
950
		pos = index_name_pos(istate, ce->name, ce->ce_flags);
951 952
		pos = -pos-1;
	}
953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968
	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;
	}
969

970
	/* Make sure the array is big enough .. */
971 972 973 974
	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 *));
975 976 977
	}

	/* Add it in.. */
978
	istate->cache_nr++;
979
	if (istate->cache_nr > pos + 1)
980 981 982
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
983
	set_index_entry(istate, pos, ce);
984
	istate->cache_changed = 1;
985 986 987
	return 0;
}

988 989 990 991 992 993 994 995 996 997 998
/*
 * "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.
 */
999
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1000
					     struct cache_entry *ce,
1001 1002
					     unsigned int options, int *err,
					     int *changed_ret)
1003 1004 1005 1006
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1007
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1008
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1009

1010 1011 1012
	if (ce_uptodate(ce))
		return ce;

1013
	/*
1014 1015 1016
	 * 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.
1017
	 */
1018 1019 1020 1021
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1022 1023 1024 1025 1026
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1027
	if (lstat(ce->name, &st) < 0) {
1028 1029
		if (err)
			*err = errno;
1030 1031
		return NULL;
	}
1032

1033
	changed = ie_match_stat(istate, ce, &st, options);
1034 1035
	if (changed_ret)
		*changed_ret = changed;
1036
	if (!changed) {
1037 1038 1039 1040 1041 1042 1043 1044
		/*
		 * 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 &&
1045
		    !(ce->ce_flags & CE_VALID))
1046
			; /* mark this one VALID again */
1047 1048 1049 1050 1051 1052
		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.
			 */
1053 1054
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1055
			return ce;
1056
		}
1057 1058
	}

1059
	if (ie_modified(istate, ce, &st, options)) {
1060 1061
		if (err)
			*err = EINVAL;
1062 1063
		return NULL;
	}
1064 1065 1066 1067 1068

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1069 1070 1071 1072 1073
	/*
	 * 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.
1074
	 */
1075
	if (!ignore_valid && assume_unchanged &&
1076 1077
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1078 1079 1080 1081

	return updated;
}

1082
static void show_file(const char * fmt, const char * name, int in_porcelain,
1083
		      int * first, const char *header_msg)
1084 1085 1086
{
	if (in_porcelain && *first && header_msg) {
		printf("%s\n", header_msg);
1087
		*first = 0;
1088 1089 1090 1091 1092
	}
	printf(fmt, name);
}

int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec,
1093
		  char *seen, const char *header_msg)
1094 1095 1096 1097 1098 1099 1100
{
	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;
1101
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1102 1103
	int first = 1;
	int in_porcelain