read-cache.c 44 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 13 14 15
#include "tree.h"
#include "commit.h"
#include "diff.h"
#include "diffcore.h"
#include "revision.h"
16
#include "blob.h"
17 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

30
struct index_state the_index;
31

Junio C Hamano's avatar
Junio C Hamano committed
32 33 34
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
35
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
36 37
}

38 39 40 41
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

42
	remove_name_hash(old);
43
	set_index_entry(istate, nr, ce);
44 45 46
	istate->cache_changed = 1;
}

47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
void rename_index_entry_at(struct index_state *istate, int nr, const char *new_name)
{
	struct cache_entry *old = istate->cache[nr], *new;
	int namelen = strlen(new_name);

	new = xmalloc(cache_entry_size(namelen));
	copy_cache_entry(new, old);
	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);
}

63 64 65 66 67 68 69
/*
 * 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)
{
70 71 72 73 74 75 76
	ce->ce_ctime = st->st_ctime;
	ce->ce_mtime = st->st_mtime;
	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
77 78

	if (assume_unchanged)
79
		ce->ce_flags |= CE_VALID;
80 81 82

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
83 84
}

Junio C Hamano's avatar
Junio C Hamano committed
85 86 87 88 89 90 91
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];
92
		if (!index_fd(sha1, fd, st, 0, OBJ_BLOB, ce->name))
93
			match = hashcmp(sha1, ce->sha1);
94
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
95 96 97 98
	}
	return match;
}

99
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
100 101 102 103
{
	int match = -1;
	void *buffer;
	unsigned long size;
104
	enum object_type type;
105
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
106

107
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
108
		return -1;
109

110
	buffer = read_sha1_file(ce->sha1, &type, &size);
111 112 113 114
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
115
	}
116
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
117 118 119
	return match;
}

120 121 122 123 124 125
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
126
	 * under GITLINK directory be a valid git directory. It
127 128 129 130 131 132 133 134 135 136
	 * 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
137 138 139 140 141 142 143 144
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:
145
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
146 147
			return DATA_CHANGED;
		break;
148
	case S_IFDIR:
149
		if (S_ISGITLINK(ce->ce_mode))
150
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
151 152 153 154 155 156
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

157
int is_empty_blob_sha1(const unsigned char *sha1)
158 159 160 161 162 163 164 165 166
{
	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
167
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
168 169 170
{
	unsigned int changed = 0;

171 172 173 174
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
175 176
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
177 178 179 180
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
181
		    (0100 & (ce->ce_mode ^ st->st_mode)))
182
			changed |= MODE_CHANGED;
183 184
		break;
	case S_IFLNK:
185 186 187
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
188
		break;
Martin Waitz's avatar
Martin Waitz committed
189
	case S_IFGITLINK:
190
		/* We ignore most of the st_xxx fields for gitlinks */
191 192 193 194
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
195
		return changed;
196
	default:
197
		die("internal error: ce_mode is %o", ce->ce_mode);
198
	}
199
	if (ce->ce_mtime != (unsigned int) st->st_mtime)
200
		changed |= MTIME_CHANGED;
201
	if (trust_ctime && ce->ce_ctime != (unsigned int) st->st_ctime)
202
		changed |= CTIME_CHANGED;
203

204 205
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
206
		changed |= OWNER_CHANGED;
207
	if (ce->ce_ino != (unsigned int) st->st_ino)
208
		changed |= INODE_CHANGED;
209 210 211 212 213 214 215

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
216
	if (ce->ce_dev != (unsigned int) st->st_dev)
217 218 219
		changed |= INODE_CHANGED;
#endif

220
	if (ce->ce_size != (unsigned int) st->st_size)
221
		changed |= DATA_CHANGED;
222

223 224 225 226 227 228
	/* 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
229 230 231
	return changed;
}

232
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
233
{
234 235
	return (!S_ISGITLINK(ce->ce_mode) &&
		istate->timestamp &&
236 237 238
		((unsigned int)istate->timestamp) <= ce->ce_mtime);
}

239
int ie_match_stat(const struct index_state *istate,
240 241
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
242
{
Junio C Hamano's avatar
Junio C Hamano committed
243
	unsigned int changed;
244 245
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
246 247 248 249 250

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
	 */
251
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
252 253
		return 0;

254 255 256 257 258 259 260 261
	/*
	 * 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
262
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
263

Junio C Hamano's avatar
Junio C Hamano committed
264 265 266 267 268 269 270 271 272 273 274 275 276 277
	/*
	 * 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
278
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
279
	 */
280
	if (!changed && is_racy_timestamp(istate, ce)) {
281 282 283 284 285
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
286

Junio C Hamano's avatar
Junio C Hamano committed
287
	return changed;
288 289
}

290
int ie_modified(const struct index_state *istate,
291
		struct cache_entry *ce, struct stat *st, unsigned int options)
292
{
Junio C Hamano's avatar
Junio C Hamano committed
293
	int changed, changed_fs;
294 295

	changed = ie_match_stat(istate, ce, st, options);
296 297 298 299 300 301 302 303 304
	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;

305 306 307 308 309 310 311 312 313 314 315 316 317
	/*
	 * 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.
318
	 */
319 320
	if ((changed & DATA_CHANGED) &&
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_size != 0))
321 322
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
323 324 325
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
326 327 328
	return 0;
}

329 330 331 332 333 334 335 336 337 338 339 340
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];
341
	if (!c1 && S_ISDIR(mode1))
342
		c1 = '/';
343
	if (!c2 && S_ISDIR(mode2))
344 345 346 347
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
/*
 * 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;
}

383
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
384
{
385 386
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
387 388 389 390 391 392 393 394 395 396
	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
397

398 399 400
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
401

402 403 404 405
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
406 407 408
	return 0;
}

409
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
410 411 412 413
{
	int first, last;

	first = 0;
414
	last = istate->cache_nr;
415 416
	while (last > first) {
		int next = (last + first) >> 1;
417
		struct cache_entry *ce = istate->cache[next];
418
		int cmp = cache_name_compare(name, namelen, ce->name, ce->ce_flags);
419
		if (!cmp)
420
			return next;
421 422 423 424 425 426
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
427
	return -first-1;
428 429
}

430
/* Remove entry, return true if there are more entries to go.. */
431
int remove_index_entry_at(struct index_state *istate, int pos)
432
{
433 434
	struct cache_entry *ce = istate->cache[pos];

435
	remove_name_hash(ce);
436 437 438
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
439
		return 0;
440 441 442
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
443 444 445
	return 1;
}

446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465
/*
 * 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;
}

466
int remove_file_from_index(struct index_state *istate, const char *path)
467
{
468
	int pos = index_name_pos(istate, path, strlen(path));
469 470
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
471
	cache_tree_invalidate_path(istate->cache_tree, path);
472 473
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
474 475 476
	return 0;
}

477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
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;
}

505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
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;
}

537 538 539 540 541 542 543 544
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);
}

545
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
546
{
547
	int size, namelen, was_same;
548
	mode_t st_mode = st->st_mode;
549
	struct cache_entry *ce, *alias;
550
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_RACY_IS_DIRTY;
551 552
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
553 554 555
	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
556

557
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
558
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
559 560

	namelen = strlen(path);
561
	if (S_ISDIR(st_mode)) {
562 563 564
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
565 566 567
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
568
	ce->ce_flags = namelen;
569 570
	if (!intent_only)
		fill_stat_cache_info(ce, st);
571 572
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
573

574
	if (trust_executable_bit && has_symlinks)
575
		ce->ce_mode = create_ce_mode(st_mode);
576
	else {
577 578
		/* 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
579
		 */
580
		struct cache_entry *ent;
581
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
582

583
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
584
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
585 586
	}

587
	alias = index_name_exists(istate, ce->name, ce_namelen(ce), ignore_case);
588
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
589 590
		/* Nothing changed, really */
		free(ce);
591
		ce_mark_uptodate(alias);
592
		alias->ce_flags |= CE_ADDED;
593 594
		return 0;
	}
595 596 597 598 599 600
	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);

601 602 603
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
604

Junio C Hamano's avatar
Junio C Hamano committed
605
	/* It was suspected to be racily clean, but it turns out to be Ok */
606 607 608 609 610 611 612
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
613
	else if (add_index_entry(istate, ce, add_option))
614
		return error("unable to add %s to index",path);
615
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
616 617 618 619
		printf("add '%s'\n", path);
	return 0;
}

620
int add_file_to_index(struct index_state *istate, const char *path, int flags)
621 622 623 624
{
	struct stat st;
	if (lstat(path, &st))
		die("%s: unable to stat (%s)", path, strerror(errno));
625
	return add_to_index(istate, path, &st, flags);
626 627
}

628 629 630 631 632 633 634
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;

635 636
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
637
		return NULL;
638
	}
639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654

	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;
}

655
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
656 657 658 659 660
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680
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;
681 682
		if (!matchlen)
			return 1;
683 684 685 686
	}
	return 0;
}

687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750
/*
 * 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.
	 */
	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;

	goto inside;
	for (;;) {
		if (!c)
			return 1;
		if (c == '/') {
inside:
			c = *path++;
			switch (c) {
			default:
				continue;
			case '/': case '\0':
				break;
			case '.':
				if (verify_dotfile(path))
					continue;
			}
			return 0;
		}
		c = *path++;
	}
}

751 752 753
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
754
 */
755 756
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
757
{
758 759
	int retval = 0;
	int len = ce_namelen(ce);
760
	int stage = ce_stage(ce);
761
	const char *name = ce->name;
762

763 764
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
765

766
		if (len >= ce_namelen(p))
767
			break;
768 769
		if (memcmp(name, p->name, len))
			break;
770 771
		if (ce_stage(p) != stage)
			continue;
772 773
		if (p->name[len] != '/')
			continue;
774
		if (p->ce_flags & CE_REMOVE)
775
			continue;
776 777 778
		retval = -1;
		if (!ok_to_replace)
			break;
779
		remove_index_entry_at(istate, --pos);
780
	}
781 782
	return retval;
}
783

784 785 786 787
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
788 789
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
790 791
{
	int retval = 0;
792
	int stage = ce_stage(ce);
793 794
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
795

796 797
	for (;;) {
		int len;
798

799 800 801 802 803 804 805
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
806

807
		pos = index_name_pos(istate, name, create_ce_flags(len, stage));
808
		if (pos >= 0) {
809 810 811 812 813 814 815 816
			/*
			 * 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.
			 */
817
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
818 819 820
				retval = -1;
				if (!ok_to_replace)
					break;
821
				remove_index_entry_at(istate, pos);
822 823
				continue;
			}
824
		}
825 826
		else
			pos = -pos-1;
827 828 829 830

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
831
		 * we're ok, and we can exit.
832
		 */
833 834
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
835 836 837 838
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
839 840 841
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
842 843 844 845 846 847
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
848
		}
849
	}
850 851 852 853 854 855 856
	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
857
 *
858 859 860 861
 * 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.
 */
862 863 864
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
865
{
866 867 868 869 870
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
871
	if (ce->ce_flags & CE_REMOVE)
872 873
		return 0;

874 875 876
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
877
	 * in the array.
878
	 */
879
	retval = has_file_name(istate, ce, pos, ok_to_replace);
880

881 882 883 884
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
885
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
886 887
}

888
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
889 890
{
	int pos;
891 892
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
893
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
894
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
895

Junio C Hamano's avatar
Junio C Hamano committed
896
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
897
	pos = index_name_pos(istate, ce->name, ce->ce_flags);
898

Junio C Hamano's avatar
Junio C Hamano committed
899
	/* existing match? Just replace it. */
900
	if (pos >= 0) {
901 902
		if (!new_only)
			replace_index_entry(istate, pos, ce);
903 904
		return 0;
	}
905
	pos = -pos-1;
906

907 908 909 910
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
911 912
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
913
			ok_to_add = 1;
914
			if (!remove_index_entry_at(istate, pos))
915 916 917 918
				break;
		}
	}

919 920
	if (!ok_to_add)
		return -1;
921
	if (!verify_path(ce->name))
922
		return error("Invalid path '%s'", ce->name);
923

Junio C Hamano's avatar
Junio C Hamano committed
924
	if (!skip_df_check &&
925
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
926
		if (!ok_to_replace)
927 928
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
929
		pos = index_name_pos(istate, ce->name, ce->ce_flags);
930 931
		pos = -pos-1;
	}
932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947
	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;
	}
948

949
	/* Make sure the array is big enough .. */
950 951 952 953
	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 *));
954 955 956
	}

	/* Add it in.. */
957
	istate->cache_nr++;
958
	if (istate->cache_nr > pos + 1)
959 960 961
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
962
	set_index_entry(istate, pos, ce);
963
	istate->cache_changed = 1;
964 965 966
	return 0;
}

967 968 969 970 971 972 973 974 975 976 977
/*
 * "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.
 */
978
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
979 980
					     struct cache_entry *ce,
					     unsigned int options, int *err)
981 982 983 984
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
985
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
986

987 988 989
	if (ce_uptodate(ce))
		return ce;

990 991 992 993 994 995 996 997 998
	/*
	 * CE_VALID means the user promised us that the change to
	 * the work tree does not matter and told us not to worry.
	 */
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

999
	if (lstat(ce->name, &st) < 0) {
1000 1001
		if (err)
			*err = errno;
1002 1003
		return NULL;
	}
1004

1005
	changed = ie_match_stat(istate, ce, &st, options);
1006
	if (!changed) {
1007 1008 1009 1010 1011 1012 1013 1014
		/*
		 * 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 &&
1015
		    !(ce->ce_flags & CE_VALID))
1016
			; /* mark this one VALID again */
1017 1018 1019 1020 1021 1022 1023
		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.
			 */
			ce_mark_uptodate(ce);
1024
			return ce;
1025
		}
1026 1027
	}

1028
	if (ie_modified(istate, ce, &st, options)) {
1029 1030
		if (err)
			*err = EINVAL;
1031 1032
		return NULL;
	}
1033 1034 1035 1036 1037

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1038 1039 1040 1041 1042
	/*
	 * 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.
1043
	 */
1044
	if (!ignore_valid && assume_unchanged &&
1045 1046
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1047 1048 1049 1050

	return updated;
}

1051
int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec, char *seen)
1052 1053 1054 1055 1056 1057 1058
{
	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;
1059
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1060
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
1061
	const char *needs_update_message;
1062

1063 1064
	needs_update_message = ((flags & REFRESH_SAY_CHANGED)
				? "locally modified" : "needs update");
1065
	for (i = 0; i < istate->cache_nr; i++) {
1066
		struct cache_entry *ce, *new;
1067 1068
		int cache_errno = 0;

1069
		ce = istate->cache[i];
1070 1071 1072
		if (ignore_submodules && S_ISGITLINK(ce->ce_mode))
			continue;

1073
		if (ce_stage(ce)) {
1074 1075
			while ((i < istate->cache_nr) &&
			       ! strcmp(istate->cache[i]->name, ce->name))
1076 1077 1078 1079 1080 1081 1082 1083 1084
				i++;
			i--;
			if (allow_unmerged)
				continue;
			printf("%s: needs merge\n", ce->name);
			has_errors = 1;
			continue;
		}

1085 1086 1087
		if (pathspec && !match_pathspec(pathspec, ce->name, strlen(ce->name), 0, seen))
			continue;

1088
		new = refresh_cache_ent(istate, ce, options, &cache_errno);
1089
		if (new == ce)
1090
			continue;
1091 1092
		if (!new) {
			if (not_new && cache_errno == ENOENT)
1093
				continue;
1094
			if (really && cache_errno == EINVAL) {
1095 1096 1097
				/* If we are doing --really-refresh that
				 * means the index is not valid anymore.
				 */
1098
				ce->ce_flags &= ~CE_VALID;
1099
				istate->cache_changed = 1;
1100 1101 1102
			}
			if (quiet)
				continue;
1103
			printf("%s: %s\n", ce->name, needs_update_message);
1104 1105 1106
			has_errors = 1;
			continue;
		}
1107 1108

		replace_index_entry(istate, i, new);
1109 1110 1111 1112
	}
	return has_errors;
}

1113 1114
struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really)
{
1115
	return refresh_cache_ent(&the_index, ce, really, NULL);
1116 1117
}

1118
static int verify_hdr(struct cache_header *hdr, unsigned long size)
1119
{
1120
	git_SHA_CTX c;
1121
	unsigned char sha1[20];
1122

1123
	if (hdr->hdr_signature != htonl(CACHE_SIGNATURE))
1124
		return error("bad signature");
Duy Nguyen's avatar
Duy Nguyen committed
1125
	if (hdr->hdr_version != htonl(2) && hdr->hdr_version != htonl(3))
1126
		return error("bad index version");
1127 1128 1129
	git_SHA1_Init(&c);
	git_SHA1_Update(&c, hdr, size - 20);
	git_SHA1_Final(sha1, &c);
1130
	if (hashcmp(sha1, (unsigned char *)hdr + size - 20))
1131
		return error("bad index file sha1 signature");
1132 1133 1134
	return 0;
}

1135 1136
static int read_index_extension(struct index_state *istate,
				const char *ext, void *data, unsigned long sz)