read-cache.c 44.8 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5
/*
 * GIT - The information manager from hell
 *
 * Copyright (C) Linus Torvalds, 2005
 */
6
#define NO_THE_INDEX_COMPATIBILITY_MACROS
7
#include "cache.h"
8
#include "cache-tree.h"
9
#include "refs.h"
10
#include "dir.h"
11 12 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
	ce->ce_ctime.sec = (unsigned int)st->st_ctime;
	ce->ce_mtime.sec = (unsigned int)st->st_mtime;
72 73
	ce->ce_ctime.nsec = ST_CTIME_NSEC(*st);
	ce->ce_mtime.nsec = ST_MTIME_NSEC(*st);
74 75 76 77 78
	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
79 80

	if (assume_unchanged)
81
		ce->ce_flags |= CE_VALID;
82 83 84

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
85 86
}

Junio C Hamano's avatar
Junio C Hamano committed
87 88 89 90 91 92 93
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];
94
		if (!index_fd(sha1, fd, st, 0, OBJ_BLOB, ce->name))
95
			match = hashcmp(sha1, ce->sha1);
96
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
97 98 99 100
	}
	return match;
}

101
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
102 103 104 105
{
	int match = -1;
	void *buffer;
	unsigned long size;
106
	enum object_type type;
107
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
108

109
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
110
		return -1;
111

112
	buffer = read_sha1_file(ce->sha1, &type, &size);
113 114 115 116
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
117
	}
118
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
119 120 121
	return match;
}

122 123 124 125 126 127
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
128
	 * under GITLINK directory be a valid git directory. It
129 130 131 132 133 134 135 136 137 138
	 * 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
139 140 141 142 143 144 145 146
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:
147
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
148 149
			return DATA_CHANGED;
		break;
150
	case S_IFDIR:
151
		if (S_ISGITLINK(ce->ce_mode))
152
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
153 154 155 156 157 158
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

159
int is_empty_blob_sha1(const unsigned char *sha1)
160 161 162 163 164 165 166 167 168
{
	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
169
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
170 171 172
{
	unsigned int changed = 0;

173 174 175 176
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
177 178
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
179 180 181 182
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
183
		    (0100 & (ce->ce_mode ^ st->st_mode)))
184
			changed |= MODE_CHANGED;
185 186
		break;
	case S_IFLNK:
187 188 189
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
190
		break;
Martin Waitz's avatar
Martin Waitz committed
191
	case S_IFGITLINK:
192
		/* We ignore most of the st_xxx fields for gitlinks */
193 194 195 196
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
197
		return changed;
198
	default:
199
		die("internal error: ce_mode is %o", ce->ce_mode);
200
	}
201
	if (ce->ce_mtime.sec != (unsigned int)st->st_mtime)
202
		changed |= MTIME_CHANGED;
203
	if (trust_ctime && ce->ce_ctime.sec != (unsigned int)st->st_ctime)
204
		changed |= CTIME_CHANGED;
205

206
#ifdef USE_NSEC
207
	if (ce->ce_mtime.nsec != ST_MTIME_NSEC(*st))
208
		changed |= MTIME_CHANGED;
209
	if (trust_ctime && ce->ce_ctime.nsec != ST_CTIME_NSEC(*st))
210 211 212
		changed |= CTIME_CHANGED;
#endif

213 214
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
215
		changed |= OWNER_CHANGED;
216
	if (ce->ce_ino != (unsigned int) st->st_ino)
217
		changed |= INODE_CHANGED;
218 219 220 221 222 223 224

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
225
	if (ce->ce_dev != (unsigned int) st->st_dev)
226 227 228
		changed |= INODE_CHANGED;
#endif

229
	if (ce->ce_size != (unsigned int) st->st_size)
230
		changed |= DATA_CHANGED;
231

232 233 234 235 236 237
	/* 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
238 239 240
	return changed;
}

241
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
242
{
243
	return (!S_ISGITLINK(ce->ce_mode) &&
244 245 246 247 248 249 250 251 252 253
		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
		 );
254 255
}

256
int ie_match_stat(const struct index_state *istate,
257 258
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
259
{
Junio C Hamano's avatar
Junio C Hamano committed
260
	unsigned int changed;
261 262
	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
263 264 265 266 267

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
	 */
268
	if (!ignore_valid && ((ce->ce_flags & CE_VALID) || ce_skip_worktree(ce)))
Junio C Hamano's avatar
Junio C Hamano committed
269 270
		return 0;

271 272 273 274 275 276 277 278
	/*
	 * 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
279
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
280

Junio C Hamano's avatar
Junio C Hamano committed
281 282 283 284 285 286 287 288 289 290 291 292 293 294
	/*
	 * 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
295
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
296
	 */
297
	if (!changed && is_racy_timestamp(istate, ce)) {
298 299 300 301 302
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
303

Junio C Hamano's avatar
Junio C Hamano committed
304
	return changed;
305 306
}

307
int ie_modified(const struct index_state *istate,
308
		struct cache_entry *ce, struct stat *st, unsigned int options)
309
{
Junio C Hamano's avatar
Junio C Hamano committed
310
	int changed, changed_fs;
311 312

	changed = ie_match_stat(istate, ce, st, options);
313 314 315 316 317 318 319 320 321
	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;

322 323 324 325 326 327 328 329 330 331 332 333 334
	/*
	 * 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.
335
	 */
336 337
	if ((changed & DATA_CHANGED) &&
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_size != 0))
338 339
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
340 341 342
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
343 344 345
	return 0;
}

346 347 348 349 350 351 352 353 354 355 356 357
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];
358
	if (!c1 && S_ISDIR(mode1))
359
		c1 = '/';
360
	if (!c2 && S_ISDIR(mode2))
361 362 363 364
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399
/*
 * 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;
}

400
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
401
{
402 403
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
404 405 406 407 408 409 410 411 412 413
	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
414

415 416 417
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
418

419 420 421 422
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
423 424 425
	return 0;
}

426
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
427 428 429 430
{
	int first, last;

	first = 0;
431
	last = istate->cache_nr;
432 433
	while (last > first) {
		int next = (last + first) >> 1;
434
		struct cache_entry *ce = istate->cache[next];
435
		int cmp = cache_name_compare(name, namelen, ce->name, ce->ce_flags);
436
		if (!cmp)
437
			return next;
438 439 440 441 442 443
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
444
	return -first-1;
445 446
}

447
/* Remove entry, return true if there are more entries to go.. */
448
int remove_index_entry_at(struct index_state *istate, int pos)
449
{
450 451
	struct cache_entry *ce = istate->cache[pos];

452
	remove_name_hash(ce);
453 454 455
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
456
		return 0;
457 458 459
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
460 461 462
	return 1;
}

463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
/*
 * 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;
}

483
int remove_file_from_index(struct index_state *istate, const char *path)
484
{
485
	int pos = index_name_pos(istate, path, strlen(path));
486 487
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
488
	cache_tree_invalidate_path(istate->cache_tree, path);
489 490
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
491 492 493
	return 0;
}

494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
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;
}

522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553
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;
}

554 555 556 557 558 559 560 561
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);
}

562
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
563
{
564
	int size, namelen, was_same;
565
	mode_t st_mode = st->st_mode;
566
	struct cache_entry *ce, *alias;
567
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_RACY_IS_DIRTY;
568 569
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
570 571 572
	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
573

574
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
575
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
576 577

	namelen = strlen(path);
578
	if (S_ISDIR(st_mode)) {
579 580 581
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
582 583 584
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
585
	ce->ce_flags = namelen;
586 587
	if (!intent_only)
		fill_stat_cache_info(ce, st);
588 589
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
590

591
	if (trust_executable_bit && has_symlinks)
592
		ce->ce_mode = create_ce_mode(st_mode);
593
	else {
594 595
		/* 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
596
		 */
597
		struct cache_entry *ent;
598
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
599

600
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
601
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
602 603
	}

604
	alias = index_name_exists(istate, ce->name, ce_namelen(ce), ignore_case);
605
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
606 607
		/* Nothing changed, really */
		free(ce);
608
		ce_mark_uptodate(alias);
609
		alias->ce_flags |= CE_ADDED;
610 611
		return 0;
	}
612 613 614 615 616 617
	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);

618 619 620
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
621

Junio C Hamano's avatar
Junio C Hamano committed
622
	/* It was suspected to be racily clean, but it turns out to be Ok */
623 624 625 626 627 628 629
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
630
	else if (add_index_entry(istate, ce, add_option))
631
		return error("unable to add %s to index",path);
632
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
633 634 635 636
		printf("add '%s'\n", path);
	return 0;
}

637
int add_file_to_index(struct index_state *istate, const char *path, int flags)
638 639 640
{
	struct stat st;
	if (lstat(path, &st))
641
		die_errno("unable to stat '%s'", path);
642
	return add_to_index(istate, path, &st, flags);
643 644
}

645 646 647 648 649 650 651
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;

652 653
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
654
		return NULL;
655
	}
656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671

	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;
}

672
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
673 674 675 676 677
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697
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;
698 699
		if (!matchlen)
			return 1;
700 701 702 703
	}
	return 0;
}

704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
/*
 * 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++;
	}
}

768 769 770
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
771
 */
772 773
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
774
{
775 776
	int retval = 0;
	int len = ce_namelen(ce);
777
	int stage = ce_stage(ce);
778
	const char *name = ce->name;
779

780 781
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
782

783
		if (len >= ce_namelen(p))
784
			break;
785 786
		if (memcmp(name, p->name, len))
			break;
787 788
		if (ce_stage(p) != stage)
			continue;
789 790
		if (p->name[len] != '/')
			continue;
791
		if (p->ce_flags & CE_REMOVE)
792
			continue;
793 794 795
		retval = -1;
		if (!ok_to_replace)
			break;
796
		remove_index_entry_at(istate, --pos);
797
	}
798 799
	return retval;
}
800

801 802 803 804
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
805 806
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
807 808
{
	int retval = 0;
809
	int stage = ce_stage(ce);
810 811
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
812

813 814
	for (;;) {
		int len;
815

816 817 818 819 820 821 822
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
823

824
		pos = index_name_pos(istate, name, create_ce_flags(len, stage));
825
		if (pos >= 0) {
826 827 828 829 830 831 832 833
			/*
			 * 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.
			 */
834
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
835 836 837
				retval = -1;
				if (!ok_to_replace)
					break;
838
				remove_index_entry_at(istate, pos);
839 840
				continue;
			}
841
		}
842 843
		else
			pos = -pos-1;
844 845 846 847

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
848
		 * we're ok, and we can exit.
849
		 */
850 851
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
852 853 854 855
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
856 857 858
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
859 860 861 862 863 864
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
865
		}
866
	}
867 868 869 870 871 872 873
	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
874
 *
875 876 877 878
 * 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.
 */
879 880 881
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
882
{
883 884 885 886 887
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
888
	if (ce->ce_flags & CE_REMOVE)
889 890
		return 0;

891 892 893
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
894
	 * in the array.
895
	 */
896
	retval = has_file_name(istate, ce, pos, ok_to_replace);
897

898 899 900 901
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
902
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
903 904
}

905
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
906 907
{
	int pos;
908 909
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
910
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
911
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
912

Junio C Hamano's avatar
Junio C Hamano committed
913
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
914
	pos = index_name_pos(istate, ce->name, ce->ce_flags);
915

Junio C Hamano's avatar
Junio C Hamano committed
916
	/* existing match? Just replace it. */
917
	if (pos >= 0) {
918 919
		if (!new_only)
			replace_index_entry(istate, pos, ce);
920 921
		return 0;
	}
922
	pos = -pos-1;
923

924 925 926 927
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
928 929
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
930
			ok_to_add = 1;
931
			if (!remove_index_entry_at(istate, pos))
932 933 934 935
				break;
		}
	}

936 937
	if (!ok_to_add)
		return -1;
938
	if (!verify_path(ce->name))
939
		return error("Invalid path '%s'", ce->name);
940

Junio C Hamano's avatar
Junio C Hamano committed
941
	if (!skip_df_check &&
942
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
943
		if (!ok_to_replace)
944 945
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
946
		pos = index_name_pos(istate, ce->name, ce->ce_flags);
947 948
		pos = -pos-1;
	}
949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964
	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;
	}
965

966
	/* Make sure the array is big enough .. */
967 968 969 970
	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 *));
971 972 973
	}

	/* Add it in.. */
974
	istate->cache_nr++;
975
	if (istate->cache_nr > pos + 1)
976 977 978
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
979
	set_index_entry(istate, pos, ce);
980
	istate->cache_changed = 1;
981 982 983
	return 0;
}

984 985 986 987 988 989 990 991 992 993 994
/*
 * "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.
 */
995
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
996 997
					     struct cache_entry *ce,
					     unsigned int options, int *err)
998 999 1000 1001
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1002
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1003

1004 1005 1006
	if (ce_uptodate(ce))
		return ce;

1007
	if (!ignore_valid && ((ce->ce_flags & CE_VALID) || ce_skip_worktree(ce))) {
1008 1009 1010 1011
		ce_mark_uptodate(ce);
		return ce;
	}

1012
	if (lstat(ce->name, &st) < 0) {
1013 1014
		if (err)
			*err = errno;
1015 1016
		return NULL;
	}
1017

1018
	changed = ie_match_stat(istate, ce, &st, options);
1019
	if (!changed) {
1020 1021 1022 1023 1024 1025 1026 1027
		/*
		 * 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 &&
1028
		    !(ce->ce_flags & CE_VALID))
1029
			; /* mark this one VALID again */
1030 1031 1032 1033 1034 1035 1036
		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);
1037
			return ce;
1038
		}
1039 1040
	}

1041
	if (ie_modified(istate, ce, &st, options)) {
1042 1043
		if (err)
			*err = EINVAL;
1044 1045
		return NULL;
	}
1046 1047 1048 1049 1050

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1051 1052 1053 1054 1055
	/*
	 * 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.
1056
	 */
1057
	if (!ignore_valid && assume_unchanged &&
1058 1059
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1060 1061 1062 1063

	return updated;
}

1064
int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec, char *seen)
1065 1066 1067 1068 1069 1070 1071
{
	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;
1072
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1073
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
1074
	const char *needs_update_message;
1075

1076 1077
	needs_update_message = ((flags & REFRESH_SAY_CHANGED)
				? "locally modified" : "needs update");
1078
	for (i = 0; i < istate->cache_nr; i++) {
1079
		struct cache_entry *ce, *new;
1080 1081
		int cache_errno = 0;

1082
		ce = istate->cache[i];
1083 1084 1085
		if (ignore_submodules && S_ISGITLINK(ce->ce_mode))
			continue;

1086
		if (ce_stage(ce)) {
1087 1088
			while ((i < istate->cache_nr) &&
			       ! strcmp(istate->cache[i]->name, ce->name))
1089 1090 1091 1092 1093 1094 1095 1096 1097
				i++;
			i--;
			if (allow_unmerged)
				continue;
			printf("%s: needs merge\n", ce->name);
			has_errors = 1;
			continue;
		}

1098 1099 1100
		if (pathspec && !match_pathspec(pathspec, ce->name, strlen(ce->name), 0, seen))
			continue;

1101
		new = refresh_cache_ent(istate, ce, options, &cache_errno);
1102
		if (new == ce)
1103
			continue;
1104 1105
		if (!new) {
			if (not_new && cache_errno == ENOENT)
1106
				continue;
1107
			if (really && cache_errno == EINVAL) {
1108 1109 1110
				/* If we are doing --really-refresh that
				 * means the index is not valid anymore.
				 */
1111
				ce->ce_flags &= ~CE_VALID;
1112
				istate->cache_changed = 1;
1113 1114 1115
			}
			if (quiet)
				continue;
1116
			printf("%s: %s\n", ce->name, needs_update_message);
1117 1118 1119
			has_errors = 1;
			continue;
		}
1120 1121

		replace_index_entry(istate, i, new);
1122 1123 1124 1125
	}
	return has_errors;
}

1126 1127
struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really)
{
1128
	return refresh_cache_ent(&the_index, ce, really, NULL);
1129 1130
}

1131
static int verify_hdr(struct cache_header *hdr, unsigned long size)
1132
{
1133
	git_SHA_CTX c;
1134
	unsigned char sha1[20];