read-cache.c 54.5 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,
					       unsigned int options);
20

21 22 23 24
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

25 26 27 28 29 30 31 32 33 34 35
/* 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" */
36
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
37

38
struct index_state the_index;
39
static const char *alternate_index_output;
40

Junio C Hamano's avatar
Junio C Hamano committed
41 42 43
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
44
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
45 46
}

47 48 49 50
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

51
	remove_name_hash(istate, old);
52
	free(old);
53
	set_index_entry(istate, nr, ce);
54 55 56
	istate->cache_changed = 1;
}

57 58 59 60 61 62 63
void rename_index_entry_at(struct index_state *istate, int nr, const char *new_name)
{
	struct cache_entry *old = istate->cache[nr], *new;
	int namelen = strlen(new_name);

	new = xmalloc(cache_entry_size(namelen));
	copy_cache_entry(new, old);
64
	new->ce_flags &= ~CE_HASHED;
65
	new->ce_namelen = namelen;
66 67 68 69 70 71 72
	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);
}

73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
void fill_stat_data(struct stat_data *sd, struct stat *st)
{
	sd->sd_ctime.sec = (unsigned int)st->st_ctime;
	sd->sd_mtime.sec = (unsigned int)st->st_mtime;
	sd->sd_ctime.nsec = ST_CTIME_NSEC(*st);
	sd->sd_mtime.nsec = ST_MTIME_NSEC(*st);
	sd->sd_dev = st->st_dev;
	sd->sd_ino = st->st_ino;
	sd->sd_uid = st->st_uid;
	sd->sd_gid = st->st_gid;
	sd->sd_size = st->st_size;
}

int match_stat_data(const struct stat_data *sd, struct stat *st)
{
	int changed = 0;

	if (sd->sd_mtime.sec != (unsigned int)st->st_mtime)
		changed |= MTIME_CHANGED;
	if (trust_ctime && check_stat &&
	    sd->sd_ctime.sec != (unsigned int)st->st_ctime)
		changed |= CTIME_CHANGED;

#ifdef USE_NSEC
	if (check_stat && sd->sd_mtime.nsec != ST_MTIME_NSEC(*st))
		changed |= MTIME_CHANGED;
	if (trust_ctime && check_stat &&
	    sd->sd_ctime.nsec != ST_CTIME_NSEC(*st))
		changed |= CTIME_CHANGED;
#endif

	if (check_stat) {
		if (sd->sd_uid != (unsigned int) st->st_uid ||
			sd->sd_gid != (unsigned int) st->st_gid)
			changed |= OWNER_CHANGED;
		if (sd->sd_ino != (unsigned int) st->st_ino)
			changed |= INODE_CHANGED;
	}

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
	if (check_stat && sd->sd_dev != (unsigned int) st->st_dev)
			changed |= INODE_CHANGED;
#endif

	if (sd->sd_size != (unsigned int) st->st_size)
		changed |= DATA_CHANGED;

	return changed;
}

128 129 130 131 132 133 134
/*
 * 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)
{
135
	fill_stat_data(&ce->ce_stat_data, st);
Junio C Hamano's avatar
Junio C Hamano committed
136 137

	if (assume_unchanged)
138
		ce->ce_flags |= CE_VALID;
139 140 141

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
142 143
}

144
static int ce_compare_data(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
145 146 147 148 149 150
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
151
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
152
			match = hashcmp(sha1, ce->sha1);
153
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
154 155 156 157
	}
	return match;
}

158
static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
159 160 161 162
{
	int match = -1;
	void *buffer;
	unsigned long size;
163
	enum object_type type;
164
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
165

166
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
167
		return -1;
168

169
	buffer = read_sha1_file(ce->sha1, &type, &size);
170 171 172 173
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
174
	}
175
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
176 177 178
	return match;
}

179
static int ce_compare_gitlink(const struct cache_entry *ce)
180 181 182 183 184
{
	unsigned char sha1[20];

	/*
	 * We don't actually require that the .git directory
Martin Waitz's avatar
Martin Waitz committed
185
	 * under GITLINK directory be a valid git directory. It
186 187 188 189 190 191 192 193 194 195
	 * 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);
}

196
static int ce_modified_check_fs(const struct cache_entry *ce, struct stat *st)
Junio C Hamano's avatar
Junio C Hamano committed
197 198 199 200 201 202 203
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
204
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
205 206
			return DATA_CHANGED;
		break;
207
	case S_IFDIR:
208
		if (S_ISGITLINK(ce->ce_mode))
209
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
210 211 212 213 214 215
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

216
static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)
217 218 219
{
	unsigned int changed = 0;

220 221 222 223
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
224 225
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
226 227 228 229
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
230
		    (0100 & (ce->ce_mode ^ st->st_mode)))
231
			changed |= MODE_CHANGED;
232 233
		break;
	case S_IFLNK:
234 235 236
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
237
		break;
Martin Waitz's avatar
Martin Waitz committed
238
	case S_IFGITLINK:
239
		/* We ignore most of the st_xxx fields for gitlinks */
240 241 242 243
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
244
		return changed;
245
	default:
246
		die("internal error: ce_mode is %o", ce->ce_mode);
247
	}
248

249
	changed |= match_stat_data(&ce->ce_stat_data, st);
250

251
	/* Racily smudged entry? */
252
	if (!ce->ce_stat_data.sd_size) {
253 254 255 256
		if (!is_empty_blob_sha1(ce->sha1))
			changed |= DATA_CHANGED;
	}

Junio C Hamano's avatar
Junio C Hamano committed
257 258 259
	return changed;
}

260 261
static int is_racy_timestamp(const struct index_state *istate,
			     const struct cache_entry *ce)
262
{
263
	return (!S_ISGITLINK(ce->ce_mode) &&
264 265 266
		istate->timestamp.sec &&
#ifdef USE_NSEC
		 /* nanosecond timestamped files can also be racy! */
267 268 269
		(istate->timestamp.sec < ce->ce_stat_data.sd_mtime.sec ||
		 (istate->timestamp.sec == ce->ce_stat_data.sd_mtime.sec &&
		  istate->timestamp.nsec <= ce->ce_stat_data.sd_mtime.nsec))
270
#else
271
		istate->timestamp.sec <= ce->ce_stat_data.sd_mtime.sec
272 273
#endif
		 );
274 275
}

276
int ie_match_stat(const struct index_state *istate,
277
		  const struct cache_entry *ce, struct stat *st,
278
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
279
{
Junio C Hamano's avatar
Junio C Hamano committed
280
	unsigned int changed;
281
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
282
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
283
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
284 285 286 287

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
288 289
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
290
	 */
291 292
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
293
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
294 295
		return 0;

296 297 298 299 300 301 302 303
	/*
	 * 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
304
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
305

Junio C Hamano's avatar
Junio C Hamano committed
306 307 308 309 310 311 312 313 314 315 316 317 318 319
	/*
	 * 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
320
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
321
	 */
322
	if (!changed && is_racy_timestamp(istate, ce)) {
323 324 325 326 327
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
328

Junio C Hamano's avatar
Junio C Hamano committed
329
	return changed;
330 331
}

332
int ie_modified(const struct index_state *istate,
333 334
		const struct cache_entry *ce,
		struct stat *st, unsigned int options)
335
{
Junio C Hamano's avatar
Junio C Hamano committed
336
	int changed, changed_fs;
337 338

	changed = ie_match_stat(istate, ce, st, options);
339 340 341 342 343 344 345 346 347
	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;

348 349 350 351 352 353 354 355 356 357 358 359 360
	/*
	 * 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.
361
	 */
362
	if ((changed & DATA_CHANGED) &&
363
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_stat_data.sd_size != 0))
364 365
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
366 367 368
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
369 370 371
	return 0;
}

372 373 374 375 376 377 378 379 380 381 382 383
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];
384
	if (!c1 && S_ISDIR(mode1))
385
		c1 = '/';
386
	if (!c2 && S_ISDIR(mode2))
387 388 389 390
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425
/*
 * 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;
}

426
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
427
{
428 429
	int len = len1 < len2 ? len1 : len2;
	int cmp;
430 431 432 433 434 435 436 437

	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
438

439
	if (stage1 < stage2)
440
		return -1;
441
	if (stage1 > stage2)
442
		return 1;
443 444 445
	return 0;
}

446 447 448 449 450
int cache_name_compare(const char *name1, int len1, const char *name2, int len2)
{
	return cache_name_stage_compare(name1, len1, 0, name2, len2, 0);
}

451
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
452 453 454 455
{
	int first, last;

	first = 0;
456
	last = istate->cache_nr;
457 458
	while (last > first) {
		int next = (last + first) >> 1;
459
		struct cache_entry *ce = istate->cache[next];
460
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
461
		if (!cmp)
462
			return next;
463 464 465 466 467 468
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
469
	return -first-1;
470 471
}

472 473 474 475 476
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

477
/* Remove entry, return true if there are more entries to go.. */
478
int remove_index_entry_at(struct index_state *istate, int pos)
479
{
480 481
	struct cache_entry *ce = istate->cache[pos];

482
	record_resolve_undo(istate, ce);
483
	remove_name_hash(istate, ce);
484
	free(ce);
485 486 487
	istate->cache_changed = 1;
	istate->cache_nr--;
	if (pos >= istate->cache_nr)
488
		return 0;
489 490 491
	memmove(istate->cache + pos,
		istate->cache + pos + 1,
		(istate->cache_nr - pos) * sizeof(struct cache_entry *));
492 493 494
	return 1;
}

495
/*
Ondřej Bílka's avatar
Ondřej Bílka committed
496
 * Remove all cache entries marked for removal, that is where
497 498 499 500 501 502 503 504 505
 * 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++) {
506
		if (ce_array[i]->ce_flags & CE_REMOVE) {
507
			remove_name_hash(istate, ce_array[i]);
508 509
			free(ce_array[i]);
		}
510 511 512
		else
			ce_array[j++] = ce_array[i];
	}
513 514
	if (j == istate->cache_nr)
		return;
515 516 517 518
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

519
int remove_file_from_index(struct index_state *istate, const char *path)
520
{
521
	int pos = index_name_pos(istate, path, strlen(path));
522 523
	if (pos < 0)
		pos = -pos-1;
Junio C Hamano's avatar
Junio C Hamano committed
524
	cache_tree_invalidate_path(istate->cache_tree, path);
525 526
	while (pos < istate->cache_nr && !strcmp(istate->cache[pos]->name, path))
		remove_index_entry_at(istate, pos);
527 528 529
	return 0;
}

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 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;
}

558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589
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;
}

590
void set_object_name_for_intent_to_add_entry(struct cache_entry *ce)
591 592 593 594 595 596 597
{
	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);
}

598
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
599
{
600
	int size, namelen, was_same;
601
	mode_t st_mode = st->st_mode;
602
	struct cache_entry *ce, *alias;
603
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
604 605
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
606 607 608
	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
609

610
	if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))
611
		return error("%s: can only add regular files, symbolic links or git-directories", path);
Johannes Schindelin's avatar
Johannes Schindelin committed
612 613

	namelen = strlen(path);
614
	if (S_ISDIR(st_mode)) {
615 616 617
		while (namelen && path[namelen-1] == '/')
			namelen--;
	}
Johannes Schindelin's avatar
Johannes Schindelin committed
618 619 620
	size = cache_entry_size(namelen);
	ce = xcalloc(1, size);
	memcpy(ce->name, path, namelen);
621
	ce->ce_namelen = namelen;
622 623
	if (!intent_only)
		fill_stat_cache_info(ce, st);
624 625
	else
		ce->ce_flags |= CE_INTENT_TO_ADD;
Johannes Schindelin's avatar
Johannes Schindelin committed
626

627
	if (trust_executable_bit && has_symlinks)
628
		ce->ce_mode = create_ce_mode(st_mode);
629
	else {
630 631
		/* 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
632
		 */
633
		struct cache_entry *ent;
634
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
635

636
		ent = (0 <= pos) ? istate->cache[pos] : NULL;
637
		ce->ce_mode = ce_mode_from_stat(ent, st_mode);
Johannes Schindelin's avatar
Johannes Schindelin committed
638 639
	}

640 641 642 643 644 645 646 647 648 649 650 651 652 653
	/* 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;
654
				foundce = index_dir_exists(istate, ce->name, ptr - ce->name - 1);
655 656 657 658 659 660 661 662
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

663
	alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);
664
	if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {
665 666
		/* Nothing changed, really */
		free(ce);
667 668
		if (!S_ISGITLINK(alias->ce_mode))
			ce_mark_uptodate(alias);
669
		alias->ce_flags |= CE_ADDED;
670 671
		return 0;
	}
672
	if (!intent_only) {
673
		if (index_path(ce->sha1, path, st, HASH_WRITE_OBJECT))
674 675
			return error("unable to index file %s", path);
	} else
676
		set_object_name_for_intent_to_add_entry(ce);
677

678 679 680
	if (ignore_case && alias && different_name(ce, alias))
		ce = create_alias_ce(ce, alias);
	ce->ce_flags |= CE_ADDED;
681

Junio C Hamano's avatar
Junio C Hamano committed
682
	/* It was suspected to be racily clean, but it turns out to be Ok */
683 684 685 686 687 688 689
	was_same = (alias &&
		    !ce_stage(alias) &&
		    !hashcmp(alias->sha1, ce->sha1) &&
		    ce->ce_mode == alias->ce_mode);

	if (pretend)
		;
690
	else if (add_index_entry(istate, ce, add_option))
691
		return error("unable to add %s to index",path);
692
	if (verbose && !was_same)
Johannes Schindelin's avatar
Johannes Schindelin committed
693 694 695 696
		printf("add '%s'\n", path);
	return 0;
}

697
int add_file_to_index(struct index_state *istate, const char *path, int flags)
698 699 700
{
	struct stat st;
	if (lstat(path, &st))
701
		die_errno("unable to stat '%s'", path);
702
	return add_to_index(istate, path, &st, flags);
703 704
}

705 706
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
707
		unsigned int refresh_options)
708 709 710 711
{
	int size, len;
	struct cache_entry *ce;

712 713
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
714
		return NULL;
715
	}
716 717 718 719 720 721 722

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

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
723 724
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
725 726
	ce->ce_mode = create_ce_mode(mode);

727
	return refresh_cache_entry(ce, refresh_options);
728 729
}

730
int ce_same_name(const struct cache_entry *a, const struct cache_entry *b)
731 732 733 734 735
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750
/*
 * We fundamentally don't like some paths: we don't want
 * dot or dot-dot anywhere, and for obvious reasons don't
 * want to recurse into ".git" either.
 *
 * Also, we don't want double slashes or slashes at the
 * end that can make pathnames ambiguous.
 */
static int verify_dotfile(const char *rest)
{
	/*
	 * The first character was '.', but that
	 * has already been discarded, we now test
	 * the rest.
	 */
751

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

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

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

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

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

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

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

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

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

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

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

853
		pos = index_name_stage_pos(istate, name, len, stage);
854
		if (pos >= 0) {
855 856 857 858 859 860 861 862
			/*
			 * Found one, but not so fast.  This could
			 * be a marker that says "I was here, but
			 * I am being removed".  Such an entry is
			 * not a part of the resulting tree, and
			 * it is Ok to have a directory at the same
			 * path.
			 */
863
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
864 865 866
				retval = -1;
				if (!ok_to_replace)
					break;
867
				remove_index_entry_at(istate, pos);
868 869
				continue;
			}
870
		}
871 872
		else
			pos = -pos-1;
873 874 875 876

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

/* We may be in a situation where we already have path/file and path
 * is being added, or we already have path and path/file is being
 * added.  Either one would result in a nonsense tree that has path
 * twice when git-write-tree tries to write it out.  Prevent it.
Junio C Hamano's avatar
Junio C Hamano committed
903
 *
904 905 906 907
 * If ok-to-replace is specified, we remove the conflicting entries
 * from the cache so the caller should recompute the insert position.
 * When this happens, we return non-zero.
 */
908 909 910
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
911
{
912 913 914 915 916
	int retval;

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

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

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

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

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

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

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

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

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

int add_index_entry(struct index_state *istate, struct cache_entry *ce, int option)
{
	int pos;

	if (option & ADD_CACHE_JUST_APPEND)
		pos = istate->cache_nr;
	else {
		int ret;
		ret = add_index_entry_with_check(istate, ce, option);
		if (ret <= 0)
			return ret;
		pos = ret - 1;
	}
994

995
	/* Make sure the array is big enough .. */
996
	ALLOC_GROW(istate->cache, istate->cache_nr + 1, istate->cache_alloc);
997 998

	/* Add it in.. */
999
	istate->cache_nr++;
1000
	if (istate->cache_nr > pos + 1)
1001 1002 1003
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
1004
	set_index_entry(istate, pos, ce);
1005
	istate->cache_changed = 1;
1006 1007 1008
	return 0;
}

1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
/*
 * "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.
 */
1020
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1021
					     struct cache_entry *ce,
1022 1023
					     unsigned int options, int *err,
					     int *changed_ret)
1024 1025 1026 1027
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1028
	int refresh = options & CE_MATCH_REFRESH;
1029
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1030
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1031
	int ignore_missing = options & CE_MATCH_IGNORE_MISSING;
1032

1033
	if (!refresh || ce_uptodate(ce))
1034 1035
		return ce;

1036
	/*
1037 1038 1039
	 * 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.
1040
	 */
1041 1042 1043 1044
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1045 1046 1047 1048 1049
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1050
	if (lstat(ce->name, &st) < 0) {
1051 1052
		if (ignore_missing && errno == ENOENT)
			return ce;
1053 1054
		if (err)
			*err = errno;
1055 1056
		return NULL;
	}
1057

1058
	changed = ie_match_stat(istate, ce, &st, options);
1059 1060
	if (changed_ret)
		*changed_ret = changed;
1061
	if (!changed) {
1062 1063 1064 1065 1066 1067 1068 1069
		/*
		 * The path is unchanged.  If we were told to ignore
		 * valid bit, then we did the actual stat check and
		 * found that the entry is unmodified.  If the entry
		 * is not marked VALID, this is the place to mark it
		 * valid again, under "assume unchanged" mode.
		 */
		if (ignore_valid && assume_unchanged &&
1070
		    !(ce->ce_flags & CE_VALID))
1071
			; /* mark this one VALID again */
1072 1073 1074 1075 1076 1077
		else {
			/*
			 * We do not mark the index itself "modified"
			 * because CE_UPTODATE flag is in-core only;
			 * we are not going to write this change out.
			 */
1078 1079
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1080
			return ce;
1081
		}
1082 1083
	}

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

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1094 1095 1096 1097 1098
	/*
	 * If ignore_valid is not set, we should leave CE_VALID bit
	 * alone.  Otherwise, paths marked with --no-assume-unchanged
	 * (i.e. things to be edited) will reacquire CE_VALID bit
	 * automatically, which is not really what we want.
1099
	 */
1100
	if (!ignore_valid && assume_unchanged &&
1101 1102
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1103 1104 1105 1106

	return updated;
}

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

1117 1118
int refresh_index(struct index_state *istate, unsigned int