read-cache.c 50.7 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5
/*
 * GIT - The information manager from hell
 *
 * Copyright (C) Linus Torvalds, 2005
 */
6
#define NO_THE_INDEX_COMPATIBILITY_MACROS
7
#include "cache.h"
8
#include "cache-tree.h"
9
#include "refs.h"
10
#include "dir.h"
11 12
#include "tree.h"
#include "commit.h"
13
#include "blob.h"
14
#include "resolve-undo.h"
15 16
#include "strbuf.h"
#include "varint.h"
17

18 19
static struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really);

20 21 22 23
/* Mask for the name length in ce_flags in the on-disk index */

#define CE_NAMEMASK  (0x0fff)

24 25 26 27 28 29 30 31 32 33 34
/* 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" */
35
#define CACHE_EXT_RESOLVE_UNDO 0x52455543 /* "REUC" */
36

37
struct index_state the_index;
38

Junio C Hamano's avatar
Junio C Hamano committed
39 40 41
static void set_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	istate->cache[nr] = ce;
42
	add_name_hash(istate, ce);
Junio C Hamano's avatar
Junio C Hamano committed
43 44
}

45 46 47 48
static void replace_index_entry(struct index_state *istate, int nr, struct cache_entry *ce)
{
	struct cache_entry *old = istate->cache[nr];

49
	remove_name_hash(old);
50
	set_index_entry(istate, nr, ce);
51 52 53
	istate->cache_changed = 1;
}

54 55 56 57 58 59 60
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);
61 62
	new->ce_flags &= ~CE_STATE_MASK;
	new->ce_namelen = namelen;
63 64 65 66 67 68 69
	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);
}

70 71 72 73 74 75 76
/*
 * 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)
{
77 78
	ce->ce_ctime.sec = (unsigned int)st->st_ctime;
	ce->ce_mtime.sec = (unsigned int)st->st_mtime;
79 80
	ce->ce_ctime.nsec = ST_CTIME_NSEC(*st);
	ce->ce_mtime.nsec = ST_MTIME_NSEC(*st);
81 82 83 84 85
	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
86 87

	if (assume_unchanged)
88
		ce->ce_flags |= CE_VALID;
89 90 91

	if (S_ISREG(st->st_mode))
		ce_mark_uptodate(ce);
92 93
}

Junio C Hamano's avatar
Junio C Hamano committed
94 95 96 97 98 99 100
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];
101
		if (!index_fd(sha1, fd, st, OBJ_BLOB, ce->name, 0))
102
			match = hashcmp(sha1, ce->sha1);
103
		/* index_fd() closed the file descriptor already */
Junio C Hamano's avatar
Junio C Hamano committed
104 105 106 107
	}
	return match;
}

108
static int ce_compare_link(struct cache_entry *ce, size_t expected_size)
Junio C Hamano's avatar
Junio C Hamano committed
109 110 111 112
{
	int match = -1;
	void *buffer;
	unsigned long size;
113
	enum object_type type;
114
	struct strbuf sb = STRBUF_INIT;
Junio C Hamano's avatar
Junio C Hamano committed
115

116
	if (strbuf_readlink(&sb, ce->name, expected_size))
Junio C Hamano's avatar
Junio C Hamano committed
117
		return -1;
118

119
	buffer = read_sha1_file(ce->sha1, &type, &size);
120 121 122 123
	if (buffer) {
		if (size == sb.len)
			match = memcmp(buffer, sb.buf, size);
		free(buffer);
Junio C Hamano's avatar
Junio C Hamano committed
124
	}
125
	strbuf_release(&sb);
Junio C Hamano's avatar
Junio C Hamano committed
126 127 128
	return match;
}

129 130 131 132 133 134
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
135
	 * under GITLINK directory be a valid git directory. It
136 137 138 139 140 141 142 143 144 145
	 * 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
146 147 148 149 150 151 152 153
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:
154
		if (ce_compare_link(ce, xsize_t(st->st_size)))
Junio C Hamano's avatar
Junio C Hamano committed
155 156
			return DATA_CHANGED;
		break;
157
	case S_IFDIR:
158
		if (S_ISGITLINK(ce->ce_mode))
159
			return ce_compare_gitlink(ce) ? DATA_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
160 161 162 163 164 165
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
166
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
167 168 169
{
	unsigned int changed = 0;

170 171 172 173
	if (ce->ce_flags & CE_REMOVE)
		return MODE_CHANGED | DATA_CHANGED | TYPE_CHANGED;

	switch (ce->ce_mode & S_IFMT) {
174 175
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
176 177 178 179
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
180
		    (0100 & (ce->ce_mode ^ st->st_mode)))
181
			changed |= MODE_CHANGED;
182 183
		break;
	case S_IFLNK:
184 185 186
		if (!S_ISLNK(st->st_mode) &&
		    (has_symlinks || !S_ISREG(st->st_mode)))
			changed |= TYPE_CHANGED;
187
		break;
Martin Waitz's avatar
Martin Waitz committed
188
	case S_IFGITLINK:
189
		/* We ignore most of the st_xxx fields for gitlinks */
190 191 192 193
		if (!S_ISDIR(st->st_mode))
			changed |= TYPE_CHANGED;
		else if (ce_compare_gitlink(ce))
			changed |= DATA_CHANGED;
194
		return changed;
195
	default:
196
		die("internal error: ce_mode is %o", ce->ce_mode);
197
	}
198
	if (ce->ce_mtime.sec != (unsigned int)st->st_mtime)
199
		changed |= MTIME_CHANGED;
200
	if (trust_ctime && ce->ce_ctime.sec != (unsigned int)st->st_ctime)
201
		changed |= CTIME_CHANGED;
202

203
#ifdef USE_NSEC
204
	if (ce->ce_mtime.nsec != ST_MTIME_NSEC(*st))
205
		changed |= MTIME_CHANGED;
206
	if (trust_ctime && ce->ce_ctime.nsec != ST_CTIME_NSEC(*st))
207 208 209
		changed |= CTIME_CHANGED;
#endif

210 211
	if (ce->ce_uid != (unsigned int) st->st_uid ||
	    ce->ce_gid != (unsigned int) st->st_gid)
212
		changed |= OWNER_CHANGED;
213
	if (ce->ce_ino != (unsigned int) st->st_ino)
214
		changed |= INODE_CHANGED;
215 216 217 218 219 220 221

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
222
	if (ce->ce_dev != (unsigned int) st->st_dev)
223 224 225
		changed |= INODE_CHANGED;
#endif

226
	if (ce->ce_size != (unsigned int) st->st_size)
227
		changed |= DATA_CHANGED;
228

229 230 231 232 233 234
	/* 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
235 236 237
	return changed;
}

238
static int is_racy_timestamp(const struct index_state *istate, struct cache_entry *ce)
239
{
240
	return (!S_ISGITLINK(ce->ce_mode) &&
241 242 243 244 245 246 247 248 249 250
		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
		 );
251 252
}

253
int ie_match_stat(const struct index_state *istate,
254 255
		  struct cache_entry *ce, struct stat *st,
		  unsigned int options)
Junio C Hamano's avatar
Junio C Hamano committed
256
{
Junio C Hamano's avatar
Junio C Hamano committed
257
	unsigned int changed;
258
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
259
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
260
	int assume_racy_is_modified = options & CE_MATCH_RACY_IS_DIRTY;
Junio C Hamano's avatar
Junio C Hamano committed
261 262 263 264

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
265 266
	 *
	 * skip-worktree has the same effect with higher precedence
Junio C Hamano's avatar
Junio C Hamano committed
267
	 */
268 269
	if (!ignore_skip_worktree && ce_skip_worktree(ce))
		return 0;
270
	if (!ignore_valid && (ce->ce_flags & CE_VALID))
Junio C Hamano's avatar
Junio C Hamano committed
271 272
		return 0;

273 274 275 276 277 278 279 280
	/*
	 * 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
281
	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
282

Junio C Hamano's avatar
Junio C Hamano committed
283 284 285 286 287 288 289 290 291 292 293 294 295 296
	/*
	 * 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
297
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
298
	 */
299
	if (!changed && is_racy_timestamp(istate, ce)) {
300 301 302 303 304
		if (assume_racy_is_modified)
			changed |= DATA_CHANGED;
		else
			changed |= ce_modified_check_fs(ce, st);
	}
305

Junio C Hamano's avatar
Junio C Hamano committed
306
	return changed;
307 308
}

309
int ie_modified(const struct index_state *istate,
310
		struct cache_entry *ce, struct stat *st, unsigned int options)
311
{
Junio C Hamano's avatar
Junio C Hamano committed
312
	int changed, changed_fs;
313 314

	changed = ie_match_stat(istate, ce, st, options);
315 316 317 318 319 320 321 322 323
	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;

324 325 326 327 328 329 330 331 332 333 334 335 336
	/*
	 * 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.
337
	 */
338 339
	if ((changed & DATA_CHANGED) &&
	    (S_ISGITLINK(ce->ce_mode) || ce->ce_size != 0))
340 341
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
342 343 344
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
345 346 347
	return 0;
}

348 349 350 351 352 353 354 355 356 357 358 359
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];
360
	if (!c1 && S_ISDIR(mode1))
361
		c1 = '/';
362
	if (!c2 && S_ISDIR(mode2))
363 364 365 366
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
/*
 * 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;
}

402
int cache_name_stage_compare(const char *name1, int len1, int stage1, const char *name2, int len2, int stage2)
403
{
404 405
	int len = len1 < len2 ? len1 : len2;
	int cmp;
406 407 408 409 410 411 412 413

	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
	if (stage1 < stage2)
416
		return -1;
417
	if (stage1 > stage2)
418
		return 1;
419 420 421
	return 0;
}

422 423 424 425 426
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);
}

427
static int index_name_stage_pos(const struct index_state *istate, const char *name, int namelen, int stage)
428 429 430 431
{
	int first, last;

	first = 0;
432
	last = istate->cache_nr;
433 434
	while (last > first) {
		int next = (last + first) >> 1;
435
		struct cache_entry *ce = istate->cache[next];
436
		int cmp = cache_name_stage_compare(name, namelen, stage, ce->name, ce_namelen(ce), ce_stage(ce));
437
		if (!cmp)
438
			return next;
439 440 441 442 443 444
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
445
	return -first-1;
446 447
}

448 449 450 451 452
int index_name_pos(const struct index_state *istate, const char *name, int namelen)
{
	return index_name_stage_pos(istate, name, namelen, 0);
}

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

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

470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489
/*
 * Remove all cache ententries marked for removal, that is where
 * CE_REMOVE is set in ce_flags.  This is much more effective than
 * calling remove_index_entry_at() for each entry to be removed.
 */
void remove_marked_cache_entries(struct index_state *istate)
{
	struct cache_entry **ce_array = istate->cache;
	unsigned int i, j;

	for (i = j = 0; i < istate->cache_nr; i++) {
		if (ce_array[i]->ce_flags & CE_REMOVE)
			remove_name_hash(ce_array[i]);
		else
			ce_array[j++] = ce_array[i];
	}
	istate->cache_changed = 1;
	istate->cache_nr = j;
}

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

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528
static int compare_name(struct cache_entry *ce, const char *path, int namelen)
{
	return namelen != ce_namelen(ce) || memcmp(path, ce->name, namelen);
}

static int index_name_pos_also_unmerged(struct index_state *istate,
	const char *path, int namelen)
{
	int pos = index_name_pos(istate, path, namelen);
	struct cache_entry *ce;

	if (pos >= 0)
		return pos;

	/* maybe unmerged? */
	pos = -1 - pos;
	if (pos >= istate->cache_nr ||
			compare_name((ce = istate->cache[pos]), path, namelen))
		return -1;

	/* order of preference: stage 2, 1, 3 */
	if (ce_stage(ce) == 1 && pos + 1 < istate->cache_nr &&
			ce_stage((ce = istate->cache[pos + 1])) == 2 &&
			!compare_name(ce, path, namelen))
		pos++;
	return pos;
}

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
static int different_name(struct cache_entry *ce, struct cache_entry *alias)
{
	int len = ce_namelen(ce);
	return ce_namelen(alias) != len || memcmp(ce->name, alias->name, len);
}

/*
 * If we add a filename that aliases in the cache, we will use the
 * name that we already have - but we don't want to update the same
 * alias twice, because that implies that there were actually two
 * different files with aliasing names!
 *
 * So we use the CE_ADDED flag to verify that the alias was an old
 * one before we accept it as
 */
static struct cache_entry *create_alias_ce(struct cache_entry *ce, struct cache_entry *alias)
{
	int len;
	struct cache_entry *new;

	if (alias->ce_flags & CE_ADDED)
		die("Will not add file alias '%s' ('%s' already exists in index)", ce->name, alias->name);

	/* Ok, create the new entry using the name of the existing alias */
	len = ce_namelen(alias);
	new = xcalloc(1, cache_entry_size(len));
	memcpy(new->name, alias->name, len);
	copy_cache_entry(new, ce);
	free(ce);
	return new;
}

561 562 563 564 565 566 567 568
static void record_intent_to_add(struct cache_entry *ce)
{
	unsigned char sha1[20];
	if (write_sha1_file("", 0, blob_type, sha1))
		die("cannot create an empty blob in the object database");
	hashcpy(ce->sha1, sha1);
}

569
int add_to_index(struct index_state *istate, const char *path, struct stat *st, int flags)
Johannes Schindelin's avatar
Johannes Schindelin committed
570
{
571
	int size, namelen, was_same;
572
	mode_t st_mode = st->st_mode;
573
	struct cache_entry *ce, *alias;
574
	unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;
575 576
	int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);
	int pretend = flags & ADD_CACHE_PRETEND;
577 578 579
	int intent_only = flags & ADD_CACHE_INTENT;
	int add_option = (ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE|
			  (intent_only ? ADD_CACHE_NEW_ONLY : 0));
Johannes Schindelin's avatar
Johannes Schindelin committed
580

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

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

598
	if (trust_executable_bit && has_symlinks)
599
		ce->ce_mode = create_ce_mode(st_mode);
600
	else {
601 602
		/* If there is an existing entry, pick the mode bits and type
		 * from it, otherwise assume unexecutable regular file.
Johannes Schindelin's avatar
Johannes Schindelin committed
603
		 */
604
		struct cache_entry *ent;
605
		int pos = index_name_pos_also_unmerged(istate, path, namelen);
606

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

611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633
	/* When core.ignorecase=true, determine if a directory of the same name but differing
	 * case already exists within the Git repository.  If it does, ensure the directory
	 * case of the file being added to the repository matches (is folded into) the existing
	 * entry's directory case.
	 */
	if (ignore_case) {
		const char *startPtr = ce->name;
		const char *ptr = startPtr;
		while (*ptr) {
			while (*ptr && *ptr != '/')
				++ptr;
			if (*ptr == '/') {
				struct cache_entry *foundce;
				++ptr;
				foundce = index_name_exists(&the_index, ce->name, ptr - ce->name, ignore_case);
				if (foundce) {
					memcpy((void *)startPtr, foundce->name + (startPtr - ce->name), ptr - startPtr);
					startPtr = ptr;
				}
			}
		}
	}

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

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

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

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

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

676 677 678 679 680 681 682
struct cache_entry *make_cache_entry(unsigned int mode,
		const unsigned char *sha1, const char *path, int stage,
		int refresh)
{
	int size, len;
	struct cache_entry *ce;

683 684
	if (!verify_path(path)) {
		error("Invalid path '%s'", path);
685
		return NULL;
686
	}
687 688 689 690 691 692 693

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

	hashcpy(ce->sha1, sha1);
	memcpy(ce->name, path, len);
694 695
	ce->ce_flags = create_ce_flags(stage);
	ce->ce_namelen = len;
696 697 698 699 700 701 702 703
	ce->ce_mode = create_ce_mode(mode);

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

	return ce;
}

704
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
705 706 707 708 709
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

710
int ce_path_match(const struct cache_entry *ce, const struct pathspec *pathspec)
711
{
712
	return match_pathspec_depth(pathspec, ce->name, ce_namelen(ce), 0, NULL);
713 714
}

715 716 717 718 719 720 721 722 723 724 725 726 727 728 729
/*
 * 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.
	 */
730

731
	/* "." is not allowed */
732
	if (*rest == '\0' || is_dir_sep(*rest))
733 734
		return 0;

735
	switch (*rest) {
736 737 738 739 740 741 742 743 744 745 746 747
	/*
	 * ".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 '.':
748
		if (rest[1] == '\0' || is_dir_sep(rest[1]))
749 750 751 752 753 754 755 756 757
			return 0;
	}
	return 1;
}

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

758 759 760
	if (has_dos_drive_prefix(path))
		return 0;

761 762 763 764
	goto inside;
	for (;;) {
		if (!c)
			return 1;
765
		if (is_dir_sep(c)) {
766 767
inside:
			c = *path++;
768 769 770
			if ((c == '.' && !verify_dotfile(path)) ||
			    is_dir_sep(c) || c == '\0')
				return 0;
771 772 773 774 775
		}
		c = *path++;
	}
}

776 777 778
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
779
 */
780 781
static int has_file_name(struct index_state *istate,
			 const struct cache_entry *ce, int pos, int ok_to_replace)
782
{
783 784
	int retval = 0;
	int len = ce_namelen(ce);
785
	int stage = ce_stage(ce);
786
	const char *name = ce->name;
787

788 789
	while (pos < istate->cache_nr) {
		struct cache_entry *p = istate->cache[pos++];
790

791
		if (len >= ce_namelen(p))
792
			break;
793 794
		if (memcmp(name, p->name, len))
			break;
795 796
		if (ce_stage(p) != stage)
			continue;
797 798
		if (p->name[len] != '/')
			continue;
799
		if (p->ce_flags & CE_REMOVE)
800
			continue;
801 802 803
		retval = -1;
		if (!ok_to_replace)
			break;
804
		remove_index_entry_at(istate, --pos);
805
	}
806 807
	return retval;
}
808

809 810 811 812
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
813 814
static int has_dir_name(struct index_state *istate,
			const struct cache_entry *ce, int pos, int ok_to_replace)
815 816
{
	int retval = 0;
817
	int stage = ce_stage(ce);
818 819
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
820

821 822
	for (;;) {
		int len;
823

824 825 826 827 828 829 830
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
831

832
		pos = index_name_stage_pos(istate, name, len, stage);
833
		if (pos >= 0) {
834 835 836 837 838 839 840 841
			/*
			 * 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.
			 */
842
			if (!(istate->cache[pos]->ce_flags & CE_REMOVE)) {
843 844 845
				retval = -1;
				if (!ok_to_replace)
					break;
846
				remove_index_entry_at(istate, pos);
847 848
				continue;
			}
849
		}
850 851
		else
			pos = -pos-1;
852 853 854 855

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
856
		 * we're ok, and we can exit.
857
		 */
858 859
		while (pos < istate->cache_nr) {
			struct cache_entry *p = istate->cache[pos];
860 861 862 863
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
864 865 866
			if (ce_stage(p) == stage && !(p->ce_flags & CE_REMOVE))
				/*
				 * p is at the same stage as our entry, and
867 868 869 870 871 872
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
873
		}
874
	}
875 876 877 878 879 880 881
	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
882
 *
883 884 885 886
 * 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.
 */
887 888 889
static int check_file_directory_conflict(struct index_state *istate,
					 const struct cache_entry *ce,
					 int pos, int ok_to_replace)
890
{
891 892 893 894 895
	int retval;

	/*
	 * When ce is an "I am going away" entry, we allow it to be added
	 */
896
	if (ce->ce_flags & CE_REMOVE)
897 898
		return 0;

899 900 901
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
902
	 * in the array.
903
	 */
904
	retval = has_file_name(istate, ce, pos, ok_to_replace);
905

906 907 908 909
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
910
	return retval + has_dir_name(istate, ce, pos, ok_to_replace);
911 912
}

913
static int add_index_entry_with_check(struct index_state *istate, struct cache_entry *ce, int option)
914 915
{
	int pos;
916 917
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
918
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
919
	int new_only = option & ADD_CACHE_NEW_ONLY;
Junio C Hamano's avatar
Junio C Hamano committed
920

Junio C Hamano's avatar
Junio C Hamano committed
921
	cache_tree_invalidate_path(istate->cache_tree, ce->name);
922
	pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
923

Junio C Hamano's avatar
Junio C Hamano committed
924
	/* existing match? Just replace it. */
925
	if (pos >= 0) {
926 927
		if (!new_only)
			replace_index_entry(istate, pos, ce);
928 929
		return 0;
	}
930
	pos = -pos-1;
931

932 933 934 935
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
936 937
	if (pos < istate->cache_nr && ce_stage(ce) == 0) {
		while (ce_same_name(istate->cache[pos], ce)) {
938
			ok_to_add = 1;
939
			if (!remove_index_entry_at(istate, pos))
940 941 942 943
				break;
		}
	}

944 945
	if (!ok_to_add)
		return -1;
946
	if (!verify_path(ce->name))
947
		return error("Invalid path '%s'", ce->name);
948

Junio C Hamano's avatar
Junio C Hamano committed
949
	if (!skip_df_check &&
950
	    check_file_directory_conflict(istate, ce, pos, ok_to_replace)) {
951
		if (!ok_to_replace)
952 953
			return error("'%s' appears as both a file and as a directory",
				     ce->name);
954
		pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
955 956
		pos = -pos-1;
	}
957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
	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;
	}
973

974
	/* Make sure the array is big enough .. */
975 976 977 978
	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 *));
979 980 981
	}

	/* Add it in.. */
982
	istate->cache_nr++;
983
	if (istate->cache_nr > pos + 1)
984 985 986
		memmove(istate->cache + pos + 1,
			istate->cache + pos,
			(istate->cache_nr - pos - 1) * sizeof(ce));
987
	set_index_entry(istate, pos, ce);
988
	istate->cache_changed = 1;
989 990 991
	return 0;
}

992 993 994 995 996 997 998 999 1000 1001 1002
/*
 * "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.
 */
1003
static struct cache_entry *refresh_cache_ent(struct index_state *istate,
1004
					     struct cache_entry *ce,
1005 1006
					     unsigned int options, int *err,
					     int *changed_ret)
1007 1008 1009 1010
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;
1011
	int ignore_valid = options & CE_MATCH_IGNORE_VALID;
1012
	int ignore_skip_worktree = options & CE_MATCH_IGNORE_SKIP_WORKTREE;
1013

1014 1015 1016
	if (ce_uptodate(ce))
		return ce;

1017
	/*
1018 1019 1020
	 * 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.
1021
	 */
1022 1023 1024 1025
	if (!ignore_skip_worktree && ce_skip_worktree(ce)) {
		ce_mark_uptodate(ce);
		return ce;
	}
1026 1027 1028 1029 1030
	if (!ignore_valid && (ce->ce_flags & CE_VALID)) {
		ce_mark_uptodate(ce);
		return ce;
	}

1031
	if (lstat(ce->name, &st) < 0) {
1032 1033
		if (err)
			*err = errno;
1034 1035
		return NULL;
	}
1036

1037
	changed = ie_match_stat(istate, ce, &st, options);
1038 1039
	if (changed_ret)
		*changed_ret = changed;
1040
	if (!changed) {
1041 1042 1043 1044 1045 1046 1047 1048
		/*
		 * 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 &&
1049
		    !(ce->ce_flags & CE_VALID))
1050
			; /* mark this one VALID again */
1051 1052 1053 1054 1055 1056
		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.
			 */
1057 1058
			if (!S_ISGITLINK(ce->ce_mode))
				ce_mark_uptodate(ce);
1059
			return ce;
1060
		}
1061 1062
	}

1063
	if (ie_modified(istate, ce, &st, options)) {
1064 1065
		if (err)
			*err = EINVAL;
1066 1067
		return NULL;
	}
1068 1069 1070 1071 1072

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);
1073 1074 1075 1076 1077
	/*
	 * 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.
1078
	 */
1079
	if (!ignore_valid && assume_unchanged &&
1080 1081
	    !(ce->ce_flags & CE_VALID))
		updated->ce_flags &= ~CE_VALID;
1082 1083 1084 1085

	return updated;
}

1086
static void show_file(const char * fmt, const char * name, int in_porcelain,
1087
		      int * first, const char *header_msg)
1088 1089 1090
{
	if (in_porcelain && *first && header_msg) {
		printf("%s\n", header_msg);
1091
		*first = 0;
1092 1093 1094 1095 1096
	}
	printf(fmt, name);
}

int refresh_index(struct index_state *istate, unsigned int flags, const char **pathspec,
1097
		  char *seen, const char *header_msg)
1098 1099 1100 1101 1102 1103 1104
{
	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;
1105
	int ignore_submodules = (flags & REFRESH_IGNORE_SUBMODULES) != 0;
1106 1107
	int first = 1;
	int in_porcelain = (flags & REFRESH_IN_PORCELAIN);
1108
	unsigned int options = really ? CE_MATCH_IGNORE_VALID : 0;
1109
	const char *modified_fmt;
1110 1111 1112
	const char *deleted_fmt;
	const char *typechange_fmt;
	const char *added_fmt;
1113
	const char *unmerged_fmt;
1114

1115
	modified_fmt = (in_porcelain ? "M\t%s\n" : "%s: needs update\n");
1116 1117 1118
	deleted_fmt = (in_porcelain ? "D\t%s\n" : "%s: needs update\n");
	typechange_fmt = (in_porcelain ? "T\t%s\n" : "%s needs update\n");
	added_fmt = (in_porcelain ? "A\t%s\n" : "%s needs update\n");
1119
	unmerged_fmt = (in_porcelain ? "U\t%s\n" : "%s: needs merge\n");
1120
	for (i = 0; i < istate->cache_nr; i++) {