fast-import.c 79.8 KB
Newer Older
1
/*
2
(See Documentation/git-fast-import.txt for maintained documentation.)
3 4 5 6 7
Format of STDIN stream:

  stream ::= cmd*;

  cmd ::= new_blob
8
        | new_commit
9
        | new_tag
10
        | reset_branch
11
        | checkpoint
12
        | progress
13 14
        ;

15
  new_blob ::= 'blob' lf
16
    mark?
17 18
    file_content;
  file_content ::= data;
19

20
  new_commit ::= 'commit' sp ref_str lf
21
    mark?
22 23
    ('author' (sp name)? sp '<' email '>' sp when lf)?
    'committer' (sp name)? sp '<' email '>' sp when lf
24
    commit_msg
25 26
    ('from' sp committish lf)?
    ('merge' sp committish lf)*
27
    file_change*
28
    lf?;
29
  commit_msg ::= data;
30

31 32 33 34 35 36
  file_change ::= file_clr
    | file_del
    | file_rnm
    | file_cpy
    | file_obm
    | file_inm;
37
  file_clr ::= 'deleteall' lf;
38
  file_del ::= 'D' sp path_str lf;
39
  file_rnm ::= 'R' sp path_str sp path_str lf;
40
  file_cpy ::= 'C' sp path_str sp path_str lf;
41 42 43
  file_obm ::= 'M' sp mode sp (hexsha1 | idnum) sp path_str lf;
  file_inm ::= 'M' sp mode sp 'inline' sp path_str lf
    data;
44 45 46
  note_obm ::= 'N' sp (hexsha1 | idnum) sp committish lf;
  note_inm ::= 'N' sp 'inline' sp committish lf
    data;
47 48

  new_tag ::= 'tag' sp tag_str lf
49
    'from' sp committish lf
50
    ('tagger' (sp name)? sp '<' email '>' sp when lf)?
51 52 53
    tag_msg;
  tag_msg ::= data;

54
  reset_branch ::= 'reset' sp ref_str lf
55
    ('from' sp committish lf)?
56
    lf?;
57

58
  checkpoint ::= 'checkpoint' lf
59
    lf?;
60

61 62 63
  progress ::= 'progress' sp not_lf* lf
    lf?;

64 65 66
     # note: the first idnum in a stream should be 1 and subsequent
     # idnums should not have gaps between values as this will cause
     # the stream parser to reserve space for the gapped values.  An
67
     # idnum can be updated in the future to a new object by issuing
68
     # a new mark directive with the old idnum.
69
     #
70
  mark ::= 'mark' sp idnum lf;
71
  data ::= (delimited_data | exact_data)
72
    lf?;
73 74 75 76 77 78

    # note: delim may be any string but must not contain lf.
    # data_line may contain any data but must not be exactly
    # delim.
  delimited_data ::= 'data' sp '<<' delim lf
    (data_line lf)*
79
    delim lf;
80 81

     # note: declen indicates the length of binary_data in bytes.
M
Mike Ralphson 已提交
82
     # declen does not include the lf preceding the binary data.
83
     #
84 85
  exact_data ::= 'data' sp declen lf
    binary_data;
86 87 88

     # note: quoted strings are C-style quoting supporting \c for
     # common escapes of 'c' (e..g \n, \t, \\, \") or \nnn where nnn
89
     # is the signed byte value in octal.  Note that the only
90 91
     # characters which must actually be escaped to protect the
     # stream formatting is: \, " and LF.  Otherwise these values
92
     # are UTF8.
93
     #
94
  committish  ::= (ref_str | hexsha1 | sha1exp_str | idnum);
95 96 97
  ref_str     ::= ref;
  sha1exp_str ::= sha1exp;
  tag_str     ::= tag;
98
  path_str    ::= path    | '"' quoted(path)    '"' ;
99 100
  mode        ::= '100644' | '644'
                | '100755' | '755'
J
Junio C Hamano 已提交
101
                | '120000'
102
                ;
103 104

  declen ::= # unsigned 32 bit value, ascii base10 notation;
105
  bigint ::= # unsigned integer value, ascii base10 notation;
106
  binary_data ::= # file content, not interpreted;
107

108 109 110 111
  when         ::= raw_when | rfc2822_when;
  raw_when     ::= ts sp tz;
  rfc2822_when ::= # Valid RFC 2822 date and time;

112 113
  sp ::= # ASCII space character;
  lf ::= # ASCII newline (LF) character;
114 115

     # note: a colon (':') must precede the numerical value assigned to
116
     # an idnum.  This is to distinguish it from a ref or tag name as
117
     # GIT does not permit ':' in ref or tag strings.
118
     #
119
  idnum   ::= ':' bigint;
120 121 122
  path    ::= # GIT style file path, e.g. "a/b/c";
  ref     ::= # GIT ref name, e.g. "refs/heads/MOZ_GECKO_EXPERIMENT";
  tag     ::= # GIT tag name, e.g. "FIREFOX_1_5";
123 124
  sha1exp ::= # Any valid GIT SHA1 expression;
  hexsha1 ::= # SHA1 in hexadecimal format;
125 126

     # note: name and email are UTF8 strings, however name must not
127
     # contain '<' or lf and email must not contain any of the
128
     # following: '<', '>', lf.
129
     #
130
  name  ::= # valid GIT author/committer name;
131
  email ::= # valid GIT author/committer email;
132 133
  ts    ::= # time since the epoch in seconds, ascii base10 notation;
  tz    ::= # GIT style timezone;
134

135 136 137 138
     # note: comments and cat requests may appear anywhere
     # in the input, except within a data command.  Any form
     # of the data command always escapes the related input
     # from comment processing.
139 140
     #
     # In case it is not clear, the '#' that starts the comment
141
     # must be the first character on that line (an lf
M
Mike Ralphson 已提交
142
     # preceded it).
143
     #
144 145
  cat_blob ::= 'cat-blob' sp (hexsha1 | idnum) lf;

146 147
  comment ::= '#' not_lf* lf;
  not_lf  ::= # Any byte that is not ASCII newline (LF);
148 149
*/

150 151 152 153
#include "builtin.h"
#include "cache.h"
#include "object.h"
#include "blob.h"
154
#include "tree.h"
155
#include "commit.h"
156 157
#include "delta.h"
#include "pack.h"
158
#include "refs.h"
159
#include "csum-file.h"
160
#include "quote.h"
161
#include "exec_cmd.h"
162
#include "dir.h"
163

164 165
#define PACK_ID_BITS 16
#define MAX_PACK_ID ((1<<PACK_ID_BITS)-1)
166 167
#define DEPTH_BITS 13
#define MAX_DEPTH ((1<<DEPTH_BITS)-1)
168

169 170
struct object_entry
{
171
	struct pack_idx_entry idx;
172
	struct object_entry *next;
173 174 175
	uint32_t type : TYPE_BITS,
		pack_id : PACK_ID_BITS,
		depth : DEPTH_BITS;
176 177
};

178
struct object_entry_pool
179
{
180
	struct object_entry_pool *next_pool;
181 182
	struct object_entry *next_free;
	struct object_entry *end;
183
	struct object_entry entries[FLEX_ARRAY]; /* more */
184 185
};

186 187 188 189 190 191
struct mark_set
{
	union {
		struct object_entry *marked[1024];
		struct mark_set *sets[1024];
	} data;
192
	unsigned int shift;
193 194
};

195 196
struct last_object
{
P
Pierre Habouzit 已提交
197
	struct strbuf data;
198
	off_t offset;
199
	unsigned int depth;
P
Pierre Habouzit 已提交
200
	unsigned no_swap : 1;
201 202
};

203 204 205 206 207
struct mem_pool
{
	struct mem_pool *next_pool;
	char *next_free;
	char *end;
208
	uintmax_t space[FLEX_ARRAY]; /* more */
209 210 211 212 213
};

struct atom_str
{
	struct atom_str *next_atom;
214
	unsigned short str_len;
215 216 217 218
	char str_dat[FLEX_ARRAY]; /* more */
};

struct tree_content;
219 220
struct tree_entry
{
221
	struct tree_content *tree;
222
	struct atom_str *name;
223 224
	struct tree_entry_ms
	{
225
		uint16_t mode;
226 227
		unsigned char sha1[20];
	} versions[2];
228 229
};

230
struct tree_content
231
{
232 233
	unsigned int entry_capacity; /* must match avail_tree_content */
	unsigned int entry_count;
234
	unsigned int delta_depth;
235 236 237 238 239 240 241
	struct tree_entry *entries[FLEX_ARRAY]; /* more */
};

struct avail_tree_content
{
	unsigned int entry_capacity; /* must match tree_content */
	struct avail_tree_content *next_avail;
242 243 244 245
};

struct branch
{
246 247
	struct branch *table_next_branch;
	struct branch *active_next_branch;
248
	const char *name;
249
	struct tree_entry branch_tree;
250
	uintmax_t last_commit;
251
	uintmax_t num_notes;
252 253
	unsigned active : 1;
	unsigned pack_id : PACK_ID_BITS;
254
	unsigned char sha1[20];
255 256
};

257 258 259 260
struct tag
{
	struct tag *next_tag;
	const char *name;
261
	unsigned int pack_id;
262 263 264
	unsigned char sha1[20];
};

265 266 267 268 269
struct hash_list
{
	struct hash_list *next;
	unsigned char sha1[20];
};
270

271 272 273
typedef enum {
	WHENSPEC_RAW = 1,
	WHENSPEC_RFC2822,
274
	WHENSPEC_NOW
275 276
} whenspec_type;

277 278 279 280 281 282 283
struct recent_command
{
	struct recent_command *prev;
	struct recent_command *next;
	char *buf;
};

284
/* Configured limits on output */
285
static unsigned long max_depth = 10;
286
static off_t max_packsize;
287
static uintmax_t big_file_threshold = 512 * 1024 * 1024;
288
static int force_update;
289 290
static int pack_compression_level = Z_DEFAULT_COMPRESSION;
static int pack_compression_seen;
291 292 293 294 295 296 297

/* Stats and misc. counters */
static uintmax_t alloc_count;
static uintmax_t marks_set_count;
static uintmax_t object_count_by_type[1 << TYPE_BITS];
static uintmax_t duplicate_count_by_type[1 << TYPE_BITS];
static uintmax_t delta_count_by_type[1 << TYPE_BITS];
298
static unsigned long object_count;
299
static unsigned long branch_count;
300
static unsigned long branch_load_count;
301
static int failure;
302
static FILE *pack_edges;
303
static unsigned int show_stats = 1;
S
Sverre Rabbelier 已提交
304 305
static int global_argc;
static const char **global_argv;
306

307 308 309 310 311
/* Memory pools */
static size_t mem_pool_alloc = 2*1024*1024 - sizeof(struct mem_pool);
static size_t total_allocd;
static struct mem_pool *mem_pool;

312
/* Atom management */
313 314 315 316 317
static unsigned int atom_table_sz = 4451;
static unsigned int atom_cnt;
static struct atom_str **atom_table;

/* The .pack file being generated */
318
static unsigned int pack_id;
319
static struct sha1file *pack_file;
320
static struct packed_git *pack_data;
321
static struct packed_git **all_packs;
322
static off_t pack_size;
323 324

/* Table of objects we've written. */
325
static unsigned int object_entry_alloc = 5000;
326 327
static struct object_entry_pool *blocks;
static struct object_entry *object_table[1 << 16];
328
static struct mark_set *marks;
329 330
static const char *export_marks_file;
static const char *import_marks_file;
331
static int import_marks_file_from_stream;
332
static int import_marks_file_ignore_missing;
333
static int relative_marks_paths;
334 335

/* Our last blob */
P
Pierre Habouzit 已提交
336
static struct last_object last_blob = { STRBUF_INIT, 0, 0, 0 };
337 338 339 340 341 342

/* Tree management */
static unsigned int tree_entry_alloc = 1000;
static void *avail_tree_entry;
static unsigned int avail_tree_table_sz = 100;
static struct avail_tree_content **avail_tree_table;
343 344
static struct strbuf old_tree = STRBUF_INIT;
static struct strbuf new_tree = STRBUF_INIT;
345

346
/* Branch data */
347 348 349
static unsigned long max_active_branches = 5;
static unsigned long cur_active_branches;
static unsigned long branch_table_sz = 1039;
350 351 352
static struct branch **branch_table;
static struct branch *active_branches;

353 354 355 356
/* Tag data */
static struct tag *first_tag;
static struct tag *last_tag;

357
/* Input stream parsing */
358
static whenspec_type whenspec = WHENSPEC_RAW;
359
static struct strbuf command_buf = STRBUF_INIT;
360
static int unread_command_buf;
361 362 363 364
static struct recent_command cmd_hist = {&cmd_hist, &cmd_hist, NULL};
static struct recent_command *cmd_tail = &cmd_hist;
static struct recent_command *rc_free;
static unsigned int cmd_save = 100;
365
static uintmax_t next_mark;
366
static struct strbuf new_data = STRBUF_INIT;
367
static int seen_data_command;
368

369 370 371
/* Signal handling */
static volatile sig_atomic_t checkpoint_requested;

372 373 374
/* Where to write output of cat-blob commands */
static int cat_blob_fd = STDOUT_FILENO;

S
Sverre Rabbelier 已提交
375
static void parse_argv(void);
376
static void parse_cat_blob(void);
377

378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403
static void write_branch_report(FILE *rpt, struct branch *b)
{
	fprintf(rpt, "%s:\n", b->name);

	fprintf(rpt, "  status      :");
	if (b->active)
		fputs(" active", rpt);
	if (b->branch_tree.tree)
		fputs(" loaded", rpt);
	if (is_null_sha1(b->branch_tree.versions[1].sha1))
		fputs(" dirty", rpt);
	fputc('\n', rpt);

	fprintf(rpt, "  tip commit  : %s\n", sha1_to_hex(b->sha1));
	fprintf(rpt, "  old tree    : %s\n", sha1_to_hex(b->branch_tree.versions[0].sha1));
	fprintf(rpt, "  cur tree    : %s\n", sha1_to_hex(b->branch_tree.versions[1].sha1));
	fprintf(rpt, "  commit clock: %" PRIuMAX "\n", b->last_commit);

	fputs("  last pack   : ", rpt);
	if (b->pack_id < MAX_PACK_ID)
		fprintf(rpt, "%u", b->pack_id);
	fputc('\n', rpt);

	fputc('\n', rpt);
}

404 405
static void dump_marks_helper(FILE *, uintmax_t, struct mark_set *);

406
static void write_crash_report(const char *err)
407
{
408
	char *loc = git_path("fast_import_crash_%"PRIuMAX, (uintmax_t) getpid());
409 410 411
	FILE *rpt = fopen(loc, "w");
	struct branch *b;
	unsigned long lu;
412
	struct recent_command *rc;
413 414 415 416 417 418 419 420 421

	if (!rpt) {
		error("can't write crash report %s: %s", loc, strerror(errno));
		return;
	}

	fprintf(stderr, "fast-import: dumping crash report to %s\n", loc);

	fprintf(rpt, "fast-import crash report:\n");
422 423
	fprintf(rpt, "    fast-import process: %"PRIuMAX"\n", (uintmax_t) getpid());
	fprintf(rpt, "    parent process     : %"PRIuMAX"\n", (uintmax_t) getppid());
424 425 426 427
	fprintf(rpt, "    at %s\n", show_date(time(NULL), 0, DATE_LOCAL));
	fputc('\n', rpt);

	fputs("fatal: ", rpt);
428
	fputs(err, rpt);
429 430
	fputc('\n', rpt);

431 432 433 434 435 436 437 438 439 440 441 442
	fputc('\n', rpt);
	fputs("Most Recent Commands Before Crash\n", rpt);
	fputs("---------------------------------\n", rpt);
	for (rc = cmd_hist.next; rc != &cmd_hist; rc = rc->next) {
		if (rc->next == &cmd_hist)
			fputs("* ", rpt);
		else
			fputs("  ", rpt);
		fputs(rc->buf, rpt);
		fputc('\n', rpt);
	}

443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
	fputc('\n', rpt);
	fputs("Active Branch LRU\n", rpt);
	fputs("-----------------\n", rpt);
	fprintf(rpt, "    active_branches = %lu cur, %lu max\n",
		cur_active_branches,
		max_active_branches);
	fputc('\n', rpt);
	fputs("  pos  clock name\n", rpt);
	fputs("  ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n", rpt);
	for (b = active_branches, lu = 0; b; b = b->active_next_branch)
		fprintf(rpt, "  %2lu) %6" PRIuMAX" %s\n",
			++lu, b->last_commit, b->name);

	fputc('\n', rpt);
	fputs("Inactive Branches\n", rpt);
	fputs("-----------------\n", rpt);
	for (lu = 0; lu < branch_table_sz; lu++) {
		for (b = branch_table[lu]; b; b = b->table_next_branch)
			write_branch_report(rpt, b);
	}

464 465 466 467 468 469 470 471 472 473 474 475 476
	if (first_tag) {
		struct tag *tg;
		fputc('\n', rpt);
		fputs("Annotated Tags\n", rpt);
		fputs("--------------\n", rpt);
		for (tg = first_tag; tg; tg = tg->next_tag) {
			fputs(sha1_to_hex(tg->sha1), rpt);
			fputc(' ', rpt);
			fputs(tg->name, rpt);
			fputc('\n', rpt);
		}
	}

477 478 479
	fputc('\n', rpt);
	fputs("Marks\n", rpt);
	fputs("-----\n", rpt);
480 481
	if (export_marks_file)
		fprintf(rpt, "  exported to %s\n", export_marks_file);
482 483 484
	else
		dump_marks_helper(rpt, 0, marks);

485 486 487 488 489 490
	fputc('\n', rpt);
	fputs("-------------------\n", rpt);
	fputs("END OF CRASH REPORT\n", rpt);
	fclose(rpt);
}

491 492 493 494
static void end_packfile(void);
static void unkeep_all_packs(void);
static void dump_marks(void);

495 496 497
static NORETURN void die_nicely(const char *err, va_list params)
{
	static int zombie;
498
	char message[2 * PATH_MAX];
499

500
	vsnprintf(message, sizeof(message), err, params);
501
	fputs("fatal: ", stderr);
502
	fputs(message, stderr);
503 504 505 506
	fputc('\n', stderr);

	if (!zombie) {
		zombie = 1;
507
		write_crash_report(message);
508 509 510
		end_packfile();
		unkeep_all_packs();
		dump_marks();
511 512 513
	}
	exit(128);
}
514

515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540
#ifndef SIGUSR1	/* Windows, for example */

static void set_checkpoint_signal(void)
{
}

#else

static void checkpoint_signal(int signo)
{
	checkpoint_requested = 1;
}

static void set_checkpoint_signal(void)
{
	struct sigaction sa;

	memset(&sa, 0, sizeof(sa));
	sa.sa_handler = checkpoint_signal;
	sigemptyset(&sa.sa_mask);
	sa.sa_flags = SA_RESTART;
	sigaction(SIGUSR1, &sa, NULL);
}

#endif

541
static void alloc_objects(unsigned int cnt)
542
{
543
	struct object_entry_pool *b;
544

545
	b = xmalloc(sizeof(struct object_entry_pool)
546
		+ cnt * sizeof(struct object_entry));
547
	b->next_pool = blocks;
548 549 550 551 552
	b->next_free = b->entries;
	b->end = b->entries + cnt;
	blocks = b;
	alloc_count += cnt;
}
553

554
static struct object_entry *new_object(unsigned char *sha1)
555
{
556
	struct object_entry *e;
557

558
	if (blocks->next_free == blocks->end)
559
		alloc_objects(object_entry_alloc);
560

561
	e = blocks->next_free++;
562
	hashcpy(e->idx.sha1, sha1);
563
	return e;
564 565
}

566
static struct object_entry *find_object(unsigned char *sha1)
567 568 569 570
{
	unsigned int h = sha1[0] << 8 | sha1[1];
	struct object_entry *e;
	for (e = object_table[h]; e; e = e->next)
571
		if (!hashcmp(sha1, e->idx.sha1))
572 573 574 575
			return e;
	return NULL;
}

576
static struct object_entry *insert_object(unsigned char *sha1)
577 578
{
	unsigned int h = sha1[0] << 8 | sha1[1];
579
	struct object_entry *e = object_table[h];
580 581

	while (e) {
582
		if (!hashcmp(sha1, e->idx.sha1))
583 584 585 586 587
			return e;
		e = e->next;
	}

	e = new_object(sha1);
588
	e->next = object_table[h];
589
	e->idx.offset = 0;
590
	object_table[h] = e;
591 592
	return e;
}
593

594 595 596 597 598 599 600 601
static unsigned int hc_str(const char *s, size_t len)
{
	unsigned int r = 0;
	while (len-- > 0)
		r = r * 31 + *s++;
	return r;
}

602
static void *pool_alloc(size_t len)
603 604 605 606
{
	struct mem_pool *p;
	void *r;

607 608 609 610
	/* round up to a 'uintmax_t' alignment */
	if (len & (sizeof(uintmax_t) - 1))
		len += sizeof(uintmax_t) - (len & (sizeof(uintmax_t) - 1));

611 612 613 614 615 616 617 618 619 620 621 622
	for (p = mem_pool; p; p = p->next_pool)
		if ((p->end - p->next_free >= len))
			break;

	if (!p) {
		if (len >= (mem_pool_alloc/2)) {
			total_allocd += len;
			return xmalloc(len);
		}
		total_allocd += sizeof(struct mem_pool) + mem_pool_alloc;
		p = xmalloc(sizeof(struct mem_pool) + mem_pool_alloc);
		p->next_pool = mem_pool;
623
		p->next_free = (char *) p->space;
624 625 626 627 628 629 630 631 632
		p->end = p->next_free + mem_pool_alloc;
		mem_pool = p;
	}

	r = p->next_free;
	p->next_free += len;
	return r;
}

633
static void *pool_calloc(size_t count, size_t size)
634 635 636 637 638 639 640
{
	size_t len = count * size;
	void *r = pool_alloc(len);
	memset(r, 0, len);
	return r;
}

641
static char *pool_strdup(const char *s)
642 643 644 645 646 647
{
	char *r = pool_alloc(strlen(s) + 1);
	strcpy(r, s);
	return r;
}

648
static void insert_mark(uintmax_t idnum, struct object_entry *oe)
649 650 651 652 653 654 655 656 657
{
	struct mark_set *s = marks;
	while ((idnum >> s->shift) >= 1024) {
		s = pool_calloc(1, sizeof(struct mark_set));
		s->shift = marks->shift + 10;
		s->data.sets[0] = marks;
		marks = s;
	}
	while (s->shift) {
658
		uintmax_t i = idnum >> s->shift;
659 660 661 662 663 664 665 666 667 668 669 670
		idnum -= i << s->shift;
		if (!s->data.sets[i]) {
			s->data.sets[i] = pool_calloc(1, sizeof(struct mark_set));
			s->data.sets[i]->shift = s->shift - 10;
		}
		s = s->data.sets[i];
	}
	if (!s->data.marked[idnum])
		marks_set_count++;
	s->data.marked[idnum] = oe;
}

671
static struct object_entry *find_mark(uintmax_t idnum)
672
{
673
	uintmax_t orig_idnum = idnum;
674 675 676 677
	struct mark_set *s = marks;
	struct object_entry *oe = NULL;
	if ((idnum >> s->shift) < 1024) {
		while (s && s->shift) {
678
			uintmax_t i = idnum >> s->shift;
679 680 681 682 683 684 685
			idnum -= i << s->shift;
			s = s->data.sets[i];
		}
		if (s)
			oe = s->data.marked[idnum];
	}
	if (!oe)
686
		die("mark :%" PRIuMAX " not declared", orig_idnum);
687 688 689
	return oe;
}

690
static struct atom_str *to_atom(const char *s, unsigned short len)
691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
{
	unsigned int hc = hc_str(s, len) % atom_table_sz;
	struct atom_str *c;

	for (c = atom_table[hc]; c; c = c->next_atom)
		if (c->str_len == len && !strncmp(s, c->str_dat, len))
			return c;

	c = pool_alloc(sizeof(struct atom_str) + len + 1);
	c->str_len = len;
	strncpy(c->str_dat, s, len);
	c->str_dat[len] = 0;
	c->next_atom = atom_table[hc];
	atom_table[hc] = c;
	atom_cnt++;
	return c;
}

709
static struct branch *lookup_branch(const char *name)
710 711 712 713 714 715 716 717 718 719
{
	unsigned int hc = hc_str(name, strlen(name)) % branch_table_sz;
	struct branch *b;

	for (b = branch_table[hc]; b; b = b->table_next_branch)
		if (!strcmp(name, b->name))
			return b;
	return NULL;
}

720
static struct branch *new_branch(const char *name)
721 722
{
	unsigned int hc = hc_str(name, strlen(name)) % branch_table_sz;
723
	struct branch *b = lookup_branch(name);
724 725 726

	if (b)
		die("Invalid attempt to create duplicate branch: %s", name);
727
	switch (check_ref_format(name)) {
728 729 730
	case 0: break; /* its valid */
	case CHECK_REF_FORMAT_ONELEVEL:
		break; /* valid, but too few '/', allow anyway */
731
	default:
732
		die("Branch name doesn't conform to GIT standards: %s", name);
733
	}
734 735 736 737

	b = pool_calloc(1, sizeof(struct branch));
	b->name = pool_strdup(name);
	b->table_next_branch = branch_table[hc];
738 739
	b->branch_tree.versions[0].mode = S_IFDIR;
	b->branch_tree.versions[1].mode = S_IFDIR;
740
	b->num_notes = 0;
741
	b->active = 0;
742
	b->pack_id = MAX_PACK_ID;
743 744 745 746 747 748 749 750 751 752 753
	branch_table[hc] = b;
	branch_count++;
	return b;
}

static unsigned int hc_entries(unsigned int cnt)
{
	cnt = cnt & 7 ? (cnt / 8) + 1 : cnt / 8;
	return cnt < avail_tree_table_sz ? cnt : avail_tree_table_sz - 1;
}

754
static struct tree_content *new_tree_content(unsigned int cnt)
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776
{
	struct avail_tree_content *f, *l = NULL;
	struct tree_content *t;
	unsigned int hc = hc_entries(cnt);

	for (f = avail_tree_table[hc]; f; l = f, f = f->next_avail)
		if (f->entry_capacity >= cnt)
			break;

	if (f) {
		if (l)
			l->next_avail = f->next_avail;
		else
			avail_tree_table[hc] = f->next_avail;
	} else {
		cnt = cnt & 7 ? ((cnt / 8) + 1) * 8 : cnt;
		f = pool_alloc(sizeof(*t) + sizeof(t->entries[0]) * cnt);
		f->entry_capacity = cnt;
	}

	t = (struct tree_content*)f;
	t->entry_count = 0;
777
	t->delta_depth = 0;
778 779 780 781 782 783 784 785
	return t;
}

static void release_tree_entry(struct tree_entry *e);
static void release_tree_content(struct tree_content *t)
{
	struct avail_tree_content *f = (struct avail_tree_content*)t;
	unsigned int hc = hc_entries(f->entry_capacity);
786 787 788 789 790 791
	f->next_avail = avail_tree_table[hc];
	avail_tree_table[hc] = f;
}

static void release_tree_content_recursive(struct tree_content *t)
{
792 793 794
	unsigned int i;
	for (i = 0; i < t->entry_count; i++)
		release_tree_entry(t->entries[i]);
795
	release_tree_content(t);
796 797
}

798
static struct tree_content *grow_tree_content(
799 800 801 802 803
	struct tree_content *t,
	int amt)
{
	struct tree_content *r = new_tree_content(t->entry_count + amt);
	r->entry_count = t->entry_count;
804
	r->delta_depth = t->delta_depth;
805 806 807 808 809
	memcpy(r->entries,t->entries,t->entry_count*sizeof(t->entries[0]));
	release_tree_content(t);
	return r;
}

810
static struct tree_entry *new_tree_entry(void)
811 812 813 814 815
{
	struct tree_entry *e;

	if (!avail_tree_entry) {
		unsigned int n = tree_entry_alloc;
816
		total_allocd += n * sizeof(struct tree_entry);
817
		avail_tree_entry = e = xmalloc(n * sizeof(struct tree_entry));
818
		while (n-- > 1) {
819 820 821
			*((void**)e) = e + 1;
			e++;
		}
822
		*((void**)e) = NULL;
823 824 825 826 827 828 829 830 831 832
	}

	e = avail_tree_entry;
	avail_tree_entry = *((void**)e);
	return e;
}

static void release_tree_entry(struct tree_entry *e)
{
	if (e->tree)
833
		release_tree_content_recursive(e->tree);
834 835 836 837
	*((void**)e) = avail_tree_entry;
	avail_tree_entry = e;
}

838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862
static struct tree_content *dup_tree_content(struct tree_content *s)
{
	struct tree_content *d;
	struct tree_entry *a, *b;
	unsigned int i;

	if (!s)
		return NULL;
	d = new_tree_content(s->entry_count);
	for (i = 0; i < s->entry_count; i++) {
		a = s->entries[i];
		b = new_tree_entry();
		memcpy(b, a, sizeof(*a));
		if (a->tree && is_null_sha1(b->versions[1].sha1))
			b->tree = dup_tree_content(a->tree);
		else
			b->tree = NULL;
		d->entries[i] = b;
	}
	d->entry_count = s->entry_count;
	d->delta_depth = s->delta_depth;

	return d;
}

863
static void start_packfile(void)
864
{
865
	static char tmpfile[PATH_MAX];
866
	struct packed_git *p;
867
	struct pack_header hdr;
868
	int pack_fd;
869

870 871
	pack_fd = odb_mkstemp(tmpfile, sizeof(tmpfile),
			      "pack/tmp_pack_XXXXXX");
872 873
	p = xcalloc(1, sizeof(*p) + strlen(tmpfile) + 2);
	strcpy(p->pack_name, tmpfile);
874
	p->pack_fd = pack_fd;
875
	pack_file = sha1fd(pack_fd, p->pack_name);
876 877 878 879

	hdr.hdr_signature = htonl(PACK_SIGNATURE);
	hdr.hdr_version = htonl(2);
	hdr.hdr_entries = 0;
880
	sha1write(pack_file, &hdr, sizeof(hdr));
881 882

	pack_data = p;
883 884
	pack_size = sizeof(hdr);
	object_count = 0;
885 886 887

	all_packs = xrealloc(all_packs, sizeof(*all_packs) * (pack_id + 1));
	all_packs[pack_id] = p;
888 889
}

890
static const char *create_index(void)
891
{
892 893 894
	const char *tmpfile;
	struct pack_idx_entry **idx, **c, **last;
	struct object_entry *e;
895 896
	struct object_entry_pool *o;

897 898
	/* Build the table of object IDs. */
	idx = xmalloc(object_count * sizeof(*idx));
899 900
	c = idx;
	for (o = blocks; o; o = o->next_pool)
901 902
		for (e = o->next_free; e-- != o->entries;)
			if (pack_id == e->pack_id)
903
				*c++ = &e->idx;
904
	last = idx + object_count;
905 906
	if (c != last)
		die("internal consistency error creating the index");
907

908
	tmpfile = write_idx_file(NULL, idx, object_count, pack_data->sha1);
909
	free(idx);
910 911 912
	return tmpfile;
}

913
static char *keep_pack(const char *curr_index_name)
914 915
{
	static char name[PATH_MAX];
916
	static const char *keep_msg = "fast-import";
917 918
	int keep_fd;

919
	keep_fd = odb_pack_keep(name, sizeof(name), pack_data->sha1);
920
	if (keep_fd < 0)
921
		die_errno("cannot create keep file");
922 923
	write_or_die(keep_fd, keep_msg, strlen(keep_msg));
	if (close(keep_fd))
924
		die_errno("failed to write keep file");
925 926 927 928 929 930 931 932 933 934

	snprintf(name, sizeof(name), "%s/pack/pack-%s.pack",
		 get_object_directory(), sha1_to_hex(pack_data->sha1));
	if (move_temp_to_file(pack_data->pack_name, name))
		die("cannot store pack file");

	snprintf(name, sizeof(name), "%s/pack/pack-%s.idx",
		 get_object_directory(), sha1_to_hex(pack_data->sha1));
	if (move_temp_to_file(curr_index_name, name))
		die("cannot store index file");
935
	free((void *)curr_index_name);
936 937 938
	return name;
}

939
static void unkeep_all_packs(void)
940 941 942 943 944 945 946 947
{
	static char name[PATH_MAX];
	int k;

	for (k = 0; k < pack_id; k++) {
		struct packed_git *p = all_packs[k];
		snprintf(name, sizeof(name), "%s/pack/pack-%s.keep",
			 get_object_directory(), sha1_to_hex(p->sha1));
948
		unlink_or_warn(name);
949
	}
950 951
}

952
static void end_packfile(void)
953
{
954 955
	struct packed_git *old_p = pack_data, *new_p;

956
	clear_delta_base_cache();
957
	if (object_count) {
958
		unsigned char cur_pack_sha1[20];
959
		char *idx_name;
960 961 962
		int i;
		struct branch *b;
		struct tag *t;
963

964
		close_pack_windows(pack_data);
965
		sha1close(pack_file, cur_pack_sha1, 0);
966
		fixup_pack_header_footer(pack_data->pack_fd, pack_data->sha1,
967
				    pack_data->pack_name, object_count,
968
				    cur_pack_sha1, pack_size);
969
		close(pack_data->pack_fd);
970
		idx_name = keep_pack(create_index());
971

M
Mike Ralphson 已提交
972
		/* Register the packfile with core git's machinery. */
973 974 975
		new_p = add_packed_git(idx_name, strlen(idx_name), 1);
		if (!new_p)
			die("core git rejected index %s", idx_name);
976
		all_packs[pack_id] = new_p;
977
		install_packed_git(new_p);
978 979

		/* Print the boundary */
980 981 982 983 984 985 986
		if (pack_edges) {
			fprintf(pack_edges, "%s:", new_p->pack_name);
			for (i = 0; i < branch_table_sz; i++) {
				for (b = branch_table[i]; b; b = b->table_next_branch) {
					if (b->pack_id == pack_id)
						fprintf(pack_edges, " %s", sha1_to_hex(b->sha1));
				}
987
			}
988 989 990 991 992 993
			for (t = first_tag; t; t = t->next_tag) {
				if (t->pack_id == pack_id)
					fprintf(pack_edges, " %s", sha1_to_hex(t->sha1));
			}
			fputc('\n', pack_edges);
			fflush(pack_edges);
994 995 996
		}

		pack_id++;
997
	}
998 999
	else {
		close(old_p->pack_fd);
1000
		unlink_or_warn(old_p->pack_name);
1001
	}
1002 1003 1004
	free(old_p);

	/* We can't carry a delta across packfiles. */
P
Pierre Habouzit 已提交
1005
	strbuf_release(&last_blob.data);
1006 1007
	last_blob.offset = 0;
	last_blob.depth = 0;
1008 1009
}

1010
static void cycle_packfile(void)
1011 1012 1013 1014 1015
{
	end_packfile();
	start_packfile();
}

1016 1017
static int store_object(
	enum object_type type,
1018
	struct strbuf *dat,
1019
	struct last_object *last,
1020
	unsigned char *sha1out,
1021
	uintmax_t mark)
1022 1023
{
	void *out, *delta;
1024 1025 1026
	struct object_entry *e;
	unsigned char hdr[96];
	unsigned char sha1[20];
1027
	unsigned long hdrlen, deltalen;
1028
	git_SHA_CTX c;
1029 1030
	z_stream s;

1031
	hdrlen = sprintf((char *)hdr,"%s %lu", typename(type),
1032
		(unsigned long)dat->len) + 1;
1033 1034 1035 1036
	git_SHA1_Init(&c);
	git_SHA1_Update(&c, hdr, hdrlen);
	git_SHA1_Update(&c, dat->buf, dat->len);
	git_SHA1_Final(sha1, &c);
1037
	if (sha1out)
1038
		hashcpy(sha1out, sha1);
1039 1040

	e = insert_object(sha1);
1041 1042
	if (mark)
		insert_mark(mark, e);
1043
	if (e->idx.offset) {
1044
		duplicate_count_by_type[type]++;
1045
		return 1;
1046 1047 1048
	} else if (find_sha1_pack(sha1, packed_git)) {
		e->type = type;
		e->pack_id = MAX_PACK_ID;
1049
		e->idx.offset = 1; /* just not zero! */
1050 1051
		duplicate_count_by_type[type]++;
		return 1;
1052
	}
1053

1054
	if (last && last->data.buf && last->depth < max_depth && dat->len > 20) {
P
Pierre Habouzit 已提交
1055
		delta = diff_delta(last->data.buf, last->data.len,
1056
			dat->buf, dat->len,
1057
			&deltalen, dat->len - 20);
1058 1059
	} else
		delta = NULL;
1060 1061

	memset(&s, 0, sizeof(s));
1062
	deflateInit(&s, pack_compression_level);
1063 1064 1065 1066
	if (delta) {
		s.next_in = delta;
		s.avail_in = deltalen;
	} else {
1067 1068
		s.next_in = (void *)dat->buf;
		s.avail_in = dat->len;
1069 1070 1071 1072 1073 1074 1075 1076
	}
	s.avail_out = deflateBound(&s, s.avail_in);
	s.next_out = out = xmalloc(s.avail_out);
	while (deflate(&s, Z_FINISH) == Z_OK)
		/* nothing */;
	deflateEnd(&s);

	/* Determine if we should auto-checkpoint. */
1077
	if ((max_packsize && (pack_size + 60 + s.total_out) > max_packsize)
1078 1079 1080 1081
		|| (pack_size + 60 + s.total_out) < pack_size) {

		/* This new object needs to *not* have the current pack_id. */
		e->pack_id = pack_id + 1;
1082
		cycle_packfile();
1083 1084 1085 1086 1087

		/* We cannot carry a delta into the new pack. */
		if (delta) {
			free(delta);
			delta = NULL;
1088 1089

			memset(&s, 0, sizeof(s));
1090
			deflateInit(&s, pack_compression_level);
1091 1092
			s.next_in = (void *)dat->buf;
			s.avail_in = dat->len;
1093 1094 1095 1096 1097
			s.avail_out = deflateBound(&s, s.avail_in);
			s.next_out = out = xrealloc(out, s.avail_out);
			while (deflate(&s, Z_FINISH) == Z_OK)
				/* nothing */;
			deflateEnd(&s);
1098 1099 1100 1101 1102
		}
	}

	e->type = type;
	e->pack_id = pack_id;
1103
	e->idx.offset = pack_size;
1104 1105
	object_count++;
	object_count_by_type[type]++;
1106

1107 1108
	crc32_begin(pack_file);

1109
	if (delta) {
1110
		off_t ofs = e->idx.offset - last->offset;
1111 1112
		unsigned pos = sizeof(hdr) - 1;

1113
		delta_count_by_type[type]++;
1114
		e->depth = last->depth + 1;
1115

1116
		hdrlen = encode_in_pack_object_header(OBJ_OFS_DELTA, deltalen, hdr);
1117
		sha1write(pack_file, hdr, hdrlen);
1118 1119 1120 1121 1122
		pack_size += hdrlen;

		hdr[pos] = ofs & 127;
		while (ofs >>= 7)
			hdr[--pos] = 128 | (--ofs & 127);
1123
		sha1write(pack_file, hdr + pos, sizeof(hdr) - pos);
1124
		pack_size += sizeof(hdr) - pos;
1125
	} else {
1126
		e->depth = 0;
1127
		hdrlen = encode_in_pack_object_header(type, dat->len, hdr);
1128
		sha1write(pack_file, hdr, hdrlen);
1129
		pack_size += hdrlen;
1130 1131
	}

1132
	sha1write(pack_file, out, s.total_out);
1133
	pack_size += s.total_out;
1134

1135 1136
	e->idx.crc32 = crc32_end(pack_file);

1137
	free(out);
1138
	free(delta);
1139
	if (last) {
P
Pierre Habouzit 已提交
1140 1141 1142
		if (last->no_swap) {
			last->data = *dat;
		} else {
1143
			strbuf_swap(&last->data, dat);
P
Pierre Habouzit 已提交
1144
		}
1145
		last->offset = e->idx.offset;
1146
		last->depth = e->depth;
1147 1148 1149 1150
	}
	return 0;
}

1151
static void truncate_pack(off_t to, git_SHA_CTX *ctx)
1152 1153 1154 1155 1156
{
	if (ftruncate(pack_data->pack_fd, to)
	 || lseek(pack_data->pack_fd, to, SEEK_SET) != to)
		die_errno("cannot truncate pack to skip duplicate");
	pack_size = to;
1157 1158 1159 1160 1161

	/* yes this is a layering violation */
	pack_file->total = to;
	pack_file->offset = 0;
	pack_file->ctx = *ctx;
1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173
}

static void stream_blob(uintmax_t len, unsigned char *sha1out, uintmax_t mark)
{
	size_t in_sz = 64 * 1024, out_sz = 64 * 1024;
	unsigned char *in_buf = xmalloc(in_sz);
	unsigned char *out_buf = xmalloc(out_sz);
	struct object_entry *e;
	unsigned char sha1[20];
	unsigned long hdrlen;
	off_t offset;
	git_SHA_CTX c;
1174
	git_SHA_CTX pack_file_ctx;
1175 1176 1177 1178
	z_stream s;
	int status = Z_OK;

	/* Determine if we should auto-checkpoint. */
1179
	if ((max_packsize && (pack_size + 60 + len) > max_packsize)
1180 1181 1182 1183 1184
		|| (pack_size + 60 + len) < pack_size)
		cycle_packfile();

	offset = pack_size;

1185 1186 1187 1188
	/* preserve the pack_file SHA1 ctx in case we have to truncate later */
	sha1flush(pack_file);
	pack_file_ctx = pack_file->ctx;

1189 1190 1191 1192 1193 1194 1195
	hdrlen = snprintf((char *)out_buf, out_sz, "blob %" PRIuMAX, len) + 1;
	if (out_sz <= hdrlen)
		die("impossibly large object header");

	git_SHA1_Init(&c);
	git_SHA1_Update(&c, out_buf, hdrlen);

1196 1197
	crc32_begin(pack_file);

1198 1199 1200
	memset(&s, 0, sizeof(s));
	deflateInit(&s, pack_compression_level);

1201
	hdrlen = encode_in_pack_object_header(OBJ_BLOB, len, out_buf);
1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
	if (out_sz <= hdrlen)
		die("impossibly large object header");

	s.next_out = out_buf + hdrlen;
	s.avail_out = out_sz - hdrlen;

	while (status != Z_STREAM_END) {
		if (0 < len && !s.avail_in) {
			size_t cnt = in_sz < len ? in_sz : (size_t)len;
			size_t n = fread(in_buf, 1, cnt, stdin);
			if (!n && feof(stdin))
				die("EOF in data (%" PRIuMAX " bytes remaining)", len);

			git_SHA1_Update(&c, in_buf, n);
			s.next_in = in_buf;
			s.avail_in = n;
			len -= n;
		}

		status = deflate(&s, len ? 0 : Z_FINISH);

		if (!s.avail_out || status == Z_STREAM_END) {
			size_t n = s.next_out - out_buf;
1225
			sha1write(pack_file, out_buf, n);
1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
			pack_size += n;
			s.next_out = out_buf;
			s.avail_out = out_sz;
		}

		switch (status) {
		case Z_OK:
		case Z_BUF_ERROR:
		case Z_STREAM_END:
			continue;
		default:
			die("unexpected deflate failure: %d", status);
		}
	}
	deflateEnd(&s);
	git_SHA1_Final(sha1, &c);

	if (sha1out)
		hashcpy(sha1out, sha1);

	e = insert_object(sha1);

	if (mark)
		insert_mark(mark, e);

1251
	if (e->idx.offset) {
1252
		duplicate_count_by_type[OBJ_BLOB]++;
1253
		truncate_pack(offset, &pack_file_ctx);
1254 1255 1256 1257

	} else if (find_sha1_pack(sha1, packed_git)) {
		e->type = OBJ_BLOB;
		e->pack_id = MAX_PACK_ID;
1258
		e->idx.offset = 1; /* just not zero! */
1259
		duplicate_count_by_type[OBJ_BLOB]++;
1260
		truncate_pack(offset, &pack_file_ctx);
1261 1262 1263 1264 1265

	} else {
		e->depth = 0;
		e->type = OBJ_BLOB;
		e->pack_id = pack_id;
1266
		e->idx.offset = offset;
1267
		e->idx.crc32 = crc32_end(pack_file);
1268 1269 1270 1271 1272 1273 1274 1275
		object_count++;
		object_count_by_type[OBJ_BLOB]++;
	}

	free(in_buf);
	free(out_buf);
}

1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293
/* All calls must be guarded by find_object() or find_mark() to
 * ensure the 'struct object_entry' passed was written by this
 * process instance.  We unpack the entry by the offset, avoiding
 * the need for the corresponding .idx file.  This unpacking rule
 * works because we only use OBJ_REF_DELTA within the packfiles
 * created by fast-import.
 *
 * oe must not be NULL.  Such an oe usually comes from giving
 * an unknown SHA-1 to find_object() or an undefined mark to
 * find_mark().  Callers must test for this condition and use
 * the standard read_sha1_file() when it happens.
 *
 * oe->pack_id must not be MAX_PACK_ID.  Such an oe is usually from
 * find_mark(), where the mark was reloaded from an existing marks
 * file and is referencing an object that this fast-import process
 * instance did not write out to a packfile.  Callers must test for
 * this condition and use read_sha1_file() instead.
 */
1294 1295 1296
static void *gfi_unpack_entry(
	struct object_entry *oe,
	unsigned long *sizep)
1297
{
1298
	enum object_type type;
1299
	struct packed_git *p = all_packs[oe->pack_id];
1300
	if (p == pack_data && p->pack_size < (pack_size + 20)) {
1301 1302 1303 1304 1305 1306 1307 1308
		/* The object is stored in the packfile we are writing to
		 * and we have modified it since the last time we scanned
		 * back to read a previously written object.  If an old
		 * window covered [p->pack_size, p->pack_size + 20) its
		 * data is stale and is not valid.  Closing all windows
		 * and updating the packfile length ensures we can read
		 * the newly written data.
		 */
1309
		close_pack_windows(p);
1310
		sha1flush(pack_file);
1311 1312 1313 1314 1315 1316 1317

		/* We have to offer 20 bytes additional on the end of
		 * the packfile as the core unpacker code assumes the
		 * footer is present at the file end and must promise
		 * at least 20 bytes within any window it maps.  But
		 * we don't actually create the footer here.
		 */
1318
		p->pack_size = pack_size + 20;
1319
	}
1320
	return unpack_entry(p, oe->idx.offset, &type, sizep);
1321 1322
}

1323
static const char *get_mode(const char *str, uint16_t *modep)
1324 1325
{
	unsigned char c;
1326
	uint16_t mode = 0;
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338

	while ((c = *str++) != ' ') {
		if (c < '0' || c > '7')
			return NULL;
		mode = (mode << 3) + (c - '0');
	}
	*modep = mode;
	return str;
}

static void load_tree(struct tree_entry *root)
{
1339
	unsigned char *sha1 = root->versions[1].sha1;
1340 1341 1342 1343 1344 1345 1346
	struct object_entry *myoe;
	struct tree_content *t;
	unsigned long size;
	char *buf;
	const char *c;

	root->tree = t = new_tree_content(8);
1347
	if (is_null_sha1(sha1))
1348 1349
		return;

1350
	myoe = find_object(sha1);
1351
	if (myoe && myoe->pack_id != MAX_PACK_ID) {
1352
		if (myoe->type != OBJ_TREE)
1353
			die("Not a tree: %s", sha1_to_hex(sha1));
1354
		t->delta_depth = myoe->depth;
1355
		buf = gfi_unpack_entry(myoe, &size);
1356 1357
		if (!buf)
			die("Can't load tree %s", sha1_to_hex(sha1));
1358
	} else {
1359 1360 1361
		enum object_type type;
		buf = read_sha1_file(sha1, &type, &size);
		if (!buf || type != OBJ_TREE)
1362
			die("Can't load tree %s", sha1_to_hex(sha1));
1363 1364 1365 1366 1367 1368 1369
	}

	c = buf;
	while (c != (buf + size)) {
		struct tree_entry *e = new_tree_entry();

		if (t->entry_count == t->entry_capacity)
1370
			root->tree = t = grow_tree_content(t, t->entry_count);
1371 1372 1373
		t->entries[t->entry_count++] = e;

		e->tree = NULL;
1374
		c = get_mode(c, &e->versions[1].mode);
1375
		if (!c)
1376 1377
			die("Corrupt mode in %s", sha1_to_hex(sha1));
		e->versions[0].mode = e->versions[1].mode;
1378
		e->name = to_atom(c, strlen(c));
1379
		c += e->name->str_len + 1;
1380 1381
		hashcpy(e->versions[0].sha1, (unsigned char *)c);
		hashcpy(e->versions[1].sha1, (unsigned char *)c);
1382 1383 1384 1385 1386
		c += 20;
	}
	free(buf);
}

1387
static int tecmp0 (const void *_a, const void *_b)
1388 1389 1390 1391
{
	struct tree_entry *a = *((struct tree_entry**)_a);
	struct tree_entry *b = *((struct tree_entry**)_b);
	return base_name_compare(
1392 1393
		a->name->str_dat, a->name->str_len, a->versions[0].mode,
		b->name->str_dat, b->name->str_len, b->versions[0].mode);
1394 1395
}

1396
static int tecmp1 (const void *_a, const void *_b)
1397
{
1398 1399 1400 1401 1402 1403 1404
	struct tree_entry *a = *((struct tree_entry**)_a);
	struct tree_entry *b = *((struct tree_entry**)_b);
	return base_name_compare(
		a->name->str_dat, a->name->str_len, a->versions[1].mode,
		b->name->str_dat, b->name->str_len, b->versions[1].mode);
}

1405
static void mktree(struct tree_content *t, int v, struct strbuf *b)
1406 1407
{
	size_t maxlen = 0;
1408 1409
	unsigned int i;

1410 1411 1412 1413
	if (!v)
		qsort(t->entries,t->entry_count,sizeof(t->entries[0]),tecmp0);
	else
		qsort(t->entries,t->entry_count,sizeof(t->entries[0]),tecmp1);
1414 1415

	for (i = 0; i < t->entry_count; i++) {
1416 1417
		if (t->entries[i]->versions[v].mode)
			maxlen += t->entries[i]->name->str_len + 34;
1418 1419
	}

1420 1421
	strbuf_reset(b);
	strbuf_grow(b, maxlen);
1422 1423
	for (i = 0; i < t->entry_count; i++) {
		struct tree_entry *e = t->entries[i];
1424 1425
		if (!e->versions[v].mode)
			continue;
1426 1427 1428
		strbuf_addf(b, "%o %s%c", (unsigned int)e->versions[v].mode,
					e->name->str_dat, '\0');
		strbuf_add(b, e->versions[v].sha1, 20);
1429
	}
1430 1431 1432 1433 1434 1435
}

static void store_tree(struct tree_entry *root)
{
	struct tree_content *t = root->tree;
	unsigned int i, j, del;
P
Pierre Habouzit 已提交
1436
	struct last_object lo = { STRBUF_INIT, 0, 0, /* no_swap */ 1 };
1437
	struct object_entry *le;
1438 1439 1440 1441 1442 1443 1444 1445 1446

	if (!is_null_sha1(root->versions[1].sha1))
		return;

	for (i = 0; i < t->entry_count; i++) {
		if (t->entries[i]->tree)
			store_tree(t->entries[i]);
	}

1447
	le = find_object(root->versions[0].sha1);
P
Pierre Habouzit 已提交
1448
	if (S_ISDIR(root->versions[0].mode) && le && le->pack_id == pack_id) {
1449
		mktree(t, 0, &old_tree);
P
Pierre Habouzit 已提交
1450
		lo.data = old_tree;
1451
		lo.offset = le->idx.offset;
1452 1453 1454
		lo.depth = t->delta_depth;
	}

1455 1456
	mktree(t, 1, &new_tree);
	store_object(OBJ_TREE, &new_tree, &lo, root->versions[1].sha1, 0);
1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470

	t->delta_depth = lo.depth;
	for (i = 0, j = 0, del = 0; i < t->entry_count; i++) {
		struct tree_entry *e = t->entries[i];
		if (e->versions[1].mode) {
			e->versions[0].mode = e->versions[1].mode;
			hashcpy(e->versions[0].sha1, e->versions[1].sha1);
			t->entries[j++] = e;
		} else {
			release_tree_entry(e);
			del++;
		}
	}
	t->entry_count -= del;
1471 1472
}

1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
static void tree_content_replace(
	struct tree_entry *root,
	const unsigned char *sha1,
	const uint16_t mode,
	struct tree_content *newtree)
{
	if (!S_ISDIR(mode))
		die("Root cannot be a non-directory");
	hashcpy(root->versions[1].sha1, sha1);
	if (root->tree)
		release_tree_content_recursive(root->tree);
	root->tree = newtree;
}

1487 1488 1489 1490
static int tree_content_set(
	struct tree_entry *root,
	const char *p,
	const unsigned char *sha1,
1491 1492
	const uint16_t mode,
	struct tree_content *subtree)
1493
{
1494
	struct tree_content *t;
1495 1496 1497 1498 1499 1500 1501 1502 1503
	const char *slash1;
	unsigned int i, n;
	struct tree_entry *e;

	slash1 = strchr(p, '/');
	if (slash1)
		n = slash1 - p;
	else
		n = strlen(p);
1504 1505
	if (!n)
		die("Empty path component found in input");
1506 1507
	if (!slash1 && !S_ISDIR(mode) && subtree)
		die("Non-directories cannot have subtrees");
1508

1509 1510 1511
	if (!root->tree)
		load_tree(root);
	t = root->tree;
1512 1513
	for (i = 0; i < t->entry_count; i++) {
		e = t->entries[i];
1514
		if (e->name->str_len == n && !strncmp_icase(p, e->name->str_dat, n)) {
1515
			if (!slash1) {
1516 1517
				if (!S_ISDIR(mode)
						&& e->versions[1].mode == mode
1518
						&& !hashcmp(e->versions[1].sha1, sha1))
1519
					return 0;
1520 1521
				e->versions[1].mode = mode;
				hashcpy(e->versions[1].sha1, sha1);
1522
				if (e->tree)
1523
					release_tree_content_recursive(e->tree);
1524
				e->tree = subtree;
1525
				hashclr(root->versions[1].sha1);
1526 1527
				return 1;
			}
1528
			if (!S_ISDIR(e->versions[1].mode)) {
1529
				e->tree = new_tree_content(8);
1530
				e->versions[1].mode = S_IFDIR;
1531 1532 1533
			}
			if (!e->tree)
				load_tree(e);
1534
			if (tree_content_set(e, slash1 + 1, sha1, mode, subtree)) {
1535
				hashclr(root->versions[1].sha1);
1536 1537 1538 1539 1540 1541 1542
				return 1;
			}
			return 0;
		}
	}

	if (t->entry_count == t->entry_capacity)
1543
		root->tree = t = grow_tree_content(t, t->entry_count);
1544
	e = new_tree_entry();
1545
	e->name = to_atom(p, n);
1546 1547
	e->versions[0].mode = 0;
	hashclr(e->versions[0].sha1);
1548 1549 1550
	t->entries[t->entry_count++] = e;
	if (slash1) {
		e->tree = new_tree_content(8);
1551
		e->versions[1].mode = S_IFDIR;
1552
		tree_content_set(e, slash1 + 1, sha1, mode, subtree);
1553
	} else {
1554
		e->tree = subtree;
1555 1556
		e->versions[1].mode = mode;
		hashcpy(e->versions[1].sha1, sha1);
1557
	}
1558
	hashclr(root->versions[1].sha1);
1559 1560 1561
	return 1;
}

1562 1563 1564 1565
static int tree_content_remove(
	struct tree_entry *root,
	const char *p,
	struct tree_entry *backup_leaf)
1566
{
1567
	struct tree_content *t;
1568 1569 1570 1571 1572 1573 1574 1575 1576 1577
	const char *slash1;
	unsigned int i, n;
	struct tree_entry *e;

	slash1 = strchr(p, '/');
	if (slash1)
		n = slash1 - p;
	else
		n = strlen(p);

1578 1579 1580
	if (!root->tree)
		load_tree(root);
	t = root->tree;
1581 1582
	for (i = 0; i < t->entry_count; i++) {
		e = t->entries[i];
1583
		if (e->name->str_len == n && !strncmp_icase(p, e->name->str_dat, n)) {
1584 1585 1586 1587 1588 1589 1590 1591
			if (slash1 && !S_ISDIR(e->versions[1].mode))
				/*
				 * If p names a file in some subdirectory, and a
				 * file or symlink matching the name of the
				 * parent directory of p exists, then p cannot
				 * exist and need not be deleted.
				 */
				return 1;
1592
			if (!slash1 || !S_ISDIR(e->versions[1].mode))
1593 1594 1595
				goto del_entry;
			if (!e->tree)
				load_tree(e);
1596
			if (tree_content_remove(e, slash1 + 1, backup_leaf)) {
1597 1598 1599 1600 1601 1602
				for (n = 0; n < e->tree->entry_count; n++) {
					if (e->tree->entries[n]->versions[1].mode) {
						hashclr(root->versions[1].sha1);
						return 1;
					}
				}
1603
				backup_leaf = NULL;
1604
				goto del_entry;
1605 1606 1607 1608 1609 1610 1611
			}
			return 0;
		}
	}
	return 0;

del_entry:
1612 1613 1614
	if (backup_leaf)
		memcpy(backup_leaf, e, sizeof(*backup_leaf));
	else if (e->tree)
1615
		release_tree_content_recursive(e->tree);
1616
	e->tree = NULL;
1617 1618 1619
	e->versions[1].mode = 0;
	hashclr(e->versions[1].sha1);
	hashclr(root->versions[1].sha1);
1620
	return 1;
1621 1622
}

1623 1624 1625 1626 1627
static int tree_content_get(
	struct tree_entry *root,
	const char *p,
	struct tree_entry *leaf)
{
1628
	struct tree_content *t;
1629 1630 1631 1632 1633 1634 1635 1636 1637 1638
	const char *slash1;
	unsigned int i, n;
	struct tree_entry *e;

	slash1 = strchr(p, '/');
	if (slash1)
		n = slash1 - p;
	else
		n = strlen(p);

1639 1640 1641
	if (!root->tree)
		load_tree(root);
	t = root->tree;
1642 1643
	for (i = 0; i < t->entry_count; i++) {
		e = t->entries[i];
1644
		if (e->name->str_len == n && !strncmp_icase(p, e->name->str_dat, n)) {
1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662
			if (!slash1) {
				memcpy(leaf, e, sizeof(*leaf));
				if (e->tree && is_null_sha1(e->versions[1].sha1))
					leaf->tree = dup_tree_content(e->tree);
				else
					leaf->tree = NULL;
				return 1;
			}
			if (!S_ISDIR(e->versions[1].mode))
				return 0;
			if (!e->tree)
				load_tree(e);
			return tree_content_get(e, slash1 + 1, leaf);
		}
	}
	return 0;
}

1663
static int update_branch(struct branch *b)
1664 1665
{
	static const char *msg = "fast-import";
1666 1667 1668
	struct ref_lock *lock;
	unsigned char old_sha1[20];

1669 1670
	if (is_null_sha1(b->sha1))
		return 0;
1671 1672
	if (read_ref(b->name, old_sha1))
		hashclr(old_sha1);
1673
	lock = lock_any_ref_for_update(b->name, old_sha1, 0);
1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685
	if (!lock)
		return error("Unable to lock %s", b->name);
	if (!force_update && !is_null_sha1(old_sha1)) {
		struct commit *old_cmit, *new_cmit;

		old_cmit = lookup_commit_reference_gently(old_sha1, 0);
		new_cmit = lookup_commit_reference_gently(b->sha1, 0);
		if (!old_cmit || !new_cmit) {
			unlock_ref(lock);
			return error("Branch %s is missing commits.", b->name);
		}

1686
		if (!in_merge_bases(old_cmit, &new_cmit, 1)) {
1687
			unlock_ref(lock);
1688
			warning("Not updating %s"
1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700
				" (new tip %s does not contain %s)",
				b->name, sha1_to_hex(b->sha1), sha1_to_hex(old_sha1));
			return -1;
		}
	}
	if (write_ref_sha1(lock, b->sha1, msg) < 0)
		return error("Unable to update %s", b->name);
	return 0;
}

static void dump_branches(void)
{
1701 1702 1703 1704
	unsigned int i;
	struct branch *b;

	for (i = 0; i < branch_table_sz; i++) {
1705 1706
		for (b = branch_table[i]; b; b = b->table_next_branch)
			failure |= update_branch(b);
1707 1708 1709
	}
}

1710
static void dump_tags(void)
1711 1712 1713 1714
{
	static const char *msg = "fast-import";
	struct tag *t;
	struct ref_lock *lock;
1715
	char ref_name[PATH_MAX];
1716 1717

	for (t = first_tag; t; t = t->next_tag) {
1718 1719
		sprintf(ref_name, "tags/%s", t->name);
		lock = lock_ref_sha1(ref_name, NULL);
1720
		if (!lock || write_ref_sha1(lock, t->sha1, msg) < 0)
1721
			failure |= error("Unable to update %s", ref_name);
1722 1723 1724
	}
}

1725
static void dump_marks_helper(FILE *f,
1726
	uintmax_t base,
1727 1728
	struct mark_set *m)
{
1729
	uintmax_t k;
1730 1731 1732
	if (m->shift) {
		for (k = 0; k < 1024; k++) {
			if (m->data.sets[k])
1733
				dump_marks_helper(f, base + (k << m->shift),
1734 1735 1736 1737 1738
					m->data.sets[k]);
		}
	} else {
		for (k = 0; k < 1024; k++) {
			if (m->data.marked[k])
1739
				fprintf(f, ":%" PRIuMAX " %s\n", base + k,
1740
					sha1_to_hex(m->data.marked[k]->idx.sha1));
1741 1742 1743 1744
		}
	}
}

1745
static void dump_marks(void)
1746
{
1747 1748 1749 1750
	static struct lock_file mark_lock;
	int mark_fd;
	FILE *f;

1751
	if (!export_marks_file)
1752 1753
		return;

1754
	mark_fd = hold_lock_file_for_update(&mark_lock, export_marks_file, 0);
1755 1756
	if (mark_fd < 0) {
		failure |= error("Unable to write marks file %s: %s",
1757
			export_marks_file, strerror(errno));
1758
		return;
1759
	}
1760 1761 1762

	f = fdopen(mark_fd, "w");
	if (!f) {
1763
		int saved_errno = errno;
1764 1765
		rollback_lock_file(&mark_lock);
		failure |= error("Unable to write marks file %s: %s",
1766
			export_marks_file, strerror(saved_errno));
1767
		return;
1768
	}
1769

B
Brandon Casey 已提交
1770
	/*
1771 1772
	 * Since the lock file was fdopen()'ed, it should not be close()'ed.
	 * Assign -1 to the lock file descriptor so that commit_lock_file()
B
Brandon Casey 已提交
1773 1774 1775
	 * won't try to close() it.
	 */
	mark_lock.fd = -1;
1776 1777 1778

	dump_marks_helper(f, 0, marks);
	if (ferror(f) || fclose(f)) {
1779
		int saved_errno = errno;
1780 1781
		rollback_lock_file(&mark_lock);
		failure |= error("Unable to write marks file %s: %s",
1782
			export_marks_file, strerror(saved_errno));
1783 1784 1785 1786
		return;
	}

	if (commit_lock_file(&mark_lock)) {
1787
		int saved_errno = errno;
1788 1789
		rollback_lock_file(&mark_lock);
		failure |= error("Unable to commit marks file %s: %s",
1790
			export_marks_file, strerror(saved_errno));
1791 1792
		return;
	}
1793 1794
}

1795 1796 1797 1798
static void read_marks(void)
{
	char line[512];
	FILE *f = fopen(import_marks_file, "r");
1799 1800 1801 1802 1803
	if (f)
		;
	else if (import_marks_file_ignore_missing && errno == ENOENT)
		return; /* Marks file does not exist */
	else
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826
		die_errno("cannot read '%s'", import_marks_file);
	while (fgets(line, sizeof(line), f)) {
		uintmax_t mark;
		char *end;
		unsigned char sha1[20];
		struct object_entry *e;

		end = strchr(line, '\n');
		if (line[0] != ':' || !end)
			die("corrupt mark line: %s", line);
		*end = 0;
		mark = strtoumax(line + 1, &end, 10);
		if (!mark || end == line + 1
			|| *end != ' ' || get_sha1(end + 1, sha1))
			die("corrupt mark line: %s", line);
		e = find_object(sha1);
		if (!e) {
			enum object_type type = sha1_object_info(sha1, NULL);
			if (type < 0)
				die("object not found: %s", sha1_to_hex(sha1));
			e = insert_object(sha1);
			e->type = type;
			e->pack_id = MAX_PACK_ID;
1827
			e->idx.offset = 1; /* just not zero! */
1828 1829 1830 1831 1832 1833 1834
		}
		insert_mark(mark, e);
	}
	fclose(f);
}


1835
static int read_next_command(void)
1836
{
1837 1838 1839 1840 1841 1842 1843
	static int stdin_eof = 0;

	if (stdin_eof) {
		unread_command_buf = 0;
		return EOF;
	}

1844
	for (;;) {
1845
		if (unread_command_buf) {
1846
			unread_command_buf = 0;
1847 1848 1849
		} else {
			struct recent_command *rc;

1850
			strbuf_detach(&command_buf, NULL);
1851 1852 1853
			stdin_eof = strbuf_getline(&command_buf, stdin, '\n');
			if (stdin_eof)
				return EOF;
1854

1855
			if (!seen_data_command
S
Sverre Rabbelier 已提交
1856 1857 1858
				&& prefixcmp(command_buf.buf, "feature ")
				&& prefixcmp(command_buf.buf, "option ")) {
				parse_argv();
1859 1860
			}

1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
			rc = rc_free;
			if (rc)
				rc_free = rc->next;
			else {
				rc = cmd_hist.next;
				cmd_hist.next = rc->next;
				cmd_hist.next->prev = &cmd_hist;
				free(rc->buf);
			}

			rc->buf = command_buf.buf;
			rc->prev = cmd_tail;
			rc->next = cmd_hist.prev;
			rc->prev->next = rc;
			cmd_tail = rc;
		}
1877 1878 1879 1880 1881 1882 1883 1884
		if (!prefixcmp(command_buf.buf, "cat-blob ")) {
			parse_cat_blob();
			continue;
		}
		if (command_buf.buf[0] == '#')
			continue;
		return 0;
	}
1885 1886
}

J
Junio C Hamano 已提交
1887
static void skip_optional_lf(void)
1888 1889 1890 1891 1892 1893
{
	int term_char = fgetc(stdin);
	if (term_char != '\n' && term_char != EOF)
		ungetc(term_char, stdin);
}

1894
static void parse_mark(void)
1895
{
1896
	if (!prefixcmp(command_buf.buf, "mark :")) {
1897
		next_mark = strtoumax(command_buf.buf + 6, NULL, 10);
1898 1899 1900
		read_next_command();
	}
	else
1901
		next_mark = 0;
1902 1903
}

1904
static int parse_data(struct strbuf *sb, uintmax_t limit, uintmax_t *len_res)
1905
{
1906
	strbuf_reset(sb);
1907

1908
	if (prefixcmp(command_buf.buf, "data "))
1909 1910
		die("Expected 'data n' command, found: %s", command_buf.buf);

1911
	if (!prefixcmp(command_buf.buf + 5, "<<")) {
1912
		char *term = xstrdup(command_buf.buf + 5 + 2);
1913 1914
		size_t term_len = command_buf.len - 5 - 2;

1915
		strbuf_detach(&command_buf, NULL);
1916
		for (;;) {
1917
			if (strbuf_getline(&command_buf, stdin, '\n') == EOF)
1918 1919 1920 1921
				die("EOF in data (terminator '%s' not found)", term);
			if (term_len == command_buf.len
				&& !strcmp(term, command_buf.buf))
				break;
1922 1923
			strbuf_addbuf(sb, &command_buf);
			strbuf_addch(sb, '\n');
1924 1925 1926 1927
		}
		free(term);
	}
	else {
1928 1929
		uintmax_t len = strtoumax(command_buf.buf + 5, NULL, 10);
		size_t n = 0, length = (size_t)len;
1930

1931 1932 1933 1934 1935 1936
		if (limit && limit < len) {
			*len_res = len;
			return 0;
		}
		if (length < len)
			die("data is too large to use in this context");
1937

1938
		while (n < length) {
1939
			size_t s = strbuf_fread(sb, length - n, stdin);
1940
			if (!s && feof(stdin))
1941 1942
				die("EOF in data (%lu bytes remaining)",
					(unsigned long)(length - n));
1943 1944
			n += s;
		}
1945 1946
	}

1947
	skip_optional_lf();
1948
	return 1;
1949 1950
}

1951 1952 1953
static int validate_raw_date(const char *src, char *result, int maxlen)
{
	const char *orig_src = src;
1954
	char *endp;
1955
	unsigned long num;
1956

1957 1958
	errno = 0;

1959 1960
	num = strtoul(src, &endp, 10);
	/* NEEDSWORK: perhaps check for reasonable values? */
1961
	if (errno || endp == src || *endp != ' ')
1962 1963 1964 1965 1966 1967
		return -1;

	src = endp + 1;
	if (*src != '-' && *src != '+')
		return -1;

1968 1969 1970
	num = strtoul(src + 1, &endp, 10);
	if (errno || endp == src + 1 || *endp || (endp - orig_src) >= maxlen ||
	    1400 < num)
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
		return -1;

	strcpy(result, orig_src);
	return 0;
}

static char *parse_ident(const char *buf)
{
	const char *gt;
	size_t name_len;
	char *ident;

	gt = strrchr(buf, '>');
	if (!gt)
		die("Missing > in ident string: %s", buf);
	gt++;
	if (*gt != ' ')
		die("Missing space after > in ident string: %s", buf);
	gt++;
	name_len = gt - buf;
	ident = xmalloc(name_len + 24);
	strncpy(ident, buf, name_len);

	switch (whenspec) {
	case WHENSPEC_RAW:
		if (validate_raw_date(gt, ident + name_len, 24) < 0)
			die("Invalid raw date \"%s\" in ident: %s", gt, buf);
		break;
	case WHENSPEC_RFC2822:
		if (parse_date(gt, ident + name_len, 24) < 0)
			die("Invalid rfc2822 date \"%s\" in ident: %s", gt, buf);
		break;
	case WHENSPEC_NOW:
		if (strcmp("now", gt))
			die("Date in ident must be 'now': %s", buf);
		datestamp(ident + name_len, 24);
		break;
	}

	return ident;
}

2013 2014 2015 2016
static void parse_and_store_blob(
	struct last_object *last,
	unsigned char *sha1out,
	uintmax_t mark)
2017
{
P
Pierre Habouzit 已提交
2018
	static struct strbuf buf = STRBUF_INIT;
2019
	uintmax_t len;
2020

2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035
	if (parse_data(&buf, big_file_threshold, &len))
		store_object(OBJ_BLOB, &buf, last, sha1out, mark);
	else {
		if (last) {
			strbuf_release(&last->data);
			last->offset = 0;
			last->depth = 0;
		}
		stream_blob(len, sha1out, mark);
		skip_optional_lf();
	}
}

static void parse_new_blob(void)
{
2036
	read_next_command();
2037
	parse_mark();
2038
	parse_and_store_blob(&last_blob, NULL, next_mark);
2039 2040
}

2041
static void unload_one_branch(void)
2042
{
2043 2044
	while (cur_active_branches
		&& cur_active_branches >= max_active_branches) {
2045
		uintmax_t min_commit = ULONG_MAX;
2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062
		struct branch *e, *l = NULL, *p = NULL;

		for (e = active_branches; e; e = e->active_next_branch) {
			if (e->last_commit < min_commit) {
				p = l;
				min_commit = e->last_commit;
			}
			l = e;
		}

		if (p) {
			e = p->active_next_branch;
			p->active_next_branch = e->active_next_branch;
		} else {
			e = active_branches;
			active_branches = e->active_next_branch;
		}
2063
		e->active = 0;
2064 2065
		e->active_next_branch = NULL;
		if (e->branch_tree.tree) {
2066
			release_tree_content_recursive(e->branch_tree.tree);
2067 2068 2069
			e->branch_tree.tree = NULL;
		}
		cur_active_branches--;
2070 2071 2072
	}
}

2073
static void load_branch(struct branch *b)
2074
{
2075
	load_tree(&b->branch_tree);
2076 2077 2078 2079 2080 2081 2082
	if (!b->active) {
		b->active = 1;
		b->active_next_branch = active_branches;
		active_branches = b;
		cur_active_branches++;
		branch_load_count++;
	}
2083 2084
}

2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
static unsigned char convert_num_notes_to_fanout(uintmax_t num_notes)
{
	unsigned char fanout = 0;
	while ((num_notes >>= 8))
		fanout++;
	return fanout;
}

static void construct_path_with_fanout(const char *hex_sha1,
		unsigned char fanout, char *path)
{
	unsigned int i = 0, j = 0;
	if (fanout >= 20)
		die("Too large fanout (%u)", fanout);
	while (fanout) {
		path[i++] = hex_sha1[j++];
		path[i++] = hex_sha1[j++];
		path[i++] = '/';
		fanout--;
	}
	memcpy(path + i, hex_sha1 + j, 40 - j);
	path[i + 40 - j] = '\0';
}

static uintmax_t do_change_note_fanout(
		struct tree_entry *orig_root, struct tree_entry *root,
		char *hex_sha1, unsigned int hex_sha1_len,
		char *fullpath, unsigned int fullpath_len,
		unsigned char fanout)
{
	struct tree_content *t = root->tree;
	struct tree_entry *e, leaf;
	unsigned int i, tmp_hex_sha1_len, tmp_fullpath_len;
	uintmax_t num_notes = 0;
	unsigned char sha1[20];
	char realpath[60];

	for (i = 0; t && i < t->entry_count; i++) {
		e = t->entries[i];
		tmp_hex_sha1_len = hex_sha1_len + e->name->str_len;
		tmp_fullpath_len = fullpath_len;

		/*
		 * We're interested in EITHER existing note entries (entries
		 * with exactly 40 hex chars in path, not including directory
		 * separators), OR directory entries that may contain note
		 * entries (with < 40 hex chars in path).
		 * Also, each path component in a note entry must be a multiple
		 * of 2 chars.
		 */
		if (!e->versions[1].mode ||
		    tmp_hex_sha1_len > 40 ||
		    e->name->str_len % 2)
			continue;

		/* This _may_ be a note entry, or a subdir containing notes */
		memcpy(hex_sha1 + hex_sha1_len, e->name->str_dat,
		       e->name->str_len);
		if (tmp_fullpath_len)
			fullpath[tmp_fullpath_len++] = '/';
		memcpy(fullpath + tmp_fullpath_len, e->name->str_dat,
		       e->name->str_len);
		tmp_fullpath_len += e->name->str_len;
		fullpath[tmp_fullpath_len] = '\0';

		if (tmp_hex_sha1_len == 40 && !get_sha1_hex(hex_sha1, sha1)) {
			/* This is a note entry */
			construct_path_with_fanout(hex_sha1, fanout, realpath);
			if (!strcmp(fullpath, realpath)) {
				/* Note entry is in correct location */
				num_notes++;
				continue;
			}

			/* Rename fullpath to realpath */
			if (!tree_content_remove(orig_root, fullpath, &leaf))
				die("Failed to remove path %s", fullpath);
			tree_content_set(orig_root, realpath,
				leaf.versions[1].sha1,
				leaf.versions[1].mode,
				leaf.tree);
		} else if (S_ISDIR(e->versions[1].mode)) {
			/* This is a subdir that may contain note entries */
			if (!e->tree)
				load_tree(e);
			num_notes += do_change_note_fanout(orig_root, e,
				hex_sha1, tmp_hex_sha1_len,
				fullpath, tmp_fullpath_len, fanout);
		}

		/* The above may have reallocated the current tree_content */
		t = root->tree;
	}
	return num_notes;
}

static uintmax_t change_note_fanout(struct tree_entry *root,
		unsigned char fanout)
{
	char hex_sha1[40], path[60];
	return do_change_note_fanout(root, root, hex_sha1, 0, path, 0, fanout);
}

2188
static void file_change_m(struct branch *b)
2189
{
2190
	const char *p = command_buf.buf + 2;
2191
	static struct strbuf uq = STRBUF_INIT;
2192
	const char *endp;
2193
	struct object_entry *oe = oe;
2194
	unsigned char sha1[20];
2195
	uint16_t mode, inline_data = 0;
2196

2197 2198 2199 2200
	p = get_mode(p, &mode);
	if (!p)
		die("Corrupt mode: %s", command_buf.buf);
	switch (mode) {
2201 2202 2203
	case 0644:
	case 0755:
		mode |= S_IFREG;
2204 2205
	case S_IFREG | 0644:
	case S_IFREG | 0755:
2206
	case S_IFLNK:
2207
	case S_IFDIR:
2208
	case S_IFGITLINK:
2209 2210 2211 2212 2213 2214
		/* ok */
		break;
	default:
		die("Corrupt mode: %s", command_buf.buf);
	}

2215 2216
	if (*p == ':') {
		char *x;
2217
		oe = find_mark(strtoumax(p + 1, &x, 10));
2218
		hashcpy(sha1, oe->idx.sha1);
2219
		p = x;
2220
	} else if (!prefixcmp(p, "inline")) {
2221 2222
		inline_data = 1;
		p += 6;
2223 2224 2225 2226 2227 2228
	} else {
		if (get_sha1_hex(p, sha1))
			die("Invalid SHA1: %s", command_buf.buf);
		oe = find_object(sha1);
		p += 40;
	}
2229 2230 2231
	if (*p++ != ' ')
		die("Missing space after SHA1: %s", command_buf.buf);

2232 2233
	strbuf_reset(&uq);
	if (!unquote_c_style(&uq, p, &endp)) {
2234 2235
		if (*endp)
			die("Garbage after path in: %s", command_buf.buf);
2236
		p = uq.buf;
2237
	}
2238

2239 2240 2241 2242 2243 2244
	/* Git does not track empty, non-toplevel directories. */
	if (S_ISDIR(mode) && !memcmp(sha1, EMPTY_TREE_SHA1_BIN, 20) && *p) {
		tree_content_remove(&b->branch_tree, p, NULL);
		return;
	}

2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258
	if (S_ISGITLINK(mode)) {
		if (inline_data)
			die("Git links cannot be specified 'inline': %s",
				command_buf.buf);
		else if (oe) {
			if (oe->type != OBJ_COMMIT)
				die("Not a commit (actually a %s): %s",
					typename(oe->type), command_buf.buf);
		}
		/*
		 * Accept the sha1 without checking; it expected to be in
		 * another repository.
		 */
	} else if (inline_data) {
2259 2260 2261
		if (S_ISDIR(mode))
			die("Directories cannot be specified 'inline': %s",
				command_buf.buf);
2262 2263 2264 2265
		if (p != uq.buf) {
			strbuf_addstr(&uq, p);
			p = uq.buf;
		}
2266
		read_next_command();
2267
		parse_and_store_blob(&last_blob, sha1, 0);
2268
	} else {
2269 2270 2271 2272
		enum object_type expected = S_ISDIR(mode) ?
						OBJ_TREE: OBJ_BLOB;
		enum object_type type = oe ? oe->type :
					sha1_object_info(sha1, NULL);
2273
		if (type < 0)
2274 2275 2276 2277 2278 2279 2280
			die("%s not found: %s",
					S_ISDIR(mode) ?  "Tree" : "Blob",
					command_buf.buf);
		if (type != expected)
			die("Not a %s (actually a %s): %s",
				typename(expected), typename(type),
				command_buf.buf);
2281
	}
2282

2283 2284 2285 2286
	if (!*p) {
		tree_content_replace(&b->branch_tree, sha1, mode, NULL);
		return;
	}
2287
	tree_content_set(&b->branch_tree, p, sha1, mode, NULL);
2288
}
2289

2290 2291
static void file_change_d(struct branch *b)
{
2292
	const char *p = command_buf.buf + 2;
2293
	static struct strbuf uq = STRBUF_INIT;
2294 2295
	const char *endp;

2296 2297
	strbuf_reset(&uq);
	if (!unquote_c_style(&uq, p, &endp)) {
2298 2299
		if (*endp)
			die("Garbage after path in: %s", command_buf.buf);
2300
		p = uq.buf;
2301
	}
2302
	tree_content_remove(&b->branch_tree, p, NULL);
2303 2304
}

2305
static void file_change_cr(struct branch *b, int rename)
2306 2307
{
	const char *s, *d;
2308 2309
	static struct strbuf s_uq = STRBUF_INIT;
	static struct strbuf d_uq = STRBUF_INIT;
2310 2311 2312 2313
	const char *endp;
	struct tree_entry leaf;

	s = command_buf.buf + 2;
2314 2315
	strbuf_reset(&s_uq);
	if (!unquote_c_style(&s_uq, s, &endp)) {
2316 2317
		if (*endp != ' ')
			die("Missing space after source: %s", command_buf.buf);
2318
	} else {
2319 2320 2321
		endp = strchr(s, ' ');
		if (!endp)
			die("Missing space after source: %s", command_buf.buf);
2322
		strbuf_add(&s_uq, s, endp - s);
2323
	}
2324
	s = s_uq.buf;
2325 2326 2327 2328 2329 2330

	endp++;
	if (!*endp)
		die("Missing dest: %s", command_buf.buf);

	d = endp;
2331 2332
	strbuf_reset(&d_uq);
	if (!unquote_c_style(&d_uq, d, &endp)) {
2333 2334
		if (*endp)
			die("Garbage after dest in: %s", command_buf.buf);
2335
		d = d_uq.buf;
2336 2337 2338
	}

	memset(&leaf, 0, sizeof(leaf));
2339 2340 2341 2342
	if (rename)
		tree_content_remove(&b->branch_tree, s, &leaf);
	else
		tree_content_get(&b->branch_tree, s, &leaf);
2343 2344
	if (!leaf.versions[1].mode)
		die("Path %s not in branch", s);
2345 2346 2347 2348 2349 2350 2351
	if (!*d) {	/* C "path/to/subdir" "" */
		tree_content_replace(&b->branch_tree,
			leaf.versions[1].sha1,
			leaf.versions[1].mode,
			leaf.tree);
		return;
	}
2352 2353 2354 2355 2356 2357
	tree_content_set(&b->branch_tree, d,
		leaf.versions[1].sha1,
		leaf.versions[1].mode,
		leaf.tree);
}

2358
static void note_change_n(struct branch *b, unsigned char old_fanout)
2359 2360 2361 2362 2363 2364
{
	const char *p = command_buf.buf + 2;
	static struct strbuf uq = STRBUF_INIT;
	struct object_entry *oe = oe;
	struct branch *s;
	unsigned char sha1[20], commit_sha1[20];
2365
	char path[60];
2366
	uint16_t inline_data = 0;
2367
	unsigned char new_fanout;
2368 2369 2370 2371 2372

	/* <dataref> or 'inline' */
	if (*p == ':') {
		char *x;
		oe = find_mark(strtoumax(p + 1, &x, 10));
2373
		hashcpy(sha1, oe->idx.sha1);
2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395
		p = x;
	} else if (!prefixcmp(p, "inline")) {
		inline_data = 1;
		p += 6;
	} else {
		if (get_sha1_hex(p, sha1))
			die("Invalid SHA1: %s", command_buf.buf);
		oe = find_object(sha1);
		p += 40;
	}
	if (*p++ != ' ')
		die("Missing space after SHA1: %s", command_buf.buf);

	/* <committish> */
	s = lookup_branch(p);
	if (s) {
		hashcpy(commit_sha1, s->sha1);
	} else if (*p == ':') {
		uintmax_t commit_mark = strtoumax(p + 1, NULL, 10);
		struct object_entry *commit_oe = find_mark(commit_mark);
		if (commit_oe->type != OBJ_COMMIT)
			die("Mark :%" PRIuMAX " not a commit", commit_mark);
2396
		hashcpy(commit_sha1, commit_oe->idx.sha1);
2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412
	} else if (!get_sha1(p, commit_sha1)) {
		unsigned long size;
		char *buf = read_object_with_reference(commit_sha1,
			commit_type, &size, commit_sha1);
		if (!buf || size < 46)
			die("Not a valid commit: %s", p);
		free(buf);
	} else
		die("Invalid ref name or SHA1 expression: %s", p);

	if (inline_data) {
		if (p != uq.buf) {
			strbuf_addstr(&uq, p);
			p = uq.buf;
		}
		read_next_command();
2413
		parse_and_store_blob(&last_blob, sha1, 0);
2414 2415 2416 2417
	} else if (oe) {
		if (oe->type != OBJ_BLOB)
			die("Not a blob (actually a %s): %s",
				typename(oe->type), command_buf.buf);
2418
	} else if (!is_null_sha1(sha1)) {
2419 2420 2421 2422 2423 2424 2425 2426
		enum object_type type = sha1_object_info(sha1, NULL);
		if (type < 0)
			die("Blob not found: %s", command_buf.buf);
		if (type != OBJ_BLOB)
			die("Not a blob (actually a %s): %s",
			    typename(type), command_buf.buf);
	}

2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437
	construct_path_with_fanout(sha1_to_hex(commit_sha1), old_fanout, path);
	if (tree_content_remove(&b->branch_tree, path, NULL))
		b->num_notes--;

	if (is_null_sha1(sha1))
		return; /* nothing to insert */

	b->num_notes++;
	new_fanout = convert_num_notes_to_fanout(b->num_notes);
	construct_path_with_fanout(sha1_to_hex(commit_sha1), new_fanout, path);
	tree_content_set(&b->branch_tree, path, sha1, S_IFREG | 0644, NULL);
2438 2439
}

2440 2441 2442 2443 2444 2445
static void file_change_deleteall(struct branch *b)
{
	release_tree_content_recursive(b->branch_tree.tree);
	hashclr(b->branch_tree.versions[0].sha1);
	hashclr(b->branch_tree.versions[1].sha1);
	load_tree(&b->branch_tree);
2446
	b->num_notes = 0;
2447 2448
}

2449
static void parse_from_commit(struct branch *b, char *buf, unsigned long size)
2450 2451 2452 2453 2454 2455 2456 2457 2458 2459
{
	if (!buf || size < 46)
		die("Not a valid commit: %s", sha1_to_hex(b->sha1));
	if (memcmp("tree ", buf, 5)
		|| get_sha1_hex(buf + 5, b->branch_tree.versions[1].sha1))
		die("The commit %s is corrupt", sha1_to_hex(b->sha1));
	hashcpy(b->branch_tree.versions[0].sha1,
		b->branch_tree.versions[1].sha1);
}

2460
static void parse_from_existing(struct branch *b)
2461 2462 2463 2464 2465 2466 2467 2468 2469 2470
{
	if (is_null_sha1(b->sha1)) {
		hashclr(b->branch_tree.versions[0].sha1);
		hashclr(b->branch_tree.versions[1].sha1);
	} else {
		unsigned long size;
		char *buf;

		buf = read_object_with_reference(b->sha1,
			commit_type, &size, b->sha1);
2471
		parse_from_commit(b, buf, size);
2472 2473 2474 2475
		free(buf);
	}
}

2476
static int parse_from(struct branch *b)
2477
{
2478
	const char *from;
2479 2480
	struct branch *s;

2481
	if (prefixcmp(command_buf.buf, "from "))
2482
		return 0;
2483

2484 2485 2486 2487
	if (b->branch_tree.tree) {
		release_tree_content_recursive(b->branch_tree.tree);
		b->branch_tree.tree = NULL;
	}
2488 2489 2490 2491 2492 2493

	from = strchr(command_buf.buf, ' ') + 1;
	s = lookup_branch(from);
	if (b == s)
		die("Can't create a branch from itself: %s", b->name);
	else if (s) {
2494
		unsigned char *t = s->branch_tree.versions[1].sha1;
2495
		hashcpy(b->sha1, s->sha1);
2496 2497
		hashcpy(b->branch_tree.versions[0].sha1, t);
		hashcpy(b->branch_tree.versions[1].sha1, t);
2498
	} else if (*from == ':') {
2499
		uintmax_t idnum = strtoumax(from + 1, NULL, 10);
2500 2501
		struct object_entry *oe = find_mark(idnum);
		if (oe->type != OBJ_COMMIT)
2502
			die("Mark :%" PRIuMAX " not a commit", idnum);
2503
		hashcpy(b->sha1, oe->idx.sha1);
2504
		if (oe->pack_id != MAX_PACK_ID) {
2505
			unsigned long size;
2506
			char *buf = gfi_unpack_entry(oe, &size);
2507
			parse_from_commit(b, buf, size);
2508
			free(buf);
2509
		} else
2510
			parse_from_existing(b);
2511
	} else if (!get_sha1(from, b->sha1))
2512
		parse_from_existing(b);
2513
	else
2514 2515 2516
		die("Invalid ref name or SHA1 expression: %s", from);

	read_next_command();
2517
	return 1;
2518 2519
}

2520
static struct hash_list *parse_merge(unsigned int *count)
2521
{
2522
	struct hash_list *list = NULL, *n, *e = e;
2523
	const char *from;
2524 2525 2526
	struct branch *s;

	*count = 0;
2527
	while (!prefixcmp(command_buf.buf, "merge ")) {
2528 2529 2530 2531 2532 2533
		from = strchr(command_buf.buf, ' ') + 1;
		n = xmalloc(sizeof(*n));
		s = lookup_branch(from);
		if (s)
			hashcpy(n->sha1, s->sha1);
		else if (*from == ':') {
2534
			uintmax_t idnum = strtoumax(from + 1, NULL, 10);
2535 2536
			struct object_entry *oe = find_mark(idnum);
			if (oe->type != OBJ_COMMIT)
2537
				die("Mark :%" PRIuMAX " not a commit", idnum);
2538
			hashcpy(n->sha1, oe->idx.sha1);
2539 2540 2541
		} else if (!get_sha1(from, n->sha1)) {
			unsigned long size;
			char *buf = read_object_with_reference(n->sha1,
S
Shawn O. Pearce 已提交
2542
				commit_type, &size, n->sha1);
2543 2544 2545 2546
			if (!buf || size < 46)
				die("Not a valid commit: %s", from);
			free(buf);
		} else
2547 2548 2549 2550 2551 2552 2553 2554
			die("Invalid ref name or SHA1 expression: %s", from);

		n->next = NULL;
		if (list)
			e->next = n;
		else
			list = n;
		e = n;
2555
		(*count)++;
2556 2557 2558 2559 2560
		read_next_command();
	}
	return list;
}

2561
static void parse_new_commit(void)
2562
{
2563
	static struct strbuf msg = STRBUF_INIT;
2564 2565 2566 2567
	struct branch *b;
	char *sp;
	char *author = NULL;
	char *committer = NULL;
2568 2569
	struct hash_list *merge_list = NULL;
	unsigned int merge_count;
2570
	unsigned char prev_fanout, new_fanout;
2571 2572 2573 2574

	/* Obtain the branch name from the rest of our command */
	sp = strchr(command_buf.buf, ' ') + 1;
	b = lookup_branch(sp);
2575
	if (!b)
2576
		b = new_branch(sp);
2577 2578

	read_next_command();
2579
	parse_mark();
2580
	if (!prefixcmp(command_buf.buf, "author ")) {
2581
		author = parse_ident(command_buf.buf + 7);
2582 2583
		read_next_command();
	}
2584
	if (!prefixcmp(command_buf.buf, "committer ")) {
2585
		committer = parse_ident(command_buf.buf + 10);
2586 2587 2588 2589
		read_next_command();
	}
	if (!committer)
		die("Expected committer but didn't get one");
2590
	parse_data(&msg, 0, NULL);
2591
	read_next_command();
2592 2593
	parse_from(b);
	merge_list = parse_merge(&merge_count);
2594 2595

	/* ensure the branch is active/loaded */
2596
	if (!b->branch_tree.tree || !max_active_branches) {
2597 2598 2599
		unload_one_branch();
		load_branch(b);
	}
2600

2601 2602
	prev_fanout = convert_num_notes_to_fanout(b->num_notes);

2603
	/* file_change* */
2604
	while (command_buf.len > 0) {
2605
		if (!prefixcmp(command_buf.buf, "M "))
2606
			file_change_m(b);
2607
		else if (!prefixcmp(command_buf.buf, "D "))
2608
			file_change_d(b);
2609
		else if (!prefixcmp(command_buf.buf, "R "))
2610 2611 2612
			file_change_cr(b, 1);
		else if (!prefixcmp(command_buf.buf, "C "))
			file_change_cr(b, 0);
2613
		else if (!prefixcmp(command_buf.buf, "N "))
2614
			note_change_n(b, prev_fanout);
2615 2616
		else if (!strcmp("deleteall", command_buf.buf))
			file_change_deleteall(b);
2617 2618 2619 2620
		else {
			unread_command_buf = 1;
			break;
		}
2621 2622
		if (read_next_command() == EOF)
			break;
2623 2624
	}

2625 2626 2627 2628
	new_fanout = convert_num_notes_to_fanout(b->num_notes);
	if (new_fanout != prev_fanout)
		b->num_notes = change_note_fanout(&b->branch_tree, new_fanout);

2629
	/* build the tree and the commit */
2630
	store_tree(&b->branch_tree);
2631 2632
	hashcpy(b->branch_tree.versions[0].sha1,
		b->branch_tree.versions[1].sha1);
2633 2634 2635

	strbuf_reset(&new_data);
	strbuf_addf(&new_data, "tree %s\n",
2636
		sha1_to_hex(b->branch_tree.versions[1].sha1));
2637
	if (!is_null_sha1(b->sha1))
2638
		strbuf_addf(&new_data, "parent %s\n", sha1_to_hex(b->sha1));
2639 2640
	while (merge_list) {
		struct hash_list *next = merge_list->next;
2641
		strbuf_addf(&new_data, "parent %s\n", sha1_to_hex(merge_list->sha1));
2642 2643 2644
		free(merge_list);
		merge_list = next;
	}
2645 2646 2647 2648 2649 2650
	strbuf_addf(&new_data,
		"author %s\n"
		"committer %s\n"
		"\n",
		author ? author : committer, committer);
	strbuf_addbuf(&new_data, &msg);
2651
	free(author);
2652 2653
	free(committer);

2654
	if (!store_object(OBJ_COMMIT, &new_data, NULL, b->sha1, next_mark))
2655
		b->pack_id = pack_id;
2656
	b->last_commit = object_count_by_type[OBJ_COMMIT];
2657 2658
}

2659
static void parse_new_tag(void)
2660
{
2661
	static struct strbuf msg = STRBUF_INIT;
2662 2663 2664 2665 2666
	char *sp;
	const char *from;
	char *tagger;
	struct branch *s;
	struct tag *t;
2667
	uintmax_t from_mark = 0;
2668
	unsigned char sha1[20];
2669
	enum object_type type;
2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683

	/* Obtain the new tag name from the rest of our command */
	sp = strchr(command_buf.buf, ' ') + 1;
	t = pool_alloc(sizeof(struct tag));
	t->next_tag = NULL;
	t->name = pool_strdup(sp);
	if (last_tag)
		last_tag->next_tag = t;
	else
		first_tag = t;
	last_tag = t;
	read_next_command();

	/* from ... */
2684
	if (prefixcmp(command_buf.buf, "from "))
2685 2686 2687 2688
		die("Expected from command, got %s", command_buf.buf);
	from = strchr(command_buf.buf, ' ') + 1;
	s = lookup_branch(from);
	if (s) {
2689
		hashcpy(sha1, s->sha1);
2690
		type = OBJ_COMMIT;
2691
	} else if (*from == ':') {
2692
		struct object_entry *oe;
2693
		from_mark = strtoumax(from + 1, NULL, 10);
2694
		oe = find_mark(from_mark);
2695
		type = oe->type;
2696
		hashcpy(sha1, oe->idx.sha1);
2697 2698 2699 2700
	} else if (!get_sha1(from, sha1)) {
		unsigned long size;
		char *buf;

2701
		buf = read_sha1_file(sha1, &type, &size);
2702 2703 2704 2705 2706 2707 2708 2709
		if (!buf || size < 46)
			die("Not a valid commit: %s", from);
		free(buf);
	} else
		die("Invalid ref name or SHA1 expression: %s", from);
	read_next_command();

	/* tagger ... */
2710 2711 2712 2713 2714
	if (!prefixcmp(command_buf.buf, "tagger ")) {
		tagger = parse_ident(command_buf.buf + 7);
		read_next_command();
	} else
		tagger = NULL;
2715 2716

	/* tag payload/message */
2717
	parse_data(&msg, 0, NULL);
2718 2719

	/* build the tag object */
2720
	strbuf_reset(&new_data);
2721

2722
	strbuf_addf(&new_data,
2723 2724 2725
		    "object %s\n"
		    "type %s\n"
		    "tag %s\n",
2726
		    sha1_to_hex(sha1), typename(type), t->name);
2727 2728 2729 2730
	if (tagger)
		strbuf_addf(&new_data,
			    "tagger %s\n", tagger);
	strbuf_addch(&new_data, '\n');
2731
	strbuf_addbuf(&new_data, &msg);
2732 2733
	free(tagger);

2734
	if (store_object(OBJ_TAG, &new_data, NULL, t->sha1, 0))
2735 2736 2737
		t->pack_id = MAX_PACK_ID;
	else
		t->pack_id = pack_id;
2738 2739
}

2740
static void parse_reset_branch(void)
2741 2742 2743 2744 2745 2746 2747 2748
{
	struct branch *b;
	char *sp;

	/* Obtain the branch name from the rest of our command */
	sp = strchr(command_buf.buf, ' ') + 1;
	b = lookup_branch(sp);
	if (b) {
2749 2750 2751
		hashclr(b->sha1);
		hashclr(b->branch_tree.versions[0].sha1);
		hashclr(b->branch_tree.versions[1].sha1);
2752 2753 2754 2755 2756
		if (b->branch_tree.tree) {
			release_tree_content_recursive(b->branch_tree.tree);
			b->branch_tree.tree = NULL;
		}
	}
2757 2758 2759
	else
		b = new_branch(sp);
	read_next_command();
2760
	parse_from(b);
2761
	if (command_buf.len > 0)
2762
		unread_command_buf = 1;
2763 2764
}

2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839
static void cat_blob_write(const char *buf, unsigned long size)
{
	if (write_in_full(cat_blob_fd, buf, size) != size)
		die_errno("Write to frontend failed");
}

static void cat_blob(struct object_entry *oe, unsigned char sha1[20])
{
	struct strbuf line = STRBUF_INIT;
	unsigned long size;
	enum object_type type = 0;
	char *buf;

	if (!oe || oe->pack_id == MAX_PACK_ID) {
		buf = read_sha1_file(sha1, &type, &size);
	} else {
		type = oe->type;
		buf = gfi_unpack_entry(oe, &size);
	}

	/*
	 * Output based on batch_one_object() from cat-file.c.
	 */
	if (type <= 0) {
		strbuf_reset(&line);
		strbuf_addf(&line, "%s missing\n", sha1_to_hex(sha1));
		cat_blob_write(line.buf, line.len);
		strbuf_release(&line);
		free(buf);
		return;
	}
	if (!buf)
		die("Can't read object %s", sha1_to_hex(sha1));
	if (type != OBJ_BLOB)
		die("Object %s is a %s but a blob was expected.",
		    sha1_to_hex(sha1), typename(type));
	strbuf_reset(&line);
	strbuf_addf(&line, "%s %s %lu\n", sha1_to_hex(sha1),
						typename(type), size);
	cat_blob_write(line.buf, line.len);
	strbuf_release(&line);
	cat_blob_write(buf, size);
	cat_blob_write("\n", 1);
	free(buf);
}

static void parse_cat_blob(void)
{
	const char *p;
	struct object_entry *oe = oe;
	unsigned char sha1[20];

	/* cat-blob SP <object> LF */
	p = command_buf.buf + strlen("cat-blob ");
	if (*p == ':') {
		char *x;
		oe = find_mark(strtoumax(p + 1, &x, 10));
		if (x == p + 1)
			die("Invalid mark: %s", command_buf.buf);
		if (!oe)
			die("Unknown mark: %s", command_buf.buf);
		if (*x)
			die("Garbage after mark: %s", command_buf.buf);
		hashcpy(sha1, oe->idx.sha1);
	} else {
		if (get_sha1_hex(p, sha1))
			die("Invalid SHA1: %s", command_buf.buf);
		if (p[40])
			die("Garbage after SHA1: %s", command_buf.buf);
		oe = find_object(sha1);
	}

	cat_blob(oe, sha1);
}

2840
static void checkpoint(void)
2841
{
2842
	checkpoint_requested = 0;
2843 2844 2845 2846 2847 2848
	if (object_count) {
		cycle_packfile();
		dump_branches();
		dump_tags();
		dump_marks();
	}
2849 2850 2851 2852 2853
}

static void parse_checkpoint(void)
{
	checkpoint_requested = 1;
2854
	skip_optional_lf();
2855 2856
}

2857
static void parse_progress(void)
2858
{
2859
	fwrite(command_buf.buf, 1, command_buf.len, stdout);
2860 2861 2862 2863 2864
	fputc('\n', stdout);
	fflush(stdout);
	skip_optional_lf();
}

2865
static char* make_fast_import_path(const char *path)
2866
{
2867
	struct strbuf abs_path = STRBUF_INIT;
2868

2869 2870 2871 2872 2873 2874
	if (!relative_marks_paths || is_absolute_path(path))
		return xstrdup(path);
	strbuf_addf(&abs_path, "%s/info/fast-import/%s", get_git_dir(), path);
	return strbuf_detach(&abs_path, NULL);
}

2875 2876
static void option_import_marks(const char *marks,
					int from_stream, int ignore_missing)
2877
{
2878 2879 2880 2881 2882 2883 2884
	if (import_marks_file) {
		if (from_stream)
			die("Only one import-marks command allowed per stream");

		/* read previous mark file */
		if(!import_marks_file_from_stream)
			read_marks();
2885
	}
2886

2887
	import_marks_file = make_fast_import_path(marks);
2888
	safe_create_leading_directories_const(import_marks_file);
2889
	import_marks_file_from_stream = from_stream;
2890
	import_marks_file_ignore_missing = ignore_missing;
2891 2892
}

2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904
static void option_date_format(const char *fmt)
{
	if (!strcmp(fmt, "raw"))
		whenspec = WHENSPEC_RAW;
	else if (!strcmp(fmt, "rfc2822"))
		whenspec = WHENSPEC_RFC2822;
	else if (!strcmp(fmt, "now"))
		whenspec = WHENSPEC_NOW;
	else
		die("unknown --date-format argument %s", fmt);
}

2905 2906 2907 2908 2909 2910 2911 2912 2913
static unsigned long ulong_arg(const char *option, const char *arg)
{
	char *endptr;
	unsigned long rv = strtoul(arg, &endptr, 0);
	if (strchr(arg, '-') || endptr == arg || *endptr)
		die("%s: argument must be a non-negative integer", option);
	return rv;
}

2914 2915
static void option_depth(const char *depth)
{
2916
	max_depth = ulong_arg("--depth", depth);
2917 2918 2919 2920 2921 2922
	if (max_depth > MAX_DEPTH)
		die("--depth cannot exceed %u", MAX_DEPTH);
}

static void option_active_branches(const char *branches)
{
2923
	max_active_branches = ulong_arg("--active-branches", branches);
2924 2925 2926 2927
}

static void option_export_marks(const char *marks)
{
2928
	export_marks_file = make_fast_import_path(marks);
2929
	safe_create_leading_directories_const(export_marks_file);
2930 2931
}

2932 2933 2934 2935 2936 2937 2938 2939
static void option_cat_blob_fd(const char *fd)
{
	unsigned long n = ulong_arg("--cat-blob-fd", fd);
	if (n > (unsigned long) INT_MAX)
		die("--cat-blob-fd cannot exceed %d", INT_MAX);
	cat_blob_fd = (int) n;
}

2940 2941 2942 2943 2944 2945 2946 2947 2948
static void option_export_pack_edges(const char *edges)
{
	if (pack_edges)
		fclose(pack_edges);
	pack_edges = fopen(edges, "a");
	if (!pack_edges)
		die_errno("Cannot open '%s'", edges);
}

S
Sverre Rabbelier 已提交
2949
static int parse_one_option(const char *option)
2950
{
S
Sverre Rabbelier 已提交
2951
	if (!prefixcmp(option, "max-pack-size=")) {
2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962
		unsigned long v;
		if (!git_parse_ulong(option + 14, &v))
			return 0;
		if (v < 8192) {
			warning("max-pack-size is now in bytes, assuming --max-pack-size=%lum", v);
			v *= 1024 * 1024;
		} else if (v < 1024 * 1024) {
			warning("minimum max-pack-size is 1 MiB");
			v = 1024 * 1024;
		}
		max_packsize = v;
2963
	} else if (!prefixcmp(option, "big-file-threshold=")) {
2964 2965 2966 2967
		unsigned long v;
		if (!git_parse_ulong(option + 19, &v))
			return 0;
		big_file_threshold = v;
2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978
	} else if (!prefixcmp(option, "depth=")) {
		option_depth(option + 6);
	} else if (!prefixcmp(option, "active-branches=")) {
		option_active_branches(option + 16);
	} else if (!prefixcmp(option, "export-pack-edges=")) {
		option_export_pack_edges(option + 18);
	} else if (!prefixcmp(option, "quiet")) {
		show_stats = 0;
	} else if (!prefixcmp(option, "stats")) {
		show_stats = 1;
	} else {
S
Sverre Rabbelier 已提交
2979
		return 0;
2980
	}
S
Sverre Rabbelier 已提交
2981 2982

	return 1;
2983 2984
}

2985
static int parse_one_feature(const char *feature, int from_stream)
2986 2987 2988 2989
{
	if (!prefixcmp(feature, "date-format=")) {
		option_date_format(feature + 12);
	} else if (!prefixcmp(feature, "import-marks=")) {
2990 2991 2992 2993
		option_import_marks(feature + 13, from_stream, 0);
	} else if (!prefixcmp(feature, "import-marks-if-exists=")) {
		option_import_marks(feature + strlen("import-marks-if-exists="),
					from_stream, 1);
2994 2995
	} else if (!prefixcmp(feature, "export-marks=")) {
		option_export_marks(feature + 13);
2996 2997
	} else if (!strcmp(feature, "cat-blob")) {
		; /* Don't die - this feature is supported */
2998 2999 3000 3001
	} else if (!prefixcmp(feature, "relative-marks")) {
		relative_marks_paths = 1;
	} else if (!prefixcmp(feature, "no-relative-marks")) {
		relative_marks_paths = 0;
3002 3003
	} else if (!prefixcmp(feature, "force")) {
		force_update = 1;
3004 3005
	} else if (!strcmp(feature, "notes")) {
		; /* do nothing; we have the feature */
3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019
	} else {
		return 0;
	}

	return 1;
}

static void parse_feature(void)
{
	char *feature = command_buf.buf + 8;

	if (seen_data_command)
		die("Got feature command '%s' after data command", feature);

3020
	if (parse_one_feature(feature, 1))
3021 3022 3023 3024 3025
		return;

	die("This version of fast-import does not support feature %s.", feature);
}

S
Sverre Rabbelier 已提交
3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036
static void parse_option(void)
{
	char *option = command_buf.buf + 11;

	if (seen_data_command)
		die("Got option command '%s' after data command", option);

	if (parse_one_option(option))
		return;

	die("This version of fast-import does not support option: %s", option);
3037 3038
}

3039
static int git_pack_config(const char *k, const char *v, void *cb)
3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056
{
	if (!strcmp(k, "pack.depth")) {
		max_depth = git_config_int(k, v);
		if (max_depth > MAX_DEPTH)
			max_depth = MAX_DEPTH;
		return 0;
	}
	if (!strcmp(k, "pack.compression")) {
		int level = git_config_int(k, v);
		if (level == -1)
			level = Z_DEFAULT_COMPRESSION;
		else if (level < 0 || level > Z_BEST_COMPRESSION)
			die("bad pack compression level %d", level);
		pack_compression_level = level;
		pack_compression_seen = 1;
		return 0;
	}
3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067
	if (!strcmp(k, "pack.indexversion")) {
		pack_idx_default_version = git_config_int(k, v);
		if (pack_idx_default_version > 2)
			die("bad pack.indexversion=%"PRIu32,
			    pack_idx_default_version);
		return 0;
	}
	if (!strcmp(k, "pack.packsizelimit")) {
		max_packsize = git_config_ulong(k, v);
		return 0;
	}
3068 3069 3070 3071
	if (!strcmp(k, "core.bigfilethreshold")) {
		long n = git_config_int(k, v);
		big_file_threshold = 0 < n ? n : 0;
	}
3072
	return git_default_config(k, v, cb);
3073 3074
}

3075
static const char fast_import_usage[] =
3076
"git fast-import [--date-format=<f>] [--max-pack-size=<n>] [--big-file-threshold=<n>] [--depth=<n>] [--active-branches=<n>] [--export-marks=<marks.file>]";
3077

S
Sverre Rabbelier 已提交
3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090
static void parse_argv(void)
{
	unsigned int i;

	for (i = 1; i < global_argc; i++) {
		const char *a = global_argv[i];

		if (*a != '-' || !strcmp(a, "--"))
			break;

		if (parse_one_option(a + 2))
			continue;

3091
		if (parse_one_feature(a + 2, 0))
S
Sverre Rabbelier 已提交
3092 3093
			continue;

3094 3095 3096 3097 3098
		if (!prefixcmp(a + 2, "cat-blob-fd=")) {
			option_cat_blob_fd(a + 2 + strlen("cat-blob-fd="));
			continue;
		}

S
Sverre Rabbelier 已提交
3099 3100 3101 3102 3103 3104 3105 3106 3107 3108
		die("unknown option %s", a);
	}
	if (i != global_argc)
		usage(fast_import_usage);

	seen_data_command = 1;
	if (import_marks_file)
		read_marks();
}

3109 3110
int main(int argc, const char **argv)
{
3111
	unsigned int i;
3112

3113 3114
	git_extract_argv0_path(argv[0]);

3115 3116 3117
	if (argc == 2 && !strcmp(argv[1], "-h"))
		usage(fast_import_usage);

3118
	setup_git_directory();
3119
	git_config(git_pack_config, NULL);
3120 3121 3122
	if (!pack_compression_seen && core_compression_seen)
		pack_compression_level = core_compression_level;

3123
	alloc_objects(object_entry_alloc);
3124
	strbuf_init(&command_buf, 0);
3125 3126 3127 3128
	atom_table = xcalloc(atom_table_sz, sizeof(struct atom_str*));
	branch_table = xcalloc(branch_table_sz, sizeof(struct branch*));
	avail_tree_table = xcalloc(avail_tree_table_sz, sizeof(struct avail_tree_content*));
	marks = pool_calloc(1, sizeof(struct mark_set));
3129

S
Sverre Rabbelier 已提交
3130 3131
	global_argc = argc;
	global_argv = argv;
3132

3133 3134 3135 3136 3137
	rc_free = pool_alloc(cmd_save * sizeof(*rc_free));
	for (i = 0; i < (cmd_save - 1); i++)
		rc_free[i].next = &rc_free[i + 1];
	rc_free[cmd_save - 1].next = NULL;

3138
	prepare_packed_git();
3139
	start_packfile();
3140
	set_die_routine(die_nicely);
3141
	set_checkpoint_signal();
3142 3143
	while (read_next_command() != EOF) {
		if (!strcmp("blob", command_buf.buf))
3144
			parse_new_blob();
3145
		else if (!prefixcmp(command_buf.buf, "commit "))
3146
			parse_new_commit();
3147
		else if (!prefixcmp(command_buf.buf, "tag "))
3148
			parse_new_tag();
3149
		else if (!prefixcmp(command_buf.buf, "reset "))
3150
			parse_reset_branch();
3151
		else if (!strcmp("checkpoint", command_buf.buf))
3152
			parse_checkpoint();
3153
		else if (!prefixcmp(command_buf.buf, "progress "))
3154
			parse_progress();
3155 3156
		else if (!prefixcmp(command_buf.buf, "feature "))
			parse_feature();
S
Sverre Rabbelier 已提交
3157 3158 3159 3160
		else if (!prefixcmp(command_buf.buf, "option git "))
			parse_option();
		else if (!prefixcmp(command_buf.buf, "option "))
			/* ignore non-git options*/;
3161 3162
		else
			die("Unsupported command: %s", command_buf.buf);
3163 3164 3165

		if (checkpoint_requested)
			checkpoint();
3166
	}
S
Sverre Rabbelier 已提交
3167 3168 3169 3170 3171

	/* argv hasn't been parsed yet, do so */
	if (!seen_data_command)
		parse_argv();

3172
	end_packfile();
3173

3174
	dump_branches();
3175
	dump_tags();
3176
	unkeep_all_packs();
3177
	dump_marks();
3178

3179 3180 3181
	if (pack_edges)
		fclose(pack_edges);

3182 3183 3184 3185 3186 3187 3188 3189 3190
	if (show_stats) {
		uintmax_t total_count = 0, duplicate_count = 0;
		for (i = 0; i < ARRAY_SIZE(object_count_by_type); i++)
			total_count += object_count_by_type[i];
		for (i = 0; i < ARRAY_SIZE(duplicate_count_by_type); i++)
			duplicate_count += duplicate_count_by_type[i];

		fprintf(stderr, "%s statistics:\n", argv[0]);
		fprintf(stderr, "---------------------------------------------------------------------\n");
3191 3192 3193 3194 3195 3196
		fprintf(stderr, "Alloc'd objects: %10" PRIuMAX "\n", alloc_count);
		fprintf(stderr, "Total objects:   %10" PRIuMAX " (%10" PRIuMAX " duplicates                  )\n", total_count, duplicate_count);
		fprintf(stderr, "      blobs  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_BLOB], duplicate_count_by_type[OBJ_BLOB], delta_count_by_type[OBJ_BLOB]);
		fprintf(stderr, "      trees  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_TREE], duplicate_count_by_type[OBJ_TREE], delta_count_by_type[OBJ_TREE]);
		fprintf(stderr, "      commits:   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_COMMIT], duplicate_count_by_type[OBJ_COMMIT], delta_count_by_type[OBJ_COMMIT]);
		fprintf(stderr, "      tags   :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_TAG], duplicate_count_by_type[OBJ_TAG], delta_count_by_type[OBJ_TAG]);
3197
		fprintf(stderr, "Total branches:  %10lu (%10lu loads     )\n", branch_count, branch_load_count);
3198
		fprintf(stderr, "      marks:     %10" PRIuMAX " (%10" PRIuMAX " unique    )\n", (((uintmax_t)1) << marks->shift) * 1024, marks_set_count);
3199
		fprintf(stderr, "      atoms:     %10u\n", atom_cnt);
3200
		fprintf(stderr, "Memory total:    %10" PRIuMAX " KiB\n", (total_allocd + alloc_count*sizeof(struct object_entry))/1024);
3201
		fprintf(stderr, "       pools:    %10lu KiB\n", (unsigned long)(total_allocd/1024));
3202
		fprintf(stderr, "     objects:    %10" PRIuMAX " KiB\n", (alloc_count*sizeof(struct object_entry))/1024);
3203 3204 3205 3206 3207
		fprintf(stderr, "---------------------------------------------------------------------\n");
		pack_report();
		fprintf(stderr, "---------------------------------------------------------------------\n");
		fprintf(stderr, "\n");
	}
3208

3209
	return failure ? 1 : 0;
3210
}