[thirdparty/git.git] / vcs-svn / svndump.c

/*
 * Parse and rearrange a svnadmin dump.
 * Create the dump with:
 * svnadmin dump --incremental -r<startrev>:<endrev> <repository> >outfile
 *
 * Licensed under a two-clause BSD-style license.
 * See LICENSE for details.
 */

#include "cache.h"
#include "repo_tree.h"
#include "fast_export.h"
#include "line_buffer.h"
#include "strbuf.h"
#include "svndump.h"

/*
 * Compare start of string to literal of equal length;
 * must be guarded by length test.
 */
#define constcmp(s, ref) memcmp(s, ref, sizeof(ref) - 1)

#define REPORT_FILENO 3

#define NODEACT_REPLACE 4
#define NODEACT_DELETE 3
#define NODEACT_ADD 2
#define NODEACT_CHANGE 1
#define NODEACT_UNKNOWN 0

/* States: */
#define DUMP_CTX 0	/* dump metadata */
#define REV_CTX  1	/* revision metadata */
#define NODE_CTX 2	/* node metadata */
#define INTERNODE_CTX 3	/* between nodes */

#define LENGTH_UNKNOWN (~0)
#define DATE_RFC2822_LEN 31

static struct line_buffer input = LINE_BUFFER_INIT;

static struct {
	uint32_t action, propLength, srcRev, type;
	off_t text_length;
	struct strbuf src, dst;
	uint32_t text_delta, prop_delta;
} node_ctx;

static struct {
	uint32_t revision;
	unsigned long timestamp;
	struct strbuf log, author;
} rev_ctx;

static struct {
	uint32_t version;
	struct strbuf uuid, url;
} dump_ctx;

static void reset_node_ctx(char *fname)
{
	node_ctx.type = 0;
	node_ctx.action = NODEACT_UNKNOWN;
	node_ctx.propLength = LENGTH_UNKNOWN;
	node_ctx.text_length = -1;
	strbuf_reset(&node_ctx.src);
	node_ctx.srcRev = 0;
	strbuf_reset(&node_ctx.dst);
	if (fname)
		strbuf_addstr(&node_ctx.dst, fname);
	node_ctx.text_delta = 0;
	node_ctx.prop_delta = 0;
}

static void reset_rev_ctx(uint32_t revision)
{
	rev_ctx.revision = revision;
	rev_ctx.timestamp = 0;
	strbuf_reset(&rev_ctx.log);
	strbuf_reset(&rev_ctx.author);
}

static void reset_dump_ctx(const char *url)
{
	strbuf_reset(&dump_ctx.url);
	if (url)
		strbuf_addstr(&dump_ctx.url, url);
	dump_ctx.version = 1;
	strbuf_reset(&dump_ctx.uuid);
}

static void handle_property(const struct strbuf *key_buf,
				struct strbuf *val,
				uint32_t *type_set)
{
	const char *key = key_buf->buf;
	size_t keylen = key_buf->len;

	switch (keylen + 1) {
	case sizeof("svn:log"):
		if (constcmp(key, "svn:log"))
			break;
		if (!val)
			die("invalid dump: unsets svn:log");
		strbuf_swap(&rev_ctx.log, val);
		break;
	case sizeof("svn:author"):
		if (constcmp(key, "svn:author"))
			break;
		if (!val)
			strbuf_reset(&rev_ctx.author);
		else
			strbuf_swap(&rev_ctx.author, val);
		break;
	case sizeof("svn:date"):
		if (constcmp(key, "svn:date"))
			break;
		if (!val)
			die("invalid dump: unsets svn:date");
		if (parse_date_basic(val->buf, &rev_ctx.timestamp, NULL))
			warning("invalid timestamp: %s", val->buf);
		break;
	case sizeof("svn:executable"):
	case sizeof("svn:special"):
		if (keylen == strlen("svn:executable") &&
		    constcmp(key, "svn:executable"))
			break;
		if (keylen == strlen("svn:special") &&
		    constcmp(key, "svn:special"))
			break;
		if (*type_set) {
			if (!val)
				return;
			die("invalid dump: sets type twice");
		}
		if (!val) {
			node_ctx.type = REPO_MODE_BLB;
			return;
		}
		*type_set = 1;
		node_ctx.type = keylen == strlen("svn:executable") ?
				REPO_MODE_EXE :
				REPO_MODE_LNK;
	}
}

static void die_short_read(void)
{
	if (buffer_ferror(&input))
		die_errno("error reading dump file");
	die("invalid dump: unexpected end of file");
}

static void read_props(void)
{
	static struct strbuf key = STRBUF_INIT;
	static struct strbuf val = STRBUF_INIT;
	const char *t;
	/*
	 * NEEDSWORK: to support simple mode changes like
	 *	K 11
	 *	svn:special
	 *	V 1
	 *	*
	 *	D 14
	 *	svn:executable
	 * we keep track of whether a mode has been set and reset to
	 * plain file only if not.  We should be keeping track of the
	 * symlink and executable bits separately instead.
	 */
	uint32_t type_set = 0;
	while ((t = buffer_read_line(&input)) && strcmp(t, "PROPS-END")) {
		uint32_t len;
		const char type = t[0];
		int ch;

		if (!type || t[1] != ' ')
			die("invalid property line: %s\n", t);
		len = atoi(&t[2]);
		strbuf_reset(&val);
		buffer_read_binary(&input, &val, len);
		if (val.len < len)
			die_short_read();

		/* Discard trailing newline. */
		ch = buffer_read_char(&input);
		if (ch == EOF)
			die_short_read();
		if (ch != '\n')
			die("invalid dump: expected newline after %s", val.buf);

		switch (type) {
		case 'K':
			strbuf_swap(&key, &val);
			continue;
		case 'D':
			handle_property(&val, NULL, &type_set);
			continue;
		case 'V':
			handle_property(&key, &val, &type_set);
			strbuf_reset(&key);
			continue;
		default:
			die("invalid property line: %s\n", t);
		}
	}
}

static void handle_node(void)
{
	const uint32_t type = node_ctx.type;
	const int have_props = node_ctx.propLength != LENGTH_UNKNOWN;
	const int have_text = node_ctx.text_length != -1;
	/*
	 * Old text for this node:
	 *  NULL	- directory or bug
	 *  empty_blob	- empty
	 *  "<dataref>"	- data retrievable from fast-import
	 */
	static const char *const empty_blob = "::empty::";
	const char *old_data = NULL;
	uint32_t old_mode = REPO_MODE_BLB;

	if (node_ctx.action == NODEACT_DELETE) {
		if (have_text || have_props || node_ctx.srcRev)
			die("invalid dump: deletion node has "
				"copyfrom info, text, or properties");
		repo_delete(node_ctx.dst.buf);
		return;
	}
	if (node_ctx.action == NODEACT_REPLACE) {
		repo_delete(node_ctx.dst.buf);
		node_ctx.action = NODEACT_ADD;
	}
	if (node_ctx.srcRev) {
		repo_copy(node_ctx.srcRev, node_ctx.src.buf, node_ctx.dst.buf);
		if (node_ctx.action == NODEACT_ADD)
			node_ctx.action = NODEACT_CHANGE;
	}
	if (have_text && type == REPO_MODE_DIR)
		die("invalid dump: directories cannot have text attached");

	/*
	 * Find old content (old_data) and decide on the new mode.
	 */
	if (node_ctx.action == NODEACT_CHANGE && !*node_ctx.dst.buf) {
		if (type != REPO_MODE_DIR)
			die("invalid dump: root of tree is not a regular file");
		old_data = NULL;
	} else if (node_ctx.action == NODEACT_CHANGE) {
		uint32_t mode;
		old_data = repo_read_path(node_ctx.dst.buf, &mode);
		if (mode == REPO_MODE_DIR && type != REPO_MODE_DIR)
			die("invalid dump: cannot modify a directory into a file");
		if (mode != REPO_MODE_DIR && type == REPO_MODE_DIR)
			die("invalid dump: cannot modify a file into a directory");
		node_ctx.type = mode;
		old_mode = mode;
	} else if (node_ctx.action == NODEACT_ADD) {
		if (type == REPO_MODE_DIR)
			old_data = NULL;
		else if (have_text)
			old_data = empty_blob;
		else
			die("invalid dump: adds node without text");
	} else {
		die("invalid dump: Node-path block lacks Node-action");
	}

	/*
	 * Adjust mode to reflect properties.
	 */
	if (have_props) {
		if (!node_ctx.prop_delta)
			node_ctx.type = type;
		if (node_ctx.propLength)
			read_props();
	}

	/*
	 * Save the result.
	 */
	if (type == REPO_MODE_DIR)	/* directories are not tracked. */
		return;
	assert(old_data);
	if (old_data == empty_blob)
		/* For the fast_export_* functions, NULL means empty. */
		old_data = NULL;
	if (!have_text) {
		fast_export_modify(node_ctx.dst.buf, node_ctx.type, old_data);
		return;
	}
	if (!node_ctx.text_delta) {
		fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
		fast_export_data(node_ctx.type, node_ctx.text_length, &input);
		return;
	}
	fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
	fast_export_blob_delta(node_ctx.type, old_mode, old_data,
				node_ctx.text_length, &input);
}

static void begin_revision(void)
{
	if (!rev_ctx.revision)	/* revision 0 gets no git commit. */
		return;
	fast_export_begin_commit(rev_ctx.revision, rev_ctx.author.buf,
		&rev_ctx.log, dump_ctx.uuid.buf, dump_ctx.url.buf,
		rev_ctx.timestamp);
}

static void end_revision(void)
{
	if (rev_ctx.revision)
		fast_export_end_commit(rev_ctx.revision);
}

void svndump_read(const char *url)
{
	char *val;
	char *t;
	uint32_t active_ctx = DUMP_CTX;
	uint32_t len;

	reset_dump_ctx(url);
	while ((t = buffer_read_line(&input))) {
		val = strchr(t, ':');
		if (!val)
			continue;
		val++;
		if (*val != ' ')
			continue;
		val++;

		/* strlen(key) + 1 */
		switch (val - t - 1) {
		case sizeof("SVN-fs-dump-format-version"):
			if (constcmp(t, "SVN-fs-dump-format-version"))
				continue;
			dump_ctx.version = atoi(val);
			if (dump_ctx.version > 3)
				die("expected svn dump format version <= 3, found %"PRIu32,
				    dump_ctx.version);
			break;
		case sizeof("UUID"):
			if (constcmp(t, "UUID"))
				continue;
			strbuf_reset(&dump_ctx.uuid);
			strbuf_addstr(&dump_ctx.uuid, val);
			break;
		case sizeof("Revision-number"):
			if (constcmp(t, "Revision-number"))
				continue;
			if (active_ctx == NODE_CTX)
				handle_node();
			if (active_ctx == REV_CTX)
				begin_revision();
			if (active_ctx != DUMP_CTX)
				end_revision();
			active_ctx = REV_CTX;
			reset_rev_ctx(atoi(val));
			break;
		case sizeof("Node-path"):
			if (prefixcmp(t, "Node-"))
				continue;
			if (!constcmp(t + strlen("Node-"), "path")) {
				if (active_ctx == NODE_CTX)
					handle_node();
				if (active_ctx == REV_CTX)
					begin_revision();
				active_ctx = NODE_CTX;
				reset_node_ctx(val);
				break;
			}
			if (constcmp(t + strlen("Node-"), "kind"))
				continue;
			if (!strcmp(val, "dir"))
				node_ctx.type = REPO_MODE_DIR;
			else if (!strcmp(val, "file"))
				node_ctx.type = REPO_MODE_BLB;
			else
				fprintf(stderr, "Unknown node-kind: %s\n", val);
			break;
		case sizeof("Node-action"):
			if (constcmp(t, "Node-action"))
				continue;
			if (!strcmp(val, "delete")) {
				node_ctx.action = NODEACT_DELETE;
			} else if (!strcmp(val, "add")) {
				node_ctx.action = NODEACT_ADD;
			} else if (!strcmp(val, "change")) {
				node_ctx.action = NODEACT_CHANGE;
			} else if (!strcmp(val, "replace")) {
				node_ctx.action = NODEACT_REPLACE;
			} else {
				fprintf(stderr, "Unknown node-action: %s\n", val);
				node_ctx.action = NODEACT_UNKNOWN;
			}
			break;
		case sizeof("Node-copyfrom-path"):
			if (constcmp(t, "Node-copyfrom-path"))
				continue;
			strbuf_reset(&node_ctx.src);
			strbuf_addstr(&node_ctx.src, val);
			break;
		case sizeof("Node-copyfrom-rev"):
			if (constcmp(t, "Node-copyfrom-rev"))
				continue;
			node_ctx.srcRev = atoi(val);
			break;
		case sizeof("Text-content-length"):
			if (!constcmp(t, "Text-content-length")) {
				char *end;
				uintmax_t textlen;

				textlen = strtoumax(val, &end, 10);
				if (!isdigit(*val) || *end)
					die("invalid dump: non-numeric length %s", val);
				if (textlen > maximum_signed_value_of_type(off_t))
					die("unrepresentable length in dump: %s", val);
				node_ctx.text_length = (off_t) textlen;
				break;
			}
			if (constcmp(t, "Prop-content-length"))
				continue;
			node_ctx.propLength = atoi(val);
			break;
		case sizeof("Text-delta"):
			if (!constcmp(t, "Text-delta")) {
				node_ctx.text_delta = !strcmp(val, "true");
				break;
			}
			if (constcmp(t, "Prop-delta"))
				continue;
			node_ctx.prop_delta = !strcmp(val, "true");
			break;
		case sizeof("Content-length"):
			if (constcmp(t, "Content-length"))
				continue;
			len = atoi(val);
			t = buffer_read_line(&input);
			if (!t)
				die_short_read();
			if (*t)
				die("invalid dump: expected blank line after content length header");
			if (active_ctx == REV_CTX) {
				read_props();
			} else if (active_ctx == NODE_CTX) {
				handle_node();
				active_ctx = INTERNODE_CTX;
			} else {
				fprintf(stderr, "Unexpected content length header: %"PRIu32"\n", len);
				if (buffer_skip_bytes(&input, len) != len)
					die_short_read();
			}
		}
	}
	if (buffer_ferror(&input))
		die_short_read();
	if (active_ctx == NODE_CTX)
		handle_node();
	if (active_ctx == REV_CTX)
		begin_revision();
	if (active_ctx != DUMP_CTX)
		end_revision();
}

int svndump_init(const char *filename)
{
	if (buffer_init(&input, filename))
		return error("cannot open %s: %s", filename, strerror(errno));
	fast_export_init(REPORT_FILENO);
	strbuf_init(&dump_ctx.uuid, 4096);
	strbuf_init(&dump_ctx.url, 4096);
	strbuf_init(&rev_ctx.log, 4096);
	strbuf_init(&rev_ctx.author, 4096);
	strbuf_init(&node_ctx.src, 4096);
	strbuf_init(&node_ctx.dst, 4096);
	reset_dump_ctx(NULL);
	reset_rev_ctx(0);
	reset_node_ctx(NULL);
	return 0;
}

void svndump_deinit(void)
{
	fast_export_deinit();
	reset_dump_ctx(NULL);
	reset_rev_ctx(0);
	reset_node_ctx(NULL);
	strbuf_release(&rev_ctx.log);
	strbuf_release(&node_ctx.src);
	strbuf_release(&node_ctx.dst);
	if (buffer_deinit(&input))
		fprintf(stderr, "Input error\n");
	if (ferror(stdout))
		fprintf(stderr, "Output error\n");
}

void svndump_reset(void)
{
	fast_export_reset();
	buffer_reset(&input);
	strbuf_release(&dump_ctx.uuid);
	strbuf_release(&dump_ctx.url);
	strbuf_release(&rev_ctx.log);
	strbuf_release(&rev_ctx.author);
}
Commit	Line	Data
21746aa3 DB	1	/*
	2	* Parse and rearrange a svnadmin dump.
	3	* Create the dump with:
	4	* svnadmin dump --incremental -r<startrev>:<endrev> <repository> >outfile
	5	*
	6	* Licensed under a two-clause BSD-style license.
	7	* See LICENSE for details.
	8	*/
	9
	10	#include "cache.h"
	11	#include "repo_tree.h"
	12	#include "fast_export.h"
	13	#include "line_buffer.h"
dce33c9c	14	#include "strbuf.h"
c5147722	15	#include "svndump.h"
21746aa3	16
044ad290 DB	17	/*
	18	* Compare start of string to literal of equal length;
	19	* must be guarded by length test.
	20	*/
	21	#define constcmp(s, ref) memcmp(s, ref, sizeof(ref) - 1)
	22
41529bbc DB	23	#define REPORT_FILENO 3
41529bbc DB	24
21746aa3 DB	25	#define NODEACT_REPLACE 4
	26	#define NODEACT_DELETE 3
	27	#define NODEACT_ADD 2
	28	#define NODEACT_CHANGE 1
	29	#define NODEACT_UNKNOWN 0
	30
7e11902c JN	31	/* States: */
	32	#define DUMP_CTX 0 /* dump metadata */
	33	#define REV_CTX 1 /* revision metadata */
	34	#define NODE_CTX 2 /* node metadata */
	35	#define INTERNODE_CTX 3 /* between nodes */
21746aa3 DB	36
	37	#define LENGTH_UNKNOWN (~0)
	38	#define DATE_RFC2822_LEN 31
	39
e5e45ca1 JN	40	static struct line_buffer input = LINE_BUFFER_INIT;
e5e45ca1 JN	41
21746aa3	42	static struct {
150f7546 JN	43	uint32_t action, propLength, srcRev, type;
150f7546 JN	44	off_t text_length;
03087971	45	struct strbuf src, dst;
1f05d07c	46	uint32_t text_delta, prop_delta;
21746aa3 DB	47	} node_ctx;
	48
	49	static struct {
7c5817d3	50	uint32_t revision;
21746aa3	51	unsigned long timestamp;
7c5817d3	52	struct strbuf log, author;
21746aa3 DB	53	} rev_ctx;
	54
	55	static struct {
7c5817d3 DB	56	uint32_t version;
7c5817d3 DB	57	struct strbuf uuid, url;
21746aa3 DB	58	} dump_ctx;
21746aa3 DB	59
21746aa3 DB	60	static void reset_node_ctx(char *fname)
	61	{
	62	node_ctx.type = 0;
	63	node_ctx.action = NODEACT_UNKNOWN;
	64	node_ctx.propLength = LENGTH_UNKNOWN;
150f7546	65	node_ctx.text_length = -1;
03087971	66	strbuf_reset(&node_ctx.src);
21746aa3	67	node_ctx.srcRev = 0;
03087971 DB	68	strbuf_reset(&node_ctx.dst);
	69	if (fname)
	70	strbuf_addstr(&node_ctx.dst, fname);
1f05d07c DB	71	node_ctx.text_delta = 0;
1f05d07c DB	72	node_ctx.prop_delta = 0;
21746aa3 DB	73	}
	74
	75	static void reset_rev_ctx(uint32_t revision)
	76	{
	77	rev_ctx.revision = revision;
	78	rev_ctx.timestamp = 0;
dce33c9c	79	strbuf_reset(&rev_ctx.log);
7c5817d3	80	strbuf_reset(&rev_ctx.author);
21746aa3 DB	81	}
21746aa3 DB	82
7c5817d3	83	static void reset_dump_ctx(const char *url)
21746aa3	84	{
7c5817d3 DB	85	strbuf_reset(&dump_ctx.url);
	86	if (url)
	87	strbuf_addstr(&dump_ctx.url, url);
b3e5bce1	88	dump_ctx.version = 1;
7c5817d3	89	strbuf_reset(&dump_ctx.uuid);
21746aa3 DB	90	}
21746aa3 DB	91
044ad290	92	static void handle_property(const struct strbuf *key_buf,
4c3169b0	93	struct strbuf *val,
6b01b676	94	uint32_t *type_set)
2a48afe1	95	{
044ad290 DB	96	const char *key = key_buf->buf;
	97	size_t keylen = key_buf->len;
	98
	99	switch (keylen + 1) {
	100	case sizeof("svn:log"):
	101	if (constcmp(key, "svn:log"))
	102	break;
6b01b676 DB	103	if (!val)
6b01b676 DB	104	die("invalid dump: unsets svn:log");
4c3169b0	105	strbuf_swap(&rev_ctx.log, val);
044ad290 DB	106	break;
	107	case sizeof("svn:author"):
	108	if (constcmp(key, "svn:author"))
	109	break;
4c3169b0 JN	110	if (!val)
	111	strbuf_reset(&rev_ctx.author);
	112	else
	113	strbuf_swap(&rev_ctx.author, val);
044ad290 DB	114	break;
	115	case sizeof("svn:date"):
	116	if (constcmp(key, "svn:date"))
	117	break;
6b01b676 DB	118	if (!val)
6b01b676 DB	119	die("invalid dump: unsets svn:date");
4c3169b0 JN	120	if (parse_date_basic(val->buf, &rev_ctx.timestamp, NULL))
4c3169b0 JN	121	warning("invalid timestamp: %s", val->buf);
044ad290 DB	122	break;
	123	case sizeof("svn:executable"):
	124	case sizeof("svn:special"):
	125	if (keylen == strlen("svn:executable") &&
	126	constcmp(key, "svn:executable"))
	127	break;
	128	if (keylen == strlen("svn:special") &&
	129	constcmp(key, "svn:special"))
	130	break;
6b01b676 DB	131	if (*type_set) {
	132	if (!val)
	133	return;
	134	die("invalid dump: sets type twice");
	135	}
	136	if (!val) {
	137	node_ctx.type = REPO_MODE_BLB;
	138	return;
	139	}
	140	*type_set = 1;
044ad290	141	node_ctx.type = keylen == strlen("svn:executable") ?
6b01b676 DB	142	REPO_MODE_EXE :
6b01b676 DB	143	REPO_MODE_LNK;
2a48afe1	144	}
21746aa3 DB	145	}
21746aa3 DB	146
c9d1c8ba JN	147	static void die_short_read(void)
	148	{
	149	if (buffer_ferror(&input))
	150	die_errno("error reading dump file");
	151	die("invalid dump: unexpected end of file");
	152	}
	153
21746aa3 DB	154	static void read_props(void)
21746aa3 DB	155	{
044ad290	156	static struct strbuf key = STRBUF_INIT;
e7d04ee1	157	static struct strbuf val = STRBUF_INIT;
6263c06d	158	const char *t;
6b01b676 DB	159	/*
	160	* NEEDSWORK: to support simple mode changes like
	161	* K 11
	162	* svn:special
	163	* V 1
	164	* *
	165	* D 14
	166	* svn:executable
	167	* we keep track of whether a mode has been set and reset to
	168	* plain file only if not. We should be keeping track of the
	169	* symlink and executable bits separately instead.
	170	*/
	171	uint32_t type_set = 0;
e5e45ca1	172	while ((t = buffer_read_line(&input)) && strcmp(t, "PROPS-END")) {
6263c06d	173	uint32_t len;
6263c06d	174	const char type = t[0];
c9d1c8ba	175	int ch;
6263c06d JN	176
	177	if (!type \|\| t[1] != ' ')
	178	die("invalid property line: %s\n", t);
	179	len = atoi(&t[2]);
e7d04ee1 JN	180	strbuf_reset(&val);
	181	buffer_read_binary(&input, &val, len);
	182	if (val.len < len)
c9d1c8ba JN	183	die_short_read();
	184
	185	/* Discard trailing newline. */
	186	ch = buffer_read_char(&input);
	187	if (ch == EOF)
	188	die_short_read();
	189	if (ch != '\n')
e7d04ee1	190	die("invalid dump: expected newline after %s", val.buf);
6263c06d JN	191
	192	switch (type) {
	193	case 'K':
e7d04ee1 JN	194	strbuf_swap(&key, &val);
e7d04ee1 JN	195	continue;
6b01b676	196	case 'D':
4c3169b0	197	handle_property(&val, NULL, &type_set);
e7d04ee1	198	continue;
6263c06d	199	case 'V':
4c3169b0	200	handle_property(&key, &val, &type_set);
044ad290	201	strbuf_reset(&key);
6263c06d JN	202	continue;
	203	default:
	204	die("invalid property line: %s\n", t);
21746aa3 DB	205	}
	206	}
	207	}
	208
	209	static void handle_node(void)
	210	{
1c7bb316	211	const uint32_t type = node_ctx.type;
d6e81a03	212	const int have_props = node_ctx.propLength != LENGTH_UNKNOWN;
150f7546	213	const int have_text = node_ctx.text_length != -1;
723b7a27 JN	214	/*
	215	* Old text for this node:
	216	* NULL - directory or bug
	217	* empty_blob - empty
	218	* "<dataref>" - data retrievable from fast-import
	219	*/
	220	static const char *const empty_blob = "::empty::";
	221	const char *old_data = NULL;
7a75e661	222	uint32_t old_mode = REPO_MODE_BLB;
21746aa3 DB	223
21746aa3 DB	224	if (node_ctx.action == NODEACT_DELETE) {
5a38b186	225	if (have_text \|\| have_props \|\| node_ctx.srcRev)
5af8fae2 JN	226	die("invalid dump: deletion node has "
5af8fae2 JN	227	"copyfrom info, text, or properties");
c19d653c	228	repo_delete(node_ctx.dst.buf);
9e113988	229	return;
5af8fae2	230	}
6ee4a9be	231	if (node_ctx.action == NODEACT_REPLACE) {
03087971	232	repo_delete(node_ctx.dst.buf);
6ee4a9be JN	233	node_ctx.action = NODEACT_ADD;
6ee4a9be JN	234	}
1c7bb316	235	if (node_ctx.srcRev) {
03087971	236	repo_copy(node_ctx.srcRev, node_ctx.src.buf, node_ctx.dst.buf);
414e569e JN	237	if (node_ctx.action == NODEACT_ADD)
414e569e JN	238	node_ctx.action = NODEACT_CHANGE;
1c7bb316	239	}
5a38b186	240	if (have_text && type == REPO_MODE_DIR)
462e1f51	241	die("invalid dump: directories cannot have text attached");
5a38b186 JN	242
5a38b186 JN	243	/*
723b7a27	244	* Find old content (old_data) and decide on the new mode.
5a38b186	245	*/
03087971	246	if (node_ctx.action == NODEACT_CHANGE && !*node_ctx.dst.buf) {
9e8c5321 JN	247	if (type != REPO_MODE_DIR)
9e8c5321 JN	248	die("invalid dump: root of tree is not a regular file");
723b7a27	249	old_data = NULL;
9e8c5321	250	} else if (node_ctx.action == NODEACT_CHANGE) {
5a38b186	251	uint32_t mode;
c19d653c	252	old_data = repo_read_path(node_ctx.dst.buf, &mode);
c7dbf35e JN	253	if (mode == REPO_MODE_DIR && type != REPO_MODE_DIR)
	254	die("invalid dump: cannot modify a directory into a file");
	255	if (mode != REPO_MODE_DIR && type == REPO_MODE_DIR)
	256	die("invalid dump: cannot modify a file into a directory");
	257	node_ctx.type = mode;
7a75e661	258	old_mode = mode;
21746aa3	259	} else if (node_ctx.action == NODEACT_ADD) {
723b7a27 JN	260	if (type == REPO_MODE_DIR)
	261	old_data = NULL;
	262	else if (have_text)
	263	old_data = empty_blob;
	264	else
c7dbf35e	265	die("invalid dump: adds node without text");
c7dbf35e	266	} else {
414e569e	267	die("invalid dump: Node-path block lacks Node-action");
21746aa3	268	}
5a38b186 JN	269
	270	/*
	271	* Adjust mode to reflect properties.
	272	*/
1c7bb316	273	if (have_props) {
6b01b676 DB	274	if (!node_ctx.prop_delta)
6b01b676 DB	275	node_ctx.type = type;
1c7bb316 JN	276	if (node_ctx.propLength)
1c7bb316 JN	277	read_props();
21746aa3	278	}
5a38b186 JN	279
	280	/*
	281	* Save the result.
	282	*/
723b7a27 JN	283	if (type == REPO_MODE_DIR) /* directories are not tracked. */
	284	return;
	285	assert(old_data);
	286	if (old_data == empty_blob)
	287	/* For the fast_export_* functions, NULL means empty. */
	288	old_data = NULL;
	289	if (!have_text) {
03087971	290	fast_export_modify(node_ctx.dst.buf, node_ctx.type, old_data);
723b7a27 JN	291	return;
723b7a27 JN	292	}
7a75e661 DB	293	if (!node_ctx.text_delta) {
7a75e661 DB	294	fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
150f7546	295	fast_export_data(node_ctx.type, node_ctx.text_length, &input);
7a75e661 DB	296	return;
7a75e661 DB	297	}
03087971	298	fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
7a75e661	299	fast_export_blob_delta(node_ctx.type, old_mode, old_data,
150f7546	300	node_ctx.text_length, &input);
21746aa3 DB	301	}
21746aa3 DB	302
7e11902c JN	303	static void begin_revision(void)
	304	{
	305	if (!rev_ctx.revision) /* revision 0 gets no git commit. */
	306	return;
9ecfa8ae JN	307	fast_export_begin_commit(rev_ctx.revision, rev_ctx.author.buf,
	308	&rev_ctx.log, dump_ctx.uuid.buf, dump_ctx.url.buf,
	309	rev_ctx.timestamp);
21746aa3 DB	310	}
21746aa3 DB	311
7e11902c	312	static void end_revision(void)
21746aa3 DB	313	{
21746aa3 DB	314	if (rev_ctx.revision)
723b7a27	315	fast_export_end_commit(rev_ctx.revision);
21746aa3 DB	316	}
	317
	318	void svndump_read(const char *url)
	319	{
	320	char *val;
	321	char *t;
	322	uint32_t active_ctx = DUMP_CTX;
	323	uint32_t len;
21746aa3	324
7c5817d3	325	reset_dump_ctx(url);
e5e45ca1	326	while ((t = buffer_read_line(&input))) {
f1602054	327	val = strchr(t, ':');
21746aa3 DB	328	if (!val)
21746aa3 DB	329	continue;
f1602054 DB	330	val++;
	331	if (*val != ' ')
	332	continue;
	333	val++;
21746aa3	334
90c0a3cf DB	335	/* strlen(key) + 1 */
	336	switch (val - t - 1) {
	337	case sizeof("SVN-fs-dump-format-version"):
	338	if (constcmp(t, "SVN-fs-dump-format-version"))
	339	continue;
b3e5bce1	340	dump_ctx.version = atoi(val);
1f05d07c	341	if (dump_ctx.version > 3)
a62bbf8f	342	die("expected svn dump format version <= 3, found %"PRIu32,
b3e5bce1	343	dump_ctx.version);
90c0a3cf DB	344	break;
	345	case sizeof("UUID"):
	346	if (constcmp(t, "UUID"))
	347	continue;
7c5817d3 DB	348	strbuf_reset(&dump_ctx.uuid);
7c5817d3 DB	349	strbuf_addstr(&dump_ctx.uuid, val);
90c0a3cf DB	350	break;
	351	case sizeof("Revision-number"):
	352	if (constcmp(t, "Revision-number"))
	353	continue;
21746aa3 DB	354	if (active_ctx == NODE_CTX)
21746aa3 DB	355	handle_node();
7e11902c JN	356	if (active_ctx == REV_CTX)
7e11902c JN	357	begin_revision();
21746aa3	358	if (active_ctx != DUMP_CTX)
7e11902c	359	end_revision();
21746aa3 DB	360	active_ctx = REV_CTX;
21746aa3 DB	361	reset_rev_ctx(atoi(val));
90c0a3cf DB	362	break;
	363	case sizeof("Node-path"):
	364	if (prefixcmp(t, "Node-"))
	365	continue;
	366	if (!constcmp(t + strlen("Node-"), "path")) {
	367	if (active_ctx == NODE_CTX)
	368	handle_node();
9ecfa8ae JN	369	if (active_ctx == REV_CTX)
9ecfa8ae JN	370	begin_revision();
90c0a3cf DB	371	active_ctx = NODE_CTX;
	372	reset_node_ctx(val);
	373	break;
	374	}
	375	if (constcmp(t + strlen("Node-"), "kind"))
	376	continue;
21746aa3 DB	377	if (!strcmp(val, "dir"))
	378	node_ctx.type = REPO_MODE_DIR;
	379	else if (!strcmp(val, "file"))
	380	node_ctx.type = REPO_MODE_BLB;
	381	else
	382	fprintf(stderr, "Unknown node-kind: %s\n", val);
90c0a3cf DB	383	break;
	384	case sizeof("Node-action"):
	385	if (constcmp(t, "Node-action"))
	386	continue;
21746aa3 DB	387	if (!strcmp(val, "delete")) {
	388	node_ctx.action = NODEACT_DELETE;
	389	} else if (!strcmp(val, "add")) {
	390	node_ctx.action = NODEACT_ADD;
	391	} else if (!strcmp(val, "change")) {
	392	node_ctx.action = NODEACT_CHANGE;
	393	} else if (!strcmp(val, "replace")) {
	394	node_ctx.action = NODEACT_REPLACE;
	395	} else {
	396	fprintf(stderr, "Unknown node-action: %s\n", val);
	397	node_ctx.action = NODEACT_UNKNOWN;
	398	}
90c0a3cf DB	399	break;
	400	case sizeof("Node-copyfrom-path"):
	401	if (constcmp(t, "Node-copyfrom-path"))
	402	continue;
03087971 DB	403	strbuf_reset(&node_ctx.src);
03087971 DB	404	strbuf_addstr(&node_ctx.src, val);
90c0a3cf DB	405	break;
	406	case sizeof("Node-copyfrom-rev"):
	407	if (constcmp(t, "Node-copyfrom-rev"))
	408	continue;
21746aa3	409	node_ctx.srcRev = atoi(val);
90c0a3cf DB	410	break;
	411	case sizeof("Text-content-length"):
	412	if (!constcmp(t, "Text-content-length")) {
150f7546 JN	413	char *end;
	414	uintmax_t textlen;
	415
	416	textlen = strtoumax(val, &end, 10);
	417	if (!isdigit(val) \|\| end)
	418	die("invalid dump: non-numeric length %s", val);
	419	if (textlen > maximum_signed_value_of_type(off_t))
	420	die("unrepresentable length in dump: %s", val);
	421	node_ctx.text_length = (off_t) textlen;
90c0a3cf DB	422	break;
	423	}
	424	if (constcmp(t, "Prop-content-length"))
	425	continue;
21746aa3	426	node_ctx.propLength = atoi(val);
90c0a3cf DB	427	break;
	428	case sizeof("Text-delta"):
	429	if (!constcmp(t, "Text-delta")) {
	430	node_ctx.text_delta = !strcmp(val, "true");
	431	break;
	432	}
	433	if (constcmp(t, "Prop-delta"))
	434	continue;
1f05d07c	435	node_ctx.prop_delta = !strcmp(val, "true");
90c0a3cf DB	436	break;
	437	case sizeof("Content-length"):
	438	if (constcmp(t, "Content-length"))
	439	continue;
21746aa3	440	len = atoi(val);
c9d1c8ba JN	441	t = buffer_read_line(&input);
	442	if (!t)
	443	die_short_read();
	444	if (*t)
	445	die("invalid dump: expected blank line after content length header");
21746aa3 DB	446	if (active_ctx == REV_CTX) {
	447	read_props();
	448	} else if (active_ctx == NODE_CTX) {
	449	handle_node();
7e11902c	450	active_ctx = INTERNODE_CTX;
21746aa3	451	} else {
5418d96d	452	fprintf(stderr, "Unexpected content length header: %"PRIu32"\n", len);
c9d1c8ba JN	453	if (buffer_skip_bytes(&input, len) != len)
c9d1c8ba JN	454	die_short_read();
21746aa3 DB	455	}
	456	}
	457	}
c9d1c8ba JN	458	if (buffer_ferror(&input))
c9d1c8ba JN	459	die_short_read();
21746aa3 DB	460	if (active_ctx == NODE_CTX)
21746aa3 DB	461	handle_node();
7e11902c JN	462	if (active_ctx == REV_CTX)
7e11902c JN	463	begin_revision();
21746aa3	464	if (active_ctx != DUMP_CTX)
7e11902c	465	end_revision();
21746aa3 DB	466	}
21746aa3 DB	467
5c28a8b0	468	int svndump_init(const char *filename)
21746aa3	469	{
a62bbf8f	470	if (buffer_init(&input, filename))
5c28a8b0	471	return error("cannot open %s: %s", filename, strerror(errno));
41529bbc	472	fast_export_init(REPORT_FILENO);
7c5817d3 DB	473	strbuf_init(&dump_ctx.uuid, 4096);
7c5817d3 DB	474	strbuf_init(&dump_ctx.url, 4096);
dce33c9c	475	strbuf_init(&rev_ctx.log, 4096);
7c5817d3	476	strbuf_init(&rev_ctx.author, 4096);
03087971 DB	477	strbuf_init(&node_ctx.src, 4096);
03087971 DB	478	strbuf_init(&node_ctx.dst, 4096);
7c5817d3	479	reset_dump_ctx(NULL);
21746aa3 DB	480	reset_rev_ctx(0);
21746aa3 DB	481	reset_node_ctx(NULL);
5c28a8b0	482	return 0;
21746aa3 DB	483	}
	484
	485	void svndump_deinit(void)
	486	{
41529bbc	487	fast_export_deinit();
7c5817d3	488	reset_dump_ctx(NULL);
21746aa3 DB	489	reset_rev_ctx(0);
21746aa3 DB	490	reset_node_ctx(NULL);
dce33c9c	491	strbuf_release(&rev_ctx.log);
03087971 DB	492	strbuf_release(&node_ctx.src);
03087971 DB	493	strbuf_release(&node_ctx.dst);
e5e45ca1	494	if (buffer_deinit(&input))
21746aa3 DB	495	fprintf(stderr, "Input error\n");
	496	if (ferror(stdout))
	497	fprintf(stderr, "Output error\n");
	498	}
	499
	500	void svndump_reset(void)
	501	{
41529bbc	502	fast_export_reset();
e5e45ca1	503	buffer_reset(&input);
7c5817d3 DB	504	strbuf_release(&dump_ctx.uuid);
	505	strbuf_release(&dump_ctx.url);
	506	strbuf_release(&rev_ctx.log);
	507	strbuf_release(&rev_ctx.author);
21746aa3	508	}