[thirdparty/git.git] / builtin / mailsplit.c

/*
 * Totally braindamaged mbox splitter program.
 *
 * It just splits a mbox into a list of files: "0001" "0002" ..
 * so you can process them further from there.
 */
#include "cache.h"
#include "builtin.h"
#include "string-list.h"
#include "strbuf.h"

static const char git_mailsplit_usage[] =
"git mailsplit [-d<prec>] [-f<n>] [-b] [--keep-cr] -o<directory> [(<mbox>|<Maildir>)...]";

static int is_from_line(const char *line, int len)
{
	const char *colon;

	if (len < 20 || memcmp("From ", line, 5))
		return 0;

	colon = line + len - 2;
	line += 5;
	for (;;) {
		if (colon < line)
			return 0;
		if (*--colon == ':')
			break;
	}

	if (!isdigit(colon[-4]) ||
	    !isdigit(colon[-2]) ||
	    !isdigit(colon[-1]) ||
	    !isdigit(colon[ 1]) ||
	    !isdigit(colon[ 2]))
		return 0;

	/* year */
	if (strtol(colon+3, NULL, 10) <= 90)
		return 0;

	/* Ok, close enough */
	return 1;
}

static struct strbuf buf = STRBUF_INIT;
static int keep_cr;
static int mboxrd;

static int is_gtfrom(const struct strbuf *buf)
{
	size_t min = strlen(">From ");
	size_t ngt;

	if (buf->len < min)
		return 0;

	ngt = strspn(buf->buf, ">");
	return ngt && starts_with(buf->buf + ngt, "From ");
}

/* Called with the first line (potentially partial)
 * already in buf[] -- normally that should begin with
 * the Unix "From " line.  Write it into the specified
 * file.
 */
static int split_one(FILE *mbox, const char *name, int allow_bare)
{
	FILE *output;
	int fd;
	int status = 0;
	int is_bare = !is_from_line(buf.buf, buf.len);

	if (is_bare && !allow_bare) {
		fprintf(stderr, "corrupt mailbox\n");
		exit(1);
	}
	fd = open(name, O_WRONLY | O_CREAT | O_EXCL, 0666);
	if (fd < 0)
		die_errno("cannot open output file '%s'", name);
	output = xfdopen(fd, "w");

	/* Copy it out, while searching for a line that begins with
	 * "From " and having something that looks like a date format.
	 */
	for (;;) {
		if (!keep_cr && buf.len > 1 && buf.buf[buf.len-1] == '\n' &&
			buf.buf[buf.len-2] == '\r') {
			strbuf_setlen(&buf, buf.len-2);
			strbuf_addch(&buf, '\n');
		}

		if (mboxrd && is_gtfrom(&buf))
			strbuf_remove(&buf, 0, 1);

		if (fwrite(buf.buf, 1, buf.len, output) != buf.len)
			die_errno("cannot write output");

		if (strbuf_getwholeline(&buf, mbox, '\n')) {
			if (feof(mbox)) {
				status = 1;
				break;
			}
			die_errno("cannot read mbox");
		}
		if (!is_bare && is_from_line(buf.buf, buf.len))
			break; /* done with one message */
	}
	fclose(output);
	return status;
}

static int populate_maildir_list(struct string_list *list, const char *path)
{
	DIR *dir;
	struct dirent *dent;
	char *name = NULL;
	char *subs[] = { "cur", "new", NULL };
	char **sub;
	int ret = -1;

	for (sub = subs; *sub; ++sub) {
		free(name);
		name = xstrfmt("%s/%s", path, *sub);
		if ((dir = opendir(name)) == NULL) {
			if (errno == ENOENT)
				continue;
			error_errno("cannot opendir %s", name);
			goto out;
		}

		while ((dent = readdir(dir)) != NULL) {
			if (dent->d_name[0] == '.')
				continue;
			free(name);
			name = xstrfmt("%s/%s", *sub, dent->d_name);
			string_list_insert(list, name);
		}

		closedir(dir);
	}

	ret = 0;

out:
	free(name);
	return ret;
}

static int maildir_filename_cmp(const char *a, const char *b)
{
	while (*a && *b) {
		if (isdigit(*a) && isdigit(*b)) {
			long int na, nb;
			na = strtol(a, (char **)&a, 10);
			nb = strtol(b, (char **)&b, 10);
			if (na != nb)
				return na - nb;
			/* strtol advanced our pointers */
		}
		else {
			if (*a != *b)
				return (unsigned char)*a - (unsigned char)*b;
			a++;
			b++;
		}
	}
	return (unsigned char)*a - (unsigned char)*b;
}

static int split_maildir(const char *maildir, const char *dir,
	int nr_prec, int skip)
{
	char *file = NULL;
	FILE *f = NULL;
	int ret = -1;
	int i;
	struct string_list list = STRING_LIST_INIT_DUP;

	list.cmp = maildir_filename_cmp;

	if (populate_maildir_list(&list, maildir) < 0)
		goto out;

	for (i = 0; i < list.nr; i++) {
		char *name;

		free(file);
		file = xstrfmt("%s/%s", maildir, list.items[i].string);

		f = fopen(file, "r");
		if (!f) {
			error_errno("cannot open mail %s", file);
			goto out;
		}

		if (strbuf_getwholeline(&buf, f, '\n')) {
			error_errno("cannot read mail %s", file);
			goto out;
		}

		name = xstrfmt("%s/%0*d", dir, nr_prec, ++skip);
		split_one(f, name, 1);
		free(name);

		fclose(f);
		f = NULL;
	}

	ret = skip;
out:
	if (f)
		fclose(f);
	free(file);
	string_list_clear(&list, 1);
	return ret;
}

static int split_mbox(const char *file, const char *dir, int allow_bare,
		      int nr_prec, int skip)
{
	int ret = -1;
	int peek;

	FILE *f = !strcmp(file, "-") ? stdin : fopen(file, "r");
	int file_done = 0;

	if (!f) {
		error_errno("cannot open mbox %s", file);
		goto out;
	}

	do {
		peek = fgetc(f);
		if (peek == EOF) {
			if (f == stdin)
				/* empty stdin is OK */
				ret = skip;
			else {
				fclose(f);
				error(_("empty mbox: '%s'"), file);
			}
			goto out;
		}
	} while (isspace(peek));
	ungetc(peek, f);

	if (strbuf_getwholeline(&buf, f, '\n')) {
		/* empty stdin is OK */
		if (f != stdin) {
			error("cannot read mbox %s", file);
			goto out;
		}
		file_done = 1;
	}

	while (!file_done) {
		char *name = xstrfmt("%s/%0*d", dir, nr_prec, ++skip);
		file_done = split_one(f, name, allow_bare);
		free(name);
	}

	if (f != stdin)
		fclose(f);

	ret = skip;
out:
	return ret;
}

int cmd_mailsplit(int argc, const char **argv, const char *prefix)
{
	int nr = 0, nr_prec = 4, num = 0;
	int allow_bare = 0;
	const char *dir = NULL;
	const char **argp;
	static const char *stdin_only[] = { "-", NULL };

	for (argp = argv+1; *argp; argp++) {
		const char *arg = *argp;

		if (arg[0] != '-')
			break;
		/* do flags here */
		if ( arg[1] == 'd' ) {
			nr_prec = strtol(arg+2, NULL, 10);
			if (nr_prec < 3 || 10 <= nr_prec)
				usage(git_mailsplit_usage);
			continue;
		} else if ( arg[1] == 'f' ) {
			nr = strtol(arg+2, NULL, 10);
		} else if ( arg[1] == 'h' ) {
			usage(git_mailsplit_usage);
		} else if ( arg[1] == 'b' && !arg[2] ) {
			allow_bare = 1;
		} else if (!strcmp(arg, "--keep-cr")) {
			keep_cr = 1;
		} else if ( arg[1] == 'o' && arg[2] ) {
			dir = arg+2;
		} else if (!strcmp(arg, "--mboxrd")) {
			mboxrd = 1;
		} else if ( arg[1] == '-' && !arg[2] ) {
			argp++;	/* -- marks end of options */
			break;
		} else {
			die("unknown option: %s", arg);
		}
	}

	if ( !dir ) {
		/* Backwards compatibility: if no -o specified, accept
		   <mbox> <dir> or just <dir> */
		switch (argc - (argp-argv)) {
		case 1:
			dir = argp[0];
			argp = stdin_only;
			break;
		case 2:
			stdin_only[0] = argp[0];
			dir = argp[1];
			argp = stdin_only;
			break;
		default:
			usage(git_mailsplit_usage);
		}
	} else {
		/* New usage: if no more argument, parse stdin */
		if ( !*argp )
			argp = stdin_only;
	}

	while (*argp) {
		const char *arg = *argp++;
		struct stat argstat;
		int ret = 0;

		if (arg[0] == '-' && arg[1] == 0) {
			ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);
			if (ret < 0) {
				error("cannot split patches from stdin");
				return 1;
			}
			num += (ret - nr);
			nr = ret;
			continue;
		}

		if (stat(arg, &argstat) == -1) {
			error_errno("cannot stat %s", arg);
			return 1;
		}

		if (S_ISDIR(argstat.st_mode))
			ret = split_maildir(arg, dir, nr_prec, nr);
		else
			ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);

		if (ret < 0) {
			error("cannot split patches from %s", arg);
			return 1;
		}
		num += (ret - nr);
		nr = ret;
	}

	printf("%d\n", num);

	return 0;
}
Commit	Line	Data
2744b234 LT	1	/*
	2	* Totally braindamaged mbox splitter program.
	3	*
	4	* It just splits a mbox into a list of files: "0001" "0002" ..
	5	* so you can process them further from there.
	6	*/
8b73edf4	7	#include "cache.h"
e690e843	8	#include "builtin.h"
c455c87c	9	#include "string-list.h"
c8f373a5	10	#include "strbuf.h"
2744b234	11
8b73edf4	12	static const char git_mailsplit_usage[] =
0adda936	13	"git mailsplit [-d<prec>] [-f<n>] [-b] [--keep-cr] -o<directory> [(<mbox>\|<Maildir>)...]";
2744b234 LT	14
	15	static int is_from_line(const char *line, int len)
	16	{
	17	const char *colon;
	18
	19	if (len < 20 \|\| memcmp("From ", line, 5))
	20	return 0;
	21
	22	colon = line + len - 2;
	23	line += 5;
	24	for (;;) {
	25	if (colon < line)
	26	return 0;
	27	if (*--colon == ':')
	28	break;
	29	}
	30
	31	if (!isdigit(colon[-4]) \|\|
	32	!isdigit(colon[-2]) \|\|
	33	!isdigit(colon[-1]) \|\|
	34	!isdigit(colon[ 1]) \|\|
	35	!isdigit(colon[ 2]))
	36	return 0;
	37
	38	/* year */
	39	if (strtol(colon+3, NULL, 10) <= 90)
	40	return 0;
	41
	42	/* Ok, close enough */
	43	return 1;
	44	}
	45
c8f373a5	46	static struct strbuf buf = STRBUF_INIT;
c2ca1d79	47	static int keep_cr;
c88098d7 EW	48	static int mboxrd;
	49
	50	static int is_gtfrom(const struct strbuf *buf)
	51	{
	52	size_t min = strlen(">From ");
	53	size_t ngt;
	54
	55	if (buf->len < min)
	56	return 0;
	57
	58	ngt = strspn(buf->buf, ">");
	59	return ngt && starts_with(buf->buf + ngt, "From ");
	60	}
8b73edf4 JH	61
	62	/* Called with the first line (potentially partial)
	63	* already in buf[] -- normally that should begin with
	64	* the Unix "From " line. Write it into the specified
	65	* file.
	66	*/
b3f041fb	67	static int split_one(FILE mbox, const char name, int allow_bare)
2744b234	68	{
13b08125	69	FILE *output;
8b73edf4 JH	70	int fd;
8b73edf4 JH	71	int status = 0;
c8f373a5	72	int is_bare = !is_from_line(buf.buf, buf.len);
2744b234	73
13b08125	74	if (is_bare && !allow_bare) {
13b08125 SB	75	fprintf(stderr, "corrupt mailbox\n");
	76	exit(1);
	77	}
8b73edf4 JH	78	fd = open(name, O_WRONLY \| O_CREAT \| O_EXCL, 0666);
8b73edf4 JH	79	if (fd < 0)
0721c314	80	die_errno("cannot open output file '%s'", name);
41698375	81	output = xfdopen(fd, "w");
2744b234	82
8b73edf4 JH	83	/* Copy it out, while searching for a line that begins with
8b73edf4 JH	84	* "From " and having something that looks like a date format.
2744b234	85	*/
8b73edf4	86	for (;;) {
c2ca1d79 JH	87	if (!keep_cr && buf.len > 1 && buf.buf[buf.len-1] == '\n' &&
	88	buf.buf[buf.len-2] == '\r') {
	89	strbuf_setlen(&buf, buf.len-2);
	90	strbuf_addch(&buf, '\n');
	91	}
	92
c88098d7 EW	93	if (mboxrd && is_gtfrom(&buf))
	94	strbuf_remove(&buf, 0, 1);
	95
c8f373a5	96	if (fwrite(buf.buf, 1, buf.len, output) != buf.len)
0721c314	97	die_errno("cannot write output");
8b73edf4	98
c8f373a5	99	if (strbuf_getwholeline(&buf, mbox, '\n')) {
8b73edf4 JH	100	if (feof(mbox)) {
	101	status = 1;
	102	break;
	103	}
0721c314	104	die_errno("cannot read mbox");
8b73edf4	105	}
c8f373a5	106	if (!is_bare && is_from_line(buf.buf, buf.len))
8b73edf4 JH	107	break; /* done with one message */
	108	}
	109	fclose(output);
	110	return status;
2744b234 LT	111	}
2744b234 LT	112
c455c87c	113	static int populate_maildir_list(struct string_list list, const char path)
2744b234	114	{
d63bd9a2 FP	115	DIR *dir;
d63bd9a2 FP	116	struct dirent *dent;
1d895f19	117	char *name = NULL;
d50a4bc4 GP	118	char *subs[] = { "cur", "new", NULL };
d50a4bc4 GP	119	char **sub;
1d895f19	120	int ret = -1;
d50a4bc4 GP	121
d50a4bc4 GP	122	for (sub = subs; *sub; ++sub) {
1d895f19 JK	123	free(name);
1d895f19 JK	124	name = xstrfmt("%s/%s", path, *sub);
d50a4bc4 GP	125	if ((dir = opendir(name)) == NULL) {
	126	if (errno == ENOENT)
	127	continue;
880c0aef	128	error_errno("cannot opendir %s", name);
1d895f19	129	goto out;
d50a4bc4	130	}
d63bd9a2	131
d50a4bc4 GP	132	while ((dent = readdir(dir)) != NULL) {
	133	if (dent->d_name[0] == '.')
	134	continue;
1d895f19 JK	135	free(name);
1d895f19 JK	136	name = xstrfmt("%s/%s", *sub, dent->d_name);
78a395d3	137	string_list_insert(list, name);
d50a4bc4	138	}
d63bd9a2	139
d50a4bc4	140	closedir(dir);
d63bd9a2 FP	141	}
d63bd9a2 FP	142
1d895f19 JK	143	ret = 0;
	144
	145	out:
	146	free(name);
	147	return ret;
d63bd9a2 FP	148	}
d63bd9a2 FP	149
18505c34 JK	150	static int maildir_filename_cmp(const char a, const char b)
	151	{
	152	while (a && b) {
	153	if (isdigit(a) && isdigit(b)) {
	154	long int na, nb;
	155	na = strtol(a, (char **)&a, 10);
	156	nb = strtol(b, (char **)&b, 10);
	157	if (na != nb)
	158	return na - nb;
	159	/* strtol advanced our pointers */
	160	}
	161	else {
	162	if (a != b)
	163	return (unsigned char)a - (unsigned char)b;
	164	a++;
	165	b++;
	166	}
	167	}
	168	return (unsigned char)a - (unsigned char)b;
	169	}
	170
d63bd9a2 FP	171	static int split_maildir(const char maildir, const char dir,
	172	int nr_prec, int skip)
	173	{
1d895f19	174	char *file = NULL;
d270d7b7	175	FILE *f = NULL;
e690e843	176	int ret = -1;
d63bd9a2	177	int i;
183113a5	178	struct string_list list = STRING_LIST_INIT_DUP;
e690e843	179
18505c34 JK	180	list.cmp = maildir_filename_cmp;
18505c34 JK	181
d50a4bc4	182	if (populate_maildir_list(&list, maildir) < 0)
d63bd9a2	183	goto out;
e690e843	184
d63bd9a2	185	for (i = 0; i < list.nr; i++) {
1d895f19 JK	186	char *name;
	187
	188	free(file);
	189	file = xstrfmt("%s/%s", maildir, list.items[i].string);
	190
d63bd9a2 FP	191	f = fopen(file, "r");
d63bd9a2 FP	192	if (!f) {
880c0aef	193	error_errno("cannot open mail %s", file);
e690e843 LS	194	goto out;
	195	}
	196
c8f373a5	197	if (strbuf_getwholeline(&buf, f, '\n')) {
880c0aef	198	error_errno("cannot read mail %s", file);
e690e843 LS	199	goto out;
	200	}
	201
1d895f19	202	name = xstrfmt("%s/%0*d", dir, nr_prec, ++skip);
d63bd9a2	203	split_one(f, name, 1);
1d895f19	204	free(name);
d63bd9a2 FP	205
d63bd9a2 FP	206	fclose(f);
d270d7b7	207	f = NULL;
d63bd9a2 FP	208	}
d63bd9a2 FP	209
d63bd9a2 FP	210	ret = skip;
d63bd9a2 FP	211	out:
d270d7b7 JK	212	if (f)
d270d7b7 JK	213	fclose(f);
1d895f19	214	free(file);
c455c87c	215	string_list_clear(&list, 1);
d63bd9a2 FP	216	return ret;
	217	}
	218
fcd056a6 JH	219	static int split_mbox(const char file, const char dir, int allow_bare,
fcd056a6 JH	220	int nr_prec, int skip)
d63bd9a2	221	{
d63bd9a2	222	int ret = -1;
f88a545a	223	int peek;
d63bd9a2 FP	224
	225	FILE *f = !strcmp(file, "-") ? stdin : fopen(file, "r");
	226	int file_done = 0;
	227
	228	if (!f) {
880c0aef	229	error_errno("cannot open mbox %s", file);
d63bd9a2 FP	230	goto out;
	231	}
	232
f88a545a SS	233	do {
f88a545a SS	234	peek = fgetc(f);
f0733c13 JS	235	if (peek == EOF) {
	236	if (f == stdin)
	237	/* empty stdin is OK */
	238	ret = skip;
	239	else {
	240	fclose(f);
	241	error(_("empty mbox: '%s'"), file);
	242	}
	243	goto out;
	244	}
f88a545a SS	245	} while (isspace(peek));
	246	ungetc(peek, f);
	247
c8f373a5	248	if (strbuf_getwholeline(&buf, f, '\n')) {
d63bd9a2 FP	249	/* empty stdin is OK */
	250	if (f != stdin) {
	251	error("cannot read mbox %s", file);
	252	goto out;
e690e843	253	}
d63bd9a2 FP	254	file_done = 1;
d63bd9a2 FP	255	}
e690e843	256
d63bd9a2	257	while (!file_done) {
1d895f19	258	char name = xstrfmt("%s/%0d", dir, nr_prec, ++skip);
d63bd9a2	259	file_done = split_one(f, name, allow_bare);
1d895f19	260	free(name);
e690e843	261	}
d63bd9a2 FP	262
	263	if (f != stdin)
	264	fclose(f);
	265
e690e843 LS	266	ret = skip;
e690e843 LS	267	out:
e690e843 LS	268	return ret;
e690e843 LS	269	}
d63bd9a2	270
a633fca0	271	int cmd_mailsplit(int argc, const char *argv, const char prefix)
e690e843	272	{
d63bd9a2	273	int nr = 0, nr_prec = 4, num = 0;
b3f041fb PA	274	int allow_bare = 0;
	275	const char *dir = NULL;
	276	const char **argp;
	277	static const char *stdin_only[] = { "-", NULL };
8b73edf4	278
b3f041fb PA	279	for (argp = argv+1; *argp; argp++) {
b3f041fb PA	280	const char arg = argp;
8b73edf4 JH	281
	282	if (arg[0] != '-')
	283	break;
	284	/* do flags here */
b3f041fb PA	285	if ( arg[1] == 'd' ) {
b3f041fb PA	286	nr_prec = strtol(arg+2, NULL, 10);
8b73edf4 JH	287	if (nr_prec < 3 \|\| 10 <= nr_prec)
	288	usage(git_mailsplit_usage);
	289	continue;
b3f041fb PA	290	} else if ( arg[1] == 'f' ) {
b3f041fb PA	291	nr = strtol(arg+2, NULL, 10);
aa481d38 JN	292	} else if ( arg[1] == 'h' ) {
aa481d38 JN	293	usage(git_mailsplit_usage);
b3f041fb PA	294	} else if ( arg[1] == 'b' && !arg[2] ) {
b3f041fb PA	295	allow_bare = 1;
c2ca1d79 JH	296	} else if (!strcmp(arg, "--keep-cr")) {
c2ca1d79 JH	297	keep_cr = 1;
b3f041fb PA	298	} else if ( arg[1] == 'o' && arg[2] ) {
b3f041fb PA	299	dir = arg+2;
c88098d7 EW	300	} else if (!strcmp(arg, "--mboxrd")) {
c88098d7 EW	301	mboxrd = 1;
b3f041fb PA	302	} else if ( arg[1] == '-' && !arg[2] ) {
	303	argp++; /* -- marks end of options */
	304	break;
	305	} else {
	306	die("unknown option: %s", arg);
8b73edf4	307	}
2744b234	308	}
8b73edf4	309
b3f041fb PA	310	if ( !dir ) {
	311	/* Backwards compatibility: if no -o specified, accept
	312	<mbox> <dir> or just <dir> */
	313	switch (argc - (argp-argv)) {
	314	case 1:
	315	dir = argp[0];
	316	argp = stdin_only;
	317	break;
	318	case 2:
	319	stdin_only[0] = argp[0];
	320	dir = argp[1];
	321	argp = stdin_only;
	322	break;
	323	default:
	324	usage(git_mailsplit_usage);
	325	}
	326	} else {
	327	/* New usage: if no more argument, parse stdin */
	328	if ( !*argp )
	329	argp = stdin_only;
2744b234	330	}
8b73edf4	331
d63bd9a2 FP	332	while (*argp) {
	333	const char arg = argp++;
	334	struct stat argstat;
	335	int ret = 0;
	336
	337	if (arg[0] == '-' && arg[1] == 0) {
	338	ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);
	339	if (ret < 0) {
	340	error("cannot split patches from stdin");
	341	return 1;
	342	}
b3327180 JH	343	num += (ret - nr);
b3327180 JH	344	nr = ret;
d63bd9a2 FP	345	continue;
	346	}
	347
	348	if (stat(arg, &argstat) == -1) {
880c0aef	349	error_errno("cannot stat %s", arg);
d63bd9a2 FP	350	return 1;
	351	}
	352
	353	if (S_ISDIR(argstat.st_mode))
	354	ret = split_maildir(arg, dir, nr_prec, nr);
	355	else
	356	ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);
	357
	358	if (ret < 0) {
	359	error("cannot split patches from %s", arg);
	360	return 1;
	361	}
b3327180 JH	362	num += (ret - nr);
b3327180 JH	363	nr = ret;
d63bd9a2 FP	364	}
	365
	366	printf("%d\n", num);
b3f041fb	367
d63bd9a2	368	return 0;
2744b234	369	}