[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 "gettext.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 = xopen(name, O_WRONLY | O_CREAT | O_EXCL, 0666);
	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))) {
			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 (isatty(fileno(f)))
		warning(_("reading patches from stdin/tty..."));

	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"
f394e093	9	#include "gettext.h"
c455c87c	10	#include "string-list.h"
c8f373a5	11	#include "strbuf.h"
2744b234	12
8b73edf4	13	static const char git_mailsplit_usage[] =
0adda936	14	"git mailsplit [-d<prec>] [-f<n>] [-b] [--keep-cr] -o<directory> [(<mbox>\|<Maildir>)...]";
2744b234 LT	15
	16	static int is_from_line(const char *line, int len)
	17	{
	18	const char *colon;
	19
	20	if (len < 20 \|\| memcmp("From ", line, 5))
	21	return 0;
	22
	23	colon = line + len - 2;
	24	line += 5;
	25	for (;;) {
	26	if (colon < line)
	27	return 0;
	28	if (*--colon == ':')
	29	break;
	30	}
	31
	32	if (!isdigit(colon[-4]) \|\|
	33	!isdigit(colon[-2]) \|\|
	34	!isdigit(colon[-1]) \|\|
	35	!isdigit(colon[ 1]) \|\|
	36	!isdigit(colon[ 2]))
	37	return 0;
	38
	39	/* year */
	40	if (strtol(colon+3, NULL, 10) <= 90)
	41	return 0;
	42
	43	/* Ok, close enough */
	44	return 1;
	45	}
	46
c8f373a5	47	static struct strbuf buf = STRBUF_INIT;
c2ca1d79	48	static int keep_cr;
c88098d7 EW	49	static int mboxrd;
	50
	51	static int is_gtfrom(const struct strbuf *buf)
	52	{
	53	size_t min = strlen(">From ");
	54	size_t ngt;
	55
	56	if (buf->len < min)
	57	return 0;
	58
	59	ngt = strspn(buf->buf, ">");
	60	return ngt && starts_with(buf->buf + ngt, "From ");
	61	}
8b73edf4 JH	62
	63	/* Called with the first line (potentially partial)
	64	* already in buf[] -- normally that should begin with
	65	* the Unix "From " line. Write it into the specified
	66	* file.
	67	*/
b3f041fb	68	static int split_one(FILE mbox, const char name, int allow_bare)
2744b234	69	{
13b08125	70	FILE *output;
8b73edf4 JH	71	int fd;
8b73edf4 JH	72	int status = 0;
c8f373a5	73	int is_bare = !is_from_line(buf.buf, buf.len);
2744b234	74
13b08125	75	if (is_bare && !allow_bare) {
13b08125 SB	76	fprintf(stderr, "corrupt mailbox\n");
	77	exit(1);
	78	}
66e905b7	79	fd = xopen(name, O_WRONLY \| O_CREAT \| O_EXCL, 0666);
41698375	80	output = xfdopen(fd, "w");
2744b234	81
8b73edf4 JH	82	/* Copy it out, while searching for a line that begins with
8b73edf4 JH	83	* "From " and having something that looks like a date format.
2744b234	84	*/
8b73edf4	85	for (;;) {
c2ca1d79 JH	86	if (!keep_cr && buf.len > 1 && buf.buf[buf.len-1] == '\n' &&
	87	buf.buf[buf.len-2] == '\r') {
	88	strbuf_setlen(&buf, buf.len-2);
	89	strbuf_addch(&buf, '\n');
	90	}
	91
c88098d7 EW	92	if (mboxrd && is_gtfrom(&buf))
	93	strbuf_remove(&buf, 0, 1);
	94
c8f373a5	95	if (fwrite(buf.buf, 1, buf.len, output) != buf.len)
0721c314	96	die_errno("cannot write output");
8b73edf4	97
c8f373a5	98	if (strbuf_getwholeline(&buf, mbox, '\n')) {
8b73edf4 JH	99	if (feof(mbox)) {
	100	status = 1;
	101	break;
	102	}
0721c314	103	die_errno("cannot read mbox");
8b73edf4	104	}
c8f373a5	105	if (!is_bare && is_from_line(buf.buf, buf.len))
8b73edf4 JH	106	break; /* done with one message */
	107	}
	108	fclose(output);
	109	return status;
2744b234 LT	110	}
2744b234 LT	111
c455c87c	112	static int populate_maildir_list(struct string_list list, const char path)
2744b234	113	{
d63bd9a2 FP	114	DIR *dir;
d63bd9a2 FP	115	struct dirent *dent;
1d895f19	116	char *name = NULL;
d50a4bc4 GP	117	char *subs[] = { "cur", "new", NULL };
d50a4bc4 GP	118	char **sub;
1d895f19	119	int ret = -1;
d50a4bc4 GP	120
d50a4bc4 GP	121	for (sub = subs; *sub; ++sub) {
1d895f19 JK	122	free(name);
1d895f19 JK	123	name = xstrfmt("%s/%s", path, *sub);
afe8a907	124	if (!(dir = opendir(name))) {
d50a4bc4 GP	125	if (errno == ENOENT)
d50a4bc4 GP	126	continue;
880c0aef	127	error_errno("cannot opendir %s", name);
1d895f19	128	goto out;
d50a4bc4	129	}
d63bd9a2	130
d50a4bc4 GP	131	while ((dent = readdir(dir)) != NULL) {
	132	if (dent->d_name[0] == '.')
	133	continue;
1d895f19 JK	134	free(name);
1d895f19 JK	135	name = xstrfmt("%s/%s", *sub, dent->d_name);
78a395d3	136	string_list_insert(list, name);
d50a4bc4	137	}
d63bd9a2	138
d50a4bc4	139	closedir(dir);
d63bd9a2 FP	140	}
d63bd9a2 FP	141
1d895f19 JK	142	ret = 0;
	143
	144	out:
	145	free(name);
	146	return ret;
d63bd9a2 FP	147	}
d63bd9a2 FP	148
18505c34 JK	149	static int maildir_filename_cmp(const char a, const char b)
	150	{
	151	while (a && b) {
	152	if (isdigit(a) && isdigit(b)) {
	153	long int na, nb;
	154	na = strtol(a, (char **)&a, 10);
	155	nb = strtol(b, (char **)&b, 10);
	156	if (na != nb)
	157	return na - nb;
	158	/* strtol advanced our pointers */
	159	}
	160	else {
	161	if (a != b)
	162	return (unsigned char)a - (unsigned char)b;
	163	a++;
	164	b++;
	165	}
	166	}
	167	return (unsigned char)a - (unsigned char)b;
	168	}
	169
d63bd9a2 FP	170	static int split_maildir(const char maildir, const char dir,
	171	int nr_prec, int skip)
	172	{
1d895f19	173	char *file = NULL;
d270d7b7	174	FILE *f = NULL;
e690e843	175	int ret = -1;
d63bd9a2	176	int i;
183113a5	177	struct string_list list = STRING_LIST_INIT_DUP;
e690e843	178
18505c34 JK	179	list.cmp = maildir_filename_cmp;
18505c34 JK	180
d50a4bc4	181	if (populate_maildir_list(&list, maildir) < 0)
d63bd9a2	182	goto out;
e690e843	183
d63bd9a2	184	for (i = 0; i < list.nr; i++) {
1d895f19 JK	185	char *name;
	186
	187	free(file);
	188	file = xstrfmt("%s/%s", maildir, list.items[i].string);
	189
d63bd9a2 FP	190	f = fopen(file, "r");
d63bd9a2 FP	191	if (!f) {
880c0aef	192	error_errno("cannot open mail %s", file);
e690e843 LS	193	goto out;
	194	}
	195
c8f373a5	196	if (strbuf_getwholeline(&buf, f, '\n')) {
880c0aef	197	error_errno("cannot read mail %s", file);
e690e843 LS	198	goto out;
	199	}
	200
1d895f19	201	name = xstrfmt("%s/%0*d", dir, nr_prec, ++skip);
d63bd9a2	202	split_one(f, name, 1);
1d895f19	203	free(name);
d63bd9a2 FP	204
d63bd9a2 FP	205	fclose(f);
d270d7b7	206	f = NULL;
d63bd9a2 FP	207	}
d63bd9a2 FP	208
d63bd9a2 FP	209	ret = skip;
d63bd9a2 FP	210	out:
d270d7b7 JK	211	if (f)
d270d7b7 JK	212	fclose(f);
1d895f19	213	free(file);
c455c87c	214	string_list_clear(&list, 1);
d63bd9a2 FP	215	return ret;
	216	}
	217
fcd056a6 JH	218	static int split_mbox(const char file, const char dir, int allow_bare,
fcd056a6 JH	219	int nr_prec, int skip)
d63bd9a2	220	{
d63bd9a2	221	int ret = -1;
f88a545a	222	int peek;
d63bd9a2 FP	223
	224	FILE *f = !strcmp(file, "-") ? stdin : fopen(file, "r");
	225	int file_done = 0;
	226
7b20af6a JH	227	if (isatty(fileno(f)))
	228	warning(_("reading patches from stdin/tty..."));
	229
d63bd9a2	230	if (!f) {
880c0aef	231	error_errno("cannot open mbox %s", file);
d63bd9a2 FP	232	goto out;
	233	}
	234
f88a545a SS	235	do {
f88a545a SS	236	peek = fgetc(f);
f0733c13 JS	237	if (peek == EOF) {
	238	if (f == stdin)
	239	/* empty stdin is OK */
	240	ret = skip;
	241	else {
	242	fclose(f);
	243	error(_("empty mbox: '%s'"), file);
	244	}
	245	goto out;
	246	}
f88a545a SS	247	} while (isspace(peek));
	248	ungetc(peek, f);
	249
c8f373a5	250	if (strbuf_getwholeline(&buf, f, '\n')) {
d63bd9a2 FP	251	/* empty stdin is OK */
	252	if (f != stdin) {
	253	error("cannot read mbox %s", file);
	254	goto out;
e690e843	255	}
d63bd9a2 FP	256	file_done = 1;
d63bd9a2 FP	257	}
e690e843	258
d63bd9a2	259	while (!file_done) {
1d895f19	260	char name = xstrfmt("%s/%0d", dir, nr_prec, ++skip);
d63bd9a2	261	file_done = split_one(f, name, allow_bare);
1d895f19	262	free(name);
e690e843	263	}
d63bd9a2 FP	264
	265	if (f != stdin)
	266	fclose(f);
	267
e690e843 LS	268	ret = skip;
e690e843 LS	269	out:
e690e843 LS	270	return ret;
e690e843 LS	271	}
d63bd9a2	272
a633fca0	273	int cmd_mailsplit(int argc, const char *argv, const char prefix)
e690e843	274	{
d63bd9a2	275	int nr = 0, nr_prec = 4, num = 0;
b3f041fb PA	276	int allow_bare = 0;
	277	const char *dir = NULL;
	278	const char **argp;
	279	static const char *stdin_only[] = { "-", NULL };
8b73edf4	280
b3f041fb PA	281	for (argp = argv+1; *argp; argp++) {
b3f041fb PA	282	const char arg = argp;
8b73edf4 JH	283
	284	if (arg[0] != '-')
	285	break;
	286	/* do flags here */
b3f041fb PA	287	if ( arg[1] == 'd' ) {
b3f041fb PA	288	nr_prec = strtol(arg+2, NULL, 10);
8b73edf4 JH	289	if (nr_prec < 3 \|\| 10 <= nr_prec)
	290	usage(git_mailsplit_usage);
	291	continue;
b3f041fb PA	292	} else if ( arg[1] == 'f' ) {
b3f041fb PA	293	nr = strtol(arg+2, NULL, 10);
aa481d38 JN	294	} else if ( arg[1] == 'h' ) {
aa481d38 JN	295	usage(git_mailsplit_usage);
b3f041fb PA	296	} else if ( arg[1] == 'b' && !arg[2] ) {
b3f041fb PA	297	allow_bare = 1;
c2ca1d79 JH	298	} else if (!strcmp(arg, "--keep-cr")) {
c2ca1d79 JH	299	keep_cr = 1;
b3f041fb PA	300	} else if ( arg[1] == 'o' && arg[2] ) {
b3f041fb PA	301	dir = arg+2;
c88098d7 EW	302	} else if (!strcmp(arg, "--mboxrd")) {
c88098d7 EW	303	mboxrd = 1;
b3f041fb PA	304	} else if ( arg[1] == '-' && !arg[2] ) {
	305	argp++; /* -- marks end of options */
	306	break;
	307	} else {
	308	die("unknown option: %s", arg);
8b73edf4	309	}
2744b234	310	}
8b73edf4	311
b3f041fb PA	312	if ( !dir ) {
	313	/* Backwards compatibility: if no -o specified, accept
	314	<mbox> <dir> or just <dir> */
	315	switch (argc - (argp-argv)) {
	316	case 1:
	317	dir = argp[0];
	318	argp = stdin_only;
	319	break;
	320	case 2:
	321	stdin_only[0] = argp[0];
	322	dir = argp[1];
	323	argp = stdin_only;
	324	break;
	325	default:
	326	usage(git_mailsplit_usage);
	327	}
	328	} else {
	329	/* New usage: if no more argument, parse stdin */
	330	if ( !*argp )
	331	argp = stdin_only;
2744b234	332	}
8b73edf4	333
d63bd9a2 FP	334	while (*argp) {
	335	const char arg = argp++;
	336	struct stat argstat;
	337	int ret = 0;
	338
	339	if (arg[0] == '-' && arg[1] == 0) {
	340	ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);
	341	if (ret < 0) {
	342	error("cannot split patches from stdin");
	343	return 1;
	344	}
b3327180 JH	345	num += (ret - nr);
b3327180 JH	346	nr = ret;
d63bd9a2 FP	347	continue;
	348	}
	349
	350	if (stat(arg, &argstat) == -1) {
880c0aef	351	error_errno("cannot stat %s", arg);
d63bd9a2 FP	352	return 1;
	353	}
	354
	355	if (S_ISDIR(argstat.st_mode))
	356	ret = split_maildir(arg, dir, nr_prec, nr);
	357	else
	358	ret = split_mbox(arg, dir, allow_bare, nr_prec, nr);
	359
	360	if (ret < 0) {
	361	error("cannot split patches from %s", arg);
	362	return 1;
	363	}
b3327180 JH	364	num += (ret - nr);
b3327180 JH	365	nr = ret;
d63bd9a2 FP	366	}
	367
	368	printf("%d\n", num);
b3f041fb	369
d63bd9a2	370	return 0;
2744b234	371	}