[thirdparty/git.git] / apply.c

/*
 * apply.c
 *
 * Copyright (C) Linus Torvalds, 2005
 *
 * This applies patches on top of some (arbitrary) version of the SCM.
 *
 * NOTE! It does all its work in the index file, and only cares about
 * the files in the working directory if you tell it to "merge" the
 * patch apply.
 *
 * Even when merging it always takes the source from the index, and
 * uses the working tree as a "branch" for a 3-way merge.
 */
#include <ctype.h>

#include "cache.h"

// We default to the merge behaviour, since that's what most people would
// expect
static int merge_patch = 1;
static const char apply_usage[] = "git-apply <patch>";

static int linenr = 1;

#define CHUNKSIZE (8192)

static void *read_patch_file(int fd, unsigned long *sizep)
{
	unsigned long size = 0, alloc = CHUNKSIZE;
	void *buffer = xmalloc(alloc);

	for (;;) {
		int nr = alloc - size;
		if (nr < 1024) {
			alloc += CHUNKSIZE;
			buffer = xrealloc(buffer, alloc);
			nr = alloc - size;
		}
		nr = read(fd, buffer + size, nr);
		if (!nr)
			break;
		if (nr < 0) {
			if (errno == EAGAIN)
				continue;
			die("git-apply: read returned %s", strerror(errno));
		}
		size += nr;
	}
	*sizep = size;
	return buffer;
}

static unsigned long linelen(char *buffer, unsigned long size)
{
	unsigned long len = 0;
	while (size--) {
		len++;
		if (*buffer++ == '\n')
			break;
	}
	return len;
}

static int match_word(const char *line, const char *match)
{
	for (;;) {
		char c = *match++;
		if (!c)
			break;
		if (*line++ != c)
			return 0;
	}
	return *line == ' ';
}

/* Verify that we recognize the lines following a git header */
static int parse_git_header(char *line, unsigned int size)
{
	unsigned long offset, len;

	for (offset = 0 ; size > 0 ; offset += len, size -= len, line += len, linenr++) {
		len = linelen(line, size);
		if (!len)
			break;
		if (line[len-1] != '\n')
			return -1;
		if (len < 4)
			break;
		if (!memcmp(line, "@@ -", 4))
			return offset;
		if (match_word(line, "new file mode"))
			continue;
		if (match_word(line, "deleted file mode"))
			continue;
		if (match_word(line, "copy"))
			continue;
		if (match_word(line, "rename"))
			continue;
		if (match_word(line, "similarity index"))
			continue;
		break;
	}

	/* We want either a patch _or_ something real */
	return offset ? :-1;
}

static int parse_num(const char *line, int len, int offset, const char *expect, unsigned long *p)
{
	char *ptr;
	int digits, ex;

	if (offset < 0 || offset >= len)
		return -1;
	line += offset;
	len -= offset;

	if (!isdigit(*line))
		return -1;
	*p = strtoul(line, &ptr, 10);

	digits = ptr - line;

	offset += digits;
	line += digits;
	len -= digits;

	ex = strlen(expect);
	if (ex > len)
		return -1;
	if (memcmp(line, expect, ex))
		return -1;

	return offset + ex;
}

/*
 * Parse a unified diff fragment header of the
 * form "@@ -a,b +c,d @@"
 */
static int parse_fragment_header(char *line, int len, unsigned long *pos)
{
	int offset;

	if (!len || line[len-1] != '\n')
		return -1;

	/* Figure out the number of lines in a fragment */
	offset = parse_num(line, len, 4, ",", pos);
	offset = parse_num(line, len, offset, " +", pos+1);
	offset = parse_num(line, len, offset, ",", pos+2);
	offset = parse_num(line, len, offset, " @@", pos+3);

	return offset;
}

static int find_header(char *line, unsigned long size, int *hdrsize)
{
	unsigned long offset, len;

	for (offset = 0; size > 0; offset += len, size -= len, line += len, linenr++) {
		unsigned long nextlen;

		len = linelen(line, size);
		if (!len)
			break;

		/* Testing this early allows us to take a few shortcuts.. */
		if (len < 6)
			continue;

		/*
		 * Make sure we don't find any unconnected patch fragmants.
		 * That's a sign that we didn't find a header, and that a
		 * patch has become corrupted/broken up.
		 */
		if (!memcmp("@@ -", line, 4)) {
			unsigned long pos[4];
			if (parse_fragment_header(line, len, pos) < 0)
				continue;
			error("patch fragment without header at line %d: %.*s", linenr, len-1, line);
		}

		if (size < len + 6)
			break;

		/*
		 * Git patch? It might not have a real patch, just a rename
		 * or mode change, so we handle that specially
		 */
		if (!memcmp("diff --git ", line, 11)) {
			int git_hdr_len = parse_git_header(line + len, size - len);
			if (git_hdr_len < 0)
				continue;

			*hdrsize = len + git_hdr_len;
			return offset;
		}

		/** --- followed by +++ ? */
		if (memcmp("--- ", line,  4) || memcmp("+++ ", line + len, 4))
			continue;

		/*
		 * We only accept unified patches, so we want it to
		 * at least have "@@ -a,b +c,d @@\n", which is 14 chars
		 * minimum
		 */
		nextlen = linelen(line + len, size - len);
		if (size < nextlen + 14 || memcmp("@@ -", line + len + nextlen, 4))
			continue;

		/* Ok, we'll consider it a patch */
		*hdrsize = len + nextlen;
		linenr += 2;
		return offset;
	}
	return -1;
}

/*
 * Parse a unified diff. Note that this really needs
 * to parse each fragment separately, since the only
 * way to know the difference between a "---" that is
 * part of a patch, and a "---" that starts the next
 * patch is to look at the line counts..
 */
static int apply_fragment(char *line, unsigned long size)
{
	int len = linelen(line, size), offset;
	unsigned long pos[4], oldlines, newlines;

	offset = parse_fragment_header(line, len, pos);
	if (offset < 0)
		return -1;
	oldlines = pos[1];
	newlines = pos[3];

	/* Parse the thing.. */
	line += len;
	size -= len;
	linenr++;
	for (offset = len; size > 0; offset += len, size -= len, line += len, linenr++) {
		if (!oldlines && !newlines)
			break;
		len = linelen(line, size);
		if (!len || line[len-1] != '\n')
			return -1;
		switch (*line) {
		default:
			return -1;
		case ' ':
			oldlines--;
			newlines--;
			break;
		case '-':
			oldlines--;
			break;
		case '+':
			newlines--;
			break;
		}
	}
	return offset;
}

static int apply_single_patch(char *line, unsigned long size)
{
	unsigned long offset = 0;

	while (size > 4 && !memcmp(line, "@@ -", 4)) {
		int len = apply_fragment(line, size);
		if (len <= 0)
			die("corrupt patch at line %d", linenr);

printf("applying fragment:\n%.*s\n\n", len, line);

		offset += len;
		line += len;
		size -= len;
	}
	return offset;
}

static int apply_chunk(char *buffer, unsigned long size)
{
	int hdrsize, patchsize;
	int offset = find_header(buffer, size, &hdrsize);
	char *header, *patch;

	if (offset < 0)
		return offset;
	header = buffer + offset;

printf("Found header:\n%.*s\n\n", hdrsize, header);

	patch = header + hdrsize;
	patchsize = apply_single_patch(patch, size - offset - hdrsize);

	return offset + hdrsize + patchsize;
}

static int apply_patch(int fd)
{
	unsigned long offset, size;
	char *buffer = read_patch_file(fd, &size);

	if (!buffer)
		return -1;
	offset = 0;
	while (size > 0) {
		int nr = apply_chunk(buffer + offset, size);
		if (nr < 0)
			break;
		offset += nr;
		size -= nr;
	}
	free(buffer);
	return 0;
}

int main(int argc, char **argv)
{
	int i;

	if (read_cache() < 0)
		die("unable to read index file");

	for (i = 1; i < argc; i++) {
		const char *arg = argv[i];
		int fd;

		if (!strcmp(arg, "-")) {
			apply_patch(0);
			continue;
		}
		if (!strcmp(arg, "--no-merge")) {
			merge_patch = 0;
			continue;
		}
		fd = open(arg, O_RDONLY);
		if (fd < 0)
			usage(apply_usage);
		apply_patch(fd);
		close(fd);
	}
	return 0;
}
Commit	Line	Data
c1bb9350 LT	1	/*
	2	* apply.c
	3	*
	4	* Copyright (C) Linus Torvalds, 2005
	5	*
	6	* This applies patches on top of some (arbitrary) version of the SCM.
	7	*
	8	* NOTE! It does all its work in the index file, and only cares about
	9	* the files in the working directory if you tell it to "merge" the
	10	* patch apply.
	11	*
	12	* Even when merging it always takes the source from the index, and
	13	* uses the working tree as a "branch" for a 3-way merge.
	14	*/
	15	#include <ctype.h>
	16
	17	#include "cache.h"
	18
	19	// We default to the merge behaviour, since that's what most people would
	20	// expect
	21	static int merge_patch = 1;
	22	static const char apply_usage[] = "git-apply <patch>";
	23
46979f56 LT	24	static int linenr = 1;
46979f56 LT	25
c1bb9350 LT	26	#define CHUNKSIZE (8192)
	27
	28	static void read_patch_file(int fd, unsigned long sizep)
	29	{
	30	unsigned long size = 0, alloc = CHUNKSIZE;
	31	void *buffer = xmalloc(alloc);
	32
	33	for (;;) {
	34	int nr = alloc - size;
	35	if (nr < 1024) {
	36	alloc += CHUNKSIZE;
	37	buffer = xrealloc(buffer, alloc);
	38	nr = alloc - size;
	39	}
	40	nr = read(fd, buffer + size, nr);
	41	if (!nr)
	42	break;
	43	if (nr < 0) {
	44	if (errno == EAGAIN)
	45	continue;
	46	die("git-apply: read returned %s", strerror(errno));
	47	}
	48	size += nr;
	49	}
	50	*sizep = size;
	51	return buffer;
	52	}
	53
	54	static unsigned long linelen(char *buffer, unsigned long size)
	55	{
	56	unsigned long len = 0;
	57	while (size--) {
	58	len++;
	59	if (*buffer++ == '\n')
	60	break;
	61	}
	62	return len;
	63	}
	64
	65	static int match_word(const char line, const char match)
	66	{
	67	for (;;) {
	68	char c = *match++;
	69	if (!c)
	70	break;
	71	if (*line++ != c)
	72	return 0;
	73	}
	74	return *line == ' ';
	75	}
	76
	77	/* Verify that we recognize the lines following a git header */
	78	static int parse_git_header(char *line, unsigned int size)
	79	{
	80	unsigned long offset, len;
	81
46979f56	82	for (offset = 0 ; size > 0 ; offset += len, size -= len, line += len, linenr++) {
c1bb9350 LT	83	len = linelen(line, size);
	84	if (!len)
	85	break;
	86	if (line[len-1] != '\n')
	87	return -1;
	88	if (len < 4)
	89	break;
	90	if (!memcmp(line, "@@ -", 4))
	91	return offset;
	92	if (match_word(line, "new file mode"))
	93	continue;
	94	if (match_word(line, "deleted file mode"))
	95	continue;
	96	if (match_word(line, "copy"))
	97	continue;
	98	if (match_word(line, "rename"))
	99	continue;
	100	if (match_word(line, "similarity index"))
	101	continue;
	102	break;
	103	}
	104
	105	/* We want either a patch _or_ something real */
	106	return offset ? :-1;
	107	}
	108
46979f56 LT	109	static int parse_num(const char line, int len, int offset, const char expect, unsigned long *p)
	110	{
	111	char *ptr;
	112	int digits, ex;
	113
	114	if (offset < 0 \|\| offset >= len)
	115	return -1;
	116	line += offset;
	117	len -= offset;
	118
	119	if (!isdigit(*line))
	120	return -1;
	121	*p = strtoul(line, &ptr, 10);
	122
	123	digits = ptr - line;
	124
	125	offset += digits;
	126	line += digits;
	127	len -= digits;
	128
	129	ex = strlen(expect);
	130	if (ex > len)
	131	return -1;
	132	if (memcmp(line, expect, ex))
	133	return -1;
	134
	135	return offset + ex;
	136	}
	137
	138	/*
	139	* Parse a unified diff fragment header of the
	140	* form "@@ -a,b +c,d @@"
	141	*/
	142	static int parse_fragment_header(char line, int len, unsigned long pos)
	143	{
	144	int offset;
	145
	146	if (!len \|\| line[len-1] != '\n')
	147	return -1;
	148
	149	/* Figure out the number of lines in a fragment */
	150	offset = parse_num(line, len, 4, ",", pos);
	151	offset = parse_num(line, len, offset, " +", pos+1);
	152	offset = parse_num(line, len, offset, ",", pos+2);
	153	offset = parse_num(line, len, offset, " @@", pos+3);
	154
	155	return offset;
	156	}
	157
c1bb9350 LT	158	static int find_header(char line, unsigned long size, int hdrsize)
	159	{
	160	unsigned long offset, len;
	161
46979f56	162	for (offset = 0; size > 0; offset += len, size -= len, line += len, linenr++) {
c1bb9350 LT	163	unsigned long nextlen;
	164
	165	len = linelen(line, size);
	166	if (!len)
	167	break;
	168
	169	/* Testing this early allows us to take a few shortcuts.. */
	170	if (len < 6)
	171	continue;
46979f56 LT	172
	173	/*
	174	* Make sure we don't find any unconnected patch fragmants.
	175	* That's a sign that we didn't find a header, and that a
	176	* patch has become corrupted/broken up.
	177	*/
	178	if (!memcmp("@@ -", line, 4)) {
	179	unsigned long pos[4];
	180	if (parse_fragment_header(line, len, pos) < 0)
	181	continue;
	182	error("patch fragment without header at line %d: %.*s", linenr, len-1, line);
	183	}
	184
c1bb9350 LT	185	if (size < len + 6)
	186	break;
	187
	188	/*
	189	* Git patch? It might not have a real patch, just a rename
	190	* or mode change, so we handle that specially
	191	*/
	192	if (!memcmp("diff --git ", line, 11)) {
	193	int git_hdr_len = parse_git_header(line + len, size - len);
	194	if (git_hdr_len < 0)
	195	continue;
	196
	197	*hdrsize = len + git_hdr_len;
	198	return offset;
	199	}
	200
	201	/** --- followed by +++ ? */
	202	if (memcmp("--- ", line, 4) \|\| memcmp("+++ ", line + len, 4))
	203	continue;
	204
	205	/*
	206	* We only accept unified patches, so we want it to
	207	* at least have "@@ -a,b +c,d @@\n", which is 14 chars
	208	* minimum
	209	*/
	210	nextlen = linelen(line + len, size - len);
	211	if (size < nextlen + 14 \|\| memcmp("@@ -", line + len + nextlen, 4))
	212	continue;
	213
	214	/* Ok, we'll consider it a patch */
	215	*hdrsize = len + nextlen;
46979f56	216	linenr += 2;
c1bb9350 LT	217	return offset;
	218	}
	219	return -1;
	220	}
	221
c1bb9350 LT	222	/*
	223	* Parse a unified diff. Note that this really needs
	224	* to parse each fragment separately, since the only
	225	* way to know the difference between a "---" that is
	226	* part of a patch, and a "---" that starts the next
	227	* patch is to look at the line counts..
	228	*/
	229	static int apply_fragment(char *line, unsigned long size)
	230	{
	231	int len = linelen(line, size), offset;
46979f56	232	unsigned long pos[4], oldlines, newlines;
c1bb9350	233
46979f56	234	offset = parse_fragment_header(line, len, pos);
c1bb9350 LT	235	if (offset < 0)
c1bb9350 LT	236	return -1;
46979f56 LT	237	oldlines = pos[1];
46979f56 LT	238	newlines = pos[3];
c1bb9350 LT	239
	240	/* Parse the thing.. */
	241	line += len;
	242	size -= len;
46979f56 LT	243	linenr++;
46979f56 LT	244	for (offset = len; size > 0; offset += len, size -= len, line += len, linenr++) {
c1bb9350 LT	245	if (!oldlines && !newlines)
	246	break;
	247	len = linelen(line, size);
	248	if (!len \|\| line[len-1] != '\n')
	249	return -1;
	250	switch (*line) {
	251	default:
	252	return -1;
	253	case ' ':
	254	oldlines--;
	255	newlines--;
	256	break;
	257	case '-':
	258	oldlines--;
	259	break;
	260	case '+':
	261	newlines--;
	262	break;
	263	}
	264	}
	265	return offset;
	266	}
	267
	268	static int apply_single_patch(char *line, unsigned long size)
	269	{
	270	unsigned long offset = 0;
	271
	272	while (size > 4 && !memcmp(line, "@@ -", 4)) {
	273	int len = apply_fragment(line, size);
	274	if (len <= 0)
46979f56	275	die("corrupt patch at line %d", linenr);
c1bb9350 LT	276
	277	printf("applying fragment:\n%.*s\n\n", len, line);
	278
	279	offset += len;
	280	line += len;
	281	size -= len;
	282	}
	283	return offset;
	284	}
	285
	286	static int apply_chunk(char *buffer, unsigned long size)
	287	{
	288	int hdrsize, patchsize;
	289	int offset = find_header(buffer, size, &hdrsize);
	290	char header, patch;
	291
	292	if (offset < 0)
	293	return offset;
	294	header = buffer + offset;
	295
	296	printf("Found header:\n%.*s\n\n", hdrsize, header);
	297
	298	patch = header + hdrsize;
	299	patchsize = apply_single_patch(patch, size - offset - hdrsize);
	300
	301	return offset + hdrsize + patchsize;
	302	}
	303
	304	static int apply_patch(int fd)
	305	{
	306	unsigned long offset, size;
	307	char *buffer = read_patch_file(fd, &size);
	308
	309	if (!buffer)
	310	return -1;
	311	offset = 0;
	312	while (size > 0) {
	313	int nr = apply_chunk(buffer + offset, size);
	314	if (nr < 0)
	315	break;
	316	offset += nr;
	317	size -= nr;
	318	}
	319	free(buffer);
	320	return 0;
	321	}
	322
	323	int main(int argc, char **argv)
	324	{
	325	int i;
	326
	327	if (read_cache() < 0)
	328	die("unable to read index file");
	329
	330	for (i = 1; i < argc; i++) {
	331	const char *arg = argv[i];
	332	int fd;
	333
	334	if (!strcmp(arg, "-")) {
	335	apply_patch(0);
	336	continue;
	337	}
	338	if (!strcmp(arg, "--no-merge")) {
	339	merge_patch = 0;
340	continue;
341	}
342	fd = open(arg, O_RDONLY);
343	if (fd < 0)
344	usage(apply_usage);
345	apply_patch(fd);
346	close(fd);
347	}
348	return 0;
349	}