[thirdparty/git.git] / xdiff-interface.c

#include "cache.h"
#include "config.h"
#include "object-store.h"
#include "xdiff-interface.h"
#include "xdiff/xtypes.h"
#include "xdiff/xdiffi.h"
#include "xdiff/xemit.h"
#include "xdiff/xmacros.h"
#include "xdiff/xutils.h"

struct xdiff_emit_state {
	xdiff_emit_hunk_fn hunk_fn;
	xdiff_emit_line_fn line_fn;
	void *consume_callback_data;
	struct strbuf remainder;
};

static int xdiff_out_hunk(void *priv_,
			  long old_begin, long old_nr,
			  long new_begin, long new_nr,
			  const char *func, long funclen)
{
	struct xdiff_emit_state *priv = priv_;

	if (priv->remainder.len)
		BUG("xdiff emitted hunk in the middle of a line");

	priv->hunk_fn(priv->consume_callback_data,
		      old_begin, old_nr, new_begin, new_nr,
		      func, funclen);
	return 0;
}

static void consume_one(void *priv_, char *s, unsigned long size)
{
	struct xdiff_emit_state *priv = priv_;
	char *ep;
	while (size) {
		unsigned long this_size;
		ep = memchr(s, '\n', size);
		this_size = (ep == NULL) ? size : (ep - s + 1);
		priv->line_fn(priv->consume_callback_data, s, this_size);
		size -= this_size;
		s += this_size;
	}
}

static int xdiff_outf(void *priv_, mmbuffer_t *mb, int nbuf)
{
	struct xdiff_emit_state *priv = priv_;
	int i;

	if (!priv->line_fn)
		return 0;

	for (i = 0; i < nbuf; i++) {
		if (mb[i].ptr[mb[i].size-1] != '\n') {
			/* Incomplete line */
			strbuf_add(&priv->remainder, mb[i].ptr, mb[i].size);
			continue;
		}

		/* we have a complete line */
		if (!priv->remainder.len) {
			consume_one(priv, mb[i].ptr, mb[i].size);
			continue;
		}
		strbuf_add(&priv->remainder, mb[i].ptr, mb[i].size);
		consume_one(priv, priv->remainder.buf, priv->remainder.len);
		strbuf_reset(&priv->remainder);
	}
	if (priv->remainder.len) {
		consume_one(priv, priv->remainder.buf, priv->remainder.len);
		strbuf_reset(&priv->remainder);
	}
	return 0;
}

/*
 * Trim down common substring at the end of the buffers,
 * but end on a complete line.
 */
static void trim_common_tail(mmfile_t *a, mmfile_t *b)
{
	const int blk = 1024;
	long trimmed = 0, recovered = 0;
	char *ap = a->ptr + a->size;
	char *bp = b->ptr + b->size;
	long smaller = (a->size < b->size) ? a->size : b->size;

	while (blk + trimmed <= smaller && !memcmp(ap - blk, bp - blk, blk)) {
		trimmed += blk;
		ap -= blk;
		bp -= blk;
	}

	while (recovered < trimmed)
		if (ap[recovered++] == '\n')
			break;
	a->size -= trimmed - recovered;
	b->size -= trimmed - recovered;
}

int xdi_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp, xdemitconf_t const *xecfg, xdemitcb_t *xecb)
{
	mmfile_t a = *mf1;
	mmfile_t b = *mf2;

	if (mf1->size > MAX_XDIFF_SIZE || mf2->size > MAX_XDIFF_SIZE)
		return -1;

	if (!xecfg->ctxlen && !(xecfg->flags & XDL_EMIT_FUNCCONTEXT))
		trim_common_tail(&a, &b);

	return xdl_diff(&a, &b, xpp, xecfg, xecb);
}

void discard_hunk_line(void *priv,
		       long ob, long on, long nb, long nn,
		       const char *func, long funclen)
{
}

int xdi_diff_outf(mmfile_t *mf1, mmfile_t *mf2,
		  xdiff_emit_hunk_fn hunk_fn,
		  xdiff_emit_line_fn line_fn,
		  void *consume_callback_data,
		  xpparam_t const *xpp, xdemitconf_t const *xecfg)
{
	int ret;
	struct xdiff_emit_state state;
	xdemitcb_t ecb;

	memset(&state, 0, sizeof(state));
	state.hunk_fn = hunk_fn;
	state.line_fn = line_fn;
	state.consume_callback_data = consume_callback_data;
	memset(&ecb, 0, sizeof(ecb));
	if (hunk_fn)
		ecb.out_hunk = xdiff_out_hunk;
	ecb.out_line = xdiff_outf;
	ecb.priv = &state;
	strbuf_init(&state.remainder, 0);
	ret = xdi_diff(mf1, mf2, xpp, xecfg, &ecb);
	strbuf_release(&state.remainder);
	return ret;
}

int read_mmfile(mmfile_t *ptr, const char *filename)
{
	struct stat st;
	FILE *f;
	size_t sz;

	if (stat(filename, &st))
		return error_errno("Could not stat %s", filename);
	if ((f = fopen(filename, "rb")) == NULL)
		return error_errno("Could not open %s", filename);
	sz = xsize_t(st.st_size);
	ptr->ptr = xmalloc(sz ? sz : 1);
	if (sz && fread(ptr->ptr, sz, 1, f) != 1) {
		fclose(f);
		return error("Could not read %s", filename);
	}
	fclose(f);
	ptr->size = sz;
	return 0;
}

void read_mmblob(mmfile_t *ptr, const struct object_id *oid)
{
	unsigned long size;
	enum object_type type;

	if (oideq(oid, &null_oid)) {
		ptr->ptr = xstrdup("");
		ptr->size = 0;
		return;
	}

	ptr->ptr = read_object_file(oid, &type, &size);
	if (!ptr->ptr || type != OBJ_BLOB)
		die("unable to read blob object %s", oid_to_hex(oid));
	ptr->size = size;
}

#define FIRST_FEW_BYTES 8000
int buffer_is_binary(const char *ptr, unsigned long size)
{
	if (FIRST_FEW_BYTES < size)
		size = FIRST_FEW_BYTES;
	return !!memchr(ptr, 0, size);
}

struct ff_regs {
	int nr;
	struct ff_reg {
		regex_t re;
		int negate;
	} *array;
};

static long ff_regexp(const char *line, long len,
		char *buffer, long buffer_size, void *priv)
{
	struct ff_regs *regs = priv;
	regmatch_t pmatch[2];
	int i;
	int result;

	/* Exclude terminating newline (and cr) from matching */
	if (len > 0 && line[len-1] == '\n') {
		if (len > 1 && line[len-2] == '\r')
			len -= 2;
		else
			len--;
	}

	for (i = 0; i < regs->nr; i++) {
		struct ff_reg *reg = regs->array + i;
		if (!regexec_buf(&reg->re, line, len, 2, pmatch, 0)) {
			if (reg->negate)
				return -1;
			break;
		}
	}
	if (regs->nr <= i)
		return -1;
	i = pmatch[1].rm_so >= 0 ? 1 : 0;
	line += pmatch[i].rm_so;
	result = pmatch[i].rm_eo - pmatch[i].rm_so;
	if (result > buffer_size)
		result = buffer_size;
	while (result > 0 && (isspace(line[result - 1])))
		result--;
	memcpy(buffer, line, result);
	return result;
}

void xdiff_set_find_func(xdemitconf_t *xecfg, const char *value, int cflags)
{
	int i;
	struct ff_regs *regs;

	xecfg->find_func = ff_regexp;
	regs = xecfg->find_func_priv = xmalloc(sizeof(struct ff_regs));
	for (i = 0, regs->nr = 1; value[i]; i++)
		if (value[i] == '\n')
			regs->nr++;
	ALLOC_ARRAY(regs->array, regs->nr);
	for (i = 0; i < regs->nr; i++) {
		struct ff_reg *reg = regs->array + i;
		const char *ep = strchr(value, '\n'), *expression;
		char *buffer = NULL;

		reg->negate = (*value == '!');
		if (reg->negate && i == regs->nr - 1)
			die("Last expression must not be negated: %s", value);
		if (*value == '!')
			value++;
		if (ep)
			expression = buffer = xstrndup(value, ep - value);
		else
			expression = value;
		if (regcomp(&reg->re, expression, cflags))
			die("Invalid regexp to look for hunk header: %s", expression);
		free(buffer);
		value = ep + 1;
	}
}

void xdiff_clear_find_func(xdemitconf_t *xecfg)
{
	if (xecfg->find_func) {
		int i;
		struct ff_regs *regs = xecfg->find_func_priv;

		for (i = 0; i < regs->nr; i++)
			regfree(&regs->array[i].re);
		free(regs->array);
		free(regs);
		xecfg->find_func = NULL;
		xecfg->find_func_priv = NULL;
	}
}

unsigned long xdiff_hash_string(const char *s, size_t len, long flags)
{
	return xdl_hash_record(&s, s + len, flags);
}

int xdiff_compare_lines(const char *l1, long s1,
			const char *l2, long s2, long flags)
{
	return xdl_recmatch(l1, s1, l2, s2, flags);
}

int git_xmerge_style = -1;

int git_xmerge_config(const char *var, const char *value, void *cb)
{
	if (!strcmp(var, "merge.conflictstyle")) {
		if (!value)
			die("'%s' is not a boolean", var);
		if (!strcmp(value, "diff3"))
			git_xmerge_style = XDL_MERGE_DIFF3;
		else if (!strcmp(value, "merge"))
			git_xmerge_style = 0;
		else
			die("unknown style '%s' given for '%s'",
			    value, var);
		return 0;
	}
	return git_default_config(var, value, cb);
}
Commit	Line	Data
d9ea73e0	1	#include "cache.h"
b2141fc1	2	#include "config.h"
cbd53a21	3	#include "object-store.h"
d9ea73e0	4	#include "xdiff-interface.h"
86295bb6 RS	5	#include "xdiff/xtypes.h"
	6	#include "xdiff/xdiffi.h"
	7	#include "xdiff/xemit.h"
	8	#include "xdiff/xmacros.h"
5ec8274b	9	#include "xdiff/xutils.h"
8a3f524b JH	10
8a3f524b JH	11	struct xdiff_emit_state {
9346d6d1 JK	12	xdiff_emit_hunk_fn hunk_fn;
9346d6d1 JK	13	xdiff_emit_line_fn line_fn;
8a3f524b JH	14	void *consume_callback_data;
	15	struct strbuf remainder;
	16	};
d9ea73e0	17
9346d6d1 JK	18	static int xdiff_out_hunk(void *priv_,
	19	long old_begin, long old_nr,
	20	long new_begin, long new_nr,
	21	const char *func, long funclen)
c1e335a4	22	{
9346d6d1	23	struct xdiff_emit_state *priv = priv_;
c1e335a4	24
9346d6d1 JK	25	if (priv->remainder.len)
9346d6d1 JK	26	BUG("xdiff emitted hunk in the middle of a line");
c1e335a4	27
9346d6d1 JK	28	priv->hunk_fn(priv->consume_callback_data,
	29	old_begin, old_nr, new_begin, new_nr,
	30	func, funclen);
	31	return 0;
c1e335a4 JH	32	}
c1e335a4 JH	33
d9ea73e0 JH	34	static void consume_one(void priv_, char s, unsigned long size)
	35	{
	36	struct xdiff_emit_state *priv = priv_;
	37	char *ep;
	38	while (size) {
	39	unsigned long this_size;
	40	ep = memchr(s, '\n', size);
	41	this_size = (ep == NULL) ? size : (ep - s + 1);
9346d6d1	42	priv->line_fn(priv->consume_callback_data, s, this_size);
d9ea73e0 JH	43	size -= this_size;
	44	s += this_size;
	45	}
	46	}
	47
c99db9d2	48	static int xdiff_outf(void priv_, mmbuffer_t mb, int nbuf)
d9ea73e0 JH	49	{
	50	struct xdiff_emit_state *priv = priv_;
	51	int i;
	52
7c61e25f JK	53	if (!priv->line_fn)
	54	return 0;
	55
d9ea73e0 JH	56	for (i = 0; i < nbuf; i++) {
	57	if (mb[i].ptr[mb[i].size-1] != '\n') {
	58	/* Incomplete line */
b4637760	59	strbuf_add(&priv->remainder, mb[i].ptr, mb[i].size);
d9ea73e0 JH	60	continue;
	61	}
	62
	63	/* we have a complete line */
b4637760	64	if (!priv->remainder.len) {
d9ea73e0 JH	65	consume_one(priv, mb[i].ptr, mb[i].size);
	66	continue;
	67	}
b4637760 BD	68	strbuf_add(&priv->remainder, mb[i].ptr, mb[i].size);
	69	consume_one(priv, priv->remainder.buf, priv->remainder.len);
	70	strbuf_reset(&priv->remainder);
d9ea73e0	71	}
b4637760 BD	72	if (priv->remainder.len) {
	73	consume_one(priv, priv->remainder.buf, priv->remainder.len);
	74	strbuf_reset(&priv->remainder);
d9ea73e0 JH	75	}
	76	return 0;
	77	}
7cab5883	78
913b45f5 JH	79	/*
913b45f5 JH	80	* Trim down common substring at the end of the buffers,
e0876bca	81	* but end on a complete line.
913b45f5	82	*/
e0876bca	83	static void trim_common_tail(mmfile_t a, mmfile_t b)
913b45f5 JH	84	{
913b45f5 JH	85	const int blk = 1024;
52499977	86	long trimmed = 0, recovered = 0;
913b45f5 JH	87	char *ap = a->ptr + a->size;
	88	char *bp = b->ptr + b->size;
	89	long smaller = (a->size < b->size) ? a->size : b->size;
	90
	91	while (blk + trimmed <= smaller && !memcmp(ap - blk, bp - blk, blk)) {
	92	trimmed += blk;
	93	ap -= blk;
	94	bp -= blk;
	95	}
	96
d2f82950	97	while (recovered < trimmed)
52499977	98	if (ap[recovered++] == '\n')
d2f82950 LT	99	break;
	100	a->size -= trimmed - recovered;
	101	b->size -= trimmed - recovered;
913b45f5 JH	102	}
913b45f5 JH	103
c279d7e9 JH	104	int xdi_diff(mmfile_t mf1, mmfile_t mf2, xpparam_t const xpp, xdemitconf_t const xecfg, xdemitcb_t *xecb)
c279d7e9 JH	105	{
913b45f5 JH	106	mmfile_t a = *mf1;
	107	mmfile_t b = *mf2;
	108
dcd1742e JK	109	if (mf1->size > MAX_XDIFF_SIZE \|\| mf2->size > MAX_XDIFF_SIZE)
	110	return -1;
	111
e0876bca RS	112	if (!xecfg->ctxlen && !(xecfg->flags & XDL_EMIT_FUNCCONTEXT))
e0876bca RS	113	trim_common_tail(&a, &b);
913b45f5 JH	114
913b45f5 JH	115	return xdl_diff(&a, &b, xpp, xecfg, xecb);
c279d7e9 JH	116	}
c279d7e9 JH	117
3b40a090 JK	118	void discard_hunk_line(void *priv,
	119	long ob, long on, long nb, long nn,
	120	const char *func, long funclen)
	121	{
	122	}
	123
c99db9d2	124	int xdi_diff_outf(mmfile_t mf1, mmfile_t mf2,
9346d6d1 JK	125	xdiff_emit_hunk_fn hunk_fn,
	126	xdiff_emit_line_fn line_fn,
	127	void *consume_callback_data,
dfea7900	128	xpparam_t const xpp, xdemitconf_t const xecfg)
c99db9d2 BD	129	{
c99db9d2 BD	130	int ret;
8a3f524b	131	struct xdiff_emit_state state;
dfea7900	132	xdemitcb_t ecb;
8a3f524b JH	133
8a3f524b JH	134	memset(&state, 0, sizeof(state));
9346d6d1 JK	135	state.hunk_fn = hunk_fn;
9346d6d1 JK	136	state.line_fn = line_fn;
8a3f524b	137	state.consume_callback_data = consume_callback_data;
dfea7900	138	memset(&ecb, 0, sizeof(ecb));
9346d6d1 JK	139	if (hunk_fn)
9346d6d1 JK	140	ecb.out_hunk = xdiff_out_hunk;
611e42a5	141	ecb.out_line = xdiff_outf;
dfea7900	142	ecb.priv = &state;
8a3f524b	143	strbuf_init(&state.remainder, 0);
dfea7900	144	ret = xdi_diff(mf1, mf2, xpp, xecfg, &ecb);
8a3f524b	145	strbuf_release(&state.remainder);
c99db9d2 BD	146	return ret;
	147	}
	148
7cab5883 JS	149	int read_mmfile(mmfile_t ptr, const char filename)
	150	{
	151	struct stat st;
	152	FILE *f;
dc49cd76	153	size_t sz;
7cab5883 JS	154
7cab5883 JS	155	if (stat(filename, &st))
5118d7f4	156	return error_errno("Could not stat %s", filename);
7cab5883	157	if ((f = fopen(filename, "rb")) == NULL)
5118d7f4	158	return error_errno("Could not open %s", filename);
dc49cd76	159	sz = xsize_t(st.st_size);
381b851c	160	ptr->ptr = xmalloc(sz ? sz : 1);
5fd89814 RS	161	if (sz && fread(ptr->ptr, sz, 1, f) != 1) {
5fd89814 RS	162	fclose(f);
7cab5883	163	return error("Could not read %s", filename);
5fd89814	164	}
7cab5883	165	fclose(f);
dc49cd76	166	ptr->size = sz;
7cab5883 JS	167	return 0;
	168	}
	169
d449347d	170	void read_mmblob(mmfile_t ptr, const struct object_id oid)
06b65939 ML	171	{
	172	unsigned long size;
	173	enum object_type type;
	174
4a7e27e9	175	if (oideq(oid, &null_oid)) {
06b65939 ML	176	ptr->ptr = xstrdup("");
	177	ptr->size = 0;
	178	return;
	179	}
	180
b4f5aca4	181	ptr->ptr = read_object_file(oid, &type, &size);
06b65939	182	if (!ptr->ptr \|\| type != OBJ_BLOB)
d449347d	183	die("unable to read blob object %s", oid_to_hex(oid));
06b65939 ML	184	ptr->size = size;
	185	}
	186
6bfce93e JS	187	#define FIRST_FEW_BYTES 8000
	188	int buffer_is_binary(const char *ptr, unsigned long size)
	189	{
	190	if (FIRST_FEW_BYTES < size)
	191	size = FIRST_FEW_BYTES;
	192	return !!memchr(ptr, 0, size);
	193	}
f258475a JH	194
	195	struct ff_regs {
	196	int nr;
	197	struct ff_reg {
	198	regex_t re;
	199	int negate;
	200	} *array;
	201	};
	202
	203	static long ff_regexp(const char *line, long len,
	204	char buffer, long buffer_size, void priv)
	205	{
f258475a JH	206	struct ff_regs *regs = priv;
f258475a JH	207	regmatch_t pmatch[2];
3d8dccd7	208	int i;
b7d36ffc	209	int result;
f258475a	210
a5a5a048 BC	211	/* Exclude terminating newline (and cr) from matching */
a5a5a048 BC	212	if (len > 0 && line[len-1] == '\n') {
46dc1b0e JH	213	if (len > 1 && line[len-2] == '\r')
	214	len -= 2;
	215	else
	216	len--;
	217	}
	218
f258475a JH	219	for (i = 0; i < regs->nr; i++) {
f258475a JH	220	struct ff_reg *reg = regs->array + i;
b7d36ffc	221	if (!regexec_buf(&reg->re, line, len, 2, pmatch, 0)) {
3d8dccd7	222	if (reg->negate)
b7d36ffc	223	return -1;
3d8dccd7	224	break;
f258475a JH	225	}
f258475a JH	226	}
3d8dccd7	227	if (regs->nr <= i)
b7d36ffc	228	return -1;
f258475a JH	229	i = pmatch[1].rm_so >= 0 ? 1 : 0;
	230	line += pmatch[i].rm_so;
	231	result = pmatch[i].rm_eo - pmatch[i].rm_so;
	232	if (result > buffer_size)
	233	result = buffer_size;
1b6ecbad BC	234	while (result > 0 && (isspace(line[result - 1])))
1b6ecbad BC	235	result--;
f258475a	236	memcpy(buffer, line, result);
f258475a JH	237	return result;
	238	}
	239
a013585b	240	void xdiff_set_find_func(xdemitconf_t xecfg, const char value, int cflags)
f258475a JH	241	{
	242	int i;
	243	struct ff_regs *regs;
	244
	245	xecfg->find_func = ff_regexp;
	246	regs = xecfg->find_func_priv = xmalloc(sizeof(struct ff_regs));
	247	for (i = 0, regs->nr = 1; value[i]; i++)
	248	if (value[i] == '\n')
	249	regs->nr++;
b32fa95f	250	ALLOC_ARRAY(regs->array, regs->nr);
f258475a JH	251	for (i = 0; i < regs->nr; i++) {
	252	struct ff_reg *reg = regs->array + i;
	253	const char ep = strchr(value, '\n'), expression;
	254	char *buffer = NULL;
	255
	256	reg->negate = (*value == '!');
	257	if (reg->negate && i == regs->nr - 1)
	258	die("Last expression must not be negated: %s", value);
	259	if (*value == '!')
	260	value++;
	261	if (ep)
	262	expression = buffer = xstrndup(value, ep - value);
	263	else
	264	expression = value;
a013585b	265	if (regcomp(&reg->re, expression, cflags))
f258475a	266	die("Invalid regexp to look for hunk header: %s", expression);
8e0f7003	267	free(buffer);
f258475a JH	268	value = ep + 1;
	269	}
	270	}
b5412484	271
8cfe5f1c RS	272	void xdiff_clear_find_func(xdemitconf_t *xecfg)
	273	{
	274	if (xecfg->find_func) {
	275	int i;
	276	struct ff_regs *regs = xecfg->find_func_priv;
	277
	278	for (i = 0; i < regs->nr; i++)
	279	regfree(&regs->array[i].re);
	280	free(regs->array);
	281	free(regs);
	282	xecfg->find_func = NULL;
	283	xecfg->find_func_priv = NULL;
	284	}
	285	}
	286
5ec8274b SB	287	unsigned long xdiff_hash_string(const char *s, size_t len, long flags)
	288	{
	289	return xdl_hash_record(&s, s + len, flags);
	290	}
	291
	292	int xdiff_compare_lines(const char *l1, long s1,
	293	const char *l2, long s2, long flags)
	294	{
	295	return xdl_recmatch(l1, s1, l2, s2, flags);
	296	}
	297
b5412484 JH	298	int git_xmerge_style = -1;
	299
	300	int git_xmerge_config(const char var, const char value, void *cb)
	301	{
8c2be75f	302	if (!strcmp(var, "merge.conflictstyle")) {
b5412484 JH	303	if (!value)
	304	die("'%s' is not a boolean", var);
	305	if (!strcmp(value, "diff3"))
	306	git_xmerge_style = XDL_MERGE_DIFF3;
	307	else if (!strcmp(value, "merge"))
	308	git_xmerge_style = 0;
	309	else
	310	die("unknown style '%s' given for '%s'",
	311	value, var);
	312	return 0;
	313	}
	314	return git_default_config(var, value, cb);
	315	}