[thirdparty/git.git] / diffcore-pickaxe.c

/*
 * Copyright (C) 2005 Junio C Hamano
 * Copyright (C) 2010 Google Inc.
 */
#include "cache.h"
#include "diff.h"
#include "diffcore.h"
#include "xdiff-interface.h"
#include "kwset.h"

typedef int (*pickaxe_fn)(struct diff_filepair *p, struct diff_options *o, regex_t *regexp, kwset_t kws);

static void pickaxe(struct diff_queue_struct *q, struct diff_options *o,
		    regex_t *regexp, kwset_t kws, pickaxe_fn fn)
{
	int i;
	struct diff_queue_struct outq;

	DIFF_QUEUE_CLEAR(&outq);

	if (o->pickaxe_opts & DIFF_PICKAXE_ALL) {
		/* Showing the whole changeset if needle exists */
		for (i = 0; i < q->nr; i++) {
			struct diff_filepair *p = q->queue[i];
			if (fn(p, o, regexp, kws))
				return; /* do not munge the queue */
		}

		/*
		 * Otherwise we will clear the whole queue by copying
		 * the empty outq at the end of this function, but
		 * first clear the current entries in the queue.
		 */
		for (i = 0; i < q->nr; i++)
			diff_free_filepair(q->queue[i]);
	} else {
		/* Showing only the filepairs that has the needle */
		for (i = 0; i < q->nr; i++) {
			struct diff_filepair *p = q->queue[i];
			if (fn(p, o, regexp, kws))
				diff_q(&outq, p);
			else
				diff_free_filepair(p);
		}
	}

	free(q->queue);
	*q = outq;
}

struct diffgrep_cb {
	regex_t *regexp;
	int hit;
};

static void diffgrep_consume(void *priv, char *line, unsigned long len)
{
	struct diffgrep_cb *data = priv;
	regmatch_t regmatch;
	int hold;

	if (line[0] != '+' && line[0] != '-')
		return;
	if (data->hit)
		/*
		 * NEEDSWORK: we should have a way to terminate the
		 * caller early.
		 */
		return;
	/* Yuck -- line ought to be "const char *"! */
	hold = line[len];
	line[len] = '\0';
	data->hit = !regexec(data->regexp, line + 1, 1, &regmatch, 0);
	line[len] = hold;
}

static void fill_one(struct diff_filespec *one,
		     mmfile_t *mf, struct userdiff_driver **textconv)
{
	if (DIFF_FILE_VALID(one)) {
		*textconv = get_textconv(one);
		mf->size = fill_textconv(*textconv, one, &mf->ptr);
	} else {
		memset(mf, 0, sizeof(*mf));
	}
}

static int diff_grep(struct diff_filepair *p, struct diff_options *o,
		     regex_t *regexp, kwset_t kws)
{
	regmatch_t regmatch;
	struct userdiff_driver *textconv_one = NULL;
	struct userdiff_driver *textconv_two = NULL;
	mmfile_t mf1, mf2;
	int hit;

	if (diff_unmodified_pair(p))
		return 0;

	fill_one(p->one, &mf1, &textconv_one);
	fill_one(p->two, &mf2, &textconv_two);

	if (!mf1.ptr) {
		if (!mf2.ptr)
			return 0; /* ignore unmerged */
		/* created "two" -- does it have what we are looking for? */
		hit = !regexec(regexp, mf2.ptr, 1, &regmatch, 0);
	} else if (!mf2.ptr) {
		/* removed "one" -- did it have what we are looking for? */
		hit = !regexec(regexp, mf1.ptr, 1, &regmatch, 0);
	} else {
		/*
		 * We have both sides; need to run textual diff and see if
		 * the pattern appears on added/deleted lines.
		 */
		struct diffgrep_cb ecbdata;
		xpparam_t xpp;
		xdemitconf_t xecfg;

		memset(&xpp, 0, sizeof(xpp));
		memset(&xecfg, 0, sizeof(xecfg));
		ecbdata.regexp = regexp;
		ecbdata.hit = 0;
		xecfg.ctxlen = o->context;
		xecfg.interhunkctxlen = o->interhunkcontext;
		xdi_diff_outf(&mf1, &mf2, diffgrep_consume, &ecbdata,
			      &xpp, &xecfg);
		hit = ecbdata.hit;
	}
	if (textconv_one)
		free(mf1.ptr);
	if (textconv_two)
		free(mf2.ptr);
	return hit;
}

static void diffcore_pickaxe_grep(struct diff_options *o)
{
	int err;
	regex_t regex;
	int cflags = REG_EXTENDED | REG_NEWLINE;

	if (DIFF_OPT_TST(o, PICKAXE_IGNORE_CASE))
		cflags |= REG_ICASE;

	err = regcomp(&regex, o->pickaxe, cflags);
	if (err) {
		char errbuf[1024];
		regerror(err, &regex, errbuf, 1024);
		regfree(&regex);
		die("invalid log-grep regex: %s", errbuf);
	}

	pickaxe(&diff_queued_diff, o, &regex, NULL, diff_grep);

	regfree(&regex);
	return;
}

static unsigned int contains(mmfile_t *mf, struct diff_options *o,
			     regex_t *regexp, kwset_t kws)
{
	unsigned int cnt;
	unsigned long sz;
	const char *data;

	sz = mf->size;
	data = mf->ptr;
	cnt = 0;

	if (regexp) {
		regmatch_t regmatch;
		int flags = 0;

		assert(data[sz] == '\0');
		while (*data && !regexec(regexp, data, 1, &regmatch, flags)) {
			flags |= REG_NOTBOL;
			data += regmatch.rm_eo;
			if (*data && regmatch.rm_so == regmatch.rm_eo)
				data++;
			cnt++;
		}

	} else { /* Classic exact string match */
		while (sz) {
			struct kwsmatch kwsm;
			size_t offset = kwsexec(kws, data, sz, &kwsm);
			const char *found;
			if (offset == -1)
				break;
			else
				found = data + offset;
			sz -= found - data + kwsm.size[0];
			data = found + kwsm.size[0];
			cnt++;
		}
	}
	return cnt;
}

static int has_changes(struct diff_filepair *p, struct diff_options *o,
		       regex_t *regexp, kwset_t kws)
{
	struct userdiff_driver *textconv_one = get_textconv(p->one);
	struct userdiff_driver *textconv_two = get_textconv(p->two);
	mmfile_t mf1, mf2;
	int ret;

	if (!o->pickaxe[0])
		return 0;

	/*
	 * If we have an unmodified pair, we know that the count will be the
	 * same and don't even have to load the blobs. Unless textconv is in
	 * play, _and_ we are using two different textconv filters (e.g.,
	 * because a pair is an exact rename with different textconv attributes
	 * for each side, which might generate different content).
	 */
	if (textconv_one == textconv_two && diff_unmodified_pair(p))
		return 0;

	fill_one(p->one, &mf1, &textconv_one);
	fill_one(p->two, &mf2, &textconv_two);

	if (!mf1.ptr) {
		if (!mf2.ptr)
			ret = 0; /* ignore unmerged */
		/* created */
		ret = contains(&mf2, o, regexp, kws) != 0;
	}
	else if (!mf2.ptr) /* removed */
		ret = contains(&mf1, o, regexp, kws) != 0;
	else
		ret = contains(&mf1, o, regexp, kws) !=
		      contains(&mf2, o, regexp, kws);

	if (textconv_one)
		free(mf1.ptr);
	if (textconv_two)
		free(mf2.ptr);
	diff_free_filespec_data(p->one);
	diff_free_filespec_data(p->two);

	return ret;
}

static void diffcore_pickaxe_count(struct diff_options *o)
{
	const char *needle = o->pickaxe;
	int opts = o->pickaxe_opts;
	unsigned long len = strlen(needle);
	regex_t regex, *regexp = NULL;
	kwset_t kws = NULL;

	if (opts & DIFF_PICKAXE_REGEX) {
		int err;
		err = regcomp(&regex, needle, REG_EXTENDED | REG_NEWLINE);
		if (err) {
			/* The POSIX.2 people are surely sick */
			char errbuf[1024];
			regerror(err, &regex, errbuf, 1024);
			regfree(&regex);
			die("invalid pickaxe regex: %s", errbuf);
		}
		regexp = &regex;
	} else {
		kws = kwsalloc(DIFF_OPT_TST(o, PICKAXE_IGNORE_CASE)
			       ? tolower_trans_tbl : NULL);
		kwsincr(kws, needle, len);
		kwsprep(kws);
	}

	pickaxe(&diff_queued_diff, o, regexp, kws, has_changes);

	if (opts & DIFF_PICKAXE_REGEX)
		regfree(&regex);
	else
		kwsfree(kws);
	return;
}

void diffcore_pickaxe(struct diff_options *o)
{
	/* Might want to warn when both S and G are on; I don't care... */
	if (o->pickaxe_opts & DIFF_PICKAXE_KIND_G)
		diffcore_pickaxe_grep(o);
	else
		diffcore_pickaxe_count(o);
}
Commit	Line	Data
52e95789 JH	1	/*
52e95789 JH	2	* Copyright (C) 2005 Junio C Hamano
f506b8e8	3	* Copyright (C) 2010 Google Inc.
52e95789 JH	4	*/
	5	#include "cache.h"
	6	#include "diff.h"
	7	#include "diffcore.h"
f506b8e8	8	#include "xdiff-interface.h"
b95c5ada	9	#include "kwset.h"
f506b8e8	10
8a94151d RS	11	typedef int (pickaxe_fn)(struct diff_filepair p, struct diff_options o, regex_t regexp, kwset_t kws);
	12
	13	static void pickaxe(struct diff_queue_struct q, struct diff_options o,
	14	regex_t *regexp, kwset_t kws, pickaxe_fn fn)
	15	{
	16	int i;
	17	struct diff_queue_struct outq;
	18
	19	DIFF_QUEUE_CLEAR(&outq);
	20
	21	if (o->pickaxe_opts & DIFF_PICKAXE_ALL) {
	22	/* Showing the whole changeset if needle exists */
	23	for (i = 0; i < q->nr; i++) {
	24	struct diff_filepair *p = q->queue[i];
	25	if (fn(p, o, regexp, kws))
	26	return; /* do not munge the queue */
	27	}
	28
	29	/*
	30	* Otherwise we will clear the whole queue by copying
	31	* the empty outq at the end of this function, but
	32	* first clear the current entries in the queue.
	33	*/
	34	for (i = 0; i < q->nr; i++)
	35	diff_free_filepair(q->queue[i]);
	36	} else {
	37	/* Showing only the filepairs that has the needle */
	38	for (i = 0; i < q->nr; i++) {
	39	struct diff_filepair *p = q->queue[i];
	40	if (fn(p, o, regexp, kws))
	41	diff_q(&outq, p);
	42	else
	43	diff_free_filepair(p);
	44	}
	45	}
	46
	47	free(q->queue);
	48	*q = outq;
	49	}
	50
f506b8e8 JH	51	struct diffgrep_cb {
	52	regex_t *regexp;
	53	int hit;
	54	};
	55
	56	static void diffgrep_consume(void priv, char line, unsigned long len)
	57	{
	58	struct diffgrep_cb *data = priv;
	59	regmatch_t regmatch;
	60	int hold;
	61
	62	if (line[0] != '+' && line[0] != '-')
	63	return;
	64	if (data->hit)
	65	/*
	66	* NEEDSWORK: we should have a way to terminate the
	67	* caller early.
	68	*/
	69	return;
	70	/* Yuck -- line ought to be "const char "! /
	71	hold = line[len];
	72	line[len] = '\0';
	73	data->hit = !regexec(data->regexp, line + 1, 1, &regmatch, 0);
	74	line[len] = hold;
	75	}
	76
	77	static void fill_one(struct diff_filespec *one,
	78	mmfile_t mf, struct userdiff_driver *textconv)
	79	{
	80	if (DIFF_FILE_VALID(one)) {
	81	*textconv = get_textconv(one);
	82	mf->size = fill_textconv(*textconv, one, &mf->ptr);
	83	} else {
	84	memset(mf, 0, sizeof(*mf));
	85	}
	86	}
	87
db99cb70 RS	88	static int diff_grep(struct diff_filepair p, struct diff_options o,
db99cb70 RS	89	regex_t *regexp, kwset_t kws)
f506b8e8 JH	90	{
	91	regmatch_t regmatch;
	92	struct userdiff_driver *textconv_one = NULL;
	93	struct userdiff_driver *textconv_two = NULL;
	94	mmfile_t mf1, mf2;
	95	int hit;
	96
	97	if (diff_unmodified_pair(p))
	98	return 0;
	99
	100	fill_one(p->one, &mf1, &textconv_one);
	101	fill_one(p->two, &mf2, &textconv_two);
	102
	103	if (!mf1.ptr) {
	104	if (!mf2.ptr)
	105	return 0; /* ignore unmerged */
	106	/* created "two" -- does it have what we are looking for? */
b1c2f57d	107	hit = !regexec(regexp, mf2.ptr, 1, &regmatch, 0);
f506b8e8 JH	108	} else if (!mf2.ptr) {
f506b8e8 JH	109	/* removed "one" -- did it have what we are looking for? */
b1c2f57d	110	hit = !regexec(regexp, mf1.ptr, 1, &regmatch, 0);
f506b8e8 JH	111	} else {
	112	/*
	113	* We have both sides; need to run textual diff and see if
	114	* the pattern appears on added/deleted lines.
	115	*/
	116	struct diffgrep_cb ecbdata;
	117	xpparam_t xpp;
	118	xdemitconf_t xecfg;
	119
	120	memset(&xpp, 0, sizeof(xpp));
	121	memset(&xecfg, 0, sizeof(xecfg));
	122	ecbdata.regexp = regexp;
	123	ecbdata.hit = 0;
	124	xecfg.ctxlen = o->context;
	125	xecfg.interhunkctxlen = o->interhunkcontext;
	126	xdi_diff_outf(&mf1, &mf2, diffgrep_consume, &ecbdata,
	127	&xpp, &xecfg);
	128	hit = ecbdata.hit;
	129	}
	130	if (textconv_one)
	131	free(mf1.ptr);
	132	if (textconv_two)
	133	free(mf2.ptr);
	134	return hit;
	135	}
	136
	137	static void diffcore_pickaxe_grep(struct diff_options *o)
	138	{
8a94151d	139	int err;
f506b8e8	140	regex_t regex;
accccde4	141	int cflags = REG_EXTENDED \| REG_NEWLINE;
f506b8e8	142
accccde4 JH	143	if (DIFF_OPT_TST(o, PICKAXE_IGNORE_CASE))
	144	cflags \|= REG_ICASE;
	145
	146	err = regcomp(&regex, o->pickaxe, cflags);
f506b8e8 JH	147	if (err) {
	148	char errbuf[1024];
	149	regerror(err, &regex, errbuf, 1024);
	150	regfree(&regex);
	151	die("invalid log-grep regex: %s", errbuf);
	152	}
	153
8a94151d	154	pickaxe(&diff_queued_diff, o, &regex, NULL, diff_grep);
f506b8e8	155
2b5f07f1	156	regfree(&regex);
f506b8e8 JH	157	return;
f506b8e8 JH	158	}
52e95789	159
ef90ab66	160	static unsigned int contains(mmfile_t mf, struct diff_options o,
b95c5ada	161	regex_t *regexp, kwset_t kws)
52e95789	162	{
2002eed6	163	unsigned int cnt;
ce163c79	164	unsigned long sz;
52e95789	165	const char *data;
2002eed6	166
ef90ab66 JK	167	sz = mf->size;
ef90ab66 JK	168	data = mf->ptr;
2002eed6 JH	169	cnt = 0;
2002eed6 JH	170
d01d8c67 PB	171	if (regexp) {
	172	regmatch_t regmatch;
	173	int flags = 0;
	174
50fd6997	175	assert(data[sz] == '\0');
d01d8c67 PB	176	while (*data && !regexec(regexp, data, 1, &regmatch, flags)) {
d01d8c67 PB	177	flags \|= REG_NOTBOL;
50fd6997 RS	178	data += regmatch.rm_eo;
	179	if (*data && regmatch.rm_so == regmatch.rm_eo)
	180	data++;
2002eed6 JH	181	cnt++;
2002eed6 JH	182	}
d01d8c67 PB	183
d01d8c67 PB	184	} else { /* Classic exact string match */
ce163c79	185	while (sz) {
5d176fb6 RS	186	struct kwsmatch kwsm;
5d176fb6 RS	187	size_t offset = kwsexec(kws, data, sz, &kwsm);
b95c5ada FK	188	const char *found;
b95c5ada FK	189	if (offset == -1)
ce163c79	190	break;
b95c5ada FK	191	else
b95c5ada FK	192	found = data + offset;
5d176fb6 RS	193	sz -= found - data + kwsm.size[0];
5d176fb6 RS	194	data = found + kwsm.size[0];
ce163c79	195	cnt++;
d01d8c67	196	}
2002eed6 JH	197	}
2002eed6 JH	198	return cnt;
52e95789 JH	199	}
52e95789 JH	200
5d176fb6 RS	201	static int has_changes(struct diff_filepair p, struct diff_options o,
5d176fb6 RS	202	regex_t *regexp, kwset_t kws)
15dafaf8	203	{
ef90ab66 JK	204	struct userdiff_driver *textconv_one = get_textconv(p->one);
	205	struct userdiff_driver *textconv_two = get_textconv(p->two);
	206	mmfile_t mf1, mf2;
	207	int ret;
	208
8fa4b09f JK	209	if (!o->pickaxe[0])
	210	return 0;
	211
ef90ab66 JK	212	/*
	213	* If we have an unmodified pair, we know that the count will be the
	214	* same and don't even have to load the blobs. Unless textconv is in
	215	* play, _and_ we are using two different textconv filters (e.g.,
	216	* because a pair is an exact rename with different textconv attributes
	217	* for each side, which might generate different content).
	218	*/
	219	if (textconv_one == textconv_two && diff_unmodified_pair(p))
	220	return 0;
	221
	222	fill_one(p->one, &mf1, &textconv_one);
	223	fill_one(p->two, &mf2, &textconv_two);
	224
	225	if (!mf1.ptr) {
	226	if (!mf2.ptr)
	227	ret = 0; /* ignore unmerged */
15dafaf8	228	/* created */
ef90ab66	229	ret = contains(&mf2, o, regexp, kws) != 0;
15dafaf8	230	}
ef90ab66 JK	231	else if (!mf2.ptr) /* removed */
	232	ret = contains(&mf1, o, regexp, kws) != 0;
	233	else
	234	ret = contains(&mf1, o, regexp, kws) !=
	235	contains(&mf2, o, regexp, kws);
	236
	237	if (textconv_one)
	238	free(mf1.ptr);
	239	if (textconv_two)
	240	free(mf2.ptr);
	241	diff_free_filespec_data(p->one);
	242	diff_free_filespec_data(p->two);
	243
	244	return ret;
15dafaf8 RS	245	}
15dafaf8 RS	246
f506b8e8	247	static void diffcore_pickaxe_count(struct diff_options *o)
52e95789	248	{
382f013b JH	249	const char *needle = o->pickaxe;
382f013b JH	250	int opts = o->pickaxe_opts;
52e95789	251	unsigned long len = strlen(needle);
d01d8c67	252	regex_t regex, *regexp = NULL;
b95c5ada	253	kwset_t kws = NULL;
52e95789	254
d01d8c67 PB	255	if (opts & DIFF_PICKAXE_REGEX) {
	256	int err;
	257	err = regcomp(&regex, needle, REG_EXTENDED \| REG_NEWLINE);
	258	if (err) {
	259	/* The POSIX.2 people are surely sick */
	260	char errbuf[1024];
	261	regerror(err, &regex, errbuf, 1024);
	262	regfree(&regex);
	263	die("invalid pickaxe regex: %s", errbuf);
	264	}
	265	regexp = &regex;
b95c5ada	266	} else {
accccde4 JH	267	kws = kwsalloc(DIFF_OPT_TST(o, PICKAXE_IGNORE_CASE)
accccde4 JH	268	? tolower_trans_tbl : NULL);
b95c5ada FK	269	kwsincr(kws, needle, len);
b95c5ada FK	270	kwsprep(kws);
d01d8c67 PB	271	}
d01d8c67 PB	272
8a94151d	273	pickaxe(&diff_queued_diff, o, regexp, kws, has_changes);
8e854b00	274
95ae69b9	275	if (opts & DIFF_PICKAXE_REGEX)
d01d8c67	276	regfree(&regex);
b95c5ada FK	277	else
b95c5ada FK	278	kwsfree(kws);
52e95789 JH	279	return;
52e95789 JH	280	}
f506b8e8 JH	281
	282	void diffcore_pickaxe(struct diff_options *o)
	283	{
	284	/* Might want to warn when both S and G are on; I don't care... */
	285	if (o->pickaxe_opts & DIFF_PICKAXE_KIND_G)
8520913c	286	diffcore_pickaxe_grep(o);
f506b8e8	287	else
8520913c	288	diffcore_pickaxe_count(o);
f506b8e8	289	}