[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"

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, regex_t *regexp, struct diff_options *o)
{
	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, p->two->data, 1, &regmatch, 0);
	} else if (!mf2.ptr) {
		/* removed "one" -- did it have what we are looking for? */
		hit = !regexec(regexp, p->one->data, 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)
{
	struct diff_queue_struct *q = &diff_queued_diff;
	int i, err;
	regex_t regex;
	struct diff_queue_struct outq;
	outq.queue = NULL;
	outq.nr = outq.alloc = 0;

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

	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 (diff_grep(p, &regex, o))
				goto out; /* 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 (diff_grep(p, &regex, o))
				diff_q(&outq, p);
			else
				diff_free_filepair(p);
		}
	}

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

 out:
	regfree(&regex);
	return;
}

static unsigned int contains(struct diff_filespec *one,
			     const char *needle, unsigned long len,
			     regex_t *regexp, kwset_t kws)
{
	unsigned int cnt;
	unsigned long sz;
	const char *data;
	if (!len)
		return 0;
	if (diff_populate_filespec(one, 0))
		return 0;

	sz = one->size;
	data = one->data;
	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) {
			size_t offset = kwsexec(kws, data, sz, NULL);
			const char *found;
			if (offset == -1)
				break;
			else
				found = data + offset;
			sz -= found - data + len;
			data = found + len;
			cnt++;
		}
	}
	diff_free_filespec_data(one);
	return cnt;
}

static int has_changes(struct diff_filepair *p, const char *needle,
		       unsigned long len, regex_t *regexp, kwset_t kws)
{
	if (!DIFF_FILE_VALID(p->one)) {
		if (!DIFF_FILE_VALID(p->two))
			return 0; /* ignore unmerged */
		/* created */
		return contains(p->two, needle, len, regexp, kws) != 0;
	}
	if (!DIFF_FILE_VALID(p->two))
		return contains(p->one, needle, len, regexp, kws) != 0;
	if (!diff_unmodified_pair(p)) {
		return contains(p->one, needle, len, regexp, kws) !=
		       contains(p->two, needle, len, regexp, kws);
	}
	return 0;
}

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

	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(NULL);
		kwsincr(kws, needle, len);
		kwsprep(kws);
	}

	if (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 (has_changes(p, needle, len, regexp, kws))
				goto out; /* 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 (has_changes(p, needle, len, regexp, kws))
				diff_q(&outq, p);
			else
				diff_free_filepair(p);
		}

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

 out:
	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 JH	10
	11	struct diffgrep_cb {
	12	regex_t *regexp;
	13	int hit;
	14	};
	15
	16	static void diffgrep_consume(void priv, char line, unsigned long len)
	17	{
	18	struct diffgrep_cb *data = priv;
	19	regmatch_t regmatch;
	20	int hold;
	21
	22	if (line[0] != '+' && line[0] != '-')
	23	return;
	24	if (data->hit)
	25	/*
	26	* NEEDSWORK: we should have a way to terminate the
	27	* caller early.
	28	*/
	29	return;
	30	/* Yuck -- line ought to be "const char "! /
	31	hold = line[len];
	32	line[len] = '\0';
	33	data->hit = !regexec(data->regexp, line + 1, 1, &regmatch, 0);
	34	line[len] = hold;
	35	}
	36
	37	static void fill_one(struct diff_filespec *one,
	38	mmfile_t mf, struct userdiff_driver *textconv)
	39	{
	40	if (DIFF_FILE_VALID(one)) {
	41	*textconv = get_textconv(one);
	42	mf->size = fill_textconv(*textconv, one, &mf->ptr);
	43	} else {
	44	memset(mf, 0, sizeof(*mf));
	45	}
	46	}
	47
	48	static int diff_grep(struct diff_filepair p, regex_t regexp, struct diff_options *o)
	49	{
	50	regmatch_t regmatch;
	51	struct userdiff_driver *textconv_one = NULL;
	52	struct userdiff_driver *textconv_two = NULL;
	53	mmfile_t mf1, mf2;
	54	int hit;
	55
	56	if (diff_unmodified_pair(p))
	57	return 0;
	58
	59	fill_one(p->one, &mf1, &textconv_one);
	60	fill_one(p->two, &mf2, &textconv_two);
	61
	62	if (!mf1.ptr) {
	63	if (!mf2.ptr)
	64	return 0; /* ignore unmerged */
	65	/* created "two" -- does it have what we are looking for? */
	66	hit = !regexec(regexp, p->two->data, 1, &regmatch, 0);
	67	} else if (!mf2.ptr) {
	68	/* removed "one" -- did it have what we are looking for? */
	69	hit = !regexec(regexp, p->one->data, 1, &regmatch, 0);
	70	} else {
	71	/*
	72	* We have both sides; need to run textual diff and see if
	73	* the pattern appears on added/deleted lines.
74	*/
75	struct diffgrep_cb ecbdata;
76	xpparam_t xpp;
77	xdemitconf_t xecfg;
78
79	memset(&xpp, 0, sizeof(xpp));
80	memset(&xecfg, 0, sizeof(xecfg));
81	ecbdata.regexp = regexp;
82	ecbdata.hit = 0;
83	xecfg.ctxlen = o->context;
84	xecfg.interhunkctxlen = o->interhunkcontext;
85	xdi_diff_outf(&mf1, &mf2, diffgrep_consume, &ecbdata,
86	&xpp, &xecfg);
87	hit = ecbdata.hit;
88	}
89	if (textconv_one)
90	free(mf1.ptr);
91	if (textconv_two)
92	free(mf2.ptr);
93	return hit;
94	}
95
96	static void diffcore_pickaxe_grep(struct diff_options *o)
97	{
98	struct diff_queue_struct *q = &diff_queued_diff;
2b5f07f1	99	int i, err;
f506b8e8 JH	100	regex_t regex;
	101	struct diff_queue_struct outq;
	102	outq.queue = NULL;
	103	outq.nr = outq.alloc = 0;
	104
	105	err = regcomp(&regex, o->pickaxe, REG_EXTENDED \| REG_NEWLINE);
	106	if (err) {
	107	char errbuf[1024];
	108	regerror(err, &regex, errbuf, 1024);
	109	regfree(&regex);
	110	die("invalid log-grep regex: %s", errbuf);
	111	}
	112
	113	if (o->pickaxe_opts & DIFF_PICKAXE_ALL) {
	114	/* Showing the whole changeset if needle exists */
2b5f07f1	115	for (i = 0; i < q->nr; i++) {
f506b8e8 JH	116	struct diff_filepair *p = q->queue[i];
f506b8e8 JH	117	if (diff_grep(p, &regex, o))
2b5f07f1	118	goto out; /* do not munge the queue */
f506b8e8	119	}
f506b8e8 JH	120
	121	/*
	122	* Otherwise we will clear the whole queue by copying
	123	* the empty outq at the end of this function, but
	124	* first clear the current entries in the queue.
	125	*/
	126	for (i = 0; i < q->nr; i++)
	127	diff_free_filepair(q->queue[i]);
	128	} else {
	129	/* Showing only the filepairs that has the needle */
	130	for (i = 0; i < q->nr; i++) {
	131	struct diff_filepair *p = q->queue[i];
	132	if (diff_grep(p, &regex, o))
	133	diff_q(&outq, p);
	134	else
	135	diff_free_filepair(p);
	136	}
	137	}
	138
f506b8e8 JH	139	free(q->queue);
f506b8e8 JH	140	*q = outq;
2b5f07f1 RS	141
	142	out:
	143	regfree(&regex);
f506b8e8 JH	144	return;
f506b8e8 JH	145	}
52e95789	146
2002eed6	147	static unsigned int contains(struct diff_filespec *one,
d01d8c67	148	const char *needle, unsigned long len,
b95c5ada	149	regex_t *regexp, kwset_t kws)
52e95789	150	{
2002eed6	151	unsigned int cnt;
ce163c79	152	unsigned long sz;
52e95789	153	const char *data;
e1b16116 JK	154	if (!len)
e1b16116 JK	155	return 0;
05ac9784 RS	156	if (diff_populate_filespec(one, 0))
05ac9784 RS	157	return 0;
2002eed6	158
52e95789 JH	159	sz = one->size;
52e95789 JH	160	data = one->data;
2002eed6 JH	161	cnt = 0;
2002eed6 JH	162
d01d8c67 PB	163	if (regexp) {
	164	regmatch_t regmatch;
	165	int flags = 0;
	166
50fd6997	167	assert(data[sz] == '\0');
d01d8c67 PB	168	while (*data && !regexec(regexp, data, 1, &regmatch, flags)) {
d01d8c67 PB	169	flags \|= REG_NOTBOL;
50fd6997 RS	170	data += regmatch.rm_eo;
	171	if (*data && regmatch.rm_so == regmatch.rm_eo)
	172	data++;
2002eed6 JH	173	cnt++;
2002eed6 JH	174	}
d01d8c67 PB	175
d01d8c67 PB	176	} else { /* Classic exact string match */
ce163c79	177	while (sz) {
b95c5ada FK	178	size_t offset = kwsexec(kws, data, sz, NULL);
	179	const char *found;
	180	if (offset == -1)
ce163c79	181	break;
b95c5ada FK	182	else
b95c5ada FK	183	found = data + offset;
ce163c79 RS	184	sz -= found - data + len;
	185	data = found + len;
	186	cnt++;
d01d8c67	187	}
2002eed6	188	}
a0cb9400	189	diff_free_filespec_data(one);
2002eed6	190	return cnt;
52e95789 JH	191	}
52e95789 JH	192
15dafaf8 RS	193	static int has_changes(struct diff_filepair p, const char needle,
	194	unsigned long len, regex_t *regexp, kwset_t kws)
	195	{
	196	if (!DIFF_FILE_VALID(p->one)) {
	197	if (!DIFF_FILE_VALID(p->two))
	198	return 0; /* ignore unmerged */
	199	/* created */
	200	return contains(p->two, needle, len, regexp, kws) != 0;
	201	}
	202	if (!DIFF_FILE_VALID(p->two))
	203	return contains(p->one, needle, len, regexp, kws) != 0;
	204	if (!diff_unmodified_pair(p)) {
	205	return contains(p->one, needle, len, regexp, kws) !=
	206	contains(p->two, needle, len, regexp, kws);
	207	}
	208	return 0;
	209	}
	210
f506b8e8	211	static void diffcore_pickaxe_count(struct diff_options *o)
52e95789	212	{
382f013b JH	213	const char *needle = o->pickaxe;
382f013b JH	214	int opts = o->pickaxe_opts;
38c6f780	215	struct diff_queue_struct *q = &diff_queued_diff;
52e95789	216	unsigned long len = strlen(needle);
15dafaf8	217	int i;
d01d8c67	218	regex_t regex, *regexp = NULL;
b95c5ada	219	kwset_t kws = NULL;
52e95789	220	struct diff_queue_struct outq;
9ca5df90	221	DIFF_QUEUE_CLEAR(&outq);
52e95789	222
d01d8c67 PB	223	if (opts & DIFF_PICKAXE_REGEX) {
	224	int err;
	225	err = regcomp(&regex, needle, REG_EXTENDED \| REG_NEWLINE);
	226	if (err) {
	227	/* The POSIX.2 people are surely sick */
	228	char errbuf[1024];
	229	regerror(err, &regex, errbuf, 1024);
	230	regfree(&regex);
	231	die("invalid pickaxe regex: %s", errbuf);
	232	}
	233	regexp = &regex;
b95c5ada FK	234	} else {
	235	kws = kwsalloc(NULL);
	236	kwsincr(kws, needle, len);
	237	kwsprep(kws);
d01d8c67 PB	238	}
d01d8c67 PB	239
367cec1c JH	240	if (opts & DIFF_PICKAXE_ALL) {
367cec1c JH	241	/* Showing the whole changeset if needle exists */
8e854b00	242	for (i = 0; i < q->nr; i++) {
367cec1c	243	struct diff_filepair *p = q->queue[i];
15dafaf8	244	if (has_changes(p, needle, len, regexp, kws))
8e854b00	245	goto out; /* do not munge the queue */
52e95789	246	}
367cec1c JH	247
	248	/* otherwise we will clear the whole queue
	249	* by copying the empty outq at the end of this
	250	* function, but first clear the current entries
	251	* in the queue.
	252	*/
	253	for (i = 0; i < q->nr; i++)
	254	diff_free_filepair(q->queue[i]);
	255	}
a6080a0a	256	else
367cec1c JH	257	/* Showing only the filepairs that has the needle */
	258	for (i = 0; i < q->nr; i++) {
	259	struct diff_filepair *p = q->queue[i];
15dafaf8	260	if (has_changes(p, needle, len, regexp, kws))
6b14d7fa	261	diff_q(&outq, p);
367cec1c JH	262	else
367cec1c JH	263	diff_free_filepair(p);
52e95789	264	}
367cec1c	265
8e854b00 RS	266	free(q->queue);
	267	*q = outq;
	268
	269	out:
95ae69b9	270	if (opts & DIFF_PICKAXE_REGEX)
d01d8c67	271	regfree(&regex);
b95c5ada FK	272	else
b95c5ada FK	273	kwsfree(kws);
52e95789 JH	274	return;
52e95789 JH	275	}
f506b8e8 JH	276
	277	void diffcore_pickaxe(struct diff_options *o)
	278	{
	279	/* Might want to warn when both S and G are on; I don't care... */
	280	if (o->pickaxe_opts & DIFF_PICKAXE_KIND_G)
8520913c	281	diffcore_pickaxe_grep(o);
f506b8e8	282	else
8520913c	283	diffcore_pickaxe_count(o);
f506b8e8	284	}