[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, has_changes, 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 = has_changes = 0; !has_changes && i < q->nr; i++) {
			struct diff_filepair *p = q->queue[i];
			if (diff_grep(p, &regex, o))
				has_changes++;
		}
		if (has_changes)
			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 (diff_grep(p, &regex, o))
				diff_q(&outq, p);
			else
				diff_free_filepair(p);
		}
	}

	regfree(&regex);

	free(q->queue);
	*q = outq;
	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 (diff_populate_filespec(one, 0))
		return 0;
	if (!len)
		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 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, has_changes;
	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 = has_changes = 0; !has_changes && i < q->nr; i++) {
			struct diff_filepair *p = q->queue[i];
			if (!DIFF_FILE_VALID(p->one)) {
				if (!DIFF_FILE_VALID(p->two))
					continue; /* ignore unmerged */
				/* created */
				if (contains(p->two, needle, len, regexp, kws))
					has_changes++;
			}
			else if (!DIFF_FILE_VALID(p->two)) {
				if (contains(p->one, needle, len, regexp, kws))
					has_changes++;
			}
			else if (!diff_unmodified_pair(p) &&
				 contains(p->one, needle, len, regexp, kws) !=
				 contains(p->two, needle, len, regexp, kws))
				has_changes++;
		}
		if (has_changes)
			return; /* 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];
			has_changes = 0;
			if (!DIFF_FILE_VALID(p->one)) {
				if (!DIFF_FILE_VALID(p->two))
					; /* ignore unmerged */
				/* created */
				else if (contains(p->two, needle, len, regexp,
						  kws))
					has_changes = 1;
			}
			else if (!DIFF_FILE_VALID(p->two)) {
				if (contains(p->one, needle, len, regexp, kws))
					has_changes = 1;
			}
			else if (!diff_unmodified_pair(p) &&
				 contains(p->one, needle, len, regexp, kws) !=
				 contains(p->two, needle, len, regexp, kws))
				has_changes = 1;

			if (has_changes)
				diff_q(&outq, p);
			else
				diff_free_filepair(p);
		}

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

	free(q->queue);
	*q = outq;
	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;
99	int i, has_changes, err;
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 */
115	for (i = has_changes = 0; !has_changes && i < q->nr; i++) {
116	struct diff_filepair *p = q->queue[i];
117	if (diff_grep(p, &regex, o))
118	has_changes++;
119	}
120	if (has_changes)
121	return; /* do not munge the queue */
122
123	/*
124	* Otherwise we will clear the whole queue by copying
125	* the empty outq at the end of this function, but
126	* first clear the current entries in the queue.
127	*/
128	for (i = 0; i < q->nr; i++)
129	diff_free_filepair(q->queue[i]);
130	} else {
131	/* Showing only the filepairs that has the needle */
132	for (i = 0; i < q->nr; i++) {
133	struct diff_filepair *p = q->queue[i];
134	if (diff_grep(p, &regex, o))
135	diff_q(&outq, p);
136	else
137	diff_free_filepair(p);
138	}
139	}
140
141	regfree(&regex);
142
143	free(q->queue);
144	*q = outq;
145	return;
146	}
52e95789	147
2002eed6	148	static unsigned int contains(struct diff_filespec *one,
d01d8c67	149	const char *needle, unsigned long len,
b95c5ada	150	regex_t *regexp, kwset_t kws)
52e95789	151	{
2002eed6	152	unsigned int cnt;
ce163c79	153	unsigned long sz;
52e95789	154	const char *data;
f0c6b2a2	155	if (diff_populate_filespec(one, 0))
52e95789	156	return 0;
e1b16116 JK	157	if (!len)
e1b16116 JK	158	return 0;
2002eed6	159
52e95789 JH	160	sz = one->size;
52e95789 JH	161	data = one->data;
2002eed6 JH	162	cnt = 0;
2002eed6 JH	163
d01d8c67 PB	164	if (regexp) {
	165	regmatch_t regmatch;
	166	int flags = 0;
	167
50fd6997	168	assert(data[sz] == '\0');
d01d8c67 PB	169	while (*data && !regexec(regexp, data, 1, &regmatch, flags)) {
d01d8c67 PB	170	flags \|= REG_NOTBOL;
50fd6997 RS	171	data += regmatch.rm_eo;
	172	if (*data && regmatch.rm_so == regmatch.rm_eo)
	173	data++;
2002eed6 JH	174	cnt++;
2002eed6 JH	175	}
d01d8c67 PB	176
d01d8c67 PB	177	} else { /* Classic exact string match */
ce163c79	178	while (sz) {
b95c5ada FK	179	size_t offset = kwsexec(kws, data, sz, NULL);
	180	const char *found;
	181	if (offset == -1)
ce163c79	182	break;
b95c5ada FK	183	else
b95c5ada FK	184	found = data + offset;
ce163c79 RS	185	sz -= found - data + len;
	186	data = found + len;
	187	cnt++;
d01d8c67	188	}
2002eed6	189	}
a0cb9400	190	diff_free_filespec_data(one);
2002eed6	191	return cnt;
52e95789 JH	192	}
52e95789 JH	193
f506b8e8	194	static void diffcore_pickaxe_count(struct diff_options *o)
52e95789	195	{
382f013b JH	196	const char *needle = o->pickaxe;
382f013b JH	197	int opts = o->pickaxe_opts;
38c6f780	198	struct diff_queue_struct *q = &diff_queued_diff;
52e95789	199	unsigned long len = strlen(needle);
367cec1c	200	int i, has_changes;
d01d8c67	201	regex_t regex, *regexp = NULL;
b95c5ada	202	kwset_t kws = NULL;
52e95789	203	struct diff_queue_struct outq;
9ca5df90	204	DIFF_QUEUE_CLEAR(&outq);
52e95789	205
d01d8c67 PB	206	if (opts & DIFF_PICKAXE_REGEX) {
	207	int err;
	208	err = regcomp(&regex, needle, REG_EXTENDED \| REG_NEWLINE);
	209	if (err) {
	210	/* The POSIX.2 people are surely sick */
	211	char errbuf[1024];
	212	regerror(err, &regex, errbuf, 1024);
	213	regfree(&regex);
	214	die("invalid pickaxe regex: %s", errbuf);
	215	}
	216	regexp = &regex;
b95c5ada FK	217	} else {
	218	kws = kwsalloc(NULL);
	219	kwsincr(kws, needle, len);
	220	kwsprep(kws);
d01d8c67 PB	221	}
d01d8c67 PB	222
367cec1c JH	223	if (opts & DIFF_PICKAXE_ALL) {
	224	/* Showing the whole changeset if needle exists */
	225	for (i = has_changes = 0; !has_changes && i < q->nr; i++) {
	226	struct diff_filepair *p = q->queue[i];
	227	if (!DIFF_FILE_VALID(p->one)) {
	228	if (!DIFF_FILE_VALID(p->two))
	229	continue; /* ignore unmerged */
	230	/* created */
b95c5ada	231	if (contains(p->two, needle, len, regexp, kws))
367cec1c JH	232	has_changes++;
	233	}
	234	else if (!DIFF_FILE_VALID(p->two)) {
b95c5ada	235	if (contains(p->one, needle, len, regexp, kws))
367cec1c JH	236	has_changes++;
	237	}
	238	else if (!diff_unmodified_pair(p) &&
b95c5ada FK	239	contains(p->one, needle, len, regexp, kws) !=
b95c5ada FK	240	contains(p->two, needle, len, regexp, kws))
367cec1c	241	has_changes++;
52e95789	242	}
367cec1c JH	243	if (has_changes)
	244	return; /* not munge the queue */
	245
	246	/* otherwise we will clear the whole queue
	247	* by copying the empty outq at the end of this
	248	* function, but first clear the current entries
	249	* in the queue.
	250	*/
	251	for (i = 0; i < q->nr; i++)
	252	diff_free_filepair(q->queue[i]);
	253	}
a6080a0a	254	else
367cec1c JH	255	/* Showing only the filepairs that has the needle */
	256	for (i = 0; i < q->nr; i++) {
	257	struct diff_filepair *p = q->queue[i];
	258	has_changes = 0;
	259	if (!DIFF_FILE_VALID(p->one)) {
	260	if (!DIFF_FILE_VALID(p->two))
	261	; /* ignore unmerged */
	262	/* created */
b95c5ada FK	263	else if (contains(p->two, needle, len, regexp,
b95c5ada FK	264	kws))
367cec1c JH	265	has_changes = 1;
	266	}
	267	else if (!DIFF_FILE_VALID(p->two)) {
b95c5ada	268	if (contains(p->one, needle, len, regexp, kws))
367cec1c JH	269	has_changes = 1;
	270	}
	271	else if (!diff_unmodified_pair(p) &&
b95c5ada FK	272	contains(p->one, needle, len, regexp, kws) !=
b95c5ada FK	273	contains(p->two, needle, len, regexp, kws))
367cec1c JH	274	has_changes = 1;
	275
	276	if (has_changes)
6b14d7fa	277	diff_q(&outq, p);
367cec1c JH	278	else
367cec1c JH	279	diff_free_filepair(p);
52e95789	280	}
367cec1c	281
95ae69b9	282	if (opts & DIFF_PICKAXE_REGEX)
d01d8c67	283	regfree(&regex);
b95c5ada FK	284	else
b95c5ada FK	285	kwsfree(kws);
d01d8c67	286
52e95789 JH	287	free(q->queue);
	288	*q = outq;
	289	return;
	290	}
f506b8e8 JH	291
	292	void diffcore_pickaxe(struct diff_options *o)
	293	{
	294	/* Might want to warn when both S and G are on; I don't care... */
	295	if (o->pickaxe_opts & DIFF_PICKAXE_KIND_G)
8520913c	296	diffcore_pickaxe_grep(o);
f506b8e8	297	else
8520913c	298	diffcore_pickaxe_count(o);
f506b8e8	299	}