[thirdparty/git.git] / grep.c

#include "cache.h"
#include "grep.h"
#include "xdiff-interface.h"

void append_header_grep_pattern(struct grep_opt *opt, enum grep_header_field field, const char *pat)
{
	struct grep_pat *p = xcalloc(1, sizeof(*p));
	p->pattern = pat;
	p->origin = "header";
	p->no = 0;
	p->token = GREP_PATTERN_HEAD;
	p->field = field;
	*opt->pattern_tail = p;
	opt->pattern_tail = &p->next;
	p->next = NULL;
}

void append_grep_pattern(struct grep_opt *opt, const char *pat,
			 const char *origin, int no, enum grep_pat_token t)
{
	struct grep_pat *p = xcalloc(1, sizeof(*p));
	p->pattern = pat;
	p->origin = origin;
	p->no = no;
	p->token = t;
	*opt->pattern_tail = p;
	opt->pattern_tail = &p->next;
	p->next = NULL;
}

static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
{
	int err = regcomp(&p->regexp, p->pattern, opt->regflags);
	if (err) {
		char errbuf[1024];
		char where[1024];
		if (p->no)
			sprintf(where, "In '%s' at %d, ",
				p->origin, p->no);
		else if (p->origin)
			sprintf(where, "%s, ", p->origin);
		else
			where[0] = 0;
		regerror(err, &p->regexp, errbuf, 1024);
		regfree(&p->regexp);
		die("%s'%s': %s", where, p->pattern, errbuf);
	}
}

static struct grep_expr *compile_pattern_or(struct grep_pat **);
static struct grep_expr *compile_pattern_atom(struct grep_pat **list)
{
	struct grep_pat *p;
	struct grep_expr *x;

	p = *list;
	switch (p->token) {
	case GREP_PATTERN: /* atom */
	case GREP_PATTERN_HEAD:
	case GREP_PATTERN_BODY:
		x = xcalloc(1, sizeof (struct grep_expr));
		x->node = GREP_NODE_ATOM;
		x->u.atom = p;
		*list = p->next;
		return x;
	case GREP_OPEN_PAREN:
		*list = p->next;
		x = compile_pattern_or(list);
		if (!x)
			return NULL;
		if (!*list || (*list)->token != GREP_CLOSE_PAREN)
			die("unmatched parenthesis");
		*list = (*list)->next;
		return x;
	default:
		return NULL;
	}
}

static struct grep_expr *compile_pattern_not(struct grep_pat **list)
{
	struct grep_pat *p;
	struct grep_expr *x;

	p = *list;
	switch (p->token) {
	case GREP_NOT:
		if (!p->next)
			die("--not not followed by pattern expression");
		*list = p->next;
		x = xcalloc(1, sizeof (struct grep_expr));
		x->node = GREP_NODE_NOT;
		x->u.unary = compile_pattern_not(list);
		if (!x->u.unary)
			die("--not followed by non pattern expression");
		return x;
	default:
		return compile_pattern_atom(list);
	}
}

static struct grep_expr *compile_pattern_and(struct grep_pat **list)
{
	struct grep_pat *p;
	struct grep_expr *x, *y, *z;

	x = compile_pattern_not(list);
	p = *list;
	if (p && p->token == GREP_AND) {
		if (!p->next)
			die("--and not followed by pattern expression");
		*list = p->next;
		y = compile_pattern_and(list);
		if (!y)
			die("--and not followed by pattern expression");
		z = xcalloc(1, sizeof (struct grep_expr));
		z->node = GREP_NODE_AND;
		z->u.binary.left = x;
		z->u.binary.right = y;
		return z;
	}
	return x;
}

static struct grep_expr *compile_pattern_or(struct grep_pat **list)
{
	struct grep_pat *p;
	struct grep_expr *x, *y, *z;

	x = compile_pattern_and(list);
	p = *list;
	if (x && p && p->token != GREP_CLOSE_PAREN) {
		y = compile_pattern_or(list);
		if (!y)
			die("not a pattern expression %s", p->pattern);
		z = xcalloc(1, sizeof (struct grep_expr));
		z->node = GREP_NODE_OR;
		z->u.binary.left = x;
		z->u.binary.right = y;
		return z;
	}
	return x;
}

static struct grep_expr *compile_pattern_expr(struct grep_pat **list)
{
	return compile_pattern_or(list);
}

void compile_grep_patterns(struct grep_opt *opt)
{
	struct grep_pat *p;

	if (opt->all_match)
		opt->extended = 1;

	for (p = opt->pattern_list; p; p = p->next) {
		switch (p->token) {
		case GREP_PATTERN: /* atom */
		case GREP_PATTERN_HEAD:
		case GREP_PATTERN_BODY:
			if (!opt->fixed)
				compile_regexp(p, opt);
			break;
		default:
			opt->extended = 1;
			break;
		}
	}

	if (!opt->extended)
		return;

	/* Then bundle them up in an expression.
	 * A classic recursive descent parser would do.
	 */
	p = opt->pattern_list;
	opt->pattern_expression = compile_pattern_expr(&p);
	if (p)
		die("incomplete pattern expression: %s", p->pattern);
}

static void free_pattern_expr(struct grep_expr *x)
{
	switch (x->node) {
	case GREP_NODE_ATOM:
		break;
	case GREP_NODE_NOT:
		free_pattern_expr(x->u.unary);
		break;
	case GREP_NODE_AND:
	case GREP_NODE_OR:
		free_pattern_expr(x->u.binary.left);
		free_pattern_expr(x->u.binary.right);
		break;
	}
	free(x);
}

void free_grep_patterns(struct grep_opt *opt)
{
	struct grep_pat *p, *n;

	for (p = opt->pattern_list; p; p = n) {
		n = p->next;
		switch (p->token) {
		case GREP_PATTERN: /* atom */
		case GREP_PATTERN_HEAD:
		case GREP_PATTERN_BODY:
			regfree(&p->regexp);
			break;
		default:
			break;
		}
		free(p);
	}

	if (!opt->extended)
		return;
	free_pattern_expr(opt->pattern_expression);
}

static char *end_of_line(char *cp, unsigned long *left)
{
	unsigned long l = *left;
	while (l && *cp != '\n') {
		l--;
		cp++;
	}
	*left = l;
	return cp;
}

static int word_char(char ch)
{
	return isalnum(ch) || ch == '_';
}

static void show_line(struct grep_opt *opt, const char *bol, const char *eol,
		      const char *name, unsigned lno, char sign)
{
	if (opt->pathname)
		printf("%s%c", name, sign);
	if (opt->linenum)
		printf("%d%c", lno, sign);
	printf("%.*s\n", (int)(eol-bol), bol);
}

static int fixmatch(const char *pattern, char *line, regmatch_t *match)
{
	char *hit = strstr(line, pattern);
	if (!hit) {
		match->rm_so = match->rm_eo = -1;
		return REG_NOMATCH;
	}
	else {
		match->rm_so = hit - line;
		match->rm_eo = match->rm_so + strlen(pattern);
		return 0;
	}
}

static int strip_timestamp(char *bol, char **eol_p)
{
	char *eol = *eol_p;
	int ch;

	while (bol < --eol) {
		if (*eol != '>')
			continue;
		*eol_p = ++eol;
		ch = *eol;
		*eol = '\0';
		return ch;
	}
	return 0;
}

static struct {
	const char *field;
	size_t len;
} header_field[] = {
	{ "author ", 7 },
	{ "committer ", 10 },
};

static int match_one_pattern(struct grep_opt *opt, struct grep_pat *p, char *bol, char *eol, enum grep_context ctx)
{
	int hit = 0;
	int at_true_bol = 1;
	int saved_ch = 0;
	regmatch_t pmatch[10];

	if ((p->token != GREP_PATTERN) &&
	    ((p->token == GREP_PATTERN_HEAD) != (ctx == GREP_CONTEXT_HEAD)))
		return 0;

	if (p->token == GREP_PATTERN_HEAD) {
		const char *field;
		size_t len;
		assert(p->field < ARRAY_SIZE(header_field));
		field = header_field[p->field].field;
		len = header_field[p->field].len;
		if (strncmp(bol, field, len))
			return 0;
		bol += len;
		saved_ch = strip_timestamp(bol, &eol);
	}

 again:
	if (!opt->fixed) {
		regex_t *exp = &p->regexp;
		hit = !regexec(exp, bol, ARRAY_SIZE(pmatch),
			       pmatch, 0);
	}
	else {
		hit = !fixmatch(p->pattern, bol, pmatch);
	}

	if (hit && opt->word_regexp) {
		if ((pmatch[0].rm_so < 0) ||
		    (eol - bol) <= pmatch[0].rm_so ||
		    (pmatch[0].rm_eo < 0) ||
		    (eol - bol) < pmatch[0].rm_eo)
			die("regexp returned nonsense");

		/* Match beginning must be either beginning of the
		 * line, or at word boundary (i.e. the last char must
		 * not be a word char).  Similarly, match end must be
		 * either end of the line, or at word boundary
		 * (i.e. the next char must not be a word char).
		 */
		if ( ((pmatch[0].rm_so == 0 && at_true_bol) ||
		      !word_char(bol[pmatch[0].rm_so-1])) &&
		     ((pmatch[0].rm_eo == (eol-bol)) ||
		      !word_char(bol[pmatch[0].rm_eo])) )
			;
		else
			hit = 0;

		if (!hit && pmatch[0].rm_so + bol + 1 < eol) {
			/* There could be more than one match on the
			 * line, and the first match might not be
			 * strict word match.  But later ones could be!
			 */
			bol = pmatch[0].rm_so + bol + 1;
			at_true_bol = 0;
			goto again;
		}
	}
	if (p->token == GREP_PATTERN_HEAD && saved_ch)
		*eol = saved_ch;
	return hit;
}

static int match_expr_eval(struct grep_opt *o,
			   struct grep_expr *x,
			   char *bol, char *eol,
			   enum grep_context ctx,
			   int collect_hits)
{
	int h = 0;

	switch (x->node) {
	case GREP_NODE_ATOM:
		h = match_one_pattern(o, x->u.atom, bol, eol, ctx);
		break;
	case GREP_NODE_NOT:
		h = !match_expr_eval(o, x->u.unary, bol, eol, ctx, 0);
		break;
	case GREP_NODE_AND:
		if (!collect_hits)
			return (match_expr_eval(o, x->u.binary.left,
						bol, eol, ctx, 0) &&
				match_expr_eval(o, x->u.binary.right,
						bol, eol, ctx, 0));
		h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
		h &= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 0);
		break;
	case GREP_NODE_OR:
		if (!collect_hits)
			return (match_expr_eval(o, x->u.binary.left,
						bol, eol, ctx, 0) ||
				match_expr_eval(o, x->u.binary.right,
						bol, eol, ctx, 0));
		h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
		x->u.binary.left->hit |= h;
		h |= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 1);
		break;
	default:
		die("Unexpected node type (internal error) %d\n", x->node);
	}
	if (collect_hits)
		x->hit |= h;
	return h;
}

static int match_expr(struct grep_opt *opt, char *bol, char *eol,
		      enum grep_context ctx, int collect_hits)
{
	struct grep_expr *x = opt->pattern_expression;
	return match_expr_eval(opt, x, bol, eol, ctx, collect_hits);
}

static int match_line(struct grep_opt *opt, char *bol, char *eol,
		      enum grep_context ctx, int collect_hits)
{
	struct grep_pat *p;
	if (opt->extended)
		return match_expr(opt, bol, eol, ctx, collect_hits);

	/* we do not call with collect_hits without being extended */
	for (p = opt->pattern_list; p; p = p->next) {
		if (match_one_pattern(opt, p, bol, eol, ctx))
			return 1;
	}
	return 0;
}

static int grep_buffer_1(struct grep_opt *opt, const char *name,
			 char *buf, unsigned long size, int collect_hits)
{
	char *bol = buf;
	unsigned long left = size;
	unsigned lno = 1;
	struct pre_context_line {
		char *bol;
		char *eol;
	} *prev = NULL, *pcl;
	unsigned last_hit = 0;
	unsigned last_shown = 0;
	int binary_match_only = 0;
	const char *hunk_mark = "";
	unsigned count = 0;
	enum grep_context ctx = GREP_CONTEXT_HEAD;

	if (buffer_is_binary(buf, size)) {
		switch (opt->binary) {
		case GREP_BINARY_DEFAULT:
			binary_match_only = 1;
			break;
		case GREP_BINARY_NOMATCH:
			return 0; /* Assume unmatch */
			break;
		default:
			break;
		}
	}

	if (opt->pre_context)
		prev = xcalloc(opt->pre_context, sizeof(*prev));
	if (opt->pre_context || opt->post_context)
		hunk_mark = "--\n";

	while (left) {
		char *eol, ch;
		int hit;

		eol = end_of_line(bol, &left);
		ch = *eol;
		*eol = 0;

		if ((ctx == GREP_CONTEXT_HEAD) && (eol == bol))
			ctx = GREP_CONTEXT_BODY;

		hit = match_line(opt, bol, eol, ctx, collect_hits);
		*eol = ch;

		if (collect_hits)
			goto next_line;

		/* "grep -v -e foo -e bla" should list lines
		 * that do not have either, so inversion should
		 * be done outside.
		 */
		if (opt->invert)
			hit = !hit;
		if (opt->unmatch_name_only) {
			if (hit)
				return 0;
			goto next_line;
		}
		if (hit) {
			count++;
			if (opt->status_only)
				return 1;
			if (binary_match_only) {
				printf("Binary file %s matches\n", name);
				return 1;
			}
			if (opt->name_only) {
				printf("%s\n", name);
				return 1;
			}
			/* Hit at this line.  If we haven't shown the
			 * pre-context lines, we would need to show them.
			 * When asked to do "count", this still show
			 * the context which is nonsense, but the user
			 * deserves to get that ;-).
			 */
			if (opt->pre_context) {
				unsigned from;
				if (opt->pre_context < lno)
					from = lno - opt->pre_context;
				else
					from = 1;
				if (from <= last_shown)
					from = last_shown + 1;
				if (last_shown && from != last_shown + 1)
					fputs(hunk_mark, stdout);
				while (from < lno) {
					pcl = &prev[lno-from-1];
					show_line(opt, pcl->bol, pcl->eol,
						  name, from, '-');
					from++;
				}
				last_shown = lno-1;
			}
			if (last_shown && lno != last_shown + 1)
				fputs(hunk_mark, stdout);
			if (!opt->count)
				show_line(opt, bol, eol, name, lno, ':');
			last_shown = last_hit = lno;
		}
		else if (last_hit &&
			 lno <= last_hit + opt->post_context) {
			/* If the last hit is within the post context,
			 * we need to show this line.
			 */
			if (last_shown && lno != last_shown + 1)
				fputs(hunk_mark, stdout);
			show_line(opt, bol, eol, name, lno, '-');
			last_shown = lno;
		}
		if (opt->pre_context) {
			memmove(prev+1, prev,
				(opt->pre_context-1) * sizeof(*prev));
			prev->bol = bol;
			prev->eol = eol;
		}

	next_line:
		bol = eol + 1;
		if (!left)
			break;
		left--;
		lno++;
	}

	free(prev);
	if (collect_hits)
		return 0;

	if (opt->status_only)
		return 0;
	if (opt->unmatch_name_only) {
		/* We did not see any hit, so we want to show this */
		printf("%s\n", name);
		return 1;
	}

	/* NEEDSWORK:
	 * The real "grep -c foo *.c" gives many "bar.c:0" lines,
	 * which feels mostly useless but sometimes useful.  Maybe
	 * make it another option?  For now suppress them.
	 */
	if (opt->count && count)
		printf("%s:%u\n", name, count);
	return !!last_hit;
}

static void clr_hit_marker(struct grep_expr *x)
{
	/* All-hit markers are meaningful only at the very top level
	 * OR node.
	 */
	while (1) {
		x->hit = 0;
		if (x->node != GREP_NODE_OR)
			return;
		x->u.binary.left->hit = 0;
		x = x->u.binary.right;
	}
}

static int chk_hit_marker(struct grep_expr *x)
{
	/* Top level nodes have hit markers.  See if they all are hits */
	while (1) {
		if (x->node != GREP_NODE_OR)
			return x->hit;
		if (!x->u.binary.left->hit)
			return 0;
		x = x->u.binary.right;
	}
}

int grep_buffer(struct grep_opt *opt, const char *name, char *buf, unsigned long size)
{
	/*
	 * we do not have to do the two-pass grep when we do not check
	 * buffer-wide "all-match".
	 */
	if (!opt->all_match)
		return grep_buffer_1(opt, name, buf, size, 0);

	/* Otherwise the toplevel "or" terms hit a bit differently.
	 * We first clear hit markers from them.
	 */
	clr_hit_marker(opt->pattern_expression);
	grep_buffer_1(opt, name, buf, size, 1);

	if (!chk_hit_marker(opt->pattern_expression))
		return 0;

	return grep_buffer_1(opt, name, buf, size, 0);
}
Commit	Line	Data
83b5d2f5	1	#include "cache.h"
83b5d2f5	2	#include "grep.h"
6bfce93e	3	#include "xdiff-interface.h"
83b5d2f5	4
a4d7d2c6 JH	5	void append_header_grep_pattern(struct grep_opt opt, enum grep_header_field field, const char pat)
	6	{
	7	struct grep_pat p = xcalloc(1, sizeof(p));
	8	p->pattern = pat;
	9	p->origin = "header";
	10	p->no = 0;
	11	p->token = GREP_PATTERN_HEAD;
	12	p->field = field;
	13	*opt->pattern_tail = p;
	14	opt->pattern_tail = &p->next;
	15	p->next = NULL;
	16	}
	17
83b5d2f5 JH	18	void append_grep_pattern(struct grep_opt opt, const char pat,
	19	const char *origin, int no, enum grep_pat_token t)
	20	{
	21	struct grep_pat p = xcalloc(1, sizeof(p));
	22	p->pattern = pat;
	23	p->origin = origin;
	24	p->no = no;
	25	p->token = t;
	26	*opt->pattern_tail = p;
	27	opt->pattern_tail = &p->next;
	28	p->next = NULL;
	29	}
	30
	31	static void compile_regexp(struct grep_pat p, struct grep_opt opt)
	32	{
	33	int err = regcomp(&p->regexp, p->pattern, opt->regflags);
	34	if (err) {
	35	char errbuf[1024];
	36	char where[1024];
	37	if (p->no)
	38	sprintf(where, "In '%s' at %d, ",
	39	p->origin, p->no);
	40	else if (p->origin)
	41	sprintf(where, "%s, ", p->origin);
	42	else
	43	where[0] = 0;
	44	regerror(err, &p->regexp, errbuf, 1024);
	45	regfree(&p->regexp);
	46	die("%s'%s': %s", where, p->pattern, errbuf);
	47	}
	48	}
	49
0ab7befa	50	static struct grep_expr compile_pattern_or(struct grep_pat *);
83b5d2f5 JH	51	static struct grep_expr compile_pattern_atom(struct grep_pat *list)
	52	{
	53	struct grep_pat *p;
	54	struct grep_expr *x;
	55
	56	p = *list;
	57	switch (p->token) {
	58	case GREP_PATTERN: /* atom */
480c1ca6 JH	59	case GREP_PATTERN_HEAD:
480c1ca6 JH	60	case GREP_PATTERN_BODY:
83b5d2f5 JH	61	x = xcalloc(1, sizeof (struct grep_expr));
	62	x->node = GREP_NODE_ATOM;
	63	x->u.atom = p;
	64	*list = p->next;
	65	return x;
	66	case GREP_OPEN_PAREN:
	67	*list = p->next;
0ab7befa	68	x = compile_pattern_or(list);
83b5d2f5 JH	69	if (!x)
	70	return NULL;
	71	if (!list \|\| (list)->token != GREP_CLOSE_PAREN)
	72	die("unmatched parenthesis");
	73	list = (list)->next;
	74	return x;
	75	default:
	76	return NULL;
	77	}
	78	}
	79
	80	static struct grep_expr compile_pattern_not(struct grep_pat *list)
	81	{
	82	struct grep_pat *p;
	83	struct grep_expr *x;
	84
	85	p = *list;
	86	switch (p->token) {
	87	case GREP_NOT:
	88	if (!p->next)
	89	die("--not not followed by pattern expression");
	90	*list = p->next;
	91	x = xcalloc(1, sizeof (struct grep_expr));
	92	x->node = GREP_NODE_NOT;
	93	x->u.unary = compile_pattern_not(list);
	94	if (!x->u.unary)
	95	die("--not followed by non pattern expression");
	96	return x;
	97	default:
	98	return compile_pattern_atom(list);
	99	}
	100	}
	101
	102	static struct grep_expr compile_pattern_and(struct grep_pat *list)
	103	{
	104	struct grep_pat *p;
	105	struct grep_expr x, y, *z;
	106
	107	x = compile_pattern_not(list);
	108	p = *list;
	109	if (p && p->token == GREP_AND) {
	110	if (!p->next)
	111	die("--and not followed by pattern expression");
	112	*list = p->next;
	113	y = compile_pattern_and(list);
	114	if (!y)
	115	die("--and not followed by pattern expression");
	116	z = xcalloc(1, sizeof (struct grep_expr));
	117	z->node = GREP_NODE_AND;
	118	z->u.binary.left = x;
	119	z->u.binary.right = y;
	120	return z;
	121	}
	122	return x;
	123	}
	124
	125	static struct grep_expr compile_pattern_or(struct grep_pat *list)
	126	{
	127	struct grep_pat *p;
	128	struct grep_expr x, y, *z;
	129
	130	x = compile_pattern_and(list);
	131	p = *list;
	132	if (x && p && p->token != GREP_CLOSE_PAREN) {
133	y = compile_pattern_or(list);
134	if (!y)
135	die("not a pattern expression %s", p->pattern);
136	z = xcalloc(1, sizeof (struct grep_expr));
137	z->node = GREP_NODE_OR;
138	z->u.binary.left = x;
139	z->u.binary.right = y;
140	return z;
141	}
142	return x;
143	}
144
145	static struct grep_expr compile_pattern_expr(struct grep_pat *list)
146	{
147	return compile_pattern_or(list);
148	}
149
150	void compile_grep_patterns(struct grep_opt *opt)
151	{
152	struct grep_pat *p;
153
0ab7befa JH	154	if (opt->all_match)
	155	opt->extended = 1;
	156
83b5d2f5	157	for (p = opt->pattern_list; p; p = p->next) {
480c1ca6 JH	158	switch (p->token) {
	159	case GREP_PATTERN: /* atom */
	160	case GREP_PATTERN_HEAD:
	161	case GREP_PATTERN_BODY:
a3f5d02e JH	162	if (!opt->fixed)
a3f5d02e JH	163	compile_regexp(p, opt);
480c1ca6 JH	164	break;
480c1ca6 JH	165	default:
83b5d2f5	166	opt->extended = 1;
480c1ca6 JH	167	break;
480c1ca6 JH	168	}
83b5d2f5 JH	169	}
	170
	171	if (!opt->extended)
	172	return;
	173
	174	/* Then bundle them up in an expression.
	175	* A classic recursive descent parser would do.
	176	*/
	177	p = opt->pattern_list;
	178	opt->pattern_expression = compile_pattern_expr(&p);
	179	if (p)
	180	die("incomplete pattern expression: %s", p->pattern);
	181	}
	182
b48fb5b6 JH	183	static void free_pattern_expr(struct grep_expr *x)
	184	{
	185	switch (x->node) {
	186	case GREP_NODE_ATOM:
	187	break;
	188	case GREP_NODE_NOT:
	189	free_pattern_expr(x->u.unary);
	190	break;
	191	case GREP_NODE_AND:
	192	case GREP_NODE_OR:
	193	free_pattern_expr(x->u.binary.left);
	194	free_pattern_expr(x->u.binary.right);
	195	break;
	196	}
	197	free(x);
	198	}
	199
	200	void free_grep_patterns(struct grep_opt *opt)
	201	{
	202	struct grep_pat p, n;
	203
	204	for (p = opt->pattern_list; p; p = n) {
	205	n = p->next;
	206	switch (p->token) {
	207	case GREP_PATTERN: /* atom */
	208	case GREP_PATTERN_HEAD:
	209	case GREP_PATTERN_BODY:
	210	regfree(&p->regexp);
	211	break;
	212	default:
	213	break;
	214	}
	215	free(p);
	216	}
	217
	218	if (!opt->extended)
	219	return;
	220	free_pattern_expr(opt->pattern_expression);
	221	}
	222
83b5d2f5 JH	223	static char end_of_line(char cp, unsigned long *left)
	224	{
	225	unsigned long l = *left;
	226	while (l && *cp != '\n') {
	227	l--;
	228	cp++;
	229	}
	230	*left = l;
	231	return cp;
	232	}
	233
	234	static int word_char(char ch)
	235	{
	236	return isalnum(ch) \|\| ch == '_';
	237	}
	238
	239	static void show_line(struct grep_opt opt, const char bol, const char *eol,
	240	const char *name, unsigned lno, char sign)
	241	{
	242	if (opt->pathname)
	243	printf("%s%c", name, sign);
	244	if (opt->linenum)
	245	printf("%d%c", lno, sign);
	246	printf("%.*s\n", (int)(eol-bol), bol);
	247	}
	248
83b5d2f5 JH	249	static int fixmatch(const char pattern, char line, regmatch_t *match)
	250	{
	251	char *hit = strstr(line, pattern);
	252	if (!hit) {
	253	match->rm_so = match->rm_eo = -1;
	254	return REG_NOMATCH;
	255	}
	256	else {
	257	match->rm_so = hit - line;
	258	match->rm_eo = match->rm_so + strlen(pattern);
	259	return 0;
	260	}
	261	}
	262
a4d7d2c6 JH	263	static int strip_timestamp(char bol, char *eol_p)
	264	{
	265	char eol = eol_p;
	266	int ch;
	267
	268	while (bol < --eol) {
	269	if (*eol != '>')
	270	continue;
	271	*eol_p = ++eol;
	272	ch = *eol;
	273	*eol = '\0';
	274	return ch;
	275	}
	276	return 0;
	277	}
	278
	279	static struct {
	280	const char *field;
	281	size_t len;
	282	} header_field[] = {
	283	{ "author ", 7 },
	284	{ "committer ", 10 },
	285	};
	286
480c1ca6	287	static int match_one_pattern(struct grep_opt opt, struct grep_pat p, char bol, char eol, enum grep_context ctx)
83b5d2f5 JH	288	{
	289	int hit = 0;
	290	int at_true_bol = 1;
a4d7d2c6	291	int saved_ch = 0;
83b5d2f5 JH	292	regmatch_t pmatch[10];
83b5d2f5 JH	293
480c1ca6 JH	294	if ((p->token != GREP_PATTERN) &&
	295	((p->token == GREP_PATTERN_HEAD) != (ctx == GREP_CONTEXT_HEAD)))
	296	return 0;
	297
a4d7d2c6 JH	298	if (p->token == GREP_PATTERN_HEAD) {
	299	const char *field;
	300	size_t len;
	301	assert(p->field < ARRAY_SIZE(header_field));
	302	field = header_field[p->field].field;
	303	len = header_field[p->field].len;
	304	if (strncmp(bol, field, len))
	305	return 0;
	306	bol += len;
	307	saved_ch = strip_timestamp(bol, &eol);
	308	}
	309
83b5d2f5 JH	310	again:
	311	if (!opt->fixed) {
	312	regex_t *exp = &p->regexp;
	313	hit = !regexec(exp, bol, ARRAY_SIZE(pmatch),
	314	pmatch, 0);
	315	}
	316	else {
	317	hit = !fixmatch(p->pattern, bol, pmatch);
	318	}
	319
	320	if (hit && opt->word_regexp) {
	321	if ((pmatch[0].rm_so < 0) \|\|
	322	(eol - bol) <= pmatch[0].rm_so \|\|
	323	(pmatch[0].rm_eo < 0) \|\|
	324	(eol - bol) < pmatch[0].rm_eo)
	325	die("regexp returned nonsense");
	326
	327	/* Match beginning must be either beginning of the
	328	* line, or at word boundary (i.e. the last char must
	329	* not be a word char). Similarly, match end must be
	330	* either end of the line, or at word boundary
	331	* (i.e. the next char must not be a word char).
	332	*/
	333	if ( ((pmatch[0].rm_so == 0 && at_true_bol) \|\|
	334	!word_char(bol[pmatch[0].rm_so-1])) &&
	335	((pmatch[0].rm_eo == (eol-bol)) \|\|
	336	!word_char(bol[pmatch[0].rm_eo])) )
	337	;
	338	else
	339	hit = 0;
	340
	341	if (!hit && pmatch[0].rm_so + bol + 1 < eol) {
	342	/* There could be more than one match on the
	343	* line, and the first match might not be
	344	* strict word match. But later ones could be!
	345	*/
	346	bol = pmatch[0].rm_so + bol + 1;
	347	at_true_bol = 0;
	348	goto again;
	349	}
	350	}
a4d7d2c6 JH	351	if (p->token == GREP_PATTERN_HEAD && saved_ch)
a4d7d2c6 JH	352	*eol = saved_ch;
83b5d2f5 JH	353	return hit;
	354	}
	355
0ab7befa	356	static int match_expr_eval(struct grep_opt *o,
83b5d2f5	357	struct grep_expr *x,
480c1ca6	358	char bol, char eol,
0ab7befa JH	359	enum grep_context ctx,
0ab7befa JH	360	int collect_hits)
83b5d2f5	361	{
0ab7befa JH	362	int h = 0;
0ab7befa JH	363
83b5d2f5 JH	364	switch (x->node) {
83b5d2f5 JH	365	case GREP_NODE_ATOM:
0ab7befa	366	h = match_one_pattern(o, x->u.atom, bol, eol, ctx);
83b5d2f5 JH	367	break;
83b5d2f5 JH	368	case GREP_NODE_NOT:
0ab7befa JH	369	h = !match_expr_eval(o, x->u.unary, bol, eol, ctx, 0);
0ab7befa JH	370	break;
83b5d2f5	371	case GREP_NODE_AND:
0ab7befa JH	372	if (!collect_hits)
	373	return (match_expr_eval(o, x->u.binary.left,
	374	bol, eol, ctx, 0) &&
	375	match_expr_eval(o, x->u.binary.right,
	376	bol, eol, ctx, 0));
	377	h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
	378	h &= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 0);
	379	break;
83b5d2f5	380	case GREP_NODE_OR:
0ab7befa JH	381	if (!collect_hits)
	382	return (match_expr_eval(o, x->u.binary.left,
	383	bol, eol, ctx, 0) \|\|
	384	match_expr_eval(o, x->u.binary.right,
	385	bol, eol, ctx, 0));
	386	h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
	387	x->u.binary.left->hit \|= h;
	388	h \|= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 1);
	389	break;
	390	default:
	391	die("Unexpected node type (internal error) %d\n", x->node);
83b5d2f5	392	}
0ab7befa JH	393	if (collect_hits)
	394	x->hit \|= h;
	395	return h;
83b5d2f5 JH	396	}
83b5d2f5 JH	397
480c1ca6	398	static int match_expr(struct grep_opt opt, char bol, char *eol,
0ab7befa	399	enum grep_context ctx, int collect_hits)
83b5d2f5 JH	400	{
83b5d2f5 JH	401	struct grep_expr *x = opt->pattern_expression;
0ab7befa	402	return match_expr_eval(opt, x, bol, eol, ctx, collect_hits);
83b5d2f5 JH	403	}
83b5d2f5 JH	404
480c1ca6	405	static int match_line(struct grep_opt opt, char bol, char *eol,
0ab7befa	406	enum grep_context ctx, int collect_hits)
83b5d2f5 JH	407	{
	408	struct grep_pat *p;
	409	if (opt->extended)
0ab7befa JH	410	return match_expr(opt, bol, eol, ctx, collect_hits);
	411
	412	/* we do not call with collect_hits without being extended */
83b5d2f5	413	for (p = opt->pattern_list; p; p = p->next) {
480c1ca6	414	if (match_one_pattern(opt, p, bol, eol, ctx))
83b5d2f5 JH	415	return 1;
	416	}
	417	return 0;
	418	}
	419
0ab7befa JH	420	static int grep_buffer_1(struct grep_opt opt, const char name,
0ab7befa JH	421	char *buf, unsigned long size, int collect_hits)
83b5d2f5 JH	422	{
	423	char *bol = buf;
	424	unsigned long left = size;
	425	unsigned lno = 1;
	426	struct pre_context_line {
	427	char *bol;
	428	char *eol;
	429	} prev = NULL, pcl;
	430	unsigned last_hit = 0;
	431	unsigned last_shown = 0;
	432	int binary_match_only = 0;
	433	const char *hunk_mark = "";
	434	unsigned count = 0;
480c1ca6	435	enum grep_context ctx = GREP_CONTEXT_HEAD;
83b5d2f5 JH	436
	437	if (buffer_is_binary(buf, size)) {
	438	switch (opt->binary) {
	439	case GREP_BINARY_DEFAULT:
	440	binary_match_only = 1;
	441	break;
	442	case GREP_BINARY_NOMATCH:
	443	return 0; /* Assume unmatch */
	444	break;
	445	default:
	446	break;
	447	}
	448	}
	449
	450	if (opt->pre_context)
	451	prev = xcalloc(opt->pre_context, sizeof(*prev));
	452	if (opt->pre_context \|\| opt->post_context)
	453	hunk_mark = "--\n";
	454
	455	while (left) {
	456	char *eol, ch;
0ab7befa	457	int hit;
83b5d2f5 JH	458
	459	eol = end_of_line(bol, &left);
	460	ch = *eol;
	461	*eol = 0;
	462
480c1ca6 JH	463	if ((ctx == GREP_CONTEXT_HEAD) && (eol == bol))
	464	ctx = GREP_CONTEXT_BODY;
	465
0ab7befa	466	hit = match_line(opt, bol, eol, ctx, collect_hits);
83b5d2f5 JH	467	*eol = ch;
83b5d2f5 JH	468
0ab7befa JH	469	if (collect_hits)
	470	goto next_line;
	471
83b5d2f5 JH	472	/* "grep -v -e foo -e bla" should list lines
	473	* that do not have either, so inversion should
	474	* be done outside.
	475	*/
	476	if (opt->invert)
	477	hit = !hit;
	478	if (opt->unmatch_name_only) {
	479	if (hit)
	480	return 0;
	481	goto next_line;
	482	}
	483	if (hit) {
	484	count++;
	485	if (opt->status_only)
	486	return 1;
	487	if (binary_match_only) {
	488	printf("Binary file %s matches\n", name);
	489	return 1;
	490	}
	491	if (opt->name_only) {
	492	printf("%s\n", name);
	493	return 1;
	494	}
	495	/* Hit at this line. If we haven't shown the
	496	* pre-context lines, we would need to show them.
	497	* When asked to do "count", this still show
	498	* the context which is nonsense, but the user
	499	* deserves to get that ;-).
	500	*/
	501	if (opt->pre_context) {
	502	unsigned from;
	503	if (opt->pre_context < lno)
	504	from = lno - opt->pre_context;
	505	else
	506	from = 1;
	507	if (from <= last_shown)
	508	from = last_shown + 1;
	509	if (last_shown && from != last_shown + 1)
9db56f71	510	fputs(hunk_mark, stdout);
83b5d2f5 JH	511	while (from < lno) {
	512	pcl = &prev[lno-from-1];
	513	show_line(opt, pcl->bol, pcl->eol,
	514	name, from, '-');
	515	from++;
	516	}
	517	last_shown = lno-1;
	518	}
	519	if (last_shown && lno != last_shown + 1)
9db56f71	520	fputs(hunk_mark, stdout);
83b5d2f5 JH	521	if (!opt->count)
	522	show_line(opt, bol, eol, name, lno, ':');
	523	last_shown = last_hit = lno;
	524	}
	525	else if (last_hit &&
	526	lno <= last_hit + opt->post_context) {
	527	/* If the last hit is within the post context,
	528	* we need to show this line.
	529	*/
	530	if (last_shown && lno != last_shown + 1)
9db56f71	531	fputs(hunk_mark, stdout);
83b5d2f5 JH	532	show_line(opt, bol, eol, name, lno, '-');
	533	last_shown = lno;
	534	}
	535	if (opt->pre_context) {
	536	memmove(prev+1, prev,
	537	(opt->pre_context-1) * sizeof(*prev));
	538	prev->bol = bol;
	539	prev->eol = eol;
	540	}
	541
	542	next_line:
	543	bol = eol + 1;
	544	if (!left)
	545	break;
	546	left--;
	547	lno++;
	548	}
	549
b48fb5b6	550	free(prev);
0ab7befa JH	551	if (collect_hits)
0ab7befa JH	552	return 0;
b48fb5b6	553
83b5d2f5 JH	554	if (opt->status_only)
	555	return 0;
	556	if (opt->unmatch_name_only) {
	557	/* We did not see any hit, so we want to show this */
	558	printf("%s\n", name);
	559	return 1;
	560	}
	561
	562	/* NEEDSWORK:
	563	* The real "grep -c foo *.c" gives many "bar.c:0" lines,
	564	* which feels mostly useless but sometimes useful. Maybe
	565	* make it another option? For now suppress them.
	566	*/
	567	if (opt->count && count)
	568	printf("%s:%u\n", name, count);
	569	return !!last_hit;
	570	}
	571
0ab7befa JH	572	static void clr_hit_marker(struct grep_expr *x)
	573	{
	574	/* All-hit markers are meaningful only at the very top level
	575	* OR node.
	576	*/
	577	while (1) {
	578	x->hit = 0;
	579	if (x->node != GREP_NODE_OR)
	580	return;
	581	x->u.binary.left->hit = 0;
	582	x = x->u.binary.right;
	583	}
	584	}
	585
	586	static int chk_hit_marker(struct grep_expr *x)
	587	{
	588	/* Top level nodes have hit markers. See if they all are hits */
	589	while (1) {
	590	if (x->node != GREP_NODE_OR)
	591	return x->hit;
	592	if (!x->u.binary.left->hit)
	593	return 0;
	594	x = x->u.binary.right;
	595	}
	596	}
	597
	598	int grep_buffer(struct grep_opt opt, const char name, char *buf, unsigned long size)
	599	{
	600	/*
	601	* we do not have to do the two-pass grep when we do not check
	602	* buffer-wide "all-match".
	603	*/
	604	if (!opt->all_match)
	605	return grep_buffer_1(opt, name, buf, size, 0);
	606
	607	/* Otherwise the toplevel "or" terms hit a bit differently.
	608	* We first clear hit markers from them.
	609	*/
	610	clr_hit_marker(opt->pattern_expression);
	611	grep_buffer_1(opt, name, buf, size, 1);
	612
	613	if (!chk_hit_marker(opt->pattern_expression))
	614	return 0;
	615
	616	return grep_buffer_1(opt, name, buf, size, 0);
	617	}