[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 int is_fixed(const char *s)
{
	while (*s && !is_regex_special(*s))
		s++;
	return !*s;
}

static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
{
	int err;

	p->word_regexp = opt->word_regexp;

	if (opt->fixed || is_fixed(p->pattern))
		p->fixed = 1;
	if (opt->regflags & REG_ICASE)
		p->fixed = 0;
	if (p->fixed)
		return;

	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;
	if (!p)
		return NULL;
	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 (!*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;
	if (!p)
		return NULL;
	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:
			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;
	if (p)
		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_name(struct grep_opt *opt, const char *name)
{
	printf("%s%c", name, opt->null_following_name ? '\0' : '\n');
}

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_pat *p, char *bol, char *eol,
			     enum grep_context ctx,
			     regmatch_t *pmatch, int eflags)
{
	int hit = 0;
	int saved_ch = 0;
	const char *start = bol;

	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 (p->fixed)
		hit = !fixmatch(p->pattern, bol, pmatch);
	else
		hit = !regexec(&p->regexp, bol, 1, pmatch, eflags);

	if (hit && p->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) ||
		      !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!
			 * Forward to the next possible start, i.e. the
			 * next position following a non-word char.
			 */
			bol = pmatch[0].rm_so + bol + 1;
			while (word_char(bol[-1]) && bol < eol)
				bol++;
			eflags |= REG_NOTBOL;
			if (bol < eol)
				goto again;
		}
	}
	if (p->token == GREP_PATTERN_HEAD && saved_ch)
		*eol = saved_ch;
	if (hit) {
		pmatch[0].rm_so += bol - start;
		pmatch[0].rm_eo += bol - start;
	}
	return hit;
}

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

	if (!x)
		die("Not a valid grep expression");
	switch (x->node) {
	case GREP_NODE_ATOM:
		h = match_one_pattern(x->u.atom, bol, eol, ctx, &match, 0);
		break;
	case GREP_NODE_NOT:
		h = !match_expr_eval(x->u.unary, bol, eol, ctx, 0);
		break;
	case GREP_NODE_AND:
		if (!match_expr_eval(x->u.binary.left, bol, eol, ctx, 0))
			return 0;
		h = match_expr_eval(x->u.binary.right, bol, eol, ctx, 0);
		break;
	case GREP_NODE_OR:
		if (!collect_hits)
			return (match_expr_eval(x->u.binary.left,
						bol, eol, ctx, 0) ||
				match_expr_eval(x->u.binary.right,
						bol, eol, ctx, 0));
		h = match_expr_eval(x->u.binary.left, bol, eol, ctx, 0);
		x->u.binary.left->hit |= h;
		h |= match_expr_eval(x->u.binary.right, bol, eol, ctx, 1);
		break;
	default:
		die("Unexpected node type (internal error) %d", 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(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;
	regmatch_t match;

	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(p, bol, eol, ctx, &match, 0))
			return 1;
	}
	return 0;
}

static int match_next_pattern(struct grep_pat *p, char *bol, char *eol,
			      enum grep_context ctx,
			      regmatch_t *pmatch, int eflags)
{
	regmatch_t match;

	if (!match_one_pattern(p, bol, eol, ctx, &match, eflags))
		return 0;
	if (match.rm_so < 0 || match.rm_eo < 0)
		return 0;
	if (pmatch->rm_so >= 0 && pmatch->rm_eo >= 0) {
		if (match.rm_so > pmatch->rm_so)
			return 1;
		if (match.rm_so == pmatch->rm_so && match.rm_eo < pmatch->rm_eo)
			return 1;
	}
	pmatch->rm_so = match.rm_so;
	pmatch->rm_eo = match.rm_eo;
	return 1;
}

static int next_match(struct grep_opt *opt, char *bol, char *eol,
		      enum grep_context ctx, regmatch_t *pmatch, int eflags)
{
	struct grep_pat *p;
	int hit = 0;

	pmatch->rm_so = pmatch->rm_eo = -1;
	if (bol < eol) {
		for (p = opt->pattern_list; p; p = p->next) {
			switch (p->token) {
			case GREP_PATTERN: /* atom */
			case GREP_PATTERN_HEAD:
			case GREP_PATTERN_BODY:
				hit |= match_next_pattern(p, bol, eol, ctx,
							  pmatch, eflags);
				break;
			default:
				break;
			}
		}
	}
	return hit;
}

static void show_line(struct grep_opt *opt, char *bol, char *eol,
		      const char *name, unsigned lno, char sign)
{
	int rest = eol - bol;

	if (opt->null_following_name)
		sign = '\0';
	if (opt->pathname)
		printf("%s%c", name, sign);
	if (opt->linenum)
		printf("%d%c", lno, sign);
	if (opt->color) {
		regmatch_t match;
		enum grep_context ctx = GREP_CONTEXT_BODY;
		int ch = *eol;
		int eflags = 0;

		*eol = '\0';
		while (next_match(opt, bol, eol, ctx, &match, eflags)) {
			if (match.rm_so == match.rm_eo)
				break;
			printf("%.*s%s%.*s%s",
			       (int)match.rm_so, bol,
			       opt->color_match,
			       (int)(match.rm_eo - match.rm_so), bol + match.rm_so,
			       GIT_COLOR_RESET);
			bol += match.rm_eo;
			rest -= match.rm_eo;
			eflags = REG_NOTBOL;
		}
		*eol = ch;
	}
	printf("%.*s\n", rest, bol);
}

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) {
				show_name(opt, 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 */
		show_name(opt, 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%c%u\n", name,
		       opt->null_following_name ? '\0' : ':', 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
c822255c RS	31	static int is_fixed(const char *s)
c822255c RS	32	{
f9b7cce6	33	while (s && !is_regex_special(s))
c822255c RS	34	s++;
	35	return !*s;
	36	}
	37
83b5d2f5 JH	38	static void compile_regexp(struct grep_pat p, struct grep_opt opt)
83b5d2f5 JH	39	{
c822255c RS	40	int err;
c822255c RS	41
d7eb527d RS	42	p->word_regexp = opt->word_regexp;
d7eb527d RS	43
c822255c RS	44	if (opt->fixed \|\| is_fixed(p->pattern))
	45	p->fixed = 1;
	46	if (opt->regflags & REG_ICASE)
	47	p->fixed = 0;
	48	if (p->fixed)
	49	return;
	50
	51	err = regcomp(&p->regexp, p->pattern, opt->regflags);
83b5d2f5 JH	52	if (err) {
	53	char errbuf[1024];
	54	char where[1024];
	55	if (p->no)
	56	sprintf(where, "In '%s' at %d, ",
	57	p->origin, p->no);
	58	else if (p->origin)
	59	sprintf(where, "%s, ", p->origin);
	60	else
	61	where[0] = 0;
	62	regerror(err, &p->regexp, errbuf, 1024);
	63	regfree(&p->regexp);
	64	die("%s'%s': %s", where, p->pattern, errbuf);
	65	}
	66	}
	67
0ab7befa	68	static struct grep_expr compile_pattern_or(struct grep_pat *);
83b5d2f5 JH	69	static struct grep_expr compile_pattern_atom(struct grep_pat *list)
	70	{
	71	struct grep_pat *p;
	72	struct grep_expr *x;
	73
	74	p = *list;
c922b01f LT	75	if (!p)
c922b01f LT	76	return NULL;
83b5d2f5 JH	77	switch (p->token) {
83b5d2f5 JH	78	case GREP_PATTERN: /* atom */
480c1ca6 JH	79	case GREP_PATTERN_HEAD:
480c1ca6 JH	80	case GREP_PATTERN_BODY:
83b5d2f5 JH	81	x = xcalloc(1, sizeof (struct grep_expr));
	82	x->node = GREP_NODE_ATOM;
	83	x->u.atom = p;
	84	*list = p->next;
	85	return x;
	86	case GREP_OPEN_PAREN:
	87	*list = p->next;
0ab7befa	88	x = compile_pattern_or(list);
83b5d2f5 JH	89	if (!list \|\| (list)->token != GREP_CLOSE_PAREN)
	90	die("unmatched parenthesis");
	91	list = (list)->next;
	92	return x;
	93	default:
	94	return NULL;
	95	}
	96	}
	97
	98	static struct grep_expr compile_pattern_not(struct grep_pat *list)
	99	{
	100	struct grep_pat *p;
	101	struct grep_expr *x;
	102
	103	p = *list;
c922b01f LT	104	if (!p)
c922b01f LT	105	return NULL;
83b5d2f5 JH	106	switch (p->token) {
	107	case GREP_NOT:
	108	if (!p->next)
	109	die("--not not followed by pattern expression");
	110	*list = p->next;
	111	x = xcalloc(1, sizeof (struct grep_expr));
	112	x->node = GREP_NODE_NOT;
	113	x->u.unary = compile_pattern_not(list);
	114	if (!x->u.unary)
	115	die("--not followed by non pattern expression");
	116	return x;
	117	default:
	118	return compile_pattern_atom(list);
	119	}
	120	}
	121
	122	static struct grep_expr compile_pattern_and(struct grep_pat *list)
	123	{
	124	struct grep_pat *p;
	125	struct grep_expr x, y, *z;
	126
	127	x = compile_pattern_not(list);
	128	p = *list;
	129	if (p && p->token == GREP_AND) {
	130	if (!p->next)
	131	die("--and not followed by pattern expression");
	132	*list = p->next;
	133	y = compile_pattern_and(list);
	134	if (!y)
	135	die("--and not followed by pattern expression");
	136	z = xcalloc(1, sizeof (struct grep_expr));
	137	z->node = GREP_NODE_AND;
	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_or(struct grep_pat *list)
	146	{
	147	struct grep_pat *p;
	148	struct grep_expr x, y, *z;
	149
	150	x = compile_pattern_and(list);
	151	p = *list;
	152	if (x && p && p->token != GREP_CLOSE_PAREN) {
	153	y = compile_pattern_or(list);
	154	if (!y)
	155	die("not a pattern expression %s", p->pattern);
	156	z = xcalloc(1, sizeof (struct grep_expr));
	157	z->node = GREP_NODE_OR;
	158	z->u.binary.left = x;
	159	z->u.binary.right = y;
	160	return z;
	161	}
	162	return x;
	163	}
	164
	165	static struct grep_expr compile_pattern_expr(struct grep_pat *list)
	166	{
	167	return compile_pattern_or(list);
	168	}
	169
170	void compile_grep_patterns(struct grep_opt *opt)
171	{
172	struct grep_pat *p;
173
0ab7befa JH	174	if (opt->all_match)
	175	opt->extended = 1;
	176
83b5d2f5	177	for (p = opt->pattern_list; p; p = p->next) {
480c1ca6 JH	178	switch (p->token) {
	179	case GREP_PATTERN: /* atom */
	180	case GREP_PATTERN_HEAD:
	181	case GREP_PATTERN_BODY:
c822255c	182	compile_regexp(p, opt);
480c1ca6 JH	183	break;
480c1ca6 JH	184	default:
83b5d2f5	185	opt->extended = 1;
480c1ca6 JH	186	break;
480c1ca6 JH	187	}
83b5d2f5 JH	188	}
	189
	190	if (!opt->extended)
	191	return;
	192
	193	/* Then bundle them up in an expression.
	194	* A classic recursive descent parser would do.
	195	*/
	196	p = opt->pattern_list;
ba150a3f MB	197	if (p)
ba150a3f MB	198	opt->pattern_expression = compile_pattern_expr(&p);
83b5d2f5 JH	199	if (p)
	200	die("incomplete pattern expression: %s", p->pattern);
	201	}
	202
b48fb5b6 JH	203	static void free_pattern_expr(struct grep_expr *x)
	204	{
	205	switch (x->node) {
	206	case GREP_NODE_ATOM:
	207	break;
	208	case GREP_NODE_NOT:
	209	free_pattern_expr(x->u.unary);
	210	break;
	211	case GREP_NODE_AND:
	212	case GREP_NODE_OR:
	213	free_pattern_expr(x->u.binary.left);
	214	free_pattern_expr(x->u.binary.right);
	215	break;
	216	}
	217	free(x);
	218	}
	219
	220	void free_grep_patterns(struct grep_opt *opt)
	221	{
	222	struct grep_pat p, n;
	223
	224	for (p = opt->pattern_list; p; p = n) {
	225	n = p->next;
	226	switch (p->token) {
	227	case GREP_PATTERN: /* atom */
	228	case GREP_PATTERN_HEAD:
	229	case GREP_PATTERN_BODY:
	230	regfree(&p->regexp);
	231	break;
	232	default:
	233	break;
	234	}
	235	free(p);
	236	}
	237
	238	if (!opt->extended)
	239	return;
	240	free_pattern_expr(opt->pattern_expression);
	241	}
	242
83b5d2f5 JH	243	static char end_of_line(char cp, unsigned long *left)
	244	{
	245	unsigned long l = *left;
	246	while (l && *cp != '\n') {
	247	l--;
	248	cp++;
	249	}
	250	*left = l;
	251	return cp;
	252	}
	253
	254	static int word_char(char ch)
	255	{
	256	return isalnum(ch) \|\| ch == '_';
	257	}
	258
83caecca RZ	259	static void show_name(struct grep_opt opt, const char name)
	260	{
	261	printf("%s%c", name, opt->null_following_name ? '\0' : '\n');
	262	}
	263
83b5d2f5 JH	264	static int fixmatch(const char pattern, char line, regmatch_t *match)
	265	{
	266	char *hit = strstr(line, pattern);
	267	if (!hit) {
	268	match->rm_so = match->rm_eo = -1;
	269	return REG_NOMATCH;
	270	}
	271	else {
	272	match->rm_so = hit - line;
	273	match->rm_eo = match->rm_so + strlen(pattern);
	274	return 0;
	275	}
	276	}
	277
a4d7d2c6 JH	278	static int strip_timestamp(char bol, char *eol_p)
	279	{
	280	char eol = eol_p;
	281	int ch;
	282
	283	while (bol < --eol) {
	284	if (*eol != '>')
	285	continue;
	286	*eol_p = ++eol;
	287	ch = *eol;
	288	*eol = '\0';
	289	return ch;
	290	}
	291	return 0;
	292	}
	293
	294	static struct {
	295	const char *field;
	296	size_t len;
	297	} header_field[] = {
	298	{ "author ", 7 },
	299	{ "committer ", 10 },
	300	};
	301
d7eb527d	302	static int match_one_pattern(struct grep_pat p, char bol, char *eol,
79212772 RS	303	enum grep_context ctx,
79212772 RS	304	regmatch_t *pmatch, int eflags)
83b5d2f5 JH	305	{
83b5d2f5 JH	306	int hit = 0;
a4d7d2c6	307	int saved_ch = 0;
e701fadb	308	const char *start = bol;
83b5d2f5	309
480c1ca6 JH	310	if ((p->token != GREP_PATTERN) &&
	311	((p->token == GREP_PATTERN_HEAD) != (ctx == GREP_CONTEXT_HEAD)))
	312	return 0;
	313
a4d7d2c6 JH	314	if (p->token == GREP_PATTERN_HEAD) {
	315	const char *field;
	316	size_t len;
	317	assert(p->field < ARRAY_SIZE(header_field));
	318	field = header_field[p->field].field;
	319	len = header_field[p->field].len;
	320	if (strncmp(bol, field, len))
	321	return 0;
	322	bol += len;
	323	saved_ch = strip_timestamp(bol, &eol);
	324	}
	325
83b5d2f5	326	again:
79212772	327	if (p->fixed)
83b5d2f5	328	hit = !fixmatch(p->pattern, bol, pmatch);
79212772 RS	329	else
79212772 RS	330	hit = !regexec(&p->regexp, bol, 1, pmatch, eflags);
83b5d2f5	331
d7eb527d	332	if (hit && p->word_regexp) {
83b5d2f5 JH	333	if ((pmatch[0].rm_so < 0) \|\|
	334	(eol - bol) <= pmatch[0].rm_so \|\|
	335	(pmatch[0].rm_eo < 0) \|\|
	336	(eol - bol) < pmatch[0].rm_eo)
	337	die("regexp returned nonsense");
	338
	339	/* Match beginning must be either beginning of the
	340	* line, or at word boundary (i.e. the last char must
	341	* not be a word char). Similarly, match end must be
	342	* either end of the line, or at word boundary
	343	* (i.e. the next char must not be a word char).
	344	*/
fb62eb7f	345	if ( ((pmatch[0].rm_so == 0) \|\|
83b5d2f5 JH	346	!word_char(bol[pmatch[0].rm_so-1])) &&
	347	((pmatch[0].rm_eo == (eol-bol)) \|\|
	348	!word_char(bol[pmatch[0].rm_eo])) )
	349	;
	350	else
	351	hit = 0;
	352
	353	if (!hit && pmatch[0].rm_so + bol + 1 < eol) {
	354	/* There could be more than one match on the
	355	* line, and the first match might not be
	356	* strict word match. But later ones could be!
fb62eb7f RS	357	* Forward to the next possible start, i.e. the
fb62eb7f RS	358	* next position following a non-word char.
83b5d2f5 JH	359	*/
83b5d2f5 JH	360	bol = pmatch[0].rm_so + bol + 1;
fb62eb7f RS	361	while (word_char(bol[-1]) && bol < eol)
fb62eb7f RS	362	bol++;
dbb6a4ad	363	eflags \|= REG_NOTBOL;
fb62eb7f RS	364	if (bol < eol)
fb62eb7f RS	365	goto again;
83b5d2f5 JH	366	}
83b5d2f5 JH	367	}
a4d7d2c6 JH	368	if (p->token == GREP_PATTERN_HEAD && saved_ch)
a4d7d2c6 JH	369	*eol = saved_ch;
e701fadb RS	370	if (hit) {
	371	pmatch[0].rm_so += bol - start;
	372	pmatch[0].rm_eo += bol - start;
	373	}
83b5d2f5 JH	374	return hit;
	375	}
	376
d7eb527d RS	377	static int match_expr_eval(struct grep_expr x, char bol, char *eol,
d7eb527d RS	378	enum grep_context ctx, int collect_hits)
83b5d2f5	379	{
0ab7befa	380	int h = 0;
79212772	381	regmatch_t match;
0ab7befa	382
c922b01f LT	383	if (!x)
c922b01f LT	384	die("Not a valid grep expression");
83b5d2f5 JH	385	switch (x->node) {
83b5d2f5 JH	386	case GREP_NODE_ATOM:
79212772	387	h = match_one_pattern(x->u.atom, bol, eol, ctx, &match, 0);
83b5d2f5 JH	388	break;
83b5d2f5 JH	389	case GREP_NODE_NOT:
d7eb527d	390	h = !match_expr_eval(x->u.unary, bol, eol, ctx, 0);
0ab7befa	391	break;
83b5d2f5	392	case GREP_NODE_AND:
d7eb527d	393	if (!match_expr_eval(x->u.binary.left, bol, eol, ctx, 0))
252d560d	394	return 0;
d7eb527d	395	h = match_expr_eval(x->u.binary.right, bol, eol, ctx, 0);
0ab7befa	396	break;
83b5d2f5	397	case GREP_NODE_OR:
0ab7befa	398	if (!collect_hits)
d7eb527d	399	return (match_expr_eval(x->u.binary.left,
0ab7befa	400	bol, eol, ctx, 0) \|\|
d7eb527d	401	match_expr_eval(x->u.binary.right,
0ab7befa	402	bol, eol, ctx, 0));
d7eb527d	403	h = match_expr_eval(x->u.binary.left, bol, eol, ctx, 0);
0ab7befa	404	x->u.binary.left->hit \|= h;
d7eb527d	405	h \|= match_expr_eval(x->u.binary.right, bol, eol, ctx, 1);
0ab7befa JH	406	break;
0ab7befa JH	407	default:
d7530708	408	die("Unexpected node type (internal error) %d", x->node);
83b5d2f5	409	}
0ab7befa JH	410	if (collect_hits)
	411	x->hit \|= h;
	412	return h;
83b5d2f5 JH	413	}
83b5d2f5 JH	414
480c1ca6	415	static int match_expr(struct grep_opt opt, char bol, char *eol,
0ab7befa	416	enum grep_context ctx, int collect_hits)
83b5d2f5 JH	417	{
83b5d2f5 JH	418	struct grep_expr *x = opt->pattern_expression;
d7eb527d	419	return match_expr_eval(x, bol, eol, ctx, collect_hits);
83b5d2f5 JH	420	}
83b5d2f5 JH	421
480c1ca6	422	static int match_line(struct grep_opt opt, char bol, char *eol,
0ab7befa	423	enum grep_context ctx, int collect_hits)
83b5d2f5 JH	424	{
83b5d2f5 JH	425	struct grep_pat *p;
79212772 RS	426	regmatch_t match;
79212772 RS	427
83b5d2f5	428	if (opt->extended)
0ab7befa JH	429	return match_expr(opt, bol, eol, ctx, collect_hits);
	430
	431	/* we do not call with collect_hits without being extended */
83b5d2f5	432	for (p = opt->pattern_list; p; p = p->next) {
79212772	433	if (match_one_pattern(p, bol, eol, ctx, &match, 0))
83b5d2f5 JH	434	return 1;
	435	}
	436	return 0;
	437	}
	438
7e8f59d5 RS	439	static int match_next_pattern(struct grep_pat p, char bol, char *eol,
	440	enum grep_context ctx,
	441	regmatch_t *pmatch, int eflags)
	442	{
	443	regmatch_t match;
	444
	445	if (!match_one_pattern(p, bol, eol, ctx, &match, eflags))
	446	return 0;
	447	if (match.rm_so < 0 \|\| match.rm_eo < 0)
	448	return 0;
	449	if (pmatch->rm_so >= 0 && pmatch->rm_eo >= 0) {
	450	if (match.rm_so > pmatch->rm_so)
	451	return 1;
	452	if (match.rm_so == pmatch->rm_so && match.rm_eo < pmatch->rm_eo)
	453	return 1;
	454	}
	455	pmatch->rm_so = match.rm_so;
	456	pmatch->rm_eo = match.rm_eo;
	457	return 1;
	458	}
	459
	460	static int next_match(struct grep_opt opt, char bol, char *eol,
	461	enum grep_context ctx, regmatch_t *pmatch, int eflags)
	462	{
	463	struct grep_pat *p;
	464	int hit = 0;
	465
	466	pmatch->rm_so = pmatch->rm_eo = -1;
	467	if (bol < eol) {
	468	for (p = opt->pattern_list; p; p = p->next) {
	469	switch (p->token) {
	470	case GREP_PATTERN: /* atom */
	471	case GREP_PATTERN_HEAD:
	472	case GREP_PATTERN_BODY:
	473	hit \|= match_next_pattern(p, bol, eol, ctx,
	474	pmatch, eflags);
	475	break;
	476	default:
	477	break;
	478	}
	479	}
	480	}
	481	return hit;
	482	}
	483
	484	static void show_line(struct grep_opt opt, char bol, char *eol,
	485	const char *name, unsigned lno, char sign)
	486	{
	487	int rest = eol - bol;
	488
	489	if (opt->null_following_name)
	490	sign = '\0';
	491	if (opt->pathname)
	492	printf("%s%c", name, sign);
	493	if (opt->linenum)
	494	printf("%d%c", lno, sign);
	495	if (opt->color) {
	496	regmatch_t match;
	497	enum grep_context ctx = GREP_CONTEXT_BODY;
	498	int ch = *eol;
	499	int eflags = 0;
	500
	501	*eol = '\0';
	502	while (next_match(opt, bol, eol, ctx, &match, eflags)) {
1f5b9cc4 RS	503	if (match.rm_so == match.rm_eo)
1f5b9cc4 RS	504	break;
7e8f59d5	505	printf("%.s%s%.s%s",
747a322b	506	(int)match.rm_so, bol,
7e8f59d5	507	opt->color_match,
747a322b	508	(int)(match.rm_eo - match.rm_so), bol + match.rm_so,
7e8f59d5 RS	509	GIT_COLOR_RESET);
	510	bol += match.rm_eo;
	511	rest -= match.rm_eo;
	512	eflags = REG_NOTBOL;
	513	}
	514	*eol = ch;
	515	}
	516	printf("%.*s\n", rest, bol);
	517	}
	518
0ab7befa JH	519	static int grep_buffer_1(struct grep_opt opt, const char name,
0ab7befa JH	520	char *buf, unsigned long size, int collect_hits)
83b5d2f5 JH	521	{
	522	char *bol = buf;
	523	unsigned long left = size;
	524	unsigned lno = 1;
	525	struct pre_context_line {
	526	char *bol;
	527	char *eol;
	528	} prev = NULL, pcl;
	529	unsigned last_hit = 0;
	530	unsigned last_shown = 0;
	531	int binary_match_only = 0;
	532	const char *hunk_mark = "";
	533	unsigned count = 0;
480c1ca6	534	enum grep_context ctx = GREP_CONTEXT_HEAD;
83b5d2f5 JH	535
	536	if (buffer_is_binary(buf, size)) {
	537	switch (opt->binary) {
	538	case GREP_BINARY_DEFAULT:
	539	binary_match_only = 1;
	540	break;
	541	case GREP_BINARY_NOMATCH:
	542	return 0; /* Assume unmatch */
	543	break;
	544	default:
	545	break;
	546	}
	547	}
	548
	549	if (opt->pre_context)
	550	prev = xcalloc(opt->pre_context, sizeof(*prev));
	551	if (opt->pre_context \|\| opt->post_context)
	552	hunk_mark = "--\n";
	553
	554	while (left) {
	555	char *eol, ch;
0ab7befa	556	int hit;
83b5d2f5 JH	557
	558	eol = end_of_line(bol, &left);
	559	ch = *eol;
	560	*eol = 0;
	561
480c1ca6 JH	562	if ((ctx == GREP_CONTEXT_HEAD) && (eol == bol))
	563	ctx = GREP_CONTEXT_BODY;
	564
0ab7befa	565	hit = match_line(opt, bol, eol, ctx, collect_hits);
83b5d2f5 JH	566	*eol = ch;
83b5d2f5 JH	567
0ab7befa JH	568	if (collect_hits)
	569	goto next_line;
	570
83b5d2f5 JH	571	/* "grep -v -e foo -e bla" should list lines
	572	* that do not have either, so inversion should
	573	* be done outside.
	574	*/
	575	if (opt->invert)
	576	hit = !hit;
	577	if (opt->unmatch_name_only) {
	578	if (hit)
	579	return 0;
	580	goto next_line;
	581	}
	582	if (hit) {
	583	count++;
	584	if (opt->status_only)
	585	return 1;
	586	if (binary_match_only) {
	587	printf("Binary file %s matches\n", name);
	588	return 1;
	589	}
	590	if (opt->name_only) {
83caecca	591	show_name(opt, name);
83b5d2f5 JH	592	return 1;
	593	}
	594	/* Hit at this line. If we haven't shown the
	595	* pre-context lines, we would need to show them.
	596	* When asked to do "count", this still show
	597	* the context which is nonsense, but the user
	598	* deserves to get that ;-).
	599	*/
	600	if (opt->pre_context) {
	601	unsigned from;
	602	if (opt->pre_context < lno)
	603	from = lno - opt->pre_context;
	604	else
	605	from = 1;
	606	if (from <= last_shown)
	607	from = last_shown + 1;
	608	if (last_shown && from != last_shown + 1)
9db56f71	609	fputs(hunk_mark, stdout);
83b5d2f5 JH	610	while (from < lno) {
	611	pcl = &prev[lno-from-1];
	612	show_line(opt, pcl->bol, pcl->eol,
	613	name, from, '-');
	614	from++;
	615	}
	616	last_shown = lno-1;
	617	}
	618	if (last_shown && lno != last_shown + 1)
9db56f71	619	fputs(hunk_mark, stdout);
83b5d2f5 JH	620	if (!opt->count)
	621	show_line(opt, bol, eol, name, lno, ':');
	622	last_shown = last_hit = lno;
	623	}
	624	else if (last_hit &&
	625	lno <= last_hit + opt->post_context) {
	626	/* If the last hit is within the post context,
	627	* we need to show this line.
	628	*/
	629	if (last_shown && lno != last_shown + 1)
9db56f71	630	fputs(hunk_mark, stdout);
83b5d2f5 JH	631	show_line(opt, bol, eol, name, lno, '-');
	632	last_shown = lno;
	633	}
	634	if (opt->pre_context) {
	635	memmove(prev+1, prev,
	636	(opt->pre_context-1) * sizeof(*prev));
	637	prev->bol = bol;
	638	prev->eol = eol;
	639	}
	640
	641	next_line:
	642	bol = eol + 1;
	643	if (!left)
	644	break;
	645	left--;
	646	lno++;
	647	}
	648
b48fb5b6	649	free(prev);
0ab7befa JH	650	if (collect_hits)
0ab7befa JH	651	return 0;
b48fb5b6	652
83b5d2f5 JH	653	if (opt->status_only)
	654	return 0;
	655	if (opt->unmatch_name_only) {
	656	/* We did not see any hit, so we want to show this */
83caecca	657	show_name(opt, name);
83b5d2f5 JH	658	return 1;
	659	}
	660
	661	/* NEEDSWORK:
	662	* The real "grep -c foo *.c" gives many "bar.c:0" lines,
	663	* which feels mostly useless but sometimes useful. Maybe
	664	* make it another option? For now suppress them.
	665	*/
	666	if (opt->count && count)
83caecca RZ	667	printf("%s%c%u\n", name,
83caecca RZ	668	opt->null_following_name ? '\0' : ':', count);
83b5d2f5 JH	669	return !!last_hit;
	670	}
	671
0ab7befa JH	672	static void clr_hit_marker(struct grep_expr *x)
	673	{
	674	/* All-hit markers are meaningful only at the very top level
	675	* OR node.
	676	*/
	677	while (1) {
	678	x->hit = 0;
	679	if (x->node != GREP_NODE_OR)
	680	return;
	681	x->u.binary.left->hit = 0;
	682	x = x->u.binary.right;
	683	}
	684	}
	685
	686	static int chk_hit_marker(struct grep_expr *x)
	687	{
	688	/* Top level nodes have hit markers. See if they all are hits */
	689	while (1) {
	690	if (x->node != GREP_NODE_OR)
	691	return x->hit;
	692	if (!x->u.binary.left->hit)
	693	return 0;
	694	x = x->u.binary.right;
	695	}
	696	}
	697
	698	int grep_buffer(struct grep_opt opt, const char name, char *buf, unsigned long size)
	699	{
	700	/*
	701	* we do not have to do the two-pass grep when we do not check
	702	* buffer-wide "all-match".
	703	*/
	704	if (!opt->all_match)
	705	return grep_buffer_1(opt, name, buf, size, 0);
	706
	707	/* Otherwise the toplevel "or" terms hit a bit differently.
	708	* We first clear hit markers from them.
	709	*/
	710	clr_hit_marker(opt->pattern_expression);
	711	grep_buffer_1(opt, name, buf, size, 1);
	712
	713	if (!chk_hit_marker(opt->pattern_expression))
	714	return 0;
	715
	716	return grep_buffer_1(opt, name, buf, size, 0);
	717	}