/*
 * Simple test program for regexp(3) stuff.  Knows about debugging hooks.
 *
 *	Copyright (c) 1986 by University of Toronto.
 *	Written by Henry Spencer.  Not derived from licensed software.
 *
 *	Permission is granted to anyone to use this software for any
 *	purpose on any computer system, and to redistribute it freely,
 *	subject to the following restrictions:
 *
 *	1. The author is not responsible for the consequences of use of
 *		this software, no matter how awful, even if they arise
 *		from defects in it.
 *
 *	2. The origin of this software must not be misrepresented, either
 *		by explicit claim or by omission.
 *
 *	3. Altered versions must be plainly marked as such, and must not
 *		be misrepresented as being the original software.
 *
 * Usage: try re [string [output [-]]]
 * The re is compiled and dumped, regexeced against the string, the result
 * is applied to output using regsub().  The - triggers a running narrative
 * from regexec().  Dumping and narrative don't happen unless DEBUG.
 *
 * If there are no arguments, stdin is assumed to be a stream of lines with
 * five fields:  a r.e., a string to match it against, a result code, a
 * source string for regsub, and the proper result.  Result codes are 'c'
 * for compile failure, 'y' for match success, 'n' for match failure.
 * Field separator is tab.
 */
#include <stdio.h>
#include <regexp.h>

/* build test strings in ++jrb */
char *test_input[] = {
    "abc	abc	y	&	abc",
    "abc	xbc	n	-	-",
    "abc	axc	n	-	-",
    "abc	abx	n	-	-",
    "abc	xabcy	y	&	abc",
    "abc	ababc	y	&	abc",
    "ab*c	abc	y	&	abc",
    "ab*bc	abc	y	&	abc",
    "ab*bc	abbc	y	&	abbc",
    "ab*bc	abbbbc	y	&	abbbbc",
    "ab+bc	abbc	y	&	abbc",
    "ab+bc	abc	n	-	-",
    "ab+bc	abq	n	-	-",
    "ab+bc	abbbbc	y	&	abbbbc",
    "ab?bc	abbc	y	&	abbc",
    "ab?bc	abc	y	&	abc",
    "ab?bc	abbbbc	n	-	-",
    "ab?c	abc	y	&	abc",
    "^abc$	abc	y	&	abc",
    "^abc$	abcc	n	-	-",
    "^abc	abcc	y	&	abc",
    "^abc$	aabc	n	-	-",
    "abc$	aabc	y	&	abc",
    "^	abc	y	&	",
    "$	abc	y	&	",
    "a.c	abc	y	&	abc",
    "a.c	axc	y	&	axc",
    "a.*c	axyzc	y	&	axyzc",
    "a.*c	axyzd	n	-	-",
    "a[bc]d	abc	n	-	-",
    "a[bc]d	abd	y	&	abd",
    "a[b-d]e	abd	n	-	-",
    "a[b-d]e	ace	y	&	ace",
    "a[b-d]	aac	y	&	ac",
    "a[-b]	a-	y	&	a-",
    "a[b-]	a-	y	&	a-",
    "a[b-a]	-	c	-	-",
    "a[]b	-	c	-	-",
    "a[	-	c	-	-",
    "a]	a]	y	&	a]",
    "a[]]b	a]b	y	&	a]b",
    "a[^bc]d	aed	y	&	aed",
    "a[^bc]d	abd	n	-	-",
    "a[^-b]c	adc	y	&	adc",
    "a[^-b]c	a-c	n	-	-",
    "a[^]b]c	a]c	n	-	-",
    "a[^]b]c	adc	y	&	adc",
    "ab|cd	abc	y	&	ab",
    "ab|cd	abcd	y	&	ab",
    "()ef	def	y	&-\\1	ef-",
    "()*	-	c	-	-",
    "*a	-	c	-	-",
    "^*	-	c	-	-",
    "$*	-	c	-	-",
    "(*)b	-	c	-	-",
    "$b	b	n	-	-",
    "a\\	-	c	-	-",
    "a\\(b	a(b	y	&-\\1	a(b-",
    "a\\(*b	ab	y	&	ab",
    "a\\(*b	a((b	y	&	a((b",
    "a\\\\b	a\\b	y	&	a\\b",
    "abc)	-	c	-	-",
    "(abc	-	c	-	-",
    "((a))	abc	y	&-\\1-\\2	a-a-a",
    "(a)b(c)	abc	y	&-\\1-\\2	abc-a-c",
    "a+b+c	aabbabc	y	&	abc",
    "a**	-	c	-	-",
    "a*?	-	c	-	-",
    "(a*)*	-	c	-	-",
    "(a*)+	-	c	-	-",
    "(a|)*	-	c	-	-",
    "(a*|b)*	-	c	-	-",
    "(a+|b)*	ab	y	&-\\1	ab-b",
    "(a+|b)+	ab	y	&-\\1	ab-b",
    "(a+|b)?	ab	y	&-\\1	a-a",
    "[^ab]*	cde	y	&	cde",
    "(^)*	-	c	-	-",
    "(ab|)*	-	c	-	-",
    ")(	-	c	-	-",
    "	abc	y	&	",
    "abc		n	-	-",
    "a*		y	&	",
    "([abc])*d	abbbcd	y	&-\\1	abbbcd-c",
    "([abc])*bcd	abcd	y	&-\\1	abcd-a",
    "a|b|c|d|e	e	y	&	e",
    "(a|b|c|d|e)f	ef	y	&-\\1	ef-e",
    "((a*|b))*	-	c	-	-",
    "abcd*efg	abcdefg	y	&	abcdefg",
    "ab*	xabyabbbz	y	&	ab",
    "ab*	xayabbbz	y	&	a",
    "(ab|cd)e	abcde	y	&-\\1	cde-cd",
    "[abhgefdc]ij	hij	y	&	hij",
    "^(ab|cd)e	abcde	n	x\\1y	xy",
    "(abc|)ef	abcdef	y	&-\\1	ef-",
    "(a|b)c*d	abcd	y	&-\\1	bcd-b",
    "(ab|ab*)bc	abc	y	&-\\1	abc-a",
    "a([bc]*)c*	abc	y	&-\\1	abc-bc",
    "a([bc]*)(c*d)	abcd	y	&-\\1-\\2	abcd-bc-d",
    "a([bc]+)(c*d)	abcd	y	&-\\1-\\2	abcd-bc-d",
    "a([bc]*)(c+d)	abcd	y	&-\\1-\\2	abcd-b-cd",
    "a[bcd]*dcdcde	adcdcde	y	&	adcdcde",
    "a[bcd]+dcdcde	adcdcde	n	-	-",
    "(ab|a)b*c	abc	y	&-\\1	abc-ab",
    "((a)(b)c)(d)	abcd	y	\\1-\\2-\\3-\\4	abc-a-b-d",
    "[a-zA-Z_][a-zA-Z0-9_]*	alpha	y	&	alpha",
    "^a(bc+|b[eh])g|.h$	abh	y	&-\\1	bh-",
    "(bc+d$|ef*g.|h?i(j|k))	effgz	y	&-\\1-\\2	effgz-effgz-",
    "(bc+d$|ef*g.|h?i(j|k))	ij	y	&-\\1-\\2	ij-ij-j",
    "(bc+d$|ef*g.|h?i(j|k))	effg	n	-	-",
    "(bc+d$|ef*g.|h?i(j|k))	bcdd	n	-	-",
    "(bc+d$|ef*g.|h?i(j|k))	reffgz	y	&-\\1-\\2	effgz-effgz-",
    "((((((((((a))))))))))	-	c	-	-",
    "(((((((((a)))))))))	a	y	&	a",
    "multiple words of text	uh-uh	n	-	-",
    "multiple words	multiple words, yeah	y	&	multiple words",
    "(.*)c(.*)	abcde	y	&-\\1-\\2	abcde-ab-de",
    "\\((.*), (.*)\\)	(a, b)	y	(\\2, \\1)	(b, a)",
    "abcd	abcd	y	&-\\&-\\\\&	abcd-&-\\abcd",
    "a(bc)d	abcd	y	\\1-\\\\1-\\\\\\1	bc-\\1-\\bc",
    "[ -~]*	abc	y	&	abc",
    "[ -~ -~]*	abc	y	&	abc",
    "[ -~ -~ -~]*	abc	y	&	abc",
    "[ -~ -~ -~ -~]*	abc	y	&	abc",
    "[ -~ -~ -~ -~ -~]*	abc	y	&	abc",
    "[ -~ -~ -~ -~ -~ -~]*	abc	y	&	abc",
    "[ -~ -~ -~ -~ -~ -~ -~]*	abc	y	&	abc",
  (char *)NULL
};

#ifdef ERRAVAIL
char *progname;
extern char *mkprogname();
#endif

#ifdef DEBUG
extern int regnarrate;
#endif

char buf[BUFSIZ];

int errreport = 0;		/* Report errors via errseen? */
char *errseen = NULL;		/* Error message. */
int status = 0;			/* Exit status. */

/* ARGSUSED */
int main(argc, argv)
int argc;
char *argv[];
{
	regexp *r;
	int i;

#ifdef ERRAVAIL
	progname = mkprogname(argv[0]);
#endif

	if (argc == 1) {
		multiple();
		return(status);
	}

	r = regcomp(argv[1]);
	if (r == NULL)
		error("regcomp failure", "");
#ifdef DEBUG
	regdump(r);
	if (argc > 4)
		regnarrate++;
#endif
	if (argc > 2) {
		i = regexec(r, argv[2], 1);
		printf("%d", i);
		for (i = 1; i < NSUBEXP; i++)
			if (r->startp[i] != NULL && r->endp[i] != NULL)
				printf(" \\%d", i);
		printf("\n");
	}
	if (argc > 3) {
		regsub(r, argv[3], buf);
		printf("%s\n", buf);
	}
	return(status);
}

void
regerror(s)
char *s;
{
	if (errreport)
		errseen = s;
	else
		error(s, "");
}

#ifndef ERRAVAIL
error(s1, s2)
char *s1;
char *s2;
{
	fprintf(stderr, "regexp: ");
	fprintf(stderr, s1, s2);
	fprintf(stderr, "\n");
	exit(1);
}
#endif

int lineno;

regexp badregexp;		/* Implicit init to 0. */

multiple()
{
	char *rbuf;
	char *field[5];
	char *scan;
	int i, j;
	regexp *r;
	extern char *strchr();

	errreport = 1;
	lineno = 0;
        for(rbuf = test_input[0], j = 0; rbuf; rbuf = test_input[++j]) {
		lineno++;
		scan = rbuf;
		for (i = 0; i < 5; i++) {
			field[i] = scan;
			if (field[i] == NULL) {
				complain("bad testfile format", "");
				exit(1);
			}
			scan = strchr(scan, '\t');
			if (scan != NULL)
				*scan++ = '\0';
		}
		try(field);
	}

	/* And finish up with some internal testing... */
	lineno = 9990;
	errseen = NULL;
	if (regcomp((char *)NULL) != NULL || errseen == NULL)
		complain("regcomp(NULL) doesn't complain", "");
	lineno = 9991;
	errseen = NULL;
	if (regexec((regexp *)NULL, "foo", 1) || errseen == NULL)
		complain("regexec(NULL, ...) doesn't complain", "");
	lineno = 9992;
	r = regcomp("foo");
	if (r == NULL) {
		complain("regcomp(\"foo\") fails", "");
		return;
	}
	lineno = 9993;
	errseen = NULL;
	if (regexec(r, (char *)NULL, 1) || errseen == NULL)
		complain("regexec(..., NULL) doesn't complain", "");
	lineno = 9994;
	errseen = NULL;
	regsub((regexp *)NULL, "foo", rbuf);
	if (errseen == NULL)
		complain("regsub(NULL, ..., ...) doesn't complain", "");
	lineno = 9995;
	errseen = NULL;
	regsub(r, (char *)NULL, rbuf);
	if (errseen == NULL)
		complain("regsub(..., NULL, ...) doesn't complain", "");
	lineno = 9996;
	errseen = NULL;
	regsub(r, "foo", (char *)NULL);
	if (errseen == NULL)
		complain("regsub(..., ..., NULL) doesn't complain", "");
	lineno = 9997;
	errseen = NULL;
	if (regexec(&badregexp, "foo", 1) || errseen == NULL)
		complain("regexec(nonsense, ...) doesn't complain", "");
	lineno = 9998;
	errseen = NULL;
	regsub(&badregexp, "foo", rbuf);
	if (errseen == NULL)
		complain("regsub(nonsense, ..., ...) doesn't complain", "");
}

try(fields)
char **fields;
{
	regexp *r;
	char dbuf[BUFSIZ];

	errseen = NULL;
	r = regcomp(fields[0]);
	if (r == NULL) {
		if (*fields[2] != 'c')
			complain("regcomp failure in `%s'", fields[0]);
		return;
	}
	if (*fields[2] == 'c') {
		complain("unexpected regcomp success in `%s'", fields[0]);
		free((char *)r);
		return;
	}
	if (!regexec(r, fields[1], 1)) {
		if (*fields[2] != 'n')
			complain("regexec failure in `%s'", "");
		free((char *)r);
		return;
	}
	if (*fields[2] == 'n') {
		complain("unexpected regexec success", "");
		free((char *)r);
		return;
	}
	errseen = NULL;
	regsub(r, fields[3], dbuf);
	if (errseen != NULL) {
		complain("regsub complaint", "");
		free((char *)r);
		return;
	}
	if (strcmp(dbuf, fields[4]) != 0)
		complain("regsub result `%s' wrong", dbuf);
	free((char *)r);
}

complain(s1, s2)
char *s1;
char *s2;
{
	fprintf(stderr, "try: %d: ", lineno);
	fprintf(stderr, s1, s2);
	fprintf(stderr, " (%s)\n", (errseen != NULL) ? errseen : "");
	status = 1;
}