src/regex/regexec.c

/*
 * the outer shell of regexec()
 *
 * This file includes engine.c *twice*, after muchos fiddling with the
 * macros that code uses.  This lets the same code operate on two different
 * representations for state sets.
 */
#include <sys/types.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <limits.h>
#include <ctype.h>
#include <regex.h>

#include "utils.h"
#include "regex2.h"

#ifdef REDEBUG
static int nope = 0;          /* for use in asserts; shuts lint up */
#endif

/* macros for manipulating states, small version */
#define	states	unsigned
#define	states1	unsigned	/* for later use in regexec() decision */
#define	CLEAR(v)	((v) = 0)
#define	SET0(v, n)	((v) &= ~((unsigned)1 << (n)))
#define	SET1(v, n)	((v) |= (unsigned)1 << (n))
#define	ISSET(v, n)	((v) & ((unsigned)1 << (n)))
#define	ASSIGN(d, s)	((d) = (s))
#define	EQ(a, b)	((a) == (b))
#define	STATEVARS	int dummy	/* dummy version */
#define	STATESETUP(m, n)	/* nothing */
#define	STATETEARDOWN(m)	/* nothing */
#define	SETUP(v)	((v) = 0)
#define	onestate	unsigned
#define	INIT(o, n)	((o) = (unsigned)1 << (n))
#define	INC(o)	((o) <<= 1)
#define	ISSTATEIN(v, o)	((v) & (o))
/* some abbreviations; note that some of these know variable names! */
/* do "if I'm here, I can also be there" etc without branches */
#define	FWD(dst, src, n)	((dst) |= ((unsigned)(src)&(here)) << (n))
#define	BACK(dst, src, n)	((dst) |= ((unsigned)(src)&(here)) >> (n))
#define	ISSETBACK(v, n)	((v) & ((unsigned)here >> (n)))
/* function names */
#define SNAMES			/* engine.c looks after details */

#include "engine.c"

/* now undo things */
#undef	states
#undef	CLEAR
#undef	SET0
#undef	SET1
#undef	ISSET
#undef	ASSIGN
#undef	EQ
#undef	STATEVARS
#undef	STATESETUP
#undef	STATETEARDOWN
#undef	SETUP
#undef	onestate
#undef	INIT
#undef	INC
#undef	ISSTATEIN
#undef	FWD
#undef	BACK
#undef	ISSETBACK
#undef	SNAMES

/* macros for manipulating states, large version */
#define	states	char *
#define	CLEAR(v)	memset(v, 0, m->g->nstates)
#define	SET0(v, n)	((v)[n] = 0)
#define	SET1(v, n)	((v)[n] = 1)
#define	ISSET(v, n)	((v)[n])
#define	ASSIGN(d, s)	memcpy(d, s, m->g->nstates)
#define	EQ(a, b)	(memcmp(a, b, m->g->nstates) == 0)
#define	STATEVARS	int vn; char *space
#define	STATESETUP(m, nv)	{ (m)->space = malloc((nv)*(m)->g->nstates); \
				if ((m)->space == NULL) return(REG_ESPACE); \
				(m)->vn = 0; }
#define	STATETEARDOWN(m)	{ free((m)->space); }
#define	SETUP(v)	((v) = &m->space[m->vn++ * m->g->nstates])
#define	onestate	int
#define	INIT(o, n)	((o) = (n))
#define	INC(o)	((o)++)
#define	ISSTATEIN(v, o)	((v)[o])
/* some abbreviations; note that some of these know variable names! */
/* do "if I'm here, I can also be there" etc without branches */
#define	FWD(dst, src, n)	((dst)[here+(n)] |= (src)[here])
#define	BACK(dst, src, n)	((dst)[here-(n)] |= (src)[here])
#define	ISSETBACK(v, n)	((v)[here - (n)])
/* function names */
#define	LNAMES			/* flag */

#include "engine.c"

/*
 - regexec - interface for matching
 = extern int regexec(const regex_t *, const char *, size_t, \
 =					regmatch_t [], int);
 = #define	REG_NOTBOL	00001
 = #define	REG_NOTEOL	00002
 = #define	REG_STARTEND	00004
 = #define	REG_TRACE	00400	// tracing of execution
 = #define	REG_LARGE	01000	// force large representation
 = #define	REG_BACKR	02000	// force use of backref code
 *
 * We put this here so we can exploit knowledge of the state representation
 * when choosing which matcher to call.  Also, by this point the matchers
 * have been prototyped.
 */
int				/* 0 success, REG_NOMATCH failure */
regexec(preg, string, nmatch, pmatch, eflags)
const regex_t *preg;
const char *string;
size_t nmatch;
regmatch_t pmatch[];
int eflags;
{
	register struct re_guts *g = preg->re_g;
#ifdef REDEBUG
#	define	GOODFLAGS(f)	(f)
#else
#	define	GOODFLAGS(f)	((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND))
#endif

	if (preg->re_magic != MAGIC1 || g->magic != MAGIC2)
		return(REG_BADPAT);
	assert(!(g->iflags&BAD));
	if (g->iflags&BAD)		/* backstop for no-debug case */
		return(REG_BADPAT);
	eflags = GOODFLAGS(eflags);

	if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
		return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
	else
		return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
}
Commit	Line	Data
	1	/*
	2	* the outer shell of regexec()
	3	*
	4	* This file includes engine.c twice, after muchos fiddling with the
	5	* macros that code uses. This lets the same code operate on two different
	6	* representations for state sets.
	7	*/
	8	#include <sys/types.h>
	9	#include <stdio.h>
	10	#include <stdlib.h>
	11	#include <string.h>
	12	#include <limits.h>
	13	#include <ctype.h>
	14	#include <regex.h>
	15
	16	#include "utils.h"
	17	#include "regex2.h"
	18
	19	#ifdef REDEBUG
	20	static int nope = 0; /* for use in asserts; shuts lint up */
	21	#endif
	22
	23	/* macros for manipulating states, small version */
	24	#define states unsigned
	25	#define states1 unsigned /* for later use in regexec() decision */
	26	#define CLEAR(v) ((v) = 0)
	27	#define SET0(v, n) ((v) &= ~((unsigned)1 << (n)))
	28	#define SET1(v, n) ((v) \|= (unsigned)1 << (n))
	29	#define ISSET(v, n) ((v) & ((unsigned)1 << (n)))
	30	#define ASSIGN(d, s) ((d) = (s))
	31	#define EQ(a, b) ((a) == (b))
	32	#define STATEVARS int dummy /* dummy version */
	33	#define STATESETUP(m, n) /* nothing */
	34	#define STATETEARDOWN(m) /* nothing */
	35	#define SETUP(v) ((v) = 0)
	36	#define onestate unsigned
	37	#define INIT(o, n) ((o) = (unsigned)1 << (n))
	38	#define INC(o) ((o) <<= 1)
	39	#define ISSTATEIN(v, o) ((v) & (o))
	40	/* some abbreviations; note that some of these know variable names! */
	41	/* do "if I'm here, I can also be there" etc without branches */
	42	#define FWD(dst, src, n) ((dst) \|= ((unsigned)(src)&(here)) << (n))
	43	#define BACK(dst, src, n) ((dst) \|= ((unsigned)(src)&(here)) >> (n))
	44	#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n)))
	45	/* function names */
	46	#define SNAMES /* engine.c looks after details */
	47
	48	#include "engine.c"
	49
	50	/* now undo things */
	51	#undef states
	52	#undef CLEAR
	53	#undef SET0
	54	#undef SET1
	55	#undef ISSET
	56	#undef ASSIGN
	57	#undef EQ
	58	#undef STATEVARS
	59	#undef STATESETUP
	60	#undef STATETEARDOWN
	61	#undef SETUP
	62	#undef onestate
	63	#undef INIT
	64	#undef INC
	65	#undef ISSTATEIN
	66	#undef FWD
	67	#undef BACK
	68	#undef ISSETBACK
	69	#undef SNAMES
	70
	71	/* macros for manipulating states, large version */
	72	#define states char *
	73	#define CLEAR(v) memset(v, 0, m->g->nstates)
	74	#define SET0(v, n) ((v)[n] = 0)
	75	#define SET1(v, n) ((v)[n] = 1)
	76	#define ISSET(v, n) ((v)[n])
	77	#define ASSIGN(d, s) memcpy(d, s, m->g->nstates)
	78	#define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0)
	79	#define STATEVARS int vn; char *space
	80	#define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \
	81	if ((m)->space == NULL) return(REG_ESPACE); \
	82	(m)->vn = 0; }
	83	#define STATETEARDOWN(m) { free((m)->space); }
	84	#define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates])
	85	#define onestate int
	86	#define INIT(o, n) ((o) = (n))
	87	#define INC(o) ((o)++)
	88	#define ISSTATEIN(v, o) ((v)[o])
	89	/* some abbreviations; note that some of these know variable names! */
	90	/* do "if I'm here, I can also be there" etc without branches */
	91	#define FWD(dst, src, n) ((dst)[here+(n)] \|= (src)[here])
	92	#define BACK(dst, src, n) ((dst)[here-(n)] \|= (src)[here])
	93	#define ISSETBACK(v, n) ((v)[here - (n)])
	94	/* function names */
	95	#define LNAMES /* flag */
	96
	97	#include "engine.c"
	98
	99	/*
	100	- regexec - interface for matching
	101	= extern int regexec(const regex_t , const char , size_t, \
	102	= regmatch_t [], int);
	103	= #define REG_NOTBOL 00001
	104	= #define REG_NOTEOL 00002
	105	= #define REG_STARTEND 00004
	106	= #define REG_TRACE 00400 // tracing of execution
	107	= #define REG_LARGE 01000 // force large representation
	108	= #define REG_BACKR 02000 // force use of backref code
	109	*
	110	* We put this here so we can exploit knowledge of the state representation
	111	* when choosing which matcher to call. Also, by this point the matchers
	112	* have been prototyped.
	113	*/
	114	int /* 0 success, REG_NOMATCH failure */
	115	regexec(preg, string, nmatch, pmatch, eflags)
	116	const regex_t *preg;
	117	const char *string;
	118	size_t nmatch;
	119	regmatch_t pmatch[];
	120	int eflags;
	121	{
	122	register struct re_guts *g = preg->re_g;
	123	#ifdef REDEBUG
	124	# define GOODFLAGS(f) (f)
	125	#else
	126	# define GOODFLAGS(f) ((f)&(REG_NOTBOL\|REG_NOTEOL\|REG_STARTEND))
	127	#endif
	128
	129	if (preg->re_magic != MAGIC1 \|\| g->magic != MAGIC2)
	130	return(REG_BADPAT);
	131	assert(!(g->iflags&BAD));
	132	if (g->iflags&BAD) /* backstop for no-debug case */
	133	return(REG_BADPAT);
	134	eflags = GOODFLAGS(eflags);
	135
	136	if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
	137	return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
	138	else
	139	return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
	140	}