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.
 */
#ifdef __MWERKS__
typedef long off_t ;
#else
#include <sys/types.h>
#endif
#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	#ifdef __MWERKS__
	9	typedef long off_t ;
	10	#else
	11	#include <sys/types.h>
	12	#endif
	13	#include <stdio.h>
	14	#include <stdlib.h>
	15	#include <string.h>
	16	#include <limits.h>
	17	#include <ctype.h>
	18	#include "regex.h"
	19
	20	#include "utils.h"
	21	#include "regex2.h"
	22
	23	#ifdef REDEBUG
	24	static int nope = 0; /* for use in asserts; shuts lint up */
	25	#endif
	26
	27	/* macros for manipulating states, small version */
	28	#define states unsigned
	29	#define states1 unsigned /* for later use in regexec() decision */
	30	#define CLEAR(v) ((v) = 0)
	31	#define SET0(v, n) ((v) &= ~((unsigned)1 << (n)))
	32	#define SET1(v, n) ((v) \|= (unsigned)1 << (n))
	33	#define ISSET(v, n) ((v) & ((unsigned)1 << (n)))
	34	#define ASSIGN(d, s) ((d) = (s))
	35	#define EQ(a, b) ((a) == (b))
	36	#define STATEVARS int dummy /* dummy version */
	37	#define STATESETUP(m, n) /* nothing */
	38	#define STATETEARDOWN(m) /* nothing */
	39	#define SETUP(v) ((v) = 0)
	40	#define onestate unsigned
	41	#define INIT(o, n) ((o) = (unsigned)1 << (n))
	42	#define INC(o) ((o) <<= 1)
	43	#define ISSTATEIN(v, o) ((v) & (o))
	44	/* some abbreviations; note that some of these know variable names! */
	45	/* do "if I'm here, I can also be there" etc without branches */
	46	#define FWD(dst, src, n) ((dst) \|= ((unsigned)(src)&(here)) << (n))
	47	#define BACK(dst, src, n) ((dst) \|= ((unsigned)(src)&(here)) >> (n))
	48	#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n)))
	49	/* function names */
	50	#define SNAMES /* engine.c looks after details */
	51
	52	#include "engine.c"
	53
	54	/* now undo things */
	55	#undef states
	56	#undef CLEAR
	57	#undef SET0
	58	#undef SET1
	59	#undef ISSET
	60	#undef ASSIGN
	61	#undef EQ
	62	#undef STATEVARS
	63	#undef STATESETUP
	64	#undef STATETEARDOWN
	65	#undef SETUP
	66	#undef onestate
	67	#undef INIT
	68	#undef INC
	69	#undef ISSTATEIN
	70	#undef FWD
	71	#undef BACK
	72	#undef ISSETBACK
	73	#undef SNAMES
	74
	75	/* macros for manipulating states, large version */
	76	#define states char *
	77	#define CLEAR(v) memset(v, 0, m->g->nstates)
	78	#define SET0(v, n) ((v)[n] = 0)
	79	#define SET1(v, n) ((v)[n] = 1)
	80	#define ISSET(v, n) ((v)[n])
	81	#define ASSIGN(d, s) memcpy(d, s, m->g->nstates)
	82	#define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0)
	83	#define STATEVARS int vn; char *space
	84	#define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \
	85	if ((m)->space == NULL) return(REG_ESPACE); \
	86	(m)->vn = 0; }
	87	#define STATETEARDOWN(m) { free((m)->space); }
	88	#define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates])
	89	#define onestate int
	90	#define INIT(o, n) ((o) = (n))
	91	#define INC(o) ((o)++)
	92	#define ISSTATEIN(v, o) ((v)[o])
	93	/* some abbreviations; note that some of these know variable names! */
	94	/* do "if I'm here, I can also be there" etc without branches */
	95	#define FWD(dst, src, n) ((dst)[here+(n)] \|= (src)[here])
	96	#define BACK(dst, src, n) ((dst)[here-(n)] \|= (src)[here])
	97	#define ISSETBACK(v, n) ((v)[here - (n)])
	98	/* function names */
	99	#define LNAMES /* flag */
	100
	101	#include "engine.c"
	102
	103	/*
	104	- regexec - interface for matching
	105	= extern int regexec(const regex_t , const char , size_t, \
	106	= regmatch_t [], int);
	107	= #define REG_NOTBOL 00001
	108	= #define REG_NOTEOL 00002
	109	= #define REG_STARTEND 00004
	110	= #define REG_TRACE 00400 // tracing of execution
	111	= #define REG_LARGE 01000 // force large representation
	112	= #define REG_BACKR 02000 // force use of backref code
	113	*
	114	* We put this here so we can exploit knowledge of the state representation
	115	* when choosing which matcher to call. Also, by this point the matchers
	116	* have been prototyped.
	117	*/
	118	int /* 0 success, REG_NOMATCH failure */
	119	regexec(preg, string, nmatch, pmatch, eflags)
	120	const regex_t *preg;
	121	const char *string;
	122	size_t nmatch;
	123	regmatch_t pmatch[];
	124	int eflags;
	125	{
	126	register struct re_guts *g = preg->re_g;
	127	#ifdef REDEBUG
	128	# define GOODFLAGS(f) (f)
	129	#else
	130	# define GOODFLAGS(f) ((f)&(REG_NOTBOL\|REG_NOTEOL\|REG_STARTEND))
	131	#endif
	132
	133	if (preg->re_magic != MAGIC1 \|\| g->magic != MAGIC2)
	134	return(REG_BADPAT);
	135	assert(!(g->iflags&BAD));
	136	if (g->iflags&BAD) /* backstop for no-debug case */
	137	return(REG_BADPAT);
	138	eflags = GOODFLAGS(eflags);
	139
	140	if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags&REG_LARGE))
	141	return(smatcher(g, (char *)string, nmatch, pmatch, eflags));
	142	else
	143	return(lmatcher(g, (char *)string, nmatch, pmatch, eflags));
	144	}