* @(#)regcomp.c 8.5 (Berkeley) 3/20/94
*/
-#if defined(LIBC_SCCS) && !defined(lint)
-static char sccsid[] = "@(#)regcomp.c 8.5 (Berkeley) 3/20/94";
+#include "postgres.h"
-#endif /* LIBC_SCCS and not lint */
-
-#include <sys/types.h>
-#include <stdio.h>
-#include <string.h>
#include <ctype.h>
#include <limits.h>
-#include <stdlib.h>
#include <assert.h>
-#include <regex/regex.h>
-#include <regex/utils.h>
-#include <regex/regex2.h>
-#include <regex/cclass.h>
-#include <regex/cname.h>
+#include "regex/regex.h"
+#include "regex/utils.h"
+#include "regex/regex2.h"
+#include "regex/cname.h"
+#include <locale.h>
+
+struct cclass
+{
+ char *name;
+ char *chars;
+ char *multis;
+};
+static struct cclass *cclasses = NULL;
+static struct cclass *cclass_init(void);
/*
* parse structure, passed up and down to avoid global variables and
*/
struct parse
{
- char *next; /* next character in RE */
- char *end; /* end of string (-> NUL normally) */
+ pg_wchar *next; /* next character in RE */
+ pg_wchar *end; /* end of string (-> NUL normally) */
int error; /* has an error been seen? */
sop *strip; /* malloced strip */
sopno ssize; /* malloced strip size (allocated) */
sopno pend[NPAREN]; /* -> ) ([0] unused) */
};
-/* ========= begin header generated by ./mkh ========= */
-#ifdef __cplusplus
-extern "C"
-{
-#endif
-
-/* === regcomp.c === */
- static void p_ere(struct parse * p, int stop);
- static void p_ere_exp(struct parse * p);
- static void p_str(struct parse * p);
- static void p_bre(struct parse * p, int end1, int end2);
- static int p_simp_re(struct parse * p, int starordinary);
- static int p_count(struct parse * p);
- static void p_bracket(struct parse * p);
- static void p_b_term(struct parse * p, cset * cs);
- static void p_b_cclass(struct parse * p, cset * cs);
- static void p_b_eclass(struct parse * p, cset * cs);
- static char p_b_symbol(struct parse * p);
- static char p_b_coll_elem(struct parse * p, int endc);
- static char othercase(int ch);
- static void bothcases(struct parse * p, int ch);
- static void ordinary(struct parse * p, int ch);
- static void nonnewline(struct parse * p);
- static void repeat(struct parse * p, sopno start, int from, int to);
- static int seterr(struct parse * p, int e);
- static cset *allocset(struct parse * p);
- static void freeset(struct parse * p, cset * cs);
- static int freezeset(struct parse * p, cset * cs);
- static int firstch(struct parse * p, cset * cs);
- static int nch(struct parse * p, cset * cs);
- static void mcadd(struct parse * p, cset * cs, char *cp);
- static void mcinvert(struct parse * p, cset * cs);
- static void mccase(struct parse * p, cset * cs);
- static int isinsets(struct re_guts * g, int c);
- static int samesets(struct re_guts * g, int c1, int c2);
- static void categorize(struct parse * p, struct re_guts * g);
- static sopno dupl(struct parse * p, sopno start, sopno finish);
- static void doemit(struct parse * p, sop op, size_t opnd);
- static void doinsert(struct parse * p, sop op, size_t opnd, sopno pos);
- static void dofwd(struct parse * p, sopno pos, sop value);
- static void enlarge(struct parse * p, sopno size);
- static void stripsnug(struct parse * p, struct re_guts * g);
- static void findmust(struct parse * p, struct re_guts * g);
- static sopno pluscount(struct parse * p, struct re_guts * g);
-
-#ifdef __cplusplus
-}
-
-#endif
-/* ========= end header generated by ./mkh ========= */
-
-static char nuls[10]; /* place to point scanner in event of
+static void p_ere(struct parse * p, int stop);
+static void p_ere_exp(struct parse * p);
+static void p_str(struct parse * p);
+static void p_bre(struct parse * p, int end1, int end2);
+static int p_simp_re(struct parse * p, int starordinary);
+static int p_count(struct parse * p);
+static void p_bracket(struct parse * p);
+static void p_b_term(struct parse * p, cset *cs);
+static void p_b_cclass(struct parse * p, cset *cs);
+static void p_b_eclass(struct parse * p, cset *cs);
+static pg_wchar p_b_symbol(struct parse * p);
+static char p_b_coll_elem(struct parse * p, int endc);
+static unsigned char othercase(int ch);
+static void bothcases(struct parse * p, int ch);
+static void ordinary(struct parse * p, int ch);
+static void nonnewline(struct parse * p);
+static void repeat(struct parse * p, sopno start, int from, int to);
+static int seterr(struct parse * p, int e);
+static cset *allocset(struct parse * p);
+static void freeset(struct parse * p, cset *cs);
+static int freezeset(struct parse * p, cset *cs);
+static int firstch(struct parse * p, cset *cs);
+static int nch(struct parse * p, cset *cs);
+static void mcadd(struct parse * p, cset *cs, char *cp);
+static void mcinvert(struct parse * p, cset *cs);
+static void mccase(struct parse * p, cset *cs);
+static int isinsets(struct re_guts * g, int c);
+static int samesets(struct re_guts * g, int c1, int c2);
+static void categorize(struct parse * p, struct re_guts * g);
+static sopno dupl(struct parse * p, sopno start, sopno finish);
+static void doemit(struct parse * p, sop op, size_t opnd);
+static void doinsert(struct parse * p, sop op, size_t opnd, sopno pos);
+static void dofwd(struct parse * p, sopno pos, sop value);
+static void enlarge(struct parse * p, sopno size);
+static void stripsnug(struct parse * p, struct re_guts * g);
+static void findmust(struct parse * p, struct re_guts * g);
+static sopno pluscount(struct parse * p, struct re_guts * g);
+static int pg_isdigit(int c);
+static int pg_isalpha(int c);
+static int pg_isalnum(int c);
+static int pg_isupper(int c);
+static int pg_islower(int c);
+static int pg_iscntrl(int c);
+static int pg_isgraph(int c);
+static int pg_isprint(int c);
+static int pg_ispunct(int c);
+
+static pg_wchar nuls[10]; /* place to point scanner in event of
* error */
/*
#endif
/*
- - regcomp - interface for parser and compilation
- = extern int regcomp(regex_t *, const char *, int);
- = #define REG_BASIC 0000
- = #define REG_EXTENDED 0001
- = #define REG_ICASE 0002
- = #define REG_NOSUB 0004
- = #define REG_NEWLINE 0010
- = #define REG_NOSPEC 0020
- = #define REG_PEND 0040
- = #define REG_DUMP 0200
+ * regcomp - interface for parser and compilation
+ * returns 0 success, otherwise REG_something
*/
-int /* 0 success, otherwise REG_something */
-pg95_regcomp(preg, pattern, cflags)
-regex_t *preg;
-const char *pattern;
-int cflags;
+int
+pg_regcomp(regex_t *preg, const char *pattern, int cflags)
{
struct parse pa;
- register struct re_guts *g;
- register struct parse *p = &pa;
- register int i;
- register size_t len;
+ struct re_guts *g;
+ struct parse *p = &pa;
+ int i;
+ size_t len;
+ pg_wchar *wcp;
+
+ if (cclasses == NULL)
+ cclasses = cclass_init();
#ifdef REDEBUG
#define GOODFLAGS(f) (f)
cflags = GOODFLAGS(cflags);
if ((cflags & REG_EXTENDED) && (cflags & REG_NOSPEC))
- return (REG_INVARG);
+ return REG_INVARG;
if (cflags & REG_PEND)
{
- if (preg->re_endp < pattern)
- return (REG_INVARG);
- len = preg->re_endp - pattern;
+ wcp = preg->patsave;
+ if (preg->re_endp < wcp)
+ return REG_INVARG;
+ len = preg->re_endp - wcp;
}
else
- len = strlen((char *) pattern);
+ {
+ wcp = (pg_wchar *) malloc((strlen(pattern) + 1) * sizeof(pg_wchar));
+ if (wcp == NULL)
+ return REG_ESPACE;
+ preg->patsave = wcp;
+ (void) pg_mb2wchar((unsigned char *) pattern, wcp);
+ len = pg_wchar_strlen(wcp);
+ }
/* do the mallocs early so failure handling is easy */
g = (struct re_guts *) malloc(sizeof(struct re_guts) +
(NC - 1) * sizeof(cat_t));
if (g == NULL)
- return (REG_ESPACE);
+ return REG_ESPACE;
p->ssize = len / (size_t) 2 *(size_t) 3 + (size_t) 1; /* ugh */
p->strip = (sop *) malloc(p->ssize * sizeof(sop));
if (p->strip == NULL)
{
free((char *) g);
- return (REG_ESPACE);
+ return REG_ESPACE;
}
/* set things up */
p->g = g;
- p->next = (char *) pattern; /* convenience; we do not modify it */
+ p->next = wcp;
p->end = p->next + len;
p->error = 0;
p->ncsalloc = 0;
/* win or lose, we're done */
if (p->error != 0) /* lose */
- pg95_regfree(preg);
- return (p->error);
+ pg_regfree(preg);
+ return p->error;
}
/*
- - p_ere - ERE parser top level, concatenation and alternation
- == static void p_ere(register struct parse *p, int stop);
+ * p_ere - ERE parser top level, concatenation and alternation
*/
static void
-p_ere(p, stop)
-register struct parse *p;
-int stop; /* character this ERE should end at */
+p_ere(struct parse * p,
+ int stop) /* character this ERE should end at */
{
- register char c;
- register sopno prevback = 0;
- register sopno prevfwd = 0;
- register sopno conc;
- register int first = 1; /* is this the first alternative? */
+ char c;
+ sopno prevback = 0;
+ sopno prevfwd = 0;
+ sopno conc;
+ int first = 1; /* is this the first alternative? */
for (;;)
{
}
/*
- - p_ere_exp - parse one subERE, an atom possibly followed by a repetition op
- == static void p_ere_exp(register struct parse *p);
+ * p_ere_exp - parse one subERE, an atom possibly followed by a repetition op
*/
static void
-p_ere_exp(p)
-register struct parse *p;
+p_ere_exp(struct parse * p)
{
- register char c;
- register sopno pos;
- register int count;
- register int count2;
- register sopno subno;
+ pg_wchar c;
+ sopno pos;
+ int count;
+ int count2;
+ sopno subno;
int wascaret = 0;
assert(MORE()); /* caller should have ensured this */
break;
case '{': /* okay as ordinary except if digit
* follows */
- REQUIRE(!MORE() || !isdigit(PEEK()), REG_BADRPT);
+ REQUIRE(!MORE() || !pg_isdigit(PEEK()), REG_BADRPT);
/* FALLTHROUGH */
default:
ordinary(p, c);
c = PEEK();
/* we call { a repetition if followed by a digit */
if (!(c == '*' || c == '+' || c == '?' ||
- (c == '{' && MORE2() && isdigit(PEEK2()))))
+ (c == '{' && MORE2() && pg_isdigit(PEEK2()))))
return; /* no repetition, we're done */
NEXT();
count = p_count(p);
if (EAT(','))
{
- if (isdigit(PEEK()))
+ if (pg_isdigit(PEEK()))
{
count2 = p_count(p);
REQUIRE(count <= count2, REG_BADBR);
return;
c = PEEK();
if (!(c == '*' || c == '+' || c == '?' ||
- (c == '{' && MORE2() && isdigit(PEEK2()))))
+ (c == '{' && MORE2() && pg_isdigit(PEEK2()))))
return;
SETERROR(REG_BADRPT);
}
/*
- - p_str - string (no metacharacters) "parser"
- == static void p_str(register struct parse *p);
+ * p_str - string (no metacharacters) "parser"
*/
static void
-p_str(p)
-register struct parse *p;
+p_str(struct parse * p)
{
REQUIRE(MORE(), REG_EMPTY);
while (MORE())
}
/*
- - p_bre - BRE parser top level, anchoring and concatenation
- == static void p_bre(register struct parse *p, register int end1, \
- == register int end2);
+ * p_bre - BRE parser top level, anchoring and concatenation
+ *
* Giving end1 as OUT essentially eliminates the end1/end2 check.
*
* This implementation is a bit of a kludge, in that a trailing $ is first
* The amount of lookahead needed to avoid this kludge is excessive.
*/
static void
-p_bre(p, end1, end2)
-register struct parse *p;
-register int end1; /* first terminating character */
-register int end2; /* second terminating character */
+p_bre(struct parse * p,
+ int end1, /* first terminating character */
+ int end2) /* second terminating character */
{
- register sopno start = HERE();
- register int first = 1; /* first subexpression? */
- register int wasdollar = 0;
+ sopno start = HERE();
+ int first = 1; /* first subexpression? */
+ int wasdollar = 0;
if (EAT('^'))
{
}
/*
- - p_simp_re - parse a simple RE, an atom possibly followed by a repetition
- == static int p_simp_re(register struct parse *p, int starordinary);
+ * p_simp_re - parse a simple RE, an atom possibly followed by a repetition
*/
static int /* was the simple RE an unbackslashed $? */
-p_simp_re(p, starordinary)
-register struct parse *p;
-int starordinary; /* is a leading * an ordinary character? */
+p_simp_re(struct parse * p,
+ int starordinary) /* is a leading * an ordinary character? */
{
- register int c;
- register int count;
- register int count2;
- register sopno pos;
- register int i;
- register sopno subno;
+ int c;
+ int count;
+ int count2;
+ sopno pos;
+ int i;
+ sopno subno;
-#define BACKSL (1<<CHAR_BIT)
+#define BACKSL (1<<24)
pos = HERE(); /* repetion op, if any, covers from here */
if (c == '\\')
{
REQUIRE(MORE(), REG_EESCAPE);
- c = BACKSL | (unsigned char) GETNEXT();
+ c = BACKSL | (pg_wchar) GETNEXT();
}
switch (c)
{
count = p_count(p);
if (EAT(','))
{
- if (MORE() && isdigit(PEEK()))
+ if (MORE() && pg_isdigit(PEEK()))
{
count2 = p_count(p);
REQUIRE(count <= count2, REG_BADBR);
}
}
else if (c == (unsigned char) '$') /* $ (but not \$) ends it */
- return (1);
+ return 1;
- return (0);
+ return 0;
}
/*
- - p_count - parse a repetition count
- == static int p_count(register struct parse *p);
+ * p_count - parse a repetition count
*/
static int /* the value */
-p_count(p)
-register struct parse *p;
+p_count(struct parse * p)
{
- register int count = 0;
- register int ndigits = 0;
+ int count = 0;
+ int ndigits = 0;
- while (MORE() && isdigit(PEEK()) && count <= DUPMAX)
+ while (MORE() && pg_isdigit(PEEK()) && count <= DUPMAX)
{
count = count * 10 + (GETNEXT() - '0');
ndigits++;
}
REQUIRE(ndigits > 0 && count <= DUPMAX, REG_BADBR);
- return (count);
+ return count;
}
/*
- - p_bracket - parse a bracketed character list
- == static void p_bracket(register struct parse *p);
+ * p_bracket - parse a bracketed character list
*
* Note a significant property of this code: if the allocset() did SETERROR,
* no set operations are done.
*/
static void
-p_bracket(p)
-register struct parse *p;
+p_bracket(struct parse * p)
{
- register cset *cs = allocset(p);
- register int invert = 0;
+ cset *cs = allocset(p);
+ int invert = 0;
+
+ pg_wchar sp1[] = {'[', ':', '<', ':', ']', ']'};
+ pg_wchar sp2[] = {'[', ':', '>', ':', ']', ']'};
/* Dept of Truly Sickening Special-Case Kludges */
- if (p->next + 5 < p->end && strncmp(p->next, "[:<:]]", 6) == 0)
+ if (p->next + 5 < p->end && pg_wchar_strncmp(p->next, sp1, 6) == 0)
{
EMIT(OBOW, 0);
NEXTn(6);
return;
}
- if (p->next + 5 < p->end && strncmp(p->next, "[:>:]]", 6) == 0)
+ if (p->next + 5 < p->end && pg_wchar_strncmp(p->next, sp2, 6) == 0)
{
EMIT(OEOW, 0);
NEXTn(6);
if (p->g->cflags & REG_ICASE)
{
- register int i;
- register int ci;
+ int i;
+ int ci;
for (i = p->g->csetsize - 1; i >= 0; i--)
- if (CHIN(cs, i) && isalpha(i))
+ if (CHIN(cs, i) && pg_isalpha(i))
{
ci = othercase(i);
if (ci != i)
}
if (invert)
{
- register int i;
+ int i;
for (i = p->g->csetsize - 1; i >= 0; i--)
if (CHIN(cs, i))
}
/*
- - p_b_term - parse one term of a bracketed character list
- == static void p_b_term(register struct parse *p, register cset *cs);
+ * p_b_term - parse one term of a bracketed character list
*/
static void
-p_b_term(p, cs)
-register struct parse *p;
-register cset *cs;
+p_b_term(struct parse * p, cset *cs)
{
- register char c;
- register char start,
+ pg_wchar c;
+ pg_wchar start,
finish;
- register int i;
+ int i;
/* classify what we've got */
switch ((MORE()) ? PEEK() : '\0')
finish = start;
/* xxx what about signed chars here... */
REQUIRE(start <= finish, REG_ERANGE);
+
+ if (CHlc(start) != CHlc(finish))
+ SETERROR(REG_ERANGE);
+
for (i = start; i <= finish; i++)
CHadd(cs, i);
break;
}
/*
- - p_b_cclass - parse a character-class name and deal with it
- == static void p_b_cclass(register struct parse *p, register cset *cs);
+ * p_b_cclass - parse a character-class name and deal with it
*/
static void
-p_b_cclass(p, cs)
-register struct parse *p;
-register cset *cs;
+p_b_cclass(struct parse * p, cset *cs)
{
- register char *sp = p->next;
- register struct cclass *cp;
- register size_t len;
- register char *u;
- register char c;
+ pg_wchar *sp = p->next;
+ struct cclass *cp;
+ size_t len;
+ char *u;
+ unsigned char c;
- while (MORE() && isalpha(PEEK()))
+ while (MORE() && pg_isalpha(PEEK()))
NEXT();
len = p->next - sp;
+
for (cp = cclasses; cp->name != NULL; cp++)
- if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0')
+ if (pg_char_and_wchar_strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0')
break;
+
if (cp->name == NULL)
{
/* oops, didn't find it */
}
/*
- - p_b_eclass - parse an equivalence-class name and deal with it
- == static void p_b_eclass(register struct parse *p, register cset *cs);
+ * p_b_eclass - parse an equivalence-class name and deal with it
*
* This implementation is incomplete. xxx
*/
static void
-p_b_eclass(p, cs)
-register struct parse *p;
-register cset *cs;
+p_b_eclass(struct parse * p, cset *cs)
{
- register char c;
+ char c;
c = p_b_coll_elem(p, '=');
CHadd(cs, c);
}
/*
- - p_b_symbol - parse a character or [..]ed multicharacter collating symbol
- == static char p_b_symbol(register struct parse *p);
+ * p_b_symbol - parse a character or [..]ed multicharacter collating symbol
*/
-static char /* value of symbol */
-p_b_symbol(p)
-register struct parse *p;
+static pg_wchar /* value of symbol */
+p_b_symbol(struct parse * p)
{
- register char value;
+ pg_wchar value;
REQUIRE(MORE(), REG_EBRACK);
if (!EATTWO('[', '.'))
- return (GETNEXT());
+ return GETNEXT();
/* collating symbol */
value = p_b_coll_elem(p, '.');
REQUIRE(EATTWO('.', ']'), REG_ECOLLATE);
- return (value);
+ return value;
}
/*
- - p_b_coll_elem - parse a collating-element name and look it up
- == static char p_b_coll_elem(register struct parse *p, int endc);
+ * p_b_coll_elem - parse a collating-element name and look it up
*/
static char /* value of collating element */
-p_b_coll_elem(p, endc)
-register struct parse *p;
-int endc; /* name ended by endc,']' */
+p_b_coll_elem(struct parse * p, int endc)
{
- register char *sp = p->next;
- register struct cname *cp;
- register int len;
+ pg_wchar *sp = p->next;
+ struct cname *cp;
+ int len;
while (MORE() && !SEETWO(endc, ']'))
NEXT();
if (!MORE())
{
SETERROR(REG_EBRACK);
- return (0);
+ return 0;
}
len = p->next - sp;
+
for (cp = cnames; cp->name != NULL; cp++)
- if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0')
- return (cp->code); /* known name */
+ if (pg_char_and_wchar_strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0')
+ return cp->code; /* known name */
+
if (len == 1)
- return (*sp); /* single character */
+ return *sp; /* single character */
SETERROR(REG_ECOLLATE); /* neither */
- return (0);
+ return 0;
}
/*
- - othercase - return the case counterpart of an alphabetic
- == static char othercase(int ch);
+ * othercase - return the case counterpart of an alphabetic
*/
-static char /* if no counterpart, return ch */
-othercase(ch)
-int ch;
+static unsigned char /* if no counterpart, return ch */
+othercase(int ch)
{
- assert(isalpha(ch));
- if (isupper(ch))
- return (tolower(ch));
- else if (islower(ch))
- return (toupper(ch));
+ assert(pg_isalpha(ch));
+ if (pg_isupper(ch))
+ return (unsigned char) tolower((unsigned char) ch);
+ else if (pg_islower(ch))
+ return (unsigned char) toupper((unsigned char) ch);
else
/* peculiar, but could happen */
- return (ch);
+ return (unsigned char) ch;
}
/*
- - bothcases - emit a dualcase version of a two-case character
- == static void bothcases(register struct parse *p, int ch);
+ * bothcases - emit a dualcase version of a two-case character
*
* Boy, is this implementation ever a kludge...
*/
static void
-bothcases(p, ch)
-register struct parse *p;
-int ch;
+bothcases(struct parse * p, int ch)
{
- register char *oldnext = p->next;
- register char *oldend = p->end;
- char bracket[3];
+ pg_wchar *oldnext = p->next;
+ pg_wchar *oldend = p->end;
+ pg_wchar bracket[3];
- assert(othercase(ch) != ch);/* p_bracket() would recurse */
+ assert(othercase(ch) != ch); /* p_bracket() would recurse */
p->next = bracket;
p->end = bracket + 2;
bracket[0] = ch;
}
/*
- - ordinary - emit an ordinary character
- == static void ordinary(register struct parse *p, register int ch);
+ * ordinary - emit an ordinary character
*/
static void
-ordinary(p, ch)
-register struct parse *p;
-register int ch;
+ordinary(struct parse * p, int ch)
{
- register cat_t *cap = p->g->categories;
+ cat_t *cap = p->g->categories;
- if ((p->g->cflags & REG_ICASE) && isalpha(ch) && othercase(ch) != ch)
+ if ((p->g->cflags & REG_ICASE) && pg_isalpha(ch) && othercase(ch) != ch)
bothcases(p, ch);
else
{
- EMIT(OCHAR, (unsigned char) ch);
- if (cap[ch] == 0)
+ EMIT(OCHAR, (pg_wchar) ch);
+ if (ch >= CHAR_MIN && ch <= CHAR_MAX && cap[ch] == 0)
cap[ch] = p->g->ncategories++;
}
}
/*
- - nonnewline - emit REG_NEWLINE version of OANY
- == static void nonnewline(register struct parse *p);
+ * nonnewline - emit REG_NEWLINE version of OANY
*
* Boy, is this implementation ever a kludge...
*/
static void
-nonnewline(p)
-register struct parse *p;
+nonnewline(struct parse * p)
{
- register char *oldnext = p->next;
- register char *oldend = p->end;
- char bracket[4];
+ pg_wchar *oldnext = p->next;
+ pg_wchar *oldend = p->end;
+ pg_wchar bracket[4];
p->next = bracket;
p->end = bracket + 3;
}
/*
- - repeat - generate code for a bounded repetition, recursively if needed
- == static void repeat(register struct parse *p, sopno start, int from, int to);
+ * repeat - generate code for a bounded repetition, recursively if needed
*/
static void
-repeat(p, start, from, to)
-register struct parse *p;
-sopno start; /* operand from here to end of strip */
-int from; /* repeated from this number */
-int to; /* to this number of times (maybe
+repeat(struct parse * p,
+ sopno start, /* operand from here to end of strip */
+ int from, /* repeated from this number */
+ int to) /* to this number of times (maybe
* INFINITY) */
{
- register sopno finish = HERE();
+ sopno finish = HERE();
#define N 2
#define INF 3
#define REP(f, t) ((f)*8 + (t))
#define MAP(n) (((n) <= 1) ? (n) : ((n) == INFINITY) ? INF : N)
- register sopno copy;
+ sopno copy;
if (p->error != 0) /* head off possible runaway recursion */
return;
case REP(0, N): /* as x{1,n}? */
case REP(0, INF): /* as x{1,}? */
/* KLUDGE: emit y? as (y|) until subtle bug gets fixed */
- INSERT(OCH_, start);/* offset is wrong... */
+ INSERT(OCH_, start); /* offset is wrong... */
repeat(p, start + 1, 1, to);
ASTERN(OOR1, start);
AHEAD(start); /* ... fix it */
}
/*
- - seterr - set an error condition
- == static int seterr(register struct parse *p, int e);
+ * seterr - set an error condition
*/
static int /* useless but makes type checking happy */
-seterr(p, e)
-register struct parse *p;
-int e;
+seterr(struct parse * p, int e)
{
if (p->error == 0) /* keep earliest error condition */
p->error = e;
p->next = nuls; /* try to bring things to a halt */
p->end = nuls;
- return (0); /* make the return value well-defined */
+ return 0; /* make the return value well-defined */
}
/*
- - allocset - allocate a set of characters for []
- == static cset *allocset(register struct parse *p);
+ * allocset - allocate a set of characters for []
*/
static cset *
-allocset(p)
-register struct parse *p;
+allocset(struct parse * p)
{
- register int no = p->g->ncsets++;
- register size_t nc;
- register size_t nbytes;
- register cset *cs;
- register size_t css = (size_t) p->g->csetsize;
- register int i;
+ int no = p->g->ncsets++;
+ size_t nc;
+ size_t nbytes;
+ cset *cs;
+ size_t css = (size_t) p->g->csetsize;
+ int i;
if (no >= p->ncsalloc)
{ /* need another column of space */
cs->smultis = 0;
cs->multis = NULL;
- return (cs);
+ return cs;
}
/*
- - freeset - free a now-unused set
- == static void freeset(register struct parse *p, register cset *cs);
+ * freeset - free a now-unused set
*/
static void
-freeset(p, cs)
-register struct parse *p;
-register cset *cs;
+freeset(struct parse * p, cset *cs)
{
- register int i;
- register cset *top = &p->g->sets[p->g->ncsets];
- register size_t css = (size_t) p->g->csetsize;
+ int i;
+ cset *top = &p->g->sets[p->g->ncsets];
+ size_t css = (size_t) p->g->csetsize;
for (i = 0; i < css; i++)
CHsub(cs, i);
}
/*
- - freezeset - final processing on a set of characters
- == static int freezeset(register struct parse *p, register cset *cs);
+ * freezeset - final processing on a set of characters
*
* The main task here is merging identical sets. This is usually a waste
* of time (although the hash code minimizes the overhead), but can win
* the same value!
*/
static int /* set number */
-freezeset(p, cs)
-register struct parse *p;
-register cset *cs;
+freezeset(struct parse * p, cset *cs)
{
- register uch h = cs->hash;
- register int i;
- register cset *top = &p->g->sets[p->g->ncsets];
- register cset *cs2;
- register size_t css = (size_t) p->g->csetsize;
+ uch h = cs->hash;
+ int i;
+ cset *top = &p->g->sets[p->g->ncsets];
+ cset *cs2;
+ size_t css = (size_t) p->g->csetsize;
/* look for an earlier one which is the same */
for (cs2 = &p->g->sets[0]; cs2 < top; cs2++)
cs = cs2;
}
- return ((int) (cs - p->g->sets));
+ return (int) (cs - p->g->sets);
}
/*
- - firstch - return first character in a set (which must have at least one)
- == static int firstch(register struct parse *p, register cset *cs);
+ * firstch - return first character in a set (which must have at least one)
*/
static int /* character; there is no "none" value */
-firstch(p, cs)
-register struct parse *p;
-register cset *cs;
+firstch(struct parse * p, cset *cs)
{
- register int i;
- register size_t css = (size_t) p->g->csetsize;
+ int i;
+ size_t css = (size_t) p->g->csetsize;
for (i = 0; i < css; i++)
if (CHIN(cs, i))
- return ((char) i);
+ return i;
assert(never);
- return (0); /* arbitrary */
+ return 0; /* arbitrary */
}
/*
- - nch - number of characters in a set
- == static int nch(register struct parse *p, register cset *cs);
+ * nch - number of characters in a set
*/
static int
-nch(p, cs)
-register struct parse *p;
-register cset *cs;
+nch(struct parse * p, cset *cs)
{
- register int i;
- register size_t css = (size_t) p->g->csetsize;
- register int n = 0;
+ int i;
+ size_t css = (size_t) p->g->csetsize;
+ int n = 0;
for (i = 0; i < css; i++)
if (CHIN(cs, i))
n++;
- return (n);
+ return n;
}
/*
- - mcadd - add a collating element to a cset
- == static void mcadd(register struct parse *p, register cset *cs, \
- == register char *cp);
+ * mcadd - add a collating element to a cset
*/
static void
-mcadd(p, cs, cp)
-register struct parse *p;
-register cset *cs;
-register char *cp;
+mcadd(struct parse * p, cset *cs, char *cp)
{
- register size_t oldend = cs->smultis;
+ size_t oldend = cs->smultis;
cs->smultis += strlen(cp) + 1;
if (cs->multis == NULL)
}
/*
- - mcsub - subtract a collating element from a cset
- == static void mcsub(register cset *cs, register char *cp);
- */
-/*
-static void
-mcsub(cs, cp)
-register cset *cs;
-register char *cp;
-{
- register char *fp = mcfind(cs, cp);
- register size_t len = strlen(fp);
-
- assert(fp != NULL);
- memmove(fp, fp + len + 1,
- cs->smultis - (fp + len + 1 - cs->multis));
- cs->smultis -= len;
-
- if (cs->smultis == 0) {
- free(cs->multis);
- cs->multis = NULL;
- return;
- }
-
- cs->multis = realloc(cs->multis, cs->smultis);
- assert(cs->multis != NULL);
-}
-*/
-
-/*
- - mcin - is a collating element in a cset?
- == static int mcin(register cset *cs, register char *cp);
- */
-/*
-static int
-mcin(cs, cp)
-register cset *cs;
-register char *cp;
-{
- return(mcfind(cs, cp) != NULL);
-}
-*/
-
-/*
- - mcfind - find a collating element in a cset
- == static char *mcfind(register cset *cs, register char *cp);
- */
-/*
-static char *
-mcfind(cs, cp)
-register cset *cs;
-register char *cp;
-{
- register char *p;
-
- if (cs->multis == NULL)
- return(NULL);
- for (p = cs->multis; *p != '\0'; p += strlen(p) + 1)
- if (strcmp(cp, p) == 0)
- return(p);
- return(NULL);
-}
-*/
-/*
- - mcinvert - invert the list of collating elements in a cset
- == static void mcinvert(register struct parse *p, register cset *cs);
+ * mcinvert - invert the list of collating elements in a cset
*
* This would have to know the set of possibilities. Implementation
* is deferred.
*/
static void
-mcinvert(p, cs)
-register struct parse *p;
-register cset *cs;
+mcinvert(struct parse * p, cset *cs)
{
assert(cs->multis == NULL); /* xxx */
}
/*
- - mccase - add case counterparts of the list of collating elements in a cset
- == static void mccase(register struct parse *p, register cset *cs);
+ * mccase - add case counterparts of the list of collating elements in a cset
*
* This would have to know the set of possibilities. Implementation
* is deferred.
*/
static void
-mccase(p, cs)
-register struct parse *p;
-register cset *cs;
+mccase(struct parse * p, cset *cs)
{
assert(cs->multis == NULL); /* xxx */
}
/*
- - isinsets - is this character in any sets?
- == static int isinsets(register struct re_guts *g, int c);
+ * isinsets - is this character in any sets?
*/
static int /* predicate */
-isinsets(g, c)
-register struct re_guts *g;
-int c;
+isinsets(struct re_guts * g, int c)
{
- register uch *col;
- register int i;
- register int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT;
- register unsigned uc = (unsigned char) c;
+ uch *col;
+ int i;
+ int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT;
+ unsigned uc = (unsigned char) c;
for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize)
if (col[uc] != 0)
- return (1);
- return (0);
+ return 1;
+ return 0;
}
/*
- - samesets - are these two characters in exactly the same sets?
- == static int samesets(register struct re_guts *g, int c1, int c2);
+ * samesets - are these two characters in exactly the same sets?
*/
static int /* predicate */
-samesets(g, c1, c2)
-register struct re_guts *g;
-int c1;
-int c2;
+samesets(struct re_guts * g, int c1, int c2)
{
- register uch *col;
- register int i;
- register int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT;
- register unsigned uc1 = (unsigned char) c1;
- register unsigned uc2 = (unsigned char) c2;
+ uch *col;
+ int i;
+ int ncols = (g->ncsets + (CHAR_BIT - 1)) / CHAR_BIT;
+ unsigned uc1 = (unsigned char) c1;
+ unsigned uc2 = (unsigned char) c2;
for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize)
if (col[uc1] != col[uc2])
- return (0);
- return (1);
+ return 0;
+ return 1;
}
/*
- - categorize - sort out character categories
- == static void categorize(struct parse *p, register struct re_guts *g);
+ * categorize - sort out character categories
*/
static void
-categorize(p, g)
-struct parse *p;
-register struct re_guts *g;
+categorize(struct parse * p, struct re_guts * g)
{
- register cat_t *cats = g->categories;
- register int c;
- register int c2;
- register cat_t cat;
+ cat_t *cats = g->categories;
+ int c;
+ int c2;
+ cat_t cat;
/* avoid making error situations worse */
if (p->error != 0)
}
/*
- - dupl - emit a duplicate of a bunch of sops
- == static sopno dupl(register struct parse *p, sopno start, sopno finish);
+ * dupl - emit a duplicate of a bunch of sops
*/
-static sopno /* start of duplicate */
-dupl(p, start, finish)
-register struct parse *p;
-sopno start; /* from here */
-sopno finish; /* to this less one */
+static sopno /* start of duplicate */
+dupl(struct parse * p,
+ sopno start, /* from here */
+ sopno finish) /* to this less one */
{
- register sopno ret = HERE();
- register sopno len = finish - start;
+ sopno ret = HERE();
+ sopno len = finish - start;
assert(finish >= start);
if (len == 0)
- return (ret);
+ return ret;
enlarge(p, p->ssize + len); /* this many unexpected additions */
assert(p->ssize >= p->slen + len);
memcpy((char *) (p->strip + p->slen),
(char *) (p->strip + start), (size_t) len * sizeof(sop));
p->slen += len;
- return (ret);
+ return ret;
}
/*
- - doemit - emit a strip operator
- == static void doemit(register struct parse *p, sop op, size_t opnd);
+ * doemit - emit a strip operator
*
* It might seem better to implement this as a macro with a function as
* hard-case backup, but it's just too big and messy unless there are
* some changes to the data structures. Maybe later.
*/
static void
-doemit(p, op, opnd)
-register struct parse *p;
-sop op;
-size_t opnd;
+doemit(struct parse * p, sop op, size_t opnd)
{
/* avoid making error situations worse */
if (p->error != 0)
}
/*
- - doinsert - insert a sop into the strip
- == static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos);
+ * doinsert - insert a sop into the strip
*/
static void
-doinsert(p, op, opnd, pos)
-register struct parse *p;
-sop op;
-size_t opnd;
-sopno pos;
+doinsert(struct parse * p, sop op, size_t opnd, sopno pos)
{
- register sopno sn;
- register sop s;
- register int i;
+ sopno sn;
+ sop s;
+ int i;
/* avoid making error situations worse */
if (p->error != 0)
for (i = 1; i < NPAREN; i++)
{
if (p->pbegin[i] >= pos)
- {
p->pbegin[i]++;
- }
if (p->pend[i] >= pos)
- {
p->pend[i]++;
- }
}
memmove((char *) &p->strip[pos + 1], (char *) &p->strip[pos],
}
/*
- - dofwd - complete a forward reference
- == static void dofwd(register struct parse *p, sopno pos, sop value);
+ * dofwd - complete a forward reference
*/
static void
-dofwd(p, pos, value)
-register struct parse *p;
-register sopno pos;
-sop value;
+dofwd(struct parse * p, sopno pos, sop value)
{
/* avoid making error situations worse */
if (p->error != 0)
}
/*
- - enlarge - enlarge the strip
- == static void enlarge(register struct parse *p, sopno size);
+ * enlarge - enlarge the strip
*/
static void
-enlarge(p, size)
-register struct parse *p;
-register sopno size;
+enlarge(struct parse * p, sopno size)
{
- register sop *sp;
+ sop *sp;
if (p->ssize >= size)
return;
}
/*
- - stripsnug - compact the strip
- == static void stripsnug(register struct parse *p, register struct re_guts *g);
+ * stripsnug - compact the strip
*/
static void
-stripsnug(p, g)
-register struct parse *p;
-register struct re_guts *g;
+stripsnug(struct parse * p, struct re_guts * g)
{
g->nstates = p->slen;
g->strip = (sop *) realloc((char *) p->strip, p->slen * sizeof(sop));
}
/*
- - findmust - fill in must and mlen with longest mandatory literal string
- == static void findmust(register struct parse *p, register struct re_guts *g);
+ * findmust - fill in must and mlen with longest mandatory literal string
*
* This algorithm could do fancy things like analyzing the operands of |
* for common subsequences. Someday. This code is simple and finds most
* Note that must and mlen got initialized during setup.
*/
static void
-findmust(p, g)
-struct parse *p;
-register struct re_guts *g;
+findmust(struct parse * p, struct re_guts * g)
{
- register sop *scan;
+ sop *scan;
sop *start = 0;
- register sop *newstart = 0;
- register sopno newlen;
- register sop s;
- register char *cp;
- register sopno i;
+ sop *newstart = 0;
+ sopno newlen;
+ sop s;
+ pg_wchar *cp;
+ sopno i;
/* avoid making error situations worse */
if (p->error != 0)
switch (OP(s))
{
case OCHAR: /* sequence member */
- if (newlen == 0)/* new sequence */
+ if (newlen == 0) /* new sequence */
newstart = scan - 1;
newlen++;
break;
return;
/* turn it into a character string */
- g->must = malloc((size_t) g->mlen + 1);
+ g->must = (pg_wchar *) malloc((size_t) (g->mlen + 1) * sizeof(pg_wchar));
if (g->must == NULL)
{ /* argh; just forget it */
g->mlen = 0;
while (OP(s = *scan++) != OCHAR)
continue;
assert(cp < g->must + g->mlen);
- *cp++ = (char) OPND(s);
+ *cp++ = (pg_wchar) OPND(s);
}
assert(cp == g->must + g->mlen);
*cp++ = '\0'; /* just on general principles */
}
/*
- - pluscount - count + nesting
- == static sopno pluscount(register struct parse *p, register struct re_guts *g);
+ * pluscount - count + nesting
*/
-static sopno /* nesting depth */
-pluscount(p, g)
-struct parse *p;
-register struct re_guts *g;
+static sopno /* nesting depth */
+pluscount(struct parse * p, struct re_guts * g)
{
- register sop *scan;
- register sop s;
- register sopno plusnest = 0;
- register sopno maxnest = 0;
+ sop *scan;
+ sop s;
+ sopno plusnest = 0;
+ sopno maxnest = 0;
if (p->error != 0)
- return (0); /* there may not be an OEND */
+ return 0; /* there may not be an OEND */
scan = g->strip + 1;
do
} while (OP(s) != OEND);
if (plusnest != 0)
g->iflags |= BAD;
- return (maxnest);
+ return maxnest;
+}
+
+/*
+ * some ctype functions with non-ascii-char guard
+ */
+static int
+pg_isdigit(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isdigit((unsigned char) c));
+}
+
+static int
+pg_isalpha(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isalpha((unsigned char) c));
+}
+
+static int
+pg_isalnum(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isalnum((unsigned char) c));
+}
+
+static int
+pg_isupper(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isupper((unsigned char) c));
+}
+
+static int
+pg_islower(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && islower((unsigned char) c));
+}
+
+static int
+pg_iscntrl(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && iscntrl((unsigned char) c));
+}
+
+static int
+pg_isgraph(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isgraph((unsigned char) c));
+}
+
+static int
+pg_isprint(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && isprint((unsigned char) c));
+}
+
+static int
+pg_ispunct(int c)
+{
+ return (c >= 0 && c <= UCHAR_MAX && ispunct((unsigned char) c));
+}
+
+static struct cclass *
+cclass_init(void)
+{
+ static struct cclass cclasses_C[] = {
+ {"alnum", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789", ""},
+ {"alpha", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", ""},
+ {"blank", " \t", ""},
+ {"cntrl", "\007\b\t\n\v\f\r\1\2\3\4\5\6\16\17\20\21\22\23\24\25\26\27\30\31\32\33\34\35\36\37\177", ""},
+ {"digit", "0123456789", ""},
+ {"graph", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", ""},
+ {"lower", "abcdefghijklmnopqrstuvwxyz", ""},
+ {"print", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ ", ""},
+ {"punct", "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", ""},
+ {"space", "\t\n\v\f\r ", ""},
+ {"upper", "ABCDEFGHIJKLMNOPQRSTUVWXYZ", ""},
+ {"xdigit", "0123456789ABCDEFabcdef", ""},
+ {NULL, NULL, ""}
+ };
+ struct cclass *cp = NULL;
+ struct cclass *classes = NULL;
+ struct cclass_factory
+ {
+ char *name;
+ int (*func) (int);
+ char *chars;
+ } cclass_factories[] =
+ {
+ {
+ "alnum", pg_isalnum, NULL
+ },
+ {
+ "alpha", pg_isalpha, NULL
+ },
+ {
+ "blank", NULL, " \t"
+ },
+ {
+ "cntrl", pg_iscntrl, NULL
+ },
+ {
+ "digit", NULL, "0123456789"
+ },
+ {
+ "graph", pg_isgraph, NULL
+ },
+ {
+ "lower", pg_islower, NULL
+ },
+ {
+ "print", pg_isprint, NULL
+ },
+ {
+ "punct", pg_ispunct, NULL
+ },
+ {
+ "space", NULL, "\t\n\v\f\r "
+ },
+ {
+ "upper", pg_isupper, NULL
+ },
+ {
+ "xdigit", NULL, "0123456789ABCDEFabcdef"
+ },
+ {
+ NULL, NULL, NULL
+ }
+ };
+ struct cclass_factory *cf = NULL;
+
+ if (strcmp(setlocale(LC_CTYPE, NULL), "C") == 0)
+ return cclasses_C;
+
+ classes = malloc(sizeof(struct cclass) * (sizeof(cclass_factories) / sizeof(struct cclass_factory)));
+ if (classes == NULL)
+ elog(ERROR, "cclass_init: out of memory");
+
+ cp = classes;
+ for (cf = cclass_factories; cf->name != NULL; cf++)
+ {
+ cp->name = strdup(cf->name);
+ if (cf->chars)
+ cp->chars = strdup(cf->chars);
+ else
+ {
+ int x = 0,
+ y = 0;
+
+ cp->chars = malloc(sizeof(char) * 256);
+ if (cp->chars == NULL)
+ elog(ERROR, "cclass_init: out of memory");
+ for (x = 0; x < 256; x++)
+ {
+ if ((cf->func) (x))
+ *(cp->chars + y++) = x;
+ }
+ *(cp->chars + y) = '\0';
+ }
+ cp->multis = "";
+ cp++;
+ }
+ cp->name = cp->chars = NULL;
+ cp->multis = "";
+
+ return classes;
}
OSDN Git Service
