/* Extended regular expression matching and search library, version
- 0.12. (Implements POSIX draft P10003.2/D11.2, except for
+ 0.12. (Implements POSIX draft P1003.2/D11.2, except for some of the
internationalization features.)
Copyright (C) 1993,94,95,96,97,98,2000 Free Software Foundation, Inc.
*/
/* AIX requires this to be the first thing in the file. */
-#if defined (_AIX) && !defined (REGEX_MALLOC)
+#if defined _AIX && !defined REGEX_MALLOC
#pragma alloca
#endif
#undef _GNU_SOURCE
#define _GNU_SOURCE
-#ifdef emacs
-/* Converts the pointer to the char to BEG-based offset from the start. */
-#define PTR_TO_OFFSET(d) POS_AS_IN_BUFFER (POINTER_TO_OFFSET (d))
-#define POS_AS_IN_BUFFER(p) ((p) + (NILP (re_match_object) || BUFFERP (re_match_object)))
-#endif
-
#ifdef HAVE_CONFIG_H
-#include <config.h>
+# include <config.h>
#endif
/* We need this for `regex.h', and perhaps for the Emacs include files. */
#include <sys/types.h>
/* This is for other GNU distributions with internationalized messages. */
-#if HAVE_LIBINTL_H || defined (_LIBC)
+#if HAVE_LIBINTL_H || defined _LIBC
# include <libintl.h>
#else
# define gettext(msgid) (msgid)
#ifndef gettext_noop
/* This define is so xgettext can find the internationalizable
strings. */
-#define gettext_noop(String) String
+# define gettext_noop(String) String
#endif
/* The `emacs' switch turns on certain matching commands
that make sense only in Emacs. */
#ifdef emacs
-#include "lisp.h"
-#include "buffer.h"
+# include "lisp.h"
+# include "buffer.h"
/* Make syntax table lookup grant data in gl_state. */
-#define SYNTAX_ENTRY_VIA_PROPERTY
+# define SYNTAX_ENTRY_VIA_PROPERTY
-#include "syntax.h"
-#include "charset.h"
-#include "category.h"
+# include "syntax.h"
+# include "charset.h"
+# include "category.h"
-#define malloc xmalloc
-#define realloc xrealloc
-#define free xfree
+# define malloc xmalloc
+# define realloc xrealloc
+# define free xfree
-#define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
-#define RE_STRING_CHAR(p, s) \
+/* Converts the pointer to the char to BEG-based offset from the start. */
+# define PTR_TO_OFFSET(d) POS_AS_IN_BUFFER (POINTER_TO_OFFSET (d))
+# define POS_AS_IN_BUFFER(p) ((p) + (NILP (re_match_object) || BUFFERP (re_match_object)))
+
+# define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
+# define RE_STRING_CHAR(p, s) \
(multibyte ? (STRING_CHAR (p, s)) : (*(p)))
-#define RE_STRING_CHAR_AND_LENGTH(p, s, len) \
+# define RE_STRING_CHAR_AND_LENGTH(p, s, len) \
(multibyte ? (STRING_CHAR_AND_LENGTH (p, s, len)) : ((len) = 1, *(p)))
/* Set C a (possibly multibyte) character before P. P points into a
string which is the virtual concatenation of STR1 (which ends at
END1) or STR2 (which ends at END2). */
-#define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
+# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
do { \
if (multibyte) \
{ \
- re_char *dtemp = (p) == (str2) ? (end1) : (p); \
- re_char *dlimit = ((p) > (str2) && (p) <= (end2)) ? (str2) : (str1); \
- while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp)); \
- c = STRING_CHAR (dtemp, (p) - dtemp); \
+ re_char *dtemp = (p) == (str2) ? (end1) : (p); \
+ re_char *dlimit = ((p) > (str2) && (p) <= (end2)) ? (str2) : (str1); \
+ while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp)); \
+ c = STRING_CHAR (dtemp, (p) - dtemp); \
} \
else \
(c = ((p) == (str2) ? (end1) : (p))[-1]); \
/* If we are not linking with Emacs proper,
we can't use the relocating allocator
even if config.h says that we can. */
-#undef REL_ALLOC
+# undef REL_ALLOC
-#if defined (STDC_HEADERS) || defined (_LIBC)
-#include <stdlib.h>
-#else
+# if defined STDC_HEADERS || defined _LIBC
+# include <stdlib.h>
+# else
char *malloc ();
char *realloc ();
-#endif
+# endif
/* When used in Emacs's lib-src, we need to get bzero and bcopy somehow.
If nothing else has been done, use the method below. */
-#ifdef INHIBIT_STRING_HEADER
-#if !(defined (HAVE_BZERO) && defined (HAVE_BCOPY))
-#if !defined (bzero) && !defined (bcopy)
-#undef INHIBIT_STRING_HEADER
-#endif
-#endif
-#endif
+# ifdef INHIBIT_STRING_HEADER
+# if !(defined HAVE_BZERO && defined HAVE_BCOPY)
+# if !defined bzero && !defined bcopy
+# undef INHIBIT_STRING_HEADER
+# endif
+# endif
+# endif
/* This is the normal way of making sure we have a bcopy and a bzero.
This is used in most programs--a few other programs avoid this
by defining INHIBIT_STRING_HEADER. */
-#ifndef INHIBIT_STRING_HEADER
-#if defined (HAVE_STRING_H) || defined (STDC_HEADERS) || defined (_LIBC)
-#include <string.h>
-#ifndef bcmp
-#define bcmp(s1, s2, n) memcmp ((s1), (s2), (n))
-#endif
-#ifndef bcopy
-#define bcopy(s, d, n) memcpy ((d), (s), (n))
-#endif
-#ifndef bzero
-#define bzero(s, n) memset ((s), 0, (n))
-#endif
-#else
-#include <strings.h>
-#endif
-#endif
+# ifndef INHIBIT_STRING_HEADER
+# if defined HAVE_STRING_H || defined STDC_HEADERS || defined _LIBC
+# include <string.h>
+# ifndef bcmp
+# define bcmp(s1, s2, n) memcmp ((s1), (s2), (n))
+# endif
+# ifndef bcopy
+# define bcopy(s, d, n) memcpy ((d), (s), (n))
+# endif
+# ifndef bzero
+# define bzero(s, n) memset ((s), 0, (n))
+# endif
+# else
+# include <strings.h>
+# endif
+# endif
/* Define the syntax stuff for \<, \>, etc. */
/* Sword must be nonzero for the wordchar pattern commands in re_match_2. */
enum syntaxcode { Swhitespace = 0, Sword = 1 };
-#ifdef SWITCH_ENUM_BUG
-#define SWITCH_ENUM_CAST(x) ((int)(x))
-#else
-#define SWITCH_ENUM_CAST(x) (x)
-#endif
-
-#ifdef SYNTAX_TABLE
-
-extern char *re_syntax_table;
-
-#else /* not SYNTAX_TABLE */
+# ifdef SWITCH_ENUM_BUG
+# define SWITCH_ENUM_CAST(x) ((int)(x))
+# else
+# define SWITCH_ENUM_CAST(x) (x)
+# endif
-/* How many characters in the character set. */
-#define CHAR_SET_SIZE 256
-
-static char re_syntax_table[CHAR_SET_SIZE];
-
-static void
-init_syntax_once ()
-{
- register int c;
- static int done = 0;
-
- if (done)
- return;
-
- bzero (re_syntax_table, sizeof re_syntax_table);
-
- for (c = 'a'; c <= 'z'; c++)
- re_syntax_table[c] = Sword;
-
- for (c = 'A'; c <= 'Z'; c++)
- re_syntax_table[c] = Sword;
-
- for (c = '0'; c <= '9'; c++)
- re_syntax_table[c] = Sword;
-
- re_syntax_table['_'] = Sword;
-
- done = 1;
-}
-
-#endif /* not SYNTAX_TABLE */
-
-#define SYNTAX(c) re_syntax_table[c]
+# define SYNTAX(c) re_syntax_table[c]
/* Dummy macros for non-Emacs environments. */
-#define BASE_LEADING_CODE_P(c) (0)
-#define CHAR_CHARSET(c) 0
-#define CHARSET_LEADING_CODE_BASE(c) 0
-#define MAX_MULTIBYTE_LENGTH 1
-#define RE_MULTIBYTE_P(x) 0
-#define WORD_BOUNDARY_P(c1, c2) (0)
-#define CHAR_HEAD_P(p) (1)
-#define SINGLE_BYTE_CHAR_P(c) (1)
-#define SAME_CHARSET_P(c1, c2) (1)
-#define MULTIBYTE_FORM_LENGTH(p, s) (1)
-#define STRING_CHAR(p, s) (*(p))
-#define RE_STRING_CHAR STRING_CHAR
-#define CHAR_STRING(c, s) (*(s) = (c), 1)
-#define STRING_CHAR_AND_LENGTH(p, s, actual_len) ((actual_len) = 1, *(p))
-#define RE_STRING_CHAR_AND_LENGTH STRING_CHAR_AND_LENGTH
-#define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
+# define BASE_LEADING_CODE_P(c) (0)
+# define CHAR_CHARSET(c) 0
+# define CHARSET_LEADING_CODE_BASE(c) 0
+# define MAX_MULTIBYTE_LENGTH 1
+# define RE_MULTIBYTE_P(x) 0
+# define WORD_BOUNDARY_P(c1, c2) (0)
+# define CHAR_HEAD_P(p) (1)
+# define SINGLE_BYTE_CHAR_P(c) (1)
+# define SAME_CHARSET_P(c1, c2) (1)
+# define MULTIBYTE_FORM_LENGTH(p, s) (1)
+# define STRING_CHAR(p, s) (*(p))
+# define RE_STRING_CHAR STRING_CHAR
+# define CHAR_STRING(c, s) (*(s) = (c), 1)
+# define STRING_CHAR_AND_LENGTH(p, s, actual_len) ((actual_len) = 1, *(p))
+# define RE_STRING_CHAR_AND_LENGTH STRING_CHAR_AND_LENGTH
+# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
(c = ((p) == (str2) ? *((end1) - 1) : *((p) - 1)))
-#define MAKE_CHAR(charset, c1, c2) (c1)
+# define MAKE_CHAR(charset, c1, c2) (c1)
#endif /* not emacs */
#ifndef RE_TRANSLATE
-#define RE_TRANSLATE(TBL, C) ((unsigned char)(TBL)[C])
-#define RE_TRANSLATE_P(TBL) (TBL)
+# define RE_TRANSLATE(TBL, C) ((unsigned char)(TBL)[C])
+# define RE_TRANSLATE_P(TBL) (TBL)
#endif
\f
/* Get the interface, including the syntax bits. */
#ifdef emacs
/* 1 if C is an ASCII character. */
-#define IS_REAL_ASCII(c) ((c) < 0200)
+# define IS_REAL_ASCII(c) ((c) < 0200)
/* 1 if C is a unibyte character. */
-#define ISUNIBYTE(c) (SINGLE_BYTE_CHAR_P ((c)))
+# define ISUNIBYTE(c) (SINGLE_BYTE_CHAR_P ((c)))
/* The Emacs definitions should not be directly affected by locales. */
/* In Emacs, these are only used for single-byte characters. */
-#define ISDIGIT(c) ((c) >= '0' && (c) <= '9')
-#define ISCNTRL(c) ((c) < ' ')
-#define ISXDIGIT(c) (((c) >= '0' && (c) <= '9') \
+# define ISDIGIT(c) ((c) >= '0' && (c) <= '9')
+# define ISCNTRL(c) ((c) < ' ')
+# define ISXDIGIT(c) (((c) >= '0' && (c) <= '9') \
|| ((c) >= 'a' && (c) <= 'f') \
|| ((c) >= 'A' && (c) <= 'F'))
/* This is only used for single-byte characters. */
-#define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
/* The rest must handle multibyte characters. */
-#define ISGRAPH(c) (SINGLE_BYTE_CHAR_P (c) \
+# define ISGRAPH(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) > ' ' && !((c) >= 0177 && (c) <= 0237) \
: 1)
-#define ISPRINT(c) (SINGLE_BYTE_CHAR_P (c) \
+# define ISPRINT(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) >= ' ' && !((c) >= 0177 && (c) <= 0237) \
: 1)
-#define ISALNUM(c) (IS_REAL_ASCII (c) \
+# define ISALNUM(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z') \
|| ((c) >= '0' && (c) <= '9')) \
: SYNTAX (c) == Sword)
-#define ISALPHA(c) (IS_REAL_ASCII (c) \
+# define ISALPHA(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z')) \
: SYNTAX (c) == Sword)
-#define ISLOWER(c) (LOWERCASEP (c))
+# define ISLOWER(c) (LOWERCASEP (c))
-#define ISPUNCT(c) (IS_REAL_ASCII (c) \
+# define ISPUNCT(c) (IS_REAL_ASCII (c) \
? ((c) > ' ' && (c) < 0177 \
&& !(((c) >= 'a' && (c) <= 'z') \
- || ((c) >= 'A' && (c) <= 'Z') \
- || ((c) >= '0' && (c) <= '9'))) \
+ || ((c) >= 'A' && (c) <= 'Z') \
+ || ((c) >= '0' && (c) <= '9'))) \
: SYNTAX (c) != Sword)
-#define ISSPACE(c) (SYNTAX (c) == Swhitespace)
+# define ISSPACE(c) (SYNTAX (c) == Swhitespace)
-#define ISUPPER(c) (UPPERCASEP (c))
+# define ISUPPER(c) (UPPERCASEP (c))
-#define ISWORD(c) (SYNTAX (c) == Sword)
+# define ISWORD(c) (SYNTAX (c) == Sword)
#else /* not emacs */
Defining isascii to 1 should let any compiler worth its salt
eliminate the && through constant folding." */
-#if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII))
-#define ISASCII(c) 1
-#else
-#define ISASCII(c) isascii(c)
-#endif
+# if defined STDC_HEADERS || (!defined isascii && !defined HAVE_ISASCII)
+# define ISASCII(c) 1
+# else
+# define ISASCII(c) isascii(c)
+# endif
/* 1 if C is an ASCII character. */
-#define IS_REAL_ASCII(c) ((c) < 0200)
+# define IS_REAL_ASCII(c) ((c) < 0200)
/* This distinction is not meaningful, except in Emacs. */
-#define ISUNIBYTE(c) 1
+# define ISUNIBYTE(c) 1
+
+# ifdef isblank
+# define ISBLANK(c) (ISASCII (c) && isblank (c))
+# else
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+# endif
+# ifdef isgraph
+# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
+# else
+# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
+# endif
+
+# define ISPRINT(c) (ISASCII (c) && isprint (c))
+# define ISDIGIT(c) (ISASCII (c) && isdigit (c))
+# define ISALNUM(c) (ISASCII (c) && isalnum (c))
+# define ISALPHA(c) (ISASCII (c) && isalpha (c))
+# define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
+# define ISLOWER(c) (ISASCII (c) && islower (c))
+# define ISPUNCT(c) (ISASCII (c) && ispunct (c))
+# define ISSPACE(c) (ISASCII (c) && isspace (c))
+# define ISUPPER(c) (ISASCII (c) && isupper (c))
+# define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
+
+# define ISWORD(c) ISALPHA(c)
+
+# ifdef SYNTAX_TABLE
-#define ISDIGIT(c) (ISASCII (c) && isdigit (c))
-#define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
-#define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
+extern char *re_syntax_table;
-#ifdef isblank
-#define ISBLANK(c) (ISASCII (c) && isblank (c))
-#else
-#define ISBLANK(c) ((c) == ' ' || (c) == '\t')
-#endif
-#ifdef isgraph
-#define ISGRAPH(c) (ISASCII (c) && isgraph (c))
-#else
-#define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
-#endif
+# else /* not SYNTAX_TABLE */
+
+/* How many characters in the character set. */
+# define CHAR_SET_SIZE 256
+
+static char re_syntax_table[CHAR_SET_SIZE];
+
+static void
+init_syntax_once ()
+{
+ register int c;
+ static int done = 0;
+
+ if (done)
+ return;
+
+ bzero (re_syntax_table, sizeof re_syntax_table);
+
+ for (c = 'a'; c <= 'z'; c++)
+ re_syntax_table[c] = Sword;
+
+ for (c = 'A'; c <= 'Z'; c++)
+ re_syntax_table[c] = Sword;
+
+ for (c = '0'; c <= '9'; c++)
+ re_syntax_table[c] = Sword;
-#define ISPRINT(c) (ISASCII (c) && isprint (c))
-#define ISDIGIT(c) (ISASCII (c) && isdigit (c))
-#define ISALNUM(c) (ISASCII (c) && isalnum (c))
-#define ISALPHA(c) (ISASCII (c) && isalpha (c))
-#define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
-#define ISLOWER(c) (ISASCII (c) && islower (c))
-#define ISPUNCT(c) (ISASCII (c) && ispunct (c))
-#define ISSPACE(c) (ISASCII (c) && isspace (c))
-#define ISUPPER(c) (ISASCII (c) && isupper (c))
-#define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
+ re_syntax_table['_'] = Sword;
-#define ISWORD(c) ISALPHA(c)
+ done = 1;
+}
+
+# endif /* not SYNTAX_TABLE */
#endif /* not emacs */
\f
#ifndef NULL
-#define NULL (void *)0
+# define NULL (void *)0
#endif
/* We remove any previous definition of `SIGN_EXTEND_CHAR',
(Per Bothner suggested the basic approach.) */
#undef SIGN_EXTEND_CHAR
#if __STDC__
-#define SIGN_EXTEND_CHAR(c) ((signed char) (c))
+# define SIGN_EXTEND_CHAR(c) ((signed char) (c))
#else /* not __STDC__ */
/* As in Harbison and Steele. */
-#define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128)
+# define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128)
#endif
\f
/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
#ifdef REGEX_MALLOC
-#define REGEX_ALLOCATE malloc
-#define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
-#define REGEX_FREE free
+# define REGEX_ALLOCATE malloc
+# define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
+# define REGEX_FREE free
#else /* not REGEX_MALLOC */
/* Emacs already defines alloca, sometimes. */
-#ifndef alloca
+# ifndef alloca
/* Make alloca work the best possible way. */
-#ifdef __GNUC__
-#define alloca __builtin_alloca
-#else /* not __GNUC__ */
-#if HAVE_ALLOCA_H
-#include <alloca.h>
-#else /* not __GNUC__ or HAVE_ALLOCA_H */
-#if 0 /* It is a bad idea to declare alloca. We always cast the result. */
-#ifndef _AIX /* Already did AIX, up at the top. */
-char *alloca ();
-#endif /* not _AIX */
-#endif
-#endif /* not HAVE_ALLOCA_H */
-#endif /* not __GNUC__ */
+# ifdef __GNUC__
+# define alloca __builtin_alloca
+# else /* not __GNUC__ */
+# if HAVE_ALLOCA_H
+# include <alloca.h>
+# endif /* HAVE_ALLOCA_H */
+# endif /* not __GNUC__ */
-#endif /* not alloca */
+# endif /* not alloca */
-#define REGEX_ALLOCATE alloca
+# define REGEX_ALLOCATE alloca
/* Assumes a `char *destination' variable. */
-#define REGEX_REALLOCATE(source, osize, nsize) \
+# define REGEX_REALLOCATE(source, osize, nsize) \
(destination = (char *) alloca (nsize), \
bcopy (source, destination, osize), \
destination)
/* No need to do anything to free, after alloca. */
-#define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */
+# define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */
#endif /* not REGEX_MALLOC */
/* Define how to allocate the failure stack. */
-#if defined (REL_ALLOC) && defined (REGEX_MALLOC)
+#if defined REL_ALLOC && defined REGEX_MALLOC
-#define REGEX_ALLOCATE_STACK(size) \
+# define REGEX_ALLOCATE_STACK(size) \
r_alloc (&failure_stack_ptr, (size))
-#define REGEX_REALLOCATE_STACK(source, osize, nsize) \
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
r_re_alloc (&failure_stack_ptr, (nsize))
-#define REGEX_FREE_STACK(ptr) \
+# define REGEX_FREE_STACK(ptr) \
r_alloc_free (&failure_stack_ptr)
#else /* not using relocating allocator */
-#ifdef REGEX_MALLOC
+# ifdef REGEX_MALLOC
-#define REGEX_ALLOCATE_STACK malloc
-#define REGEX_REALLOCATE_STACK(source, osize, nsize) realloc (source, nsize)
-#define REGEX_FREE_STACK free
+# define REGEX_ALLOCATE_STACK malloc
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) realloc (source, nsize)
+# define REGEX_FREE_STACK free
-#else /* not REGEX_MALLOC */
+# else /* not REGEX_MALLOC */
-#define REGEX_ALLOCATE_STACK alloca
+# define REGEX_ALLOCATE_STACK alloca
-#define REGEX_REALLOCATE_STACK(source, osize, nsize) \
+# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
REGEX_REALLOCATE (source, osize, nsize)
/* No need to explicitly free anything. */
-#define REGEX_FREE_STACK(arg) ((void)0)
+# define REGEX_FREE_STACK(arg) ((void)0)
-#endif /* not REGEX_MALLOC */
+# endif /* not REGEX_MALLOC */
#endif /* not using relocating allocator */
If the length byte has the 0x80 bit set, then that stuff
is followed by a range table:
2 bytes of flags for character sets (low 8 bits, high 8 bits)
- See RANGE_TABLE_WORK_BITS below.
+ See RANGE_TABLE_WORK_BITS below.
2 bytes, the number of pairs that follow
pairs, each 2 multibyte characters,
- each multibyte character represented as 3 bytes. */
+ each multibyte character represented as 3 bytes. */
charset,
/* Same parameters as charset, but match any character that is
by a `no_op'. */
on_failure_jump_nastyloop,
- /* A smart `on_failure_jump' used for greedy * and + operators.
+ /* A smart `on_failure_jump' used for greedy * and + operators.
It analyses the loop before which it is put and if the
loop does not require backtracking, it changes itself to
`on_failure_keep_string_jump' and short-circuits the loop,
*dest += temp << 8;
}
-#ifndef EXTRACT_MACROS /* To debug the macros. */
-#undef EXTRACT_NUMBER
-#define EXTRACT_NUMBER(dest, src) extract_number (&dest, src)
-#endif /* not EXTRACT_MACROS */
+# ifndef EXTRACT_MACROS /* To debug the macros. */
+# undef EXTRACT_NUMBER
+# define EXTRACT_NUMBER(dest, src) extract_number (&dest, src)
+# endif /* not EXTRACT_MACROS */
#endif /* DEBUG */
*source += 2;
}
-#ifndef EXTRACT_MACROS
-#undef EXTRACT_NUMBER_AND_INCR
-#define EXTRACT_NUMBER_AND_INCR(dest, src) \
+# ifndef EXTRACT_MACROS
+# undef EXTRACT_NUMBER_AND_INCR
+# define EXTRACT_NUMBER_AND_INCR(dest, src) \
extract_number_and_incr (&dest, &src)
-#endif /* not EXTRACT_MACROS */
+# endif /* not EXTRACT_MACROS */
#endif /* DEBUG */
\f
#ifdef DEBUG
/* We use standard I/O for debugging. */
-#include <stdio.h>
+# include <stdio.h>
/* It is useful to test things that ``must'' be true when debugging. */
-#include <assert.h>
static int debug = -100000;
-#define DEBUG_STATEMENT(e) e
-#define DEBUG_PRINT1(x) if (debug > 0) printf (x)
-#define DEBUG_PRINT2(x1, x2) if (debug > 0) printf (x1, x2)
-#define DEBUG_PRINT3(x1, x2, x3) if (debug > 0) printf (x1, x2, x3)
-#define DEBUG_PRINT4(x1, x2, x3, x4) if (debug > 0) printf (x1, x2, x3, x4)
-#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
+# define DEBUG_STATEMENT(e) e
+# define DEBUG_PRINT1(x) if (debug > 0) printf (x)
+# define DEBUG_PRINT2(x1, x2) if (debug > 0) printf (x1, x2)
+# define DEBUG_PRINT3(x1, x2, x3) if (debug > 0) printf (x1, x2, x3)
+# define DEBUG_PRINT4(x1, x2, x3, x4) if (debug > 0) printf (x1, x2, x3, x4)
+# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
if (debug > 0) print_partial_compiled_pattern (s, e)
-#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
+# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
if (debug > 0) print_double_string (w, s1, sz1, s2, sz2)
printf ("/%d", mcnt);
break;
-#ifdef emacs
+# ifdef emacs
case before_dot:
printf ("/before_dot");
break;
mcnt = *p++;
printf ("/%d", mcnt);
break;
-#endif /* emacs */
+# endif /* emacs */
case begbuf:
printf ("/begbuf");
#else /* not DEBUG */
-#undef assert
-#define assert(e)
+# undef assert
+# define assert(e)
-#define DEBUG_STATEMENT(e)
-#define DEBUG_PRINT1(x)
-#define DEBUG_PRINT2(x1, x2)
-#define DEBUG_PRINT3(x1, x2, x3)
-#define DEBUG_PRINT4(x1, x2, x3, x4)
-#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
-#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
+# define DEBUG_STATEMENT(e)
+# define DEBUG_PRINT1(x)
+# define DEBUG_PRINT2(x1, x2)
+# define DEBUG_PRINT3(x1, x2, x3)
+# define DEBUG_PRINT4(x1, x2, x3, x4)
+# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
+# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
#endif /* not DEBUG */
\f
/* When using GNU C, we are not REALLY using the C alloca, no matter
what config.h may say. So don't take precautions for it. */
#ifdef __GNUC__
-#undef C_ALLOCA
+# undef C_ALLOCA
#endif
/* The match routines may not allocate if (1) they would do it with malloc
Note that if REL_ALLOC is defined, matching would not use malloc for the
failure stack, but we would still use it for the register vectors;
so REL_ALLOC should not affect this. */
-#if (defined (C_ALLOCA) || defined (REGEX_MALLOC)) && defined (emacs)
-#undef MATCH_MAY_ALLOCATE
+#if (defined C_ALLOCA || defined REGEX_MALLOC) && defined emacs
+# undef MATCH_MAY_ALLOCATE
#endif
\f
when matching. If this number is exceeded, we allocate more
space, so it is not a hard limit. */
#ifndef INIT_FAILURE_ALLOC
-#define INIT_FAILURE_ALLOC 20
+# define INIT_FAILURE_ALLOC 20
#endif
/* Roughly the maximum number of failure points on the stack. Would be
exactly that if always used TYPICAL_FAILURE_SIZE items each time we failed.
This is a variable only so users of regex can assign to it; we never
change it ourselves. */
-#if defined (MATCH_MAY_ALLOCATE)
+#if defined MATCH_MAY_ALLOCATE
/* Note that 4400 is enough to cause a crash on Alpha OSF/1,
whose default stack limit is 2mb. In order for a larger
value to work reliably, you have to try to make it accord
Do `return -2' if the alloc fails. */
#ifdef MATCH_MAY_ALLOCATE
-#define INIT_FAIL_STACK() \
+# define INIT_FAIL_STACK() \
do { \
fail_stack.stack = (fail_stack_elt_t *) \
REGEX_ALLOCATE_STACK (INIT_FAILURE_ALLOC * TYPICAL_FAILURE_SIZE \
fail_stack.frame = 0; \
} while (0)
-#define RESET_FAIL_STACK() REGEX_FREE_STACK (fail_stack.stack)
+# define RESET_FAIL_STACK() REGEX_FREE_STACK (fail_stack.stack)
#else
-#define INIT_FAIL_STACK() \
+# define INIT_FAIL_STACK() \
do { \
fail_stack.avail = 0; \
fail_stack.frame = 0; \
} while (0)
-#define RESET_FAIL_STACK() ((void)0)
+# define RESET_FAIL_STACK() ((void)0)
#endif
`char *', to avoid warnings when a string constant is passed. But
when we use a character as a subscript we must make it unsigned. */
#ifndef TRANSLATE
-#define TRANSLATE(d) \
+# define TRANSLATE(d) \
(RE_TRANSLATE_P (translate) ? RE_TRANSLATE (translate, (d)) : (d))
#endif
|| STREQ (string, "unibyte") || STREQ (string, "multibyte"))
/* QUIT is only used on NTemacs. */
-#if !defined (WINDOWSNT) || !defined (emacs)
-#undef QUIT
-#define QUIT
+#if !defined WINDOWSNT || !defined emacs
+# undef QUIT
+# define QUIT
#endif
\f
#ifndef MATCH_MAY_ALLOCATE
/* Always count groups, whether or not bufp->no_sub is set. */
bufp->re_nsub = 0;
-#if !defined (emacs) && !defined (SYNTAX_TABLE)
+#if !defined emacs && !defined SYNTAX_TABLE
/* Initialize the syntax table. */
init_syntax_once ();
#endif
for (;;)
{
- if (!(syntax & RE_ALL_GREEDY)
+ if ((syntax & RE_FRUGAL)
&& c == '?' && (zero_times_ok || many_times_ok))
greedy = 0;
else
char *destination;
#endif
-#if defined (REL_ALLOC) && defined (REGEX_MALLOC)
+#if defined REL_ALLOC && defined REGEX_MALLOC
/* This holds the pointer to the failure stack, when
it is allocated relocatably. */
fail_stack_elt_t *failure_stack_ptr;
int c, count;
/* Make P points the range table. `+ 2' is to skip flag
- bits for a character class. */
+ bits for a character class. */
p += CHARSET_BITMAP_SIZE (&p[-2]) + 2;
/* Extract the number of ranges in range table into COUNT. */
val = re_match_2_internal (bufp, string1, size1, string2, size2,
startpos, regs, stop);
#ifndef REGEX_MALLOC
-#ifdef C_ALLOCA
+# ifdef C_ALLOCA
alloca (0);
-#endif
+# endif
#endif
if (val >= 0)
/* Free everything we malloc. */
#ifdef MATCH_MAY_ALLOCATE
-#define FREE_VAR(var) if (var) { REGEX_FREE (var); var = NULL; } else
-#define FREE_VARIABLES() \
+# define FREE_VAR(var) if (var) { REGEX_FREE (var); var = NULL; } else
+# define FREE_VARIABLES() \
do { \
REGEX_FREE_STACK (fail_stack.stack); \
FREE_VAR (regstart); \
FREE_VAR (best_regend); \
} while (0)
#else
-#define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning. */
+# define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning. */
#endif /* not MATCH_MAY_ALLOCATE */
\f
{
int result = re_match_2_internal (bufp, NULL, 0, string, size,
pos, regs, size);
-#if defined (C_ALLOCA) && !defined (REGEX_MALLOC)
+# if defined C_ALLOCA && !defined REGEX_MALLOC
alloca (0);
-#endif
+# endif
return result;
}
#endif /* not emacs */
result = re_match_2_internal (bufp, string1, size1, string2, size2,
pos, regs, stop);
-#if defined (C_ALLOCA) && !defined (REGEX_MALLOC)
+#if defined C_ALLOCA && !defined REGEX_MALLOC
alloca (0);
#endif
return result;
unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
#endif
-#if defined (REL_ALLOC) && defined (REGEX_MALLOC)
+#if defined REL_ALLOC && defined REGEX_MALLOC
/* This holds the pointer to the failure stack, when
it is allocated relocatably. */
fail_stack_elt_t *failure_stack_ptr;
assert (!REG_UNSET (regstart[*p]));
/* Strictly speaking, there should be code such as:
- assert (REG_UNSET (regend[*p]));
+ assert (REG_UNSET (regend[*p]));
PUSH_FAILURE_REGSTOP ((unsigned int)*p);
But the only info to be pushed is regend[*p] and it is known to
#endif /* emacs */
- default:
- abort ();
+ default:
+ abort ();
}
continue; /* Successfully executed one pattern command; keep going. */
re_char *str;
unsigned char *pat;
/* A restart point is known. Restore to that state. */
- DEBUG_PRINT1 ("\nFAIL:\n");
- POP_FAILURE_POINT (str, pat);
+ DEBUG_PRINT1 ("\nFAIL:\n");
+ POP_FAILURE_POINT (str, pat);
switch (SWITCH_ENUM_CAST ((re_opcode_t) *pat++))
{
case on_failure_keep_string_jump:
assert (p >= bufp->buffer && p <= pend);
- if (d >= string1 && d <= end1)
+ if (d >= string1 && d <= end1)
dend = end_match_1;
- }
+ }
else
- break; /* Matching at this starting point really fails. */
+ break; /* Matching at this starting point really fails. */
} /* for (;;) */
if (best_regs_set)
const char *
re_compile_pattern (pattern, length, bufp)
const char *pattern;
- int length;
+ size_t length;
struct re_pattern_buffer *bufp;
{
reg_errcode_t ret;
/* Entry points compatible with 4.2 BSD regex library. We don't define
them unless specifically requested. */
-#if defined (_REGEX_RE_COMP) || defined (_LIBC)
+#if defined _REGEX_RE_COMP || defined _LIBC
/* BSD has one and only one pattern buffer. */
static struct re_pattern_buffer re_comp_buf;
char *
-#ifdef _LIBC
+# ifdef _LIBC
/* Make these definitions weak in libc, so POSIX programs can redefine
these names if they don't use our functions, and still use
regcomp/regexec below without link errors. */
weak_function
-#endif
+# endif
re_comp (s)
const char *s;
{
if (!s)
{
if (!re_comp_buf.buffer)
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
+ /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
return (char *) gettext ("No previous regular expression");
return 0;
}
{
re_comp_buf.buffer = (unsigned char *) malloc (200);
if (re_comp_buf.buffer == NULL)
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
- return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
+ /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
+ return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
re_comp_buf.allocated = 200;
re_comp_buf.fastmap = (char *) malloc (1 << BYTEWIDTH);
int
-#ifdef _LIBC
+# ifdef _LIBC
weak_function
-#endif
+# endif
re_exec (s)
const char *s;
{
= (RE_TRANSLATE_TYPE) malloc (CHAR_SET_SIZE
* sizeof (*(RE_TRANSLATE_TYPE)0));
if (preg->translate == NULL)
- return (int) REG_ESPACE;
+ return (int) REG_ESPACE;
/* Map uppercase characters to corresponding lowercase ones. */
for (i = 0; i < CHAR_SET_SIZE; i++)
- preg->translate[i] = ISUPPER (i) ? tolower (i) : i;
+ preg->translate[i] = ISUPPER (i) ? tolower (i) : i;
}
else
preg->translate = NULL;
regs.start = TALLOC (nmatch, regoff_t);
regs.end = TALLOC (nmatch, regoff_t);
if (regs.start == NULL || regs.end == NULL)
- return (int) REG_NOMATCH;
+ return (int) REG_NOMATCH;
}
/* Perform the searching operation. */
ret = re_search (&private_preg, string, len,
- /* start: */ 0, /* range: */ len,
- want_reg_info ? ®s : (struct re_registers *) 0);
+ /* start: */ 0, /* range: */ len,
+ want_reg_info ? ®s : (struct re_registers *) 0);
/* Copy the register information to the POSIX structure. */
if (want_reg_info)
{
if (ret >= 0)
- {
- unsigned r;
+ {
+ unsigned r;
- for (r = 0; r < nmatch; r++)
- {
- pmatch[r].rm_so = regs.start[r];
- pmatch[r].rm_eo = regs.end[r];
- }
- }
+ for (r = 0; r < nmatch; r++)
+ {
+ pmatch[r].rm_so = regs.start[r];
+ pmatch[r].rm_eo = regs.end[r];
+ }
+ }
/* If we needed the temporary register info, free the space now. */
free (regs.start);
if (errbuf_size != 0)
{
if (msg_size > errbuf_size)
- {
- strncpy (errbuf, msg, errbuf_size - 1);
- errbuf[errbuf_size - 1] = 0;
- }
+ {
+ strncpy (errbuf, msg, errbuf_size - 1);
+ errbuf[errbuf_size - 1] = 0;
+ }
else
- strcpy (errbuf, msg);
+ strcpy (errbuf, msg);
}
return msg_size;
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
USA. */
-#ifndef __REGEXP_LIBRARY_H__
-#define __REGEXP_LIBRARY_H__
+#ifndef _REGEX_H
+#define _REGEX_H 1
+
+/* Allow the use in C++ code. */
+#ifdef __cplusplus
+extern "C" {
+#endif
/* POSIX says that <sys/types.h> must be included (by the caller) before
<regex.h>. */
-#if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS)
+#if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
should be there. */
-#include <stddef.h>
+# include <stddef.h>
#endif
-
/* The following bits are used to determine the regexp syntax we
recognize. The set/not-set meanings where historically chosen so
that Emacs syntax had the value 0.
The bits are given in alphabetical order, and
the definitions shifted by one from the previous bit; thus, when we
add or remove a bit, only one other definition need change. */
-typedef unsigned reg_syntax_t;
-
-/* If this bit is set, then even *?, +? and ?? match greedily. */
-#define RE_ALL_GREEDY (1)
+typedef unsigned long int reg_syntax_t;
/* If this bit is not set, then \ inside a bracket expression is literal.
If set, then such a \ quotes the following character. */
-#define RE_BACKSLASH_ESCAPE_IN_LISTS (RE_ALL_GREEDY << 1)
+#define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
/* If this bit is not set, then + and ? are operators, and \+ and \? are
- literals.
+ literals.
If set, then \+ and \? are operators and + and ? are literals. */
#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
^ is an anchor if it is at the beginning of a regular
expression or after an open-group or an alternation operator;
$ is an anchor if it is at the end of a regular expression, or
- before a close-group or an alternation operator.
+ before a close-group or an alternation operator.
This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
POSIX draft 11.2 says that * etc. in leading positions is undefined.
/* If this bit is set, then special characters are always special
regardless of where they are in the pattern.
If this bit is not set, then special characters are special only in
- some contexts; otherwise they are ordinary. Specifically,
+ some contexts; otherwise they are ordinary. Specifically,
* + ? and intervals are only special when not after the beginning,
open-group, or alternation operator. */
#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
/* If this bit is set, either \{...\} or {...} defines an
- interval, depending on RE_NO_BK_BRACES.
+ interval, depending on RE_NO_BK_BRACES.
If not set, \{, \}, {, and } are literals. */
#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
If not set, then \<digit> is a back-reference. */
#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
-/* If this bit is set, then | is an alternation operator, and \| is literal.
+/* If this bit is set, then | is an alternation operator, and \| is literal.
If not set, then \| is an alternation operator, and | is literal. */
#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
starting range point, the range is ignored. */
#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
+/* If this bit is set, then an unmatched ) is ordinary.
+ If not set, then an unmatched ) is invalid. */
+#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
+
/* If this bit is set, succeed as soon as we match the whole pattern,
without further backtracking. */
-#define RE_NO_POSIX_BACKTRACKING (RE_NO_EMPTY_RANGES << 1)
+#define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
-/* If this bit is set, then (?:...) is treated as a shy group. */
-#define RE_SHY_GROUPS (RE_NO_POSIX_BACKTRACKING << 1)
+/* If this bit is set, do not process the GNU regex operators.
+ If not set, then the GNU regex operators are recognized. */
+#define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
-/* If this bit is set, then an unmatched ) is ordinary.
- If not set, then an unmatched ) is invalid. */
-#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_SHY_GROUPS << 1)
+/* If this bit is set, turn on internal regex debugging.
+ If not set, and debugging was on, turn it off.
+ This only works if regex.c is compiled -DDEBUG.
+ We define this bit always, so that all that's needed to turn on
+ debugging is to recompile regex.c; the calling code can always have
+ this bit set, and it won't affect anything in the normal case. */
+#define RE_DEBUG (RE_NO_GNU_OPS << 1)
+
+/* If this bit is set, then *?, +? and ?? match non greedily. */
+#define RE_FRUGAL (RE_DEBUG << 1)
+
+/* If this bit is set, then (?:...) is treated as a shy group. */
+#define RE_SHY_GROUPS (RE_FRUGAL << 1)
/* This global variable defines the particular regexp syntax to use (for
some interfaces). When a regexp is compiled, the syntax used is
(The [[[ comments delimit what gets put into the Texinfo file, so
don't delete them!) */
/* [[[begin syntaxes]]] */
-#define RE_SYNTAX_EMACS (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS)
+#define RE_SYNTAX_EMACS \
+ (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
#define RE_SYNTAX_AWK \
- (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
- | RE_NO_BK_PARENS | RE_NO_BK_REFS \
- | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
- | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_ALL_GREEDY)
+ (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
+ | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
+
+#define RE_SYNTAX_GNU_AWK \
+ ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
+ & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
#define RE_SYNTAX_POSIX_AWK \
- (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
+ (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
+ | RE_INTERVALS | RE_NO_GNU_OPS)
#define RE_SYNTAX_GREP \
(RE_BK_PLUS_QM | RE_CHAR_CLASSES \
| RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
- | RE_NEWLINE_ALT | RE_ALL_GREEDY)
+ | RE_NEWLINE_ALT)
#define RE_SYNTAX_EGREP \
(RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
| RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
| RE_NEWLINE_ALT | RE_NO_BK_PARENS \
- | RE_NO_BK_VBAR | RE_ALL_GREEDY)
+ | RE_NO_BK_VBAR)
#define RE_SYNTAX_POSIX_EGREP \
(RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
/* Syntax bits common to both basic and extended POSIX regex syntax. */
#define _RE_SYNTAX_POSIX_COMMON \
(RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
- | RE_INTERVALS | RE_NO_EMPTY_RANGES | RE_ALL_GREEDY)
+ | RE_INTERVALS | RE_NO_EMPTY_RANGES)
#define RE_SYNTAX_POSIX_BASIC \
(_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
(_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
#define RE_SYNTAX_POSIX_EXTENDED \
- (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
- | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
- | RE_UNMATCHED_RIGHT_PAREN_ORD)
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
+ | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
-/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
- replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
+/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
+ removed and RE_NO_BK_REFS is added. */
#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
| RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
(erroneously) define this in other header files, but we want our
value, so remove any previous define. */
#ifdef RE_DUP_MAX
-#undef RE_DUP_MAX
+# undef RE_DUP_MAX
#endif
-#define RE_DUP_MAX ((1 << 15) - 1)
+/* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
+#define RE_DUP_MAX (0x7fff)
/* POSIX `cflags' bits (i.e., information for `regcomp'). */
/* If this bit is set, then ignore case when matching.
If not set, then case is significant. */
#define REG_ICASE (REG_EXTENDED << 1)
-
+
/* If this bit is set, then anchors do not match at newline
characters in the string.
If not set, then anchors do match at newlines. */
`re_error_msg' table in regex.c. */
typedef enum
{
+#ifdef _XOPEN_SOURCE
+ REG_ENOSYS = -1, /* This will never happen for this implementation. */
+#endif
+
REG_NOERROR = 0, /* Success. */
REG_NOMATCH, /* Didn't find a match (for regexec). */
REG_EESCAPE, /* Trailing backslash. */
REG_ESUBREG, /* Invalid back reference. */
REG_EBRACK, /* Unmatched left bracket. */
- REG_EPAREN, /* Parenthesis imbalance. */
+ REG_EPAREN, /* Parenthesis imbalance. */
REG_EBRACE, /* Unmatched \{. */
REG_BADBR, /* Invalid contents of \{\}. */
REG_ERANGE, /* Invalid range end. */
private to the regex routines. */
#ifndef RE_TRANSLATE_TYPE
-#define RE_TRANSLATE_TYPE char *
+# define RE_TRANSLATE_TYPE char *
#endif
struct re_pattern_buffer
unsigned char *buffer;
/* Number of bytes to which `buffer' points. */
- unsigned long allocated;
+ unsigned long int allocated;
/* Number of bytes actually used in `buffer'. */
- unsigned long used;
+ unsigned long int used;
/* Syntax setting with which the pattern was compiled. */
reg_syntax_t syntax;
unsigned no_sub : 1;
/* If set, a beginning-of-line anchor doesn't match at the
- beginning of the string. */
+ beginning of the string. */
unsigned not_bol : 1;
/* Similarly for an end-of-line anchor. */
`re_match_2' returns information about at least this many registers
the first time a `regs' structure is passed. */
#ifndef RE_NREGS
-#define RE_NREGS 30
+# define RE_NREGS 30
#endif
#if defined __STDC__ || defined PROTOTYPES
-#define _RE_ARGS(args) args
+# define _RE_ARGS(args) args
#else /* not __STDC__ || PROTOTYPES */
-#define _RE_ARGS(args) ()
+# define _RE_ARGS(args) ()
#endif /* not __STDC__ || PROTOTYPES */
and syntax given by the global `re_syntax_options', into the buffer
BUFFER. Return NULL if successful, and an error string if not. */
extern const char *re_compile_pattern
- _RE_ARGS ((const char *pattern, int length,
+ _RE_ARGS ((const char *pattern, size_t length,
struct re_pattern_buffer *buffer));
/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
-extern int re_match_2
+extern int re_match_2
_RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
int length1, const char *string2, int length2,
int start, struct re_registers *regs, int stop));
_RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
unsigned num_regs, regoff_t *starts, regoff_t *ends));
-#ifdef _REGEX_RE_COMP
+#if defined _REGEX_RE_COMP || defined _LIBC
+# ifndef _CRAY
/* 4.2 bsd compatibility. */
extern char *re_comp _RE_ARGS ((const char *));
extern int re_exec _RE_ARGS ((const char *));
+# endif
+#endif
+
+/* GCC 2.95 and later have "__restrict"; C99 compilers have
+ "restrict", and "configure" may have defined "restrict". */
+#ifndef __restrict
+# if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
+# if defined restrict || 199901L <= __STDC_VERSION__
+# define __restrict restrict
+# else
+# define __restrict
+# endif
+# endif
#endif
+/* For now unconditionally define __restrict_arr to expand to nothing.
+ Ideally we would have a test for the compiler which allows defining
+ it to restrict. */
+#define __restrict_arr
/* POSIX compatibility. */
-extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
-extern int regexec
- _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
- regmatch_t pmatch[], int eflags));
-extern size_t regerror
- _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
- size_t errbuf_size));
-extern void regfree _RE_ARGS ((regex_t *preg));
-
-#endif /* not __REGEXP_LIBRARY_H__ */
+extern int regcomp _RE_ARGS ((regex_t *__restrict __preg,
+ const char *__restrict __pattern,
+ int __cflags));
+
+extern int regexec _RE_ARGS ((const regex_t *__restrict __preg,
+ const char *__restrict __string, size_t __nmatch,
+ regmatch_t __pmatch[__restrict_arr],
+ int __eflags));
+
+extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg,
+ char *__errbuf, size_t __errbuf_size));
+
+extern void regfree _RE_ARGS ((regex_t *__preg));
+
+
+#ifdef __cplusplus
+}
+#endif /* C++ */
+
+#endif /* regex.h */
\f
/*
Local variables:
projects / gnulib.git / commitdiff