# include <config.h>
#endif
-#if defined STDC_HEADERS && !defined emacs
-# include <stddef.h>
-#else
+#include <stddef.h>
+
+#ifdef emacs
/* We need this for `regex.h', and perhaps for the Emacs include files. */
# include <sys/types.h>
#endif
# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
# define regerror(err_code, preg, errbuf, errbuf_size) \
- __regerror(err_code, preg, errbuf, errbuf_size)
+ __regerror (err_code, preg, errbuf, errbuf_size)
# define re_set_registers(bu, re, nu, st, en) \
__re_set_registers (bu, re, nu, st, en)
# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
# endif
# define realloc xrealloc
-/* This is the normal way of making sure we have memcpy, memcmp and memset. */
-# if defined HAVE_STRING_H || defined STDC_HEADERS || defined _LIBC
-# include <string.h>
-# else
-# include <strings.h>
-# ifndef memcmp
-# define memcmp(s1, s2, n) bcmp (s1, s2, n)
-# endif
-# ifndef memcpy
-# define memcpy(d, s, n) (bcopy (s, d, n), (d))
-# endif
-# endif
+# include <string.h>
/* Define the syntax stuff for \<, \>, etc. */
#else /* not emacs */
-/* Jim Meyering writes:
-
- "... Some ctype macros are valid only for character codes that
- isascii says are ASCII (SGI's IRIX-4.0.5 is one such system --when
- using /bin/cc or gcc but without giving an ansi option). So, all
- ctype uses should be through macros like ISPRINT... If
- STDC_HEADERS is defined, then autoconf has verified that the ctype
- macros don't need to be guarded with references to isascii. ...
- Defining isascii to 1 should let any compiler worth its salt
- eliminate the && through constant folding."
- Solaris defines some of these symbols so we must undefine them first. */
-
-# undef ISASCII
-# 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 ISUNIBYTE(c) 1
# ifdef isblank
-# define ISBLANK(c) (ISASCII (c) && isblank (c))
+# define ISBLANK(c) isblank (c)
# else
# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
# endif
# ifdef isgraph
-# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
+# define ISGRAPH(c) isgraph (c)
# else
-# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
+# define ISGRAPH(c) (isprint (c) && !isspace (c))
# endif
+/* Solaris defines ISPRINT so we must undefine it first. */
# undef ISPRINT
-# 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)
+# define ISPRINT(c) isprint (c)
+# define ISDIGIT(c) isdigit (c)
+# define ISALNUM(c) isalnum (c)
+# define ISALPHA(c) isalpha (c)
+# define ISCNTRL(c) iscntrl (c)
+# define ISLOWER(c) islower (c)
+# define ISPUNCT(c) ispunct (c)
+# define ISSPACE(c) isspace (c)
+# define ISUPPER(c) isupper (c)
+# define ISXDIGIT(c) isxdigit (c)
+
+# define ISWORD(c) ISALPHA (c)
# ifdef _tolower
-# define TOLOWER(c) _tolower(c)
+# define TOLOWER(c) _tolower (c)
# else
-# define TOLOWER(c) tolower(c)
+# define TOLOWER(c) tolower (c)
# endif
/* How many characters in the character set. */
#endif /* not emacs */
\f
-#ifndef NULL
-# define NULL (void *)0
-#endif
-
/* We remove any previous definition of `SIGN_EXTEND_CHAR',
since ours (we hope) works properly with all combinations of
machines, compilers, `char' and `unsigned char' argument types.
#define MIN(a, b) ((a) < (b) ? (a) : (b))
/* Type of source-pattern and string chars. */
+#ifdef _MSC_VER
+typedef unsigned char re_char;
+#else
typedef const unsigned char re_char;
+#endif
typedef char boolean;
#define false 0
#define true 1
-static int re_match_2_internal _RE_ARGS ((struct re_pattern_buffer *bufp,
- re_char *string1, int size1,
- re_char *string2, int size2,
- int pos,
- struct re_registers *regs,
- int stop));
+static regoff_t re_match_2_internal _RE_ARGS ((struct re_pattern_buffer *bufp,
+ re_char *string1, size_t size1,
+ re_char *string2, size_t size2,
+ ssize_t pos,
+ struct re_registers *regs,
+ ssize_t stop));
\f
/* These are the command codes that appear in compiled regular
expressions. Some opcodes are followed by argument bytes. A
on_failure_jump_nastyloop,
/* A smart `on_failure_jump' used for greedy * and + operators.
- It analyses the loop before which it is put and if the
+ It analyzes 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,
else it just defaults to changing itself into `on_failure_jump'.
re_char *where;
re_char *string1;
re_char *string2;
- int size1;
- int size2;
+ ssize_t size1;
+ ssize_t size2;
{
- int this_char;
+ ssize_t this_char;
if (where == NULL)
printf ("(null)");
#endif /* not DEBUG */
\f
+/* Use this to suppress gcc's `...may be used before initialized' warnings. */
+#ifdef lint
+# define IF_LINT(Code) Code
+#else
+# define IF_LINT(Code) /* empty */
+#endif
+\f
/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
also be assigned to arbitrarily: each pattern buffer stores its own
syntax, so it can be changed between regex compilations. */
/* Pop a saved register off the stack. */
#define POP_FAILURE_REG_OR_COUNT() \
do { \
- int pfreg = POP_FAILURE_INT (); \
+ long pfreg = POP_FAILURE_INT (); \
if (pfreg == -1) \
{ \
/* It's a counter. */ \
/* Check that we are not stuck in an infinite loop. */
#define CHECK_INFINITE_LOOP(pat_cur, string_place) \
do { \
- int failure = TOP_FAILURE_HANDLE (); \
+ ssize_t failure = TOP_FAILURE_HANDLE (); \
/* Check for infinite matching loops */ \
while (failure > 0 \
&& (FAILURE_STR (failure) == string_place \
typedef struct
{
compile_stack_elt_t *stack;
- unsigned size;
- unsigned avail; /* Offset of next open position. */
+ size_t size;
+ size_t avail; /* Offset of next open position. */
} compile_stack_type;
{
switch (cc)
{
- case RECC_ALNUM: return ISALNUM (ch);
- case RECC_ALPHA: return ISALPHA (ch);
- case RECC_BLANK: return ISBLANK (ch);
- case RECC_CNTRL: return ISCNTRL (ch);
- case RECC_DIGIT: return ISDIGIT (ch);
- case RECC_GRAPH: return ISGRAPH (ch);
- case RECC_LOWER: return ISLOWER (ch);
- case RECC_PRINT: return ISPRINT (ch);
- case RECC_PUNCT: return ISPUNCT (ch);
- case RECC_SPACE: return ISSPACE (ch);
- case RECC_UPPER: return ISUPPER (ch);
- case RECC_XDIGIT: return ISXDIGIT (ch);
- case RECC_ASCII: return IS_REAL_ASCII (ch);
+ case RECC_ALNUM: return ISALNUM (ch) != 0;
+ case RECC_ALPHA: return ISALPHA (ch) != 0;
+ case RECC_BLANK: return ISBLANK (ch) != 0;
+ case RECC_CNTRL: return ISCNTRL (ch) != 0;
+ case RECC_DIGIT: return ISDIGIT (ch) != 0;
+ case RECC_GRAPH: return ISGRAPH (ch) != 0;
+ case RECC_LOWER: return ISLOWER (ch) != 0;
+ case RECC_PRINT: return ISPRINT (ch) != 0;
+ case RECC_PUNCT: return ISPUNCT (ch) != 0;
+ case RECC_SPACE: return ISSPACE (ch) != 0;
+ case RECC_UPPER: return ISUPPER (ch) != 0;
+ case RECC_XDIGIT: return ISXDIGIT (ch) != 0;
+ case RECC_ASCII: return IS_REAL_ASCII (ch) != 0;
case RECC_NONASCII: return !IS_REAL_ASCII (ch);
- case RECC_UNIBYTE: return ISUNIBYTE (ch);
+ case RECC_UNIBYTE: return ISUNIBYTE (ch) != 0;
case RECC_MULTIBYTE: return !ISUNIBYTE (ch);
- case RECC_WORD: return ISWORD (ch);
+ case RECC_WORD: return ISWORD (ch) != 0;
case RECC_ERROR: return false;
default:
- abort();
+ abort ();
}
}
case RECC_ASCII: case RECC_DIGIT: case RECC_XDIGIT: case RECC_CNTRL:
case RECC_BLANK: case RECC_UNIBYTE: case RECC_ERROR: return 0;
default:
- abort();
+ abort ();
}
}
#endif
Returns -1 if successful, REG_ESPACE if ran out of space. */
static int
-set_image_of_range_1 (work_area, start, end, translate)
- RE_TRANSLATE_TYPE translate;
- struct range_table_work_area *work_area;
- re_wchar_t start, end;
+set_image_of_range_1 (struct range_table_work_area *work_area,
+ re_wchar_t start, re_wchar_t end,
+ RE_TRANSLATE_TYPE translate)
{
/* `one_case' indicates a character, or a run of characters,
each of which is an isolate (no case-equivalents).
Returns -1 if successful, REG_ESPACE if ran out of space. */
static int
-set_image_of_range (work_area, start, end, translate)
- RE_TRANSLATE_TYPE translate;
- struct range_table_work_area *work_area;
- re_wchar_t start, end;
+set_image_of_range (struct range_table_work_area *work_area,
+ re_wchar_t start, re_wchar_t end,
+ RE_TRANSLATE_TYPE translate)
{
re_wchar_t cmin, cmax;
but don't make them smaller. */
static
-regex_grow_registers (num_regs)
- int num_regs;
+regex_grow_registers (int num_regs)
{
if (num_regs > regs_allocated_size)
{
if (many_times_ok)
{
boolean simple = skip_one_char (laststart) == b;
- unsigned int startoffset = 0;
+ size_t startoffset = 0;
re_opcode_t ofj =
/* Check if the loop can match the empty string. */
(simple || !analyse_first (laststart, b, NULL, 0))
_____ _____
| | | |
| v | v
- a | b | c
+ a | b | c
If we are at `b', then fixup_alt_jump right now points to a
three-byte space after `a'. We'll put in the jump, set
static boolean
group_in_compile_stack (compile_stack_type compile_stack, regnum_t regnum)
{
- int this_element;
+ ssize_t this_element;
for (this_element = compile_stack.avail - 1;
this_element >= 0;
/* Like re_search_2, below, but only one string is specified, and
doesn't let you say where to stop matching. */
-int
-re_search (struct re_pattern_buffer *bufp, const char *string, int size, int startpos, int range, struct re_registers *regs)
+regoff_t
+re_search (struct re_pattern_buffer *bufp, const char *string, size_t size,
+ ssize_t startpos, ssize_t range, struct re_registers *regs)
{
return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
regs, size);
found, -1 if no match, or -2 if error (such as failure
stack overflow). */
-int
-re_search_2 (struct re_pattern_buffer *bufp, const char *str1, int size1, const char *str2, int size2, int startpos, int range, struct re_registers *regs, int stop)
+regoff_t
+re_search_2 (struct re_pattern_buffer *bufp, const char *str1, size_t size1,
+ const char *str2, size_t size2, ssize_t startpos, ssize_t range,
+ struct re_registers *regs, ssize_t stop)
{
- int val;
+ regoff_t val;
re_char *string1 = (re_char*) str1;
re_char *string2 = (re_char*) str2;
register char *fastmap = bufp->fastmap;
register RE_TRANSLATE_TYPE translate = bufp->translate;
- int total_size = size1 + size2;
- int endpos = startpos + range;
+ size_t total_size = size1 + size2;
+ ssize_t endpos = startpos + range;
boolean anchored_start;
/* Nonzero if we are searching multibyte string. */
const boolean multibyte = RE_TARGET_MULTIBYTE_P (bufp);
#ifdef emacs
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
{
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (startpos));
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (startpos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
}
if (range > 0) /* Searching forwards. */
{
register int lim = 0;
- int irange = range;
+ ssize_t irange = range;
if (startpos < size1 && startpos + range >= size1)
lim = range - (size1 - startpos);
\f
/* Declarations and macros for re_match_2. */
-static int bcmp_translate _RE_ARGS((re_char *s1, re_char *s2,
- register int len,
- RE_TRANSLATE_TYPE translate,
- const int multibyte));
+static int bcmp_translate _RE_ARGS ((re_char *s1, re_char *s2,
+ register ssize_t len,
+ RE_TRANSLATE_TYPE translate,
+ const int multibyte));
/* This converts PTR, a pointer into one of the search strings `string1'
and `string2' into an offset from the beginning of that string. */
&& ((p2[2 + idx] & ~ p1[2 + idx]) == 0))))
break;
- if (idx == p2[1])
- {
- DEBUG_PRINT1 (" No match => fast loop.\n");
- return 1;
- }
+ if (idx == p2[1])
+ {
+ DEBUG_PRINT1 (" No match => fast loop.\n");
+ return 1;
+ }
}
}
}
#ifndef emacs /* Emacs never uses this. */
/* re_match is like re_match_2 except it takes only a single string. */
-int
+regoff_t
re_match (struct re_pattern_buffer *bufp, const char *string,
- int size, int pos, struct re_registers *regs)
+ size_t size, ssize_t pos, struct re_registers *regs)
{
- int result = re_match_2_internal (bufp, NULL, 0, (re_char*) string, size,
- pos, regs, size);
+ regoff_t result = re_match_2_internal (bufp, NULL, 0, (re_char*) string,
+ size, pos, regs, size);
return result;
}
WEAK_ALIAS (__re_match, re_match)
failure stack overflowing). Otherwise, we return the length of the
matched substring. */
-int
-re_match_2 (struct re_pattern_buffer *bufp, const char *string1, int size1, const char *string2, int size2, int pos, struct re_registers *regs, int stop)
+regoff_t
+re_match_2 (struct re_pattern_buffer *bufp, const char *string1,
+ size_t size1, const char *string2, size_t size2, ssize_t pos,
+ struct re_registers *regs, ssize_t stop)
{
- int result;
+ regoff_t result;
#ifdef emacs
- int charpos;
+ ssize_t charpos;
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (pos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
/* This is a separate function so that we can force an alloca cleanup
afterwards. */
-static int
-re_match_2_internal (struct re_pattern_buffer *bufp, const re_char *string1, int size1, const re_char *string2, int size2, int pos, struct re_registers *regs, int stop)
+static regoff_t
+re_match_2_internal (struct re_pattern_buffer *bufp, const re_char *string1,
+ size_t size1, const re_char *string2, size_t size2,
+ ssize_t pos, struct re_registers *regs, ssize_t stop)
{
/* General temporaries. */
- int mcnt;
+ ssize_t mcnt;
size_t reg;
/* Just past the end of the corresponding string. */
int s1, s2;
int dummy;
#ifdef emacs
- int offset = PTR_TO_OFFSET (d - 1);
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d - 1);
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
int s1, s2;
int dummy;
#ifdef emacs
- int offset = PTR_TO_OFFSET (d);
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d);
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
#endif
PREFETCH ();
int s1, s2;
int dummy;
#ifdef emacs
- int offset = PTR_TO_OFFSET (d) - 1;
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d) - 1;
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
re_wchar_t c1, c2;
int s1, s2;
#ifdef emacs
- int offset = PTR_TO_OFFSET (d);
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d);
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
#endif
PREFETCH ();
re_wchar_t c1, c2;
int s1, s2;
#ifdef emacs
- int offset = PTR_TO_OFFSET (d) - 1;
- int charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d) - 1;
+ ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
PREFETCH ();
#ifdef emacs
{
- int offset = PTR_TO_OFFSET (d);
- int pos1 = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ssize_t offset = PTR_TO_OFFSET (d);
+ ssize_t pos1 = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (pos1);
}
#endif
goto fail;
default:
- abort();
+ abort ();
}
assert (p >= bufp->buffer && p <= pend);
bytes; nonzero otherwise. */
static int
-bcmp_translate (const re_char *s1, const re_char *s2, register int len,
+bcmp_translate (const re_char *s1, const re_char *s2, register ssize_t len,
RE_TRANSLATE_TYPE translate, const int target_multibyte)
{
register re_char *p1 = s1, *p2 = s2;
We call regex_compile to do the actual compilation. */
const char *
-re_compile_pattern (const char *pattern, size_t length, struct re_pattern_buffer *bufp)
+re_compile_pattern (const char *pattern, size_t length,
+ struct re_pattern_buffer *bufp)
{
reg_errcode_t ret;
regcomp/regexec below without link errors. */
weak_function
# endif
-re_comp (s)
- const char *s;
+re_comp (const char *s)
{
reg_errcode_t ret;
# ifdef _LIBC
weak_function
# endif
-re_exec (s)
- const char *s;
+re_exec (const char *s)
{
- const int len = strlen (s);
+ const size_t len = strlen (s);
return
0 <= re_search (&re_comp_buf, s, len, 0, len, (struct re_registers *) 0);
}
It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
the return codes and their meanings.) */
-int
+reg_errcode_t
regcomp (regex_t *__restrict preg, const char *__restrict pattern,
int cflags)
{
preg->fastmap = NULL;
}
}
- return (int) ret;
+ return ret;
}
WEAK_ALIAS (__regcomp, regcomp)
We return 0 if we find a match and REG_NOMATCH if not. */
-int
+reg_errcode_t
regexec (const regex_t *__restrict preg, const char *__restrict string,
size_t nmatch, regmatch_t pmatch[__restrict_arr], int eflags)
{
- int ret;
+ regoff_t ret;
struct re_registers regs;
regex_t private_preg;
- int len = strlen (string);
+ size_t len = strlen (string);
boolean want_reg_info = !preg->no_sub && nmatch > 0 && pmatch;
private_preg = *preg;
regs.num_regs = nmatch;
regs.start = TALLOC (nmatch * 2, regoff_t);
if (regs.start == NULL)
- return (int) REG_NOMATCH;
+ return REG_NOMATCH;
regs.end = regs.start + nmatch;
}
}
/* We want zero return to mean success, unlike `re_search'. */
- return ret >= 0 ? (int) REG_NOERROR : (int) REG_NOMATCH;
+ return ret >= 0 ? REG_NOERROR : REG_NOMATCH;
}
WEAK_ALIAS (__regexec, regexec)