/* Definitions for data structures and routines for the regular
expression library.
Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993, 1995, 1996, 1997, 1998,
2000, 2001, 2002, 2003, 2005, 2006, 2009, 2010 Free Software Foundation,
Inc.
This file is part of the GNU C Library.
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3, or (at your option)
any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation,
Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
#ifndef _REGEX_H
/* Allow the use in C++ code. */
#ifdef __cplusplus
extern "C" {
#endif
/* Define __USE_GNU_REGEX to declare GNU extensions that violate the
POSIX name space rules. */
#if (defined _GNU_SOURCE \
|| (!defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE \
&& !defined _XOPEN_SOURCE))
#endif
#ifdef _REGEX_LARGE_OFFSETS
/* Use types and values that are wide enough to represent signed and
unsigned byte offsets in memory. This currently works only when
the regex code is used outside of the GNU C library; it is not yet
supported within glibc itself, and glibc users should not define
_REGEX_LARGE_OFFSETS. */
/* The type of the offset of a byte within a string.
For historical reasons POSIX 1003.1-2004 requires that regoff_t be
at least as wide as off_t. However, many common POSIX platforms set
regoff_t to the more-sensible ssize_t and the Open Group has
signalled its intention to change the requirement to be that
regoff_t be at least as wide as ptrdiff_t and ssize_t; see XBD ERN
60 (2005-08-25). We don't know of any hosts where ssize_t or
ptrdiff_t is wider than ssize_t, so ssize_t is safe. */
/* The type of nonnegative object indexes. Traditionally, GNU regex
uses 'int' for these. Code that uses __re_idx_t should work
regardless of whether the type is signed. */
/* The type of object sizes. */
/* The type of object sizes, in places where the traditional code
uses unsigned long int. */
#else
/* Use types that are binary-compatible with the traditional GNU regex
implementation, which mishandles strings longer than INT_MAX. */
typedef int regoff_t;
typedef int __re_idx_t;
typedef unsigned int __re_size_t;
typedef unsigned long int __re_long_size_t;
#endif
/* The following two types have to be signed and unsigned integer type
wide enough to hold a value of a pointer. For most ANSI compilers
ptrdiff_t and size_t should be likely OK. Still size of these two
types is 2 for Microsoft C. Ugh... */
typedef long int s_reg_t;
typedef unsigned long int active_reg_t;
/* The following bits are used to determine the regexp syntax we
remains 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 long int reg_syntax_t;
#ifdef __USE_GNU_REGEX
/* If this bit is not set, then \ inside a bracket expression is literal.
If set, then such a \ quotes the following character. */
/* If this bit is not set, then + and ? are operators, and \+ and \? are
literals.
If set, then \+ and \? are operators and + and ? are literals. */
/* If this bit is set, then character classes are supported. They are:
[:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
[:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
If not set, then character classes are not supported. */
/* If this bit is set, then ^ and $ are always anchors (outside bracket
expressions, of course).
If this bit is not set, then it depends:
^ 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.
This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
POSIX draft 11.2 says that * etc. in leading positions is undefined.
We already implemented a previous draft which made those constructs
invalid, though, so we haven't changed the code back. */
/* 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,
* + ? and intervals are only special when not after the beginning,
open-group, or alternation operator. */
/* If this bit is set, then *, +, ?, and { cannot be first in an re or
immediately after an alternation or begin-group operator. */
/* If this bit is set, then . matches newline.
If not set, then it doesn't. */
/* If this bit is set, then . doesn't match NUL.
If not set, then it does. */
/* If this bit is set, nonmatching lists [^...] do not match newline.
If not set, they do. */
/* If this bit is set, either \{...\} or {...} defines an
interval, depending on RE_NO_BK_BRACES.
If not set, \{, \}, {, and } are literals. */
/* If this bit is set, +, ? and | aren't recognized as operators.
If not set, they are. */
/* If this bit is set, newline is an alternation operator.
If not set, newline is literal. */
/* If this bit is set, then `{...}' defines an interval, and \{ and \}
are literals.
If not set, then `\{...\}' defines an interval. */
/* If this bit is set, (...) defines a group, and \( and \) are literals.
If not set, \(...\) defines a group, and ( and ) are literals. */
/* If this bit is set, then \<digit> matches <digit>.
If not set, then \<digit> is a back-reference. */
/* If this bit is set, then | is an alternation operator, and \| is literal.
If not set, then \| is an alternation operator, and | is literal. */
/* If this bit is set, then an ending range point collating higher
than the starting range point, as in [z-a], is invalid.
If not set, then when ending range point collates higher than the
starting range point, the range is ignored. */
/* If this bit is set, then an unmatched ) is ordinary.
If not set, then an unmatched ) is invalid. */
/* If this bit is set, succeed as soon as we match the whole pattern,
without further backtracking. */
/* If this bit is set, do not process the GNU regex operators.
If not set, then the GNU regex operators are recognized. */
/* 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. */
/* If this bit is set, a syntactically invalid interval is treated as
a string of ordinary characters. For example, the ERE 'a{1' is
treated as 'a\{1'. */
/* If this bit is set, then ignore case when matching.
If not set, then case is significant. */
/* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
for ^, because it is difficult to scan the regex backwards to find
whether ^ should be special. */
/* If this bit is set, then \{ cannot be first in a regex or
immediately after an alternation, open-group or \} operator. */
/* If this bit is set, then no_sub will be set to 1 during
re_compile_pattern. */
#endif /* defined __USE_GNU_REGEX */
/* This global variable defines the particular regexp syntax to use (for
some interfaces). When a regexp is compiled, the syntax used is
stored in the pattern buffer, so changing this does not affect
already-compiled regexps. */
extern reg_syntax_t re_syntax_options;
#ifdef __USE_GNU_REGEX
/* Define combinations of the above bits for the standard possibilities.
(The [[[ comments delimit what gets put into the Texinfo file, so
don't delete them!) */
/* [[[begin syntaxes]]] */
# define RE_SYNTAX_EMACS 0
# define RE_SYNTAX_AWK \
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
# define RE_SYNTAX_GNU_AWK \
# define RE_SYNTAX_POSIX_AWK \
# define RE_SYNTAX_GREP \
# define RE_SYNTAX_EGREP \
# define RE_SYNTAX_POSIX_EGREP \
/* Syntax bits common to both basic and extended POSIX regex syntax. */
# define _RE_SYNTAX_POSIX_COMMON \
# define RE_SYNTAX_POSIX_BASIC \
/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
isn't minimal, since other operators, such as \`, aren't disabled. */
# define RE_SYNTAX_POSIX_MINIMAL_BASIC \
# define RE_SYNTAX_POSIX_EXTENDED \
| RE_NO_BK_PARENS | RE_NO_BK_VBAR \
/* 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_NO_BK_PARENS | RE_NO_BK_REFS \
/* [[[end syntaxes]]] */
#endif /* defined __USE_GNU_REGEX */
#ifdef __USE_GNU_REGEX
/* Maximum number of duplicates an interval can allow. POSIX-conforming
systems might define this in <limits.h>, but we want our
value, so remove any previous define. */
# ifdef RE_DUP_MAX
# endif
/* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored
the counter as a 2-byte signed integer. This is no longer true, so
RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to
((SIZE_MAX - 2) / 10 - 1) if _REGEX_LARGE_OFFSETS is defined.
However, there would be a huge performance problem if someone
actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains
its historical value. */
#endif /* defined __USE_GNU_REGEX */
/* POSIX `cflags' bits (i.e., information for `regcomp'). */
/* If this bit is set, then use extended regular expression syntax.
If not set, then use basic regular expression syntax. */
/* If this bit is set, then ignore case when matching.
If not set, then case is significant. */
/* 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. */
/* If this bit is set, then report only success or fail in regexec.
If not set, then returns differ between not matching and errors. */
/* POSIX `eflags' bits (i.e., information for regexec). */
/* If this bit is set, then the beginning-of-line operator doesn't match
the beginning of the string (presumably because it's not the
beginning of a line).
If not set, then the beginning-of-line operator does match the
beginning of the string. */
/* Like REG_NOTBOL, except for the end-of-line. */
/* Use PMATCH[0] to delimit the start and end of the search in the
buffer. */
/* If any error codes are removed, changed, or added, update the
`__re_error_msgid' table in regcomp.c. */
typedef enum
{
/* POSIX regcomp return error codes. (In the order listed in the
standard.) */
/* Error codes we've added. */
#ifdef _XOPEN_SOURCE
#endif
/* struct re_pattern_buffer normally uses member names like `buffer'
that POSIX does not allow. In POSIX mode these members have names
with leading `re_' (e.g., `re_buffer'). */
#ifdef __USE_GNU_REGEX
#else
#endif
/* The user can specify the type of the re_translate member by
defining the macro RE_TRANSLATE_TYPE, which defaults to unsigned
char *. This pollutes the POSIX name space, so in POSIX mode just
use unsigned char *. */
#ifdef __USE_GNU_REGEX
# ifndef RE_TRANSLATE_TYPE
# define RE_TRANSLATE_TYPE unsigned char *
# endif
#else
# define REG_TRANSLATE_TYPE unsigned char *
#endif
/* This data structure represents a compiled pattern. Before calling
the pattern compiler, the fields `buffer', `allocated', `fastmap',
`translate', and `no_sub' can be set. After the pattern has been
compiled, the `re_nsub' field is available. All other fields are
private to the regex routines. */
struct re_pattern_buffer
{
/* Space that holds the compiled pattern. It is declared as
`unsigned char *' because its elements are sometimes used as
array indexes. */
unsigned char *_REG_RE_NAME (buffer);
/* Number of bytes to which `buffer' points. */
/* Number of bytes actually used in `buffer'. */
/* Syntax setting with which the pattern was compiled. */
/* Pointer to a fastmap, if any, otherwise zero. re_search uses the
fastmap, if there is one, to skip over impossible starting points
for matches. */
char *_REG_RE_NAME (fastmap);
/* Either a translate table to apply to all characters before
comparing them, or zero for no translation. The translation is
applied to a pattern when it is compiled and to a string when it
is matched. */
/* Number of subexpressions found by the compiler. */
/* Zero if this pattern cannot match the empty string, one else.
Well, in truth it's used only in `re_search_2', to see whether or
not we should use the fastmap, so we don't set this absolutely
perfectly; see `re_compile_fastmap' (the `duplicate' case). */
/* If REGS_UNALLOCATED, allocate space in the `regs' structure
for `max (RE_NREGS, re_nsub + 1)' groups.
If REGS_REALLOCATE, reallocate space if necessary.
If REGS_FIXED, use what's there. */
#ifdef __USE_GNU_REGEX
# define REGS_UNALLOCATED 0
#endif
/* Set to zero when `re_compile_pattern' compiles a pattern; set to
one by `re_compile_fastmap' if it updates the fastmap. */
/* If set, `re_match_2' does not return information about
subexpressions. */
/* If set, a beginning-of-line anchor doesn't match at the beginning
of the string. */
/* Similarly for an end-of-line anchor. */
/* If true, an anchor at a newline matches. */
/* [[[end pattern_buffer]]] */
};
/* This is the structure we store register match data in. See
regex.texinfo for a full description of what registers match. */
struct re_registers
{
};
/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
`re_match_2' returns information about at least this many registers
the first time a `regs' structure is passed. */
#if !defined RE_NREGS && defined __USE_GNU_REGEX
#endif
/* POSIX specification for registers. Aside from the different names than
`re_registers', POSIX uses an array of structures, instead of a
structure of arrays. */
typedef struct
{
} regmatch_t;
/* Declarations for routines. */
/* Sets the current default syntax to SYNTAX, and return the old syntax.
You can also simply assign to the `re_syntax_options' variable. */
/* Compile the regular expression PATTERN, with length LENGTH
and syntax given by the global `re_syntax_options', into the buffer
BUFFER. Return NULL if successful, and an error string if not. */
struct re_pattern_buffer *__buffer);
/* Compile a fastmap for the compiled pattern in BUFFER; used to
accelerate searches. Return 0 if successful and -2 if was an
internal error. */
/* Search in the string STRING (with length LENGTH) for the pattern
compiled into BUFFER. Start searching at position START, for RANGE
characters. Return the starting position of the match, -1 for no
match, or -2 for an internal error. Also return register
information in REGS (if REGS and BUFFER->no_sub are nonzero). */
struct re_registers *__regs);
/* Like `re_search', but search in the concatenation of STRING1 and
STRING2. Also, stop searching at index START + STOP. */
struct re_registers *__regs,
/* Like `re_search', but return how many characters in STRING the regexp
in BUFFER matched, starting at position START. */
/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
ENDS. Subsequent matches using BUFFER and REGS will use this memory
for recording register information. STARTS and ENDS must be
allocated with malloc, and must each be at least `NUM_REGS * sizeof
(regoff_t)' bytes long.
If NUM_REGS == 0, then subsequent matches should allocate their own
register data.
Unless this function is called, the first search or match using
BUFFER will allocate its own register data, without freeing the old
data. */
struct re_registers *__regs,
#if defined _REGEX_RE_COMP || defined _LIBC
# ifndef _CRAY
/* 4.2 bsd compatibility. */
extern char *re_comp (const char *);
extern int re_exec (const char *);
# endif
#endif
/* GCC 2.95 and later have "__restrict"; C99 compilers have
"restrict", and "configure" may have defined "restrict".
Other compilers use __restrict, __restrict__, and _Restrict, and
'configure' might #define 'restrict' to those words, so pick a
different name. */
#ifndef _Restrict_
# if 199901L <= __STDC_VERSION__
# define _Restrict_ restrict
# else
# define _Restrict_
# endif
#endif
/* gcc 3.1 and up support the [restrict] syntax. Don't trust
mishandles gcc -ansi -pedantic. */
#ifndef _Restrict_arr_
# if ((199901L <= __STDC_VERSION__ \
&& !__STRICT_ANSI__)) \
&& !defined __GNUG__)
# else
# define _Restrict_arr_
# endif
#endif
/* POSIX compatibility. */
const char *_Restrict_ __pattern,
int __cflags);
int __eflags);
#ifdef __cplusplus
}
#endif /* C++ */
#endif /* regex.h */