misc/regexp.h

   1 /* Copyright (C) 1996, 1997, 1998, 1999, 2004, 2008
   2    Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Ulrich Drepper <drepper@cygnus.com>, 1996.
   5
   6    The GNU C Library is free software; you can redistribute it and/or
   7    modify it under the terms of the GNU Lesser General Public
   8    License as published by the Free Software Foundation; either
   9    version 2.1 of the License, or (at your option) any later version.
  10
  11    The GNU C Library is distributed in the hope that it will be useful,
  12    but WITHOUT ANY WARRANTY; without even the implied warranty of
  13    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14    Lesser General Public License for more details.
  15
  16    You should have received a copy of the GNU Lesser General Public
  17    License along with the GNU C Library; if not, write to the Free
  18    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  19    02111-1307 USA.  */
  20
  21 #ifndef _REGEXP_H
  22 #define _REGEXP_H       1
  23
  24 /* The contents of this header file was first standardized in X/Open
  25    System Interface and Headers Issue 2, originally coming from SysV.
  26    In issue 4, version 2, it is marked as TO BE WITDRAWN, and it has
  27    been withdrawn in SUSv3.
  28
  29    This code shouldn't be used in any newly written code.  It is
  30    included only for compatibility reasons.  Use the POSIX definition
  31    in <regex.h> for portable applications and a reasonable interface.  */
  32
  33 #include <features.h>
  34 #include <alloca.h>
  35 #include <regex.h>
  36 #include <stdlib.h>
  37 #include <string.h>
  38
  39 /* The implementation provided here emulates the needed functionality
  40    by mapping to the POSIX regular expression matcher.  The interface
  41    for the here included function is weird (this really is a harmless
  42    word).
  43
  44    The user has to provide six macros before this header file can be
  45    included:
  46
  47    INIT         Declarations vor variables which can be used by the
  48                 other macros.
  49
  50    GETC()       Return the value of the next character in the regular
  51                 expression pattern.  Successive calls should return
  52                 successive characters.
  53
  54    PEEKC()      Return the value of the next character in the regular
  55                 expression pattern.  Immediately successive calls to
  56                 PEEKC() should return the same character which should
  57                 also be the next character returned by GETC().
  58
  59    UNGETC(c)    Cause `c' to be returned by the next call to GETC() and
  60                 PEEKC().
  61
  62    RETURN(ptr)  Used for normal exit of the `compile' function.  `ptr'
  63                 is a pointer to the character after the last character of
  64                 the compiled regular expression.
  65
  66    ERROR(val)   Used for abnormal return from `compile'.  `val' is the
  67                 error number.  The error codes are:
  68                 11      Range endpoint too large.
  69                 16      Bad number.
  70                 25      \digit out of range.
  71                 36      Illegal or missing delimiter.
  72                 41      No remembered search string.
  73                 42      \( \) imbalance.
  74                 43      Too many \(.
  75                 44      More tan two numbers given in \{ \}.
  76                 45      } expected after \.
  77                 46      First number exceeds second in \{ \}.
  78                 49      [ ] imbalance.
  79                 50      Regular expression overflow.
  80
  81   */
  82
  83 __BEGIN_DECLS
  84
  85 /* Interface variables.  They contain the results of the successful
  86    calls to `setp' and `advance'.  */
  87 extern char *loc1;
  88 extern char *loc2;
  89
  90 /* The use of this variable in the `advance' function is not
  91    supported.  */
  92 extern char *locs;
  93
  94
  95 #ifndef __DO_NOT_DEFINE_COMPILE
  96 /* Get and compile the user supplied pattern up to end of line or
  97    string or until EOF is seen, whatever happens first.  The result is
  98    placed in the buffer starting at EXPBUF and delimited by ENDBUF.
  99
 100    This function cannot be defined in the libc itself since it depends
 101    on the macros.  */
 102 char *
 103 compile (char *__restrict instring, char *__restrict expbuf,
 104          __const char *__restrict endbuf, int eof)
 105 {
 106   char *__input_buffer = NULL;
 107   size_t __input_size = 0;
 108   size_t __current_size = 0;
 109   int __ch;
 110   int __error;
 111   INIT
 112
 113   /* Align the expression buffer according to the needs for an object
 114      of type `regex_t'.  Then check for minimum size of the buffer for
 115      the compiled regular expression.  */
 116   regex_t *__expr_ptr;
 117 # if defined __GNUC__ && __GNUC__ >= 2
 118   const size_t __req = __alignof__ (regex_t *);
 119 # else
 120   /* How shall we find out?  We simply guess it and can change it is
 121      this really proofs to be wrong.  */
 122   const size_t __req = 8;
 123 # endif
 124   expbuf += __req;
 125   expbuf -= (expbuf - ((char *) 0)) % __req;
 126   if (endbuf < expbuf + sizeof (regex_t))
 127     {
 128       ERROR (50);
 129     }
 130   __expr_ptr = (regex_t *) expbuf;
 131   /* The remaining space in the buffer can be used for the compiled
 132      pattern.  */
 133   __expr_ptr->__REPB_PREFIX (buffer) = expbuf + sizeof (regex_t);
 134   __expr_ptr->__REPB_PREFIX (allocated)
 135     = endbuf - (char *) __expr_ptr->__REPB_PREFIX (buffer);
 136
 137   while ((__ch = (GETC ())) != eof)
 138     {
 139       if (__ch == '\0' || __ch == '\n')
 140         {
 141           UNGETC (__ch);
 142           break;
 143         }
 144
 145       if (__current_size + 1 >= __input_size)
 146         {
 147           size_t __new_size = __input_size ? 2 * __input_size : 128;
 148           char *__new_room = (char *) alloca (__new_size);
 149           /* See whether we can use the old buffer.  */
 150           if (__new_room + __new_size == __input_buffer)
 151             {
 152               __input_size += __new_size;
 153               __input_buffer = (char *) memcpy (__new_room, __input_buffer,
 154                                                __current_size);
 155             }
 156           else if (__input_buffer + __input_size == __new_room)
 157             __input_size += __new_size;
 158           else
 159             {
 160               __input_size = __new_size;
 161               __input_buffer = (char *) memcpy (__new_room, __input_buffer,
 162                                                 __current_size);
 163             }
 164         }
 165       __input_buffer[__current_size++] = __ch;
 166     }
 167   if (__current_size)
 168     __input_buffer[__current_size++] = '\0';
 169   else
 170     __input_buffer = "";
 171
 172   /* Now compile the pattern.  */
 173   __error = regcomp (__expr_ptr, __input_buffer, REG_NEWLINE);
 174   if (__error != 0)
 175     /* Oh well, we have to translate POSIX error codes.  */
 176     switch (__error)
 177       {
 178       case REG_BADPAT:
 179       case REG_ECOLLATE:
 180       case REG_ECTYPE:
 181       case REG_EESCAPE:
 182       case REG_BADRPT:
 183       case REG_EEND:
 184       case REG_ERPAREN:
 185       default:
 186         /* There is no matching error code.  */
 187         RETURN (36);
 188       case REG_ESUBREG:
 189         RETURN (25);
 190       case REG_EBRACK:
 191         RETURN (49);
 192       case REG_EPAREN:
 193         RETURN (42);
 194       case REG_EBRACE:
 195         RETURN (44);
 196       case REG_BADBR:
 197         RETURN (46);
 198       case REG_ERANGE:
 199         RETURN (11);
 200       case REG_ESPACE:
 201       case REG_ESIZE:
 202         ERROR (50);
 203       }
 204
 205   /* Everything is ok.  */
 206   RETURN ((char *) (__expr_ptr->__REPB_PREFIX (buffer)
 207                     + __expr_ptr->__REPB_PREFIX (used)));
 208 }
 209 #endif
 210
 211
 212 /* Find the next match in STRING.  The compiled regular expression is
 213    found in the buffer starting at EXPBUF.  `loc1' will return the
 214    first character matched and `loc2' points to the next unmatched
 215    character.  */
 216 extern int step (__const char *__restrict __string,
 217                  __const char *__restrict __expbuf) __THROW;
 218
 219 /* Match the beginning of STRING with the compiled regular expression
 220    in EXPBUF.  If the match is successful `loc2' will contain the
 221    position of the first unmatched character.  */
 222 extern int advance (__const char *__restrict __string,
 223                     __const char *__restrict __expbuf) __THROW;
 224
 225
 226 __END_DECLS
 227
 228 #endif /* regexp.h */