libguile/regex-posix.c

   1 /*      Copyright (C) 1997, 1998, 1999 Free Software Foundation, Inc.
   2  *
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2, or (at your option)
   6  * any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License
  14  * along with this software; see the file COPYING.  If not, write to
  15  * the Free Software Foundation, Inc., 59 Temple Place, Suite 330,
  16  * Boston, MA 02111-1307 USA
  17  *
  18  * As a special exception, the Free Software Foundation gives permission
  19  * for additional uses of the text contained in its release of GUILE.
  20  *
  21  * The exception is that, if you link the GUILE library with other files
  22  * to produce an executable, this does not by itself cause the
  23  * resulting executable to be covered by the GNU General Public License.
  24  * Your use of that executable is in no way restricted on account of
  25  * linking the GUILE library code into it.
  26  *
  27  * This exception does not however invalidate any other reasons why
  28  * the executable file might be covered by the GNU General Public License.
  29  *
  30  * This exception applies only to the code released by the
  31  * Free Software Foundation under the name GUILE.  If you copy
  32  * code from other Free Software Foundation releases into a copy of
  33  * GUILE, as the General Public License permits, the exception does
  34  * not apply to the code that you add in this way.  To avoid misleading
  35  * anyone as to the status of such modified files, you must delete
  36  * this exception notice from them.
  37  *
  38  * If you write modifications of your own for GUILE, it is your choice
  39  * whether to permit this exception to apply to your modifications.
  40  * If you do not wish that, delete this exception notice.
  41  */
  42
  43 /* Software engineering face-lift by Greg J. Badros, 11-Dec-1999,
  44    gjb@cs.washington.edu, http://www.cs.washington.edu/homes/gjb */
  45
  46 \f
  47
  48 /* regex-posix.c -- POSIX regular expression support.
  49
  50    This code was written against Henry Spencer's famous regex package.
  51    The principal reference for POSIX behavior was the man page for this
  52    library, not the 1003.2 document itself.  Ergo, other `POSIX'
  53    libraries which do not agree with the Spencer implementation may
  54    produce varying behavior.  Sigh. */
  55
  56 #include <stdio.h>
  57 #include <sys/types.h>
  58
  59 #include "_scm.h"
  60
  61 /* Supposedly, this file is never compiled unless we know we have
  62    POSIX regular expressions.  But we still put this in an #ifdef so
  63    the file is CPP'able (for dependency scanning) even on systems that
  64    don't have a <regex.h> header.  */
  65 #ifdef HAVE_REGCOMP
  66 #ifdef HAVE_REGEX_H
  67 #include <regex.h>
  68 #else
  69 #ifdef HAVE_RXPOSIX_H
  70 #include <rxposix.h>            /* GNU Rx library */
  71 #else
  72 #ifdef HAVE_RX_RXPOSIX_H
  73 #include <rx/rxposix.h>         /* GNU Rx library on Linux */
  74 #endif
  75 #endif
  76 #endif
  77 #endif
  78
  79 #include "smob.h"
  80 #include "symbols.h"
  81 #include "vectors.h"
  82 #include "strports.h"
  83 #include "ports.h"
  84 #include "feature.h"
  85
  86 #include "scm_validate.h"
  87 #include "regex-posix.h"
  88
  89 /* This is defined by some regex libraries and omitted by others. */
  90 #ifndef REG_BASIC
  91 #define REG_BASIC 0
  92 #endif
  93
  94 long scm_tc16_regex;
  95
  96 static scm_sizet
  97 free_regex (SCM obj)
  98 {
  99   regfree (SCM_RGX (obj));
 100   free (SCM_RGX (obj));
 101   return sizeof(regex_t);
 102 }
 103
 104 \f
 105
 106 SCM_SYMBOL (scm_regexp_error_key, "regular-expression-syntax");
 107
 108 static char *
 109 scm_regexp_error_msg (int regerrno, regex_t *rx)
 110 {
 111   SCM errmsg;
 112   int l;
 113
 114   /* FIXME: must we wrap any external calls in SCM_DEFER_INTS...SCM_ALLOW_INTS?
 115      Or are these only necessary when a SCM object may be left in an
 116      undetermined state (half-formed)?  If the latter then I believe we
 117      may do without the critical section code. -twp */
 118
 119   /* We could simply make errmsg a char pointer, and allocate space with
 120      malloc.  But since we are about to pass the pointer to scm_error, which
 121      never returns, we would never have the opportunity to free it.  Creating
 122      it as a SCM object means that the system will GC it at some point. */
 123
 124   errmsg = scm_make_string (SCM_MAKINUM (80), SCM_UNDEFINED);
 125   SCM_DEFER_INTS;
 126   l = regerror (regerrno, rx, SCM_CHARS (errmsg), 80);
 127   if (l > 80)
 128     {
 129       errmsg = scm_make_string (SCM_MAKINUM (l), SCM_UNDEFINED);
 130       regerror (regerrno, rx, SCM_CHARS (errmsg), l);
 131     }
 132   SCM_ALLOW_INTS;
 133   return SCM_CHARS (errmsg);
 134 }
 135
 136 GUILE_PROC (scm_regexp_p, "regexp?", 1, 0, 0,
 137             (SCM x),
 138 "Return @code{#t} if @var{obj} is a compiled regular expression, or
 139 @code{#f} otherwise.")
 140 #define FUNC_NAME s_scm_regexp_p
 141 {
 142   return SCM_BOOL(SCM_NIMP (x) && SCM_RGXP (x));
 143 }
 144 #undef FUNC_NAME
 145
 146 GUILE_PROC (scm_make_regexp, "make-regexp", 1, 0, 1,
 147             (SCM pat, SCM flags),
 148 "Compile the regular expression described by @var{str}, and return the
 149 compiled regexp structure.  If @var{str} does not describe a legal
 150 regular expression, @code{make-regexp} throws a
 151 @code{regular-expression-syntax} error.
 152
 153 The @var{flag} arguments change the behavior of the compiled regexp.
 154 The following flags may be supplied:
 155
 156 @table @code
 157 @item regexp/icase
 158 Consider uppercase and lowercase letters to be the same when matching.
 159
 160 @item regexp/newline
 161 If a newline appears in the target string, then permit the @samp{^} and
 162 @samp{$} operators to match immediately after or immediately before the
 163 newline, respectively.  Also, the @samp{.} and @samp{[^...]} operators
 164 will never match a newline character.  The intent of this flag is to
 165 treat the target string as a buffer containing many lines of text, and
 166 the regular expression as a pattern that may match a single one of those
 167 lines.
 168
 169 @item regexp/basic
 170 Compile a basic (``obsolete'') regexp instead of the extended
 171 (``modern'') regexps that are the default.  Basic regexps do not
 172 consider @samp{|}, @samp{+} or @samp{?} to be special characters, and
 173 require the @samp{@{...@}} and @samp{(...)} metacharacters to be
 174 backslash-escaped (@pxref{Backslash Escapes}).  There are several other
 175 differences between basic and extended regular expressions, but these
 176 are the most significant.
 177
 178 @item regexp/extended
 179 Compile an extended regular expression rather than a basic regexp.  This
 180 is the default behavior; this flag will not usually be needed.  If a
 181 call to @code{make-regexp} includes both @code{regexp/basic} and
 182 @code{regexp/extended} flags, the one which comes last will override
 183 the earlier one.
 184 @end table
 185 ")
 186 #define FUNC_NAME s_scm_make_regexp
 187 {
 188   SCM flag;
 189   regex_t *rx;
 190   int status, cflags;
 191
 192   SCM_VALIDATE_ROSTRING(1,pat);
 193   SCM_COERCE_SUBSTR (pat);
 194
 195   /* Examine list of regexp flags.  If REG_BASIC is supplied, then
 196      turn off REG_EXTENDED flag (on by default). */
 197   cflags = REG_EXTENDED;
 198   flag = flags;
 199   while (SCM_NNULLP (flag))
 200     {
 201       if (SCM_INUM (SCM_CAR (flag)) == REG_BASIC)
 202         cflags &= ~REG_EXTENDED;
 203       else
 204         cflags |= SCM_INUM (SCM_CAR (flag));
 205       flag = SCM_CDR (flag);
 206     }
 207
 208   rx = SCM_MUST_MALLOC_TYPE(regex_t);
 209   status = regcomp (rx, SCM_ROCHARS (pat),
 210                     /* Make sure they're not passing REG_NOSUB;
 211                        regexp-exec assumes we're getting match data.  */
 212                     cflags & ~REG_NOSUB);
 213   if (status != 0)
 214     {
 215       scm_error (scm_regexp_error_key,
 216                  FUNC_NAME,
 217                  scm_regexp_error_msg (status, rx),
 218                  SCM_BOOL_F,
 219                  SCM_BOOL_F);
 220       /* never returns */
 221     }
 222   SCM_RETURN_NEWSMOB (scm_tc16_regex, rx);
 223 }
 224 #undef FUNC_NAME
 225
 226 GUILE_PROC (scm_regexp_exec, "regexp-exec", 2, 2, 0,
 227             (SCM rx, SCM str, SCM start, SCM flags),
 228 "Match the compiled regular expression @var{regexp} against @code{str}.
 229 If the optional integer @var{start} argument is provided, begin matching
 230 from that position in the string.  Return a match structure describing
 231 the results of the match, or @code{#f} if no match could be found.")
 232 #define FUNC_NAME s_scm_regexp_exec
 233 {
 234   int status, nmatches, offset;
 235   regmatch_t *matches;
 236   SCM mvec = SCM_BOOL_F;
 237
 238   SCM_VALIDATE_RGXP(1,rx);
 239   SCM_VALIDATE_ROSTRING(2,str);
 240   SCM_VALIDATE_INT_DEF_COPY(3,start,0,offset);
 241   SCM_ASSERT_RANGE (3,start,offset >= 0 && (unsigned) offset <= SCM_LENGTH (str));
 242   if (SCM_UNBNDP (flags))
 243     flags = SCM_INUM0;
 244   SCM_VALIDATE_INT(4,flags);
 245   SCM_COERCE_SUBSTR (str);
 246
 247   /* re_nsub doesn't account for the `subexpression' representing the
 248      whole regexp, so add 1 to nmatches. */
 249
 250   nmatches = SCM_RGX(rx)->re_nsub + 1;
 251   SCM_DEFER_INTS;
 252   matches = SCM_MUST_MALLOC_TYPE_NUM (regmatch_t,nmatches);
 253   status = regexec (SCM_RGX (rx), SCM_ROCHARS (str) + offset,
 254                     nmatches, matches,
 255                     SCM_INUM (flags));
 256   if (!status)
 257     {
 258       int i;
 259       /* The match vector must include a cell for the string that was matched,
 260          so add 1. */
 261       mvec = scm_make_vector (SCM_MAKINUM (nmatches + 1), SCM_UNSPECIFIED);
 262       SCM_VELTS(mvec)[0] = str;
 263       for (i = 0; i < nmatches; ++i)
 264         if (matches[i].rm_so == -1)
 265           SCM_VELTS(mvec)[i+1] = scm_cons (SCM_MAKINUM (-1), SCM_MAKINUM (-1));
 266         else
 267           SCM_VELTS(mvec)[i+1]
 268             = scm_cons(SCM_MAKINUM(matches[i].rm_so + offset),
 269                        SCM_MAKINUM(matches[i].rm_eo + offset));
 270     }
 271   scm_must_free ((char *) matches);
 272   SCM_ALLOW_INTS;
 273
 274   if (status != 0 && status != REG_NOMATCH)
 275     scm_error (scm_regexp_error_key,
 276                FUNC_NAME,
 277                scm_regexp_error_msg (status, SCM_RGX (rx)),
 278                SCM_BOOL_F,
 279                SCM_BOOL_F);
 280   return mvec;
 281 }
 282 #undef FUNC_NAME
 283
 284 void
 285 scm_init_regex_posix ()
 286 {
 287   scm_tc16_regex = scm_make_smob_type_mfpe ("regexp", sizeof (regex_t),
 288                                             NULL, free_regex, NULL, NULL);
 289
 290   /* Compilation flags.  */
 291   scm_sysintern ("regexp/basic", scm_long2num (REG_BASIC));
 292   scm_sysintern ("regexp/extended", scm_long2num (REG_EXTENDED));
 293   scm_sysintern ("regexp/icase", scm_long2num (REG_ICASE));
 294   scm_sysintern ("regexp/newline", scm_long2num (REG_NEWLINE));
 295
 296   /* Execution flags.  */
 297   scm_sysintern ("regexp/notbol", scm_long2num (REG_NOTBOL));
 298   scm_sysintern ("regexp/noteol", scm_long2num (REG_NOTEOL));
 299
 300 #include "regex-posix.x"
 301
 302   scm_add_feature ("regex");
 303 }