libguile/regex-posix.c

   1 /*      Copyright (C) 1997, 1998, 1999, 2000 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 <sys/types.h>
  57
  58 #include "libguile/_scm.h"
  59
  60 /* Supposedly, this file is never compiled unless we know we have
  61    POSIX regular expressions.  But we still put this in an #ifdef so
  62    the file is CPP'able (for dependency scanning) even on systems that
  63    don't have a <regex.h> header.  */
  64 #ifdef HAVE_REGCOMP
  65 #ifdef HAVE_REGEX_H
  66 #include <regex.h>
  67 #else
  68 #ifdef HAVE_RXPOSIX_H
  69 #include <rxposix.h>            /* GNU Rx library */
  70 #else
  71 #ifdef HAVE_RX_RXPOSIX_H
  72 #include <rx/rxposix.h>         /* GNU Rx library on Linux */
  73 #endif
  74 #endif
  75 #endif
  76 #endif
  77
  78 #include "libguile/smob.h"
  79 #include "libguile/symbols.h"
  80 #include "libguile/vectors.h"
  81 #include "libguile/strports.h"
  82 #include "libguile/ports.h"
  83 #include "libguile/feature.h"
  84 #include "libguile/strings.h"
  85
  86 #include "libguile/validate.h"
  87 #include "libguile/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 scm_bits_t scm_tc16_regex;
  95
  96 static scm_sizet
  97 regex_free (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_STRING_CHARS (errmsg), 80);
 127   if (l > 80)
 128     {
 129       errmsg = scm_make_string (SCM_MAKINUM (l), SCM_UNDEFINED);
 130       regerror (regerrno, rx, SCM_STRING_CHARS (errmsg), l);
 131     }
 132   SCM_ALLOW_INTS;
 133   return SCM_STRING_CHARS (errmsg);
 134 }
 135
 136 SCM_DEFINE (scm_regexp_p, "regexp?", 1, 0, 0,
 137             (SCM x),
 138             "Return @code{#t} if @var{obj} is a compiled regular expression, or\n"
 139             "@code{#f} otherwise.")
 140 #define FUNC_NAME s_scm_regexp_p
 141 {
 142   return SCM_BOOL(SCM_RGXP (x));
 143 }
 144 #undef FUNC_NAME
 145
 146 SCM_DEFINE (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\n"
 149             "compiled regexp structure.  If @var{str} does not describe a legal\n"
 150             "regular expression, @code{make-regexp} throws a\n"
 151             "@code{regular-expression-syntax} error.\n\n"
 152             "The @var{flag} arguments change the behavior of the compiled regexp.\n"
 153             "The following flags may be supplied:\n\n"
 154             "@table @code\n"
 155             "@item regexp/icase\n"
 156             "Consider uppercase and lowercase letters to be the same when matching.\n\n"
 157             "@item regexp/newline\n"
 158             "If a newline appears in the target string, then permit the @samp{^} and\n"
 159             "@samp{$} operators to match immediately after or immediately before the\n"
 160             "newline, respectively.  Also, the @samp{.} and @samp{[^...]} operators\n"
 161             "will never match a newline character.  The intent of this flag is to\n"
 162             "treat the target string as a buffer containing many lines of text, and\n"
 163             "the regular expression as a pattern that may match a single one of those\n"
 164             "lines.\n\n"
 165             "@item regexp/basic\n"
 166             "Compile a basic (``obsolete'') regexp instead of the extended\n"
 167             "(``modern'') regexps that are the default.  Basic regexps do not\n"
 168             "consider @samp{|}, @samp{+} or @samp{?} to be special characters, and\n"
 169             "require the @samp{@{...@}} and @samp{(...)} metacharacters to be\n"
 170             "backslash-escaped (@pxref{Backslash Escapes}).  There are several other\n"
 171             "differences between basic and extended regular expressions, but these\n"
 172             "are the most significant.\n\n"
 173             "@item regexp/extended\n"
 174             "Compile an extended regular expression rather than a basic regexp.  This\n"
 175             "is the default behavior; this flag will not usually be needed.  If a\n"
 176             "call to @code{make-regexp} includes both @code{regexp/basic} and\n"
 177             "@code{regexp/extended} flags, the one which comes last will override\n"
 178             "the earlier one.\n"
 179             "@end table\n")
 180 #define FUNC_NAME s_scm_make_regexp
 181 {
 182   SCM flag;
 183   regex_t *rx;
 184   int status, cflags;
 185
 186   SCM_VALIDATE_STRING (1, pat);
 187   SCM_VALIDATE_REST_ARGUMENT (flags);
 188   SCM_STRING_COERCE_0TERMINATION_X (pat);
 189
 190   /* Examine list of regexp flags.  If REG_BASIC is supplied, then
 191      turn off REG_EXTENDED flag (on by default). */
 192   cflags = REG_EXTENDED;
 193   flag = flags;
 194   while (!SCM_NULLP (flag))
 195     {
 196       if (SCM_INUM (SCM_CAR (flag)) == REG_BASIC)
 197         cflags &= ~REG_EXTENDED;
 198       else
 199         cflags |= SCM_INUM (SCM_CAR (flag));
 200       flag = SCM_CDR (flag);
 201     }
 202
 203   rx = SCM_MUST_MALLOC_TYPE (regex_t);
 204   status = regcomp (rx, SCM_STRING_CHARS (pat),
 205                     /* Make sure they're not passing REG_NOSUB;
 206                        regexp-exec assumes we're getting match data.  */
 207                     cflags & ~REG_NOSUB);
 208   if (status != 0)
 209     {
 210       scm_error (scm_regexp_error_key,
 211                  FUNC_NAME,
 212                  scm_regexp_error_msg (status, rx),
 213                  SCM_BOOL_F,
 214                  SCM_BOOL_F);
 215       /* never returns */
 216     }
 217   SCM_RETURN_NEWSMOB (scm_tc16_regex, rx);
 218 }
 219 #undef FUNC_NAME
 220
 221 SCM_DEFINE (scm_regexp_exec, "regexp-exec", 2, 2, 0,
 222             (SCM rx, SCM str, SCM start, SCM flags),
 223             "Match the compiled regular expression @var{regexp} against @code{str}.\n"
 224             "If the optional integer @var{start} argument is provided, begin matching\n"
 225             "from that position in the string.  Return a match structure describing\n"
 226             "the results of the match, or @code{#f} if no match could be found.")
 227 #define FUNC_NAME s_scm_regexp_exec
 228 {
 229   int status, nmatches, offset;
 230   regmatch_t *matches;
 231   SCM mvec = SCM_BOOL_F;
 232
 233   SCM_VALIDATE_RGXP (1,rx);
 234   SCM_VALIDATE_STRING (2, str);
 235   SCM_VALIDATE_INUM_DEF_COPY (3,start,0,offset);
 236   SCM_ASSERT_RANGE (3,start, offset >= 0 && offset <= SCM_STRING_LENGTH (str));
 237   if (SCM_UNBNDP (flags))
 238     flags = SCM_INUM0;
 239   SCM_VALIDATE_INUM (4,flags);
 240   SCM_STRING_COERCE_0TERMINATION_X (str);
 241
 242   /* re_nsub doesn't account for the `subexpression' representing the
 243      whole regexp, so add 1 to nmatches. */
 244
 245   nmatches = SCM_RGX(rx)->re_nsub + 1;
 246   SCM_DEFER_INTS;
 247   matches = SCM_MUST_MALLOC_TYPE_NUM (regmatch_t,nmatches);
 248   status = regexec (SCM_RGX (rx), SCM_STRING_CHARS (str) + offset,
 249                     nmatches, matches,
 250                     SCM_INUM (flags));
 251   if (!status)
 252     {
 253       int i;
 254       /* The match vector must include a cell for the string that was matched,
 255          so add 1. */
 256       mvec = scm_c_make_vector (nmatches + 1, SCM_UNSPECIFIED);
 257       SCM_VELTS(mvec)[0] = str;
 258       for (i = 0; i < nmatches; ++i)
 259         if (matches[i].rm_so == -1)
 260           SCM_VELTS(mvec)[i+1] = scm_cons (SCM_MAKINUM (-1), SCM_MAKINUM (-1));
 261         else
 262           SCM_VELTS(mvec)[i+1]
 263             = scm_cons(SCM_MAKINUM(matches[i].rm_so + offset),
 264                        SCM_MAKINUM(matches[i].rm_eo + offset));
 265     }
 266   scm_must_free ((char *) matches);
 267   SCM_ALLOW_INTS;
 268
 269   if (status != 0 && status != REG_NOMATCH)
 270     scm_error (scm_regexp_error_key,
 271                FUNC_NAME,
 272                scm_regexp_error_msg (status, SCM_RGX (rx)),
 273                SCM_BOOL_F,
 274                SCM_BOOL_F);
 275   return mvec;
 276 }
 277 #undef FUNC_NAME
 278
 279 void
 280 scm_init_regex_posix ()
 281 {
 282   scm_tc16_regex = scm_make_smob_type ("regexp", sizeof (regex_t));
 283   scm_set_smob_free (scm_tc16_regex, regex_free);
 284
 285   /* Compilation flags.  */
 286   scm_sysintern ("regexp/basic", scm_long2num (REG_BASIC));
 287   scm_sysintern ("regexp/extended", scm_long2num (REG_EXTENDED));
 288   scm_sysintern ("regexp/icase", scm_long2num (REG_ICASE));
 289   scm_sysintern ("regexp/newline", scm_long2num (REG_NEWLINE));
 290
 291   /* Execution flags.  */
 292   scm_sysintern ("regexp/notbol", scm_long2num (REG_NOTBOL));
 293   scm_sysintern ("regexp/noteol", scm_long2num (REG_NOTEOL));
 294
 295 #ifndef SCM_MAGIC_SNARFER
 296 #include "libguile/regex-posix.x"
 297 #endif
 298
 299   scm_add_feature ("regex");
 300 }
 301
 302 /*
 303   Local Variables:
 304   c-file-style: "gnu"
 305   End:
 306 */