libguile/regex-posix.c

   1 /*      Copyright (C) 1997, 1998, 1999, 2000, 2001, 2004 Free Software Foundation, Inc.
   2  *
   3  * This library is free software; you can redistribute it and/or
   4  * modify it under the terms of the GNU Lesser General Public
   5  * License as published by the Free Software Foundation; either
   6  * version 2.1 of the License, or (at your option) any later version.
   7  *
   8  * This library 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 GNU
  11  * Lesser General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU Lesser General Public
  14  * License along with this library; if not, write to the Free Software
  15  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
  16  */
  17
  18
  19 \f
  20
  21 /* regex-posix.c -- POSIX regular expression support.
  22
  23    This code was written against Henry Spencer's famous regex package.
  24    The principal reference for POSIX behavior was the man page for this
  25    library, not the 1003.2 document itself.  Ergo, other `POSIX'
  26    libraries which do not agree with the Spencer implementation may
  27    produce varying behavior.  Sigh. */
  28
  29 #if HAVE_CONFIG_H
  30 #  include <config.h>
  31 #endif
  32
  33 #include <sys/types.h>
  34
  35 #include "libguile/_scm.h"
  36
  37 /* Supposedly, this file is never compiled unless we know we have
  38    POSIX regular expressions.  But we still put this in an #ifdef so
  39    the file is CPP'able (for dependency scanning) even on systems that
  40    don't have a <regex.h> header.  */
  41 #ifdef HAVE_REGCOMP
  42 #ifdef HAVE_REGEX_H
  43 #include <regex.h>
  44 #else
  45 #ifdef HAVE_RXPOSIX_H
  46 #include <rxposix.h>            /* GNU Rx library */
  47 #else
  48 #ifdef HAVE_RX_RXPOSIX_H
  49 #include <rx/rxposix.h>         /* GNU Rx library on Linux */
  50 #endif
  51 #endif
  52 #endif
  53 #endif
  54
  55 #include "libguile/smob.h"
  56 #include "libguile/symbols.h"
  57 #include "libguile/vectors.h"
  58 #include "libguile/strports.h"
  59 #include "libguile/ports.h"
  60 #include "libguile/feature.h"
  61 #include "libguile/strings.h"
  62
  63 #include "libguile/validate.h"
  64 #include "libguile/regex-posix.h"
  65
  66 /* This is defined by some regex libraries and omitted by others. */
  67 #ifndef REG_BASIC
  68 #define REG_BASIC 0
  69 #endif
  70
  71 scm_t_bits scm_tc16_regex;
  72
  73 static size_t
  74 regex_free (SCM obj)
  75 {
  76   regfree (SCM_RGX (obj));
  77   scm_gc_free (SCM_RGX (obj), sizeof(regex_t), "regex");
  78   return 0;
  79 }
  80
  81 \f
  82
  83 SCM_SYMBOL (scm_regexp_error_key, "regular-expression-syntax");
  84
  85 static SCM
  86 scm_regexp_error_msg (int regerrno, regex_t *rx)
  87 {
  88   char *errmsg;
  89   int l;
  90
  91   errmsg = scm_malloc (80);
  92   l = regerror (regerrno, rx, errmsg, 80);
  93   if (l > 80)
  94     {
  95       free (errmsg);
  96       errmsg = scm_malloc (l);
  97       regerror (regerrno, rx, errmsg, l);
  98     }
  99   return scm_take_locale_string (errmsg);
 100 }
 101
 102 SCM_DEFINE (scm_regexp_p, "regexp?", 1, 0, 0,
 103             (SCM obj),
 104             "Return @code{#t} if @var{obj} is a compiled regular expression,\n"
 105             "or @code{#f} otherwise.")
 106 #define FUNC_NAME s_scm_regexp_p
 107 {
 108   return scm_from_bool(SCM_RGXP (obj));
 109 }
 110 #undef FUNC_NAME
 111
 112 SCM_DEFINE (scm_make_regexp, "make-regexp", 1, 0, 1,
 113             (SCM pat, SCM flags),
 114             "Compile the regular expression described by @var{pat}, and\n"
 115             "return the compiled regexp structure.  If @var{pat} does not\n"
 116             "describe a legal regular expression, @code{make-regexp} throws\n"
 117             "a @code{regular-expression-syntax} error.\n"
 118             "\n"
 119             "The @var{flags} arguments change the behavior of the compiled\n"
 120             "regular expression.  The following flags may be supplied:\n"
 121             "\n"
 122             "@table @code\n"
 123             "@item regexp/icase\n"
 124             "Consider uppercase and lowercase letters to be the same when\n"
 125             "matching.\n"
 126             "@item regexp/newline\n"
 127             "If a newline appears in the target string, then permit the\n"
 128             "@samp{^} and @samp{$} operators to match immediately after or\n"
 129             "immediately before the newline, respectively.  Also, the\n"
 130             "@samp{.} and @samp{[^...]} operators will never match a newline\n"
 131             "character.  The intent of this flag is to treat the target\n"
 132             "string as a buffer containing many lines of text, and the\n"
 133             "regular expression as a pattern that may match a single one of\n"
 134             "those lines.\n"
 135             "@item regexp/basic\n"
 136             "Compile a basic (``obsolete'') regexp instead of the extended\n"
 137             "(``modern'') regexps that are the default.  Basic regexps do\n"
 138             "not consider @samp{|}, @samp{+} or @samp{?} to be special\n"
 139             "characters, and require the @samp{@{...@}} and @samp{(...)}\n"
 140             "metacharacters to be backslash-escaped (@pxref{Backslash\n"
 141             "Escapes}).  There are several other differences between basic\n"
 142             "and extended regular expressions, but these are the most\n"
 143             "significant.\n"
 144             "@item regexp/extended\n"
 145             "Compile an extended regular expression rather than a basic\n"
 146             "regexp.  This is the default behavior; this flag will not\n"
 147             "usually be needed.  If a call to @code{make-regexp} includes\n"
 148             "both @code{regexp/basic} and @code{regexp/extended} flags, the\n"
 149             "one which comes last will override the earlier one.\n"
 150             "@end table")
 151 #define FUNC_NAME s_scm_make_regexp
 152 {
 153   SCM flag;
 154   regex_t *rx;
 155   int status, cflags;
 156   char *c_pat;
 157
 158   SCM_VALIDATE_STRING (1, pat);
 159   SCM_VALIDATE_REST_ARGUMENT (flags);
 160
 161   /* Examine list of regexp flags.  If REG_BASIC is supplied, then
 162      turn off REG_EXTENDED flag (on by default). */
 163   cflags = REG_EXTENDED;
 164   flag = flags;
 165   while (!scm_is_null (flag))
 166     {
 167       if (scm_to_int (SCM_CAR (flag)) == REG_BASIC)
 168         cflags &= ~REG_EXTENDED;
 169       else
 170         cflags |= scm_to_int (SCM_CAR (flag));
 171       flag = SCM_CDR (flag);
 172     }
 173
 174   rx = scm_gc_malloc (sizeof(regex_t), "regex");
 175   c_pat = scm_to_locale_string (pat);
 176   status = regcomp (rx, c_pat,
 177                     /* Make sure they're not passing REG_NOSUB;
 178                        regexp-exec assumes we're getting match data.  */
 179                     cflags & ~REG_NOSUB);
 180   free (c_pat);
 181   if (status != 0)
 182     {
 183       SCM errmsg = scm_regexp_error_msg (status, rx);
 184       scm_gc_free (rx, sizeof(regex_t), "regex");
 185       scm_error_scm (scm_regexp_error_key,
 186                      scm_from_locale_string (FUNC_NAME),
 187                      errmsg,
 188                      SCM_BOOL_F,
 189                      SCM_BOOL_F);
 190       /* never returns */
 191     }
 192   SCM_RETURN_NEWSMOB (scm_tc16_regex, rx);
 193 }
 194 #undef FUNC_NAME
 195
 196 SCM_DEFINE (scm_regexp_exec, "regexp-exec", 2, 2, 0,
 197             (SCM rx, SCM str, SCM start, SCM flags),
 198             "Match the compiled regular expression @var{rx} against\n"
 199             "@code{str}.  If the optional integer @var{start} argument is\n"
 200             "provided, begin matching from that position in the string.\n"
 201             "Return a match structure describing the results of the match,\n"
 202             "or @code{#f} if no match could be found.\n"
 203             "\n"
 204             "The @var{flags} arguments change the matching behavior.\n"
 205             "The following flags may be supplied:\n"
 206             "\n"
 207             "@table @code\n"
 208             "@item regexp/notbol\n"
 209             "Operator @samp{^} always fails (unless @code{regexp/newline}\n"
 210             "is used).  Use this when the beginning of the string should\n"
 211             "not be considered the beginning of a line.\n"
 212             "@item regexp/noteol\n"
 213             "Operator @samp{$} always fails (unless @code{regexp/newline}\n"
 214             "is used).  Use this when the end of the string should not be\n"
 215             "considered the end of a line.\n"
 216             "@end table")
 217 #define FUNC_NAME s_scm_regexp_exec
 218 {
 219   int status, nmatches, offset;
 220   regmatch_t *matches;
 221   char *c_str;
 222   SCM mvec = SCM_BOOL_F;
 223
 224   SCM_VALIDATE_RGXP (1, rx);
 225   SCM_VALIDATE_STRING (2, str);
 226
 227   if (SCM_UNBNDP (start))
 228     offset = 0;
 229   else
 230     {
 231       str = scm_substring (str, start, SCM_UNDEFINED);
 232       offset = scm_to_int (start);
 233     }
 234
 235   if (SCM_UNBNDP (flags))
 236     flags = SCM_INUM0;
 237
 238   /* re_nsub doesn't account for the `subexpression' representing the
 239      whole regexp, so add 1 to nmatches. */
 240
 241   nmatches = SCM_RGX(rx)->re_nsub + 1;
 242   SCM_DEFER_INTS;
 243   matches = scm_malloc (sizeof (regmatch_t) * nmatches);
 244   c_str = scm_to_locale_string (str);
 245   status = regexec (SCM_RGX (rx), c_str, nmatches, matches,
 246                     scm_to_int (flags));
 247   free (c_str);
 248
 249   if (!status)
 250     {
 251       int i;
 252       /* The match vector must include a cell for the string that was matched,
 253          so add 1. */
 254       mvec = scm_c_make_vector (nmatches + 1, SCM_UNSPECIFIED);
 255       SCM_VECTOR_SET(mvec,0, str);
 256       for (i = 0; i < nmatches; ++i)
 257         if (matches[i].rm_so == -1)
 258           SCM_VECTOR_SET(mvec, i+1,
 259                          scm_cons (scm_from_int (-1), scm_from_int (-1)));
 260         else
 261           SCM_VECTOR_SET(mvec, i+1,
 262                          scm_cons (scm_from_long (matches[i].rm_so + offset),
 263                                    scm_from_long (matches[i].rm_eo + offset)));
 264     }
 265   free (matches);
 266   SCM_ALLOW_INTS;
 267
 268   if (status != 0 && status != REG_NOMATCH)
 269     scm_error_scm (scm_regexp_error_key,
 270                    scm_from_locale_string (FUNC_NAME),
 271                    scm_regexp_error_msg (status, SCM_RGX (rx)),
 272                    SCM_BOOL_F,
 273                    SCM_BOOL_F);
 274   return mvec;
 275 }
 276 #undef FUNC_NAME
 277
 278 void
 279 scm_init_regex_posix ()
 280 {
 281   scm_tc16_regex = scm_make_smob_type ("regexp", sizeof (regex_t));
 282   scm_set_smob_free (scm_tc16_regex, regex_free);
 283
 284   /* Compilation flags.  */
 285   scm_c_define ("regexp/basic", scm_from_long (REG_BASIC));
 286   scm_c_define ("regexp/extended", scm_from_long (REG_EXTENDED));
 287   scm_c_define ("regexp/icase", scm_from_long (REG_ICASE));
 288   scm_c_define ("regexp/newline", scm_from_long (REG_NEWLINE));
 289
 290   /* Execution flags.  */
 291   scm_c_define ("regexp/notbol", scm_from_long (REG_NOTBOL));
 292   scm_c_define ("regexp/noteol", scm_from_long (REG_NOTEOL));
 293
 294 #include "libguile/regex-posix.x"
 295
 296   scm_add_feature ("regex");
 297 }
 298
 299 /*
 300   Local Variables:
 301   c-file-style: "gnu"
 302   End:
 303 */