1 /* Copyright (C) 1997 Free Software Foundation, Inc.
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)
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.
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, 675 Mass Ave, Cambridge, MA 02139, USA.
17 * As a special exception, the Free Software Foundation gives permission
18 * for additional uses of the text contained in its release of GUILE.
20 * The exception is that, if you link the GUILE library with other files
21 * to produce an executable, this does not by itself cause the
22 * resulting executable to be covered by the GNU General Public License.
23 * Your use of that executable is in no way restricted on account of
24 * linking the GUILE library code into it.
26 * This exception does not however invalidate any other reasons why
27 * the executable file might be covered by the GNU General Public License.
29 * This exception applies only to the code released by the
30 * Free Software Foundation under the name GUILE. If you copy
31 * code from other Free Software Foundation releases into a copy of
32 * GUILE, as the General Public License permits, the exception does
33 * not apply to the code that you add in this way. To avoid misleading
34 * anyone as to the status of such modified files, you must delete
35 * this exception notice from them.
37 * If you write modifications of your own for GUILE, it is your choice
38 * whether to permit this exception to apply to your modifications.
39 * If you do not wish that, delete this exception notice.
43 /* regex-posix.c -- POSIX regular expression support.
45 This code was written against Henry Spencer's famous regex package.
46 The principal reference for POSIX behavior was the man page for this
47 library, not the 1003.2 document itself. Ergo, other `POSIX'
48 libraries which do not agree with the Spencer implementation may
49 produce varying behavior. Sigh. */
52 #include <sys/types.h>
56 /* Supposedly, this file is never compiled unless we know we have
57 POSIX regular expressions. But we still put this in an #ifdef so
58 the file is CPP'able (for dependency scanning) even on systems that
59 don't have a <regex.h> header. */
71 #include "regex-posix.h"
73 long scm_tc16_regex_t
;
76 scm_free_regex_t (obj
)
79 regfree (SCM_RGX (obj
));
85 scm_print_regex_t (obj
, port
, pstate
)
88 scm_print_state
*pstate
;
92 scm_gen_puts (scm_regular_string
, "#<rgx ", port
);
93 scm_intprint (obj
, 16, port
);
94 scm_gen_puts (scm_regular_string
, ">", port
);
99 static scm_smobfuns regex_t_smob
=
100 { scm_mark0
, scm_free_regex_t
, scm_print_regex_t
, 0 };
103 SCM_SYMBOL (scm_regexp_error_key
, "regular-expression-syntax");
106 scm_regexp_error_msg (regerrno
, rx
)
113 /* FIXME: must we wrap any external calls in SCM_DEFER_INTS...SCM_ALLOW_INTS?
114 Or are these only necessary when a SCM object may be left in an
115 undetermined state (half-formed)? If the latter then I believe we
116 may do without the critical section code. -twp */
118 /* We could simply make errmsg a char pointer, and allocate space with
119 malloc. But since we are about to pass the pointer to scm_error, which
120 never returns, we would never have the opportunity to free it. Creating
121 it as a SCM object means that the system will GC it at some point. */
123 errmsg
= scm_make_string (SCM_MAKINUM (80), SCM_UNDEFINED
);
125 l
= regerror (regerrno
, SCM_RGX (rx
), SCM_CHARS (errmsg
), 80);
128 errmsg
= scm_make_string (SCM_MAKINUM (l
), SCM_UNDEFINED
);
129 regerror (regerrno
, SCM_RGX (rx
), SCM_CHARS (errmsg
), l
);
132 return SCM_CHARS (errmsg
);
135 SCM_PROC (s_regexp_p
, "regexp?", 1, 0, 0, scm_regexp_p
);
141 return (SCM_NIMP (x
) && SCM_RGXP (x
) ? SCM_BOOL_T
: SCM_BOOL_F
);
144 SCM_PROC (s_make_regexp
, "make-regexp", 1, 1, 0, scm_make_regexp
);
147 scm_make_regexp (SCM pat
, SCM flags
)
153 SCM_ASSERT (SCM_NIMP(pat
) && SCM_ROSTRINGP(pat
), pat
, SCM_ARG1
,
155 SCM_COERCE_SUBSTR (pat
);
156 if (SCM_UNBNDP (flags
))
157 flags
= SCM_MAKINUM (REG_EXTENDED
);
158 SCM_ASSERT (SCM_INUMP (flags
), flags
, SCM_ARG2
, s_make_regexp
);
161 rx
= (regex_t
*) scm_must_malloc (sizeof (regex_t
), s_make_regexp
);
162 status
= regcomp (rx
, SCM_ROCHARS (pat
),
163 /* Make sure they're not passing REG_NOSUB;
164 regexp-exec assumes we're getting match data. */
165 (SCM_INUM (flags
) & ~REG_NOSUB
));
169 scm_error (scm_regexp_error_key
,
171 scm_regexp_error_msg (status
, rx
),
176 SCM_NEWCELL (result
);
177 SCM_SETCAR (result
, scm_tc16_regex_t
);
178 SCM_SETCDR (result
, rx
);
183 SCM_PROC (s_regexp_exec
, "regexp-exec", 2, 2, 0, scm_regexp_exec
);
186 scm_regexp_exec (SCM rx
, SCM str
, SCM start
, SCM flags
)
188 int status
, nmatches
, offset
;
190 SCM mvec
= SCM_BOOL_F
;
192 SCM_ASSERT (SCM_NIMP (rx
) && SCM_RGXP (rx
), rx
, SCM_ARG1
, s_regexp_exec
);
193 SCM_ASSERT (SCM_NIMP (str
) && SCM_ROSTRINGP (str
), str
, SCM_ARG2
,
196 if (SCM_UNBNDP (start
))
200 SCM_ASSERT (SCM_INUMP (start
), start
, SCM_ARG3
, s_regexp_exec
);
201 offset
= SCM_INUM (start
);
202 SCM_ASSERT (offset
>= 0 && offset
<= SCM_LENGTH (str
), start
,
203 SCM_OUTOFRANGE
, s_regexp_exec
);
206 if (SCM_UNBNDP (flags
))
208 SCM_ASSERT (SCM_INUMP (flags
), flags
, SCM_ARG2
, s_regexp_exec
);
210 SCM_COERCE_SUBSTR (str
);
212 /* re_nsub doesn't account for the `subexpression' representing the
213 whole regexp, so add 1 to nmatches. */
215 nmatches
= SCM_RGX(rx
)->re_nsub
+ 1;
217 matches
= (regmatch_t
*) scm_must_malloc (sizeof (regmatch_t
) * nmatches
,
219 status
= regexec (SCM_RGX (rx
), SCM_ROCHARS (str
) + offset
,
225 /* The match vector must include a cell for the string that was matched,
227 mvec
= scm_make_vector (SCM_MAKINUM (nmatches
+ 1), SCM_UNSPECIFIED
,
229 SCM_VELTS(mvec
)[0] = str
;
230 for (i
= 0; i
< nmatches
; ++i
)
231 SCM_VELTS(mvec
)[i
+1] = scm_cons(SCM_MAKINUM(matches
[i
].rm_so
+ offset
),
232 SCM_MAKINUM(matches
[i
].rm_eo
+ offset
));
236 if (status
!= 0 && status
!= REG_NOMATCH
)
237 scm_error (scm_regexp_error_key
,
239 scm_regexp_error_msg (status
),
246 scm_init_regex_posix ()
248 scm_tc16_regex_t
= scm_newsmob (®ex_t_smob
);
250 /* Compilation flags. */
251 scm_sysintern ("regexp/extended", scm_long2num (REG_EXTENDED
));
252 scm_sysintern ("regexp/icase", scm_long2num (REG_ICASE
));
253 scm_sysintern ("regexp/newline", scm_long2num (REG_NEWLINE
));
255 /* Execution flags. */
256 scm_sysintern ("regexp/notbol", scm_long2num (REG_NOTBOL
));
257 scm_sysintern ("regexp/noteol", scm_long2num (REG_NOTEOL
));
259 #include "regex-posix.x"
261 scm_add_feature ("regex");