-const char pcrs_rcs[] = "$Id: pcrs.c,v 1.11 2001/08/15 15:32:03 oes Exp $";
+const char pcrs_rcs[] = "$Id: pcrs.c,v 1.18 2002/03/08 14:17:14 oes Exp $";
/*********************************************************************
*
* File : $Source: /cvsroot/ijbswa/current/pcrs.c,v $
*
- * Purpose : pcrs is a supplement to the brilliant pcre library by Philip
- * Hazel <ph10@cam.ac.uk> and adds Perl-style substitution. That
- * is, it mimics Perl's 's' operator.
+ * Purpose : pcrs is a supplement to the pcre library by Philip Hazel
+ * <ph10@cam.ac.uk> and adds Perl-style substitution. That
+ * is, it mimics Perl's 's' operator. See pcrs(3) for details.
*
- * Currently, there's no documentation besides comments and the
- * source itself :-(
- *
- * Note: In addition to perl's options, 'U' for ungreedy and 'T'
- * for trivial (i.e.: ignore backrefs in the substitute) are
- * supported.
*
* Copyright : Written and Copyright (C) 2000, 2001 by Andreas S. Oesterhelt
* <andreas@oesterhelt.org>
*
* This program is free software; you can redistribute it
- * and/or modify it under the terms of the GNU General
- * Public License as published by the Free Software
- * Foundation; either version 2 of the License, or (at
- * your option) any later version.
+ * and/or modify it under the terms of the GNU Lesser
+ * General Public License (LGPL), version 2.1, which should
+ * be included in this distribution (see LICENSE.txt), with
+ * the exception that the permission to replace that license
+ * with the GNU General Public License (GPL) given in section
+ * 3 is restricted to version 2 of the GPL.
*
* This program is distributed in the hope that it will
* be useful, but WITHOUT ANY WARRANTY; without even the
* implied warranty of MERCHANTABILITY or FITNESS FOR A
- * PARTICULAR PURPOSE. See the GNU General Public
- * License for more details.
+ * PARTICULAR PURPOSE. See the license for more details.
*
- * The GNU General Public License should be included with
- * this file. If not, you can view it at
- * http://www.gnu.org/copyleft/gpl.html
+ * The GNU Lesser General Public License should be included
+ * with this file. If not, you can view it at
+ * http://www.gnu.org/licenses/lgpl.html
* or write to the Free Software Foundation, Inc., 59
* Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
* Revisions :
* $Log: pcrs.c,v $
+ * Revision 1.18 2002/03/08 14:17:14 oes
+ * Fixing -Wconversion warnings
+ *
+ * Revision 1.17 2002/03/08 13:45:48 oes
+ * Hiding internal functions
+ *
+ * Revision 1.16 2001/11/30 21:32:14 jongfoster
+ * Fixing signed/unsigned comparison (Andreas please check this!)
+ * One tab->space
+ *
+ * Revision 1.15 2001/09/20 16:11:06 steudten
+ *
+ * Add casting for some string functions.
+ *
+ * Revision 1.14 2001/09/09 21:41:57 oes
+ * Fixing yet another silly bug
+ *
+ * Revision 1.13 2001/09/06 14:05:59 oes
+ * Fixed silly bug
+ *
+ * Revision 1.12 2001/08/18 11:35:00 oes
+ * - Introduced pcrs_strerror()
+ * - made some NULL arguments non-fatal
+ * - added support for \n \r \e \b \t \f \a \0 in substitute
+ * - made quoting adhere to standard rules
+ * - added warning for bad backrefs
+ * - added pcrs_execute_list()
+ * - fixed comments
+ * - bugfix & cosmetics
+ *
* Revision 1.11 2001/08/15 15:32:03 oes
* - Added support for Perl's special variables $+, $' and $`
* - Improved the substitute parser
#include <pcre.h>
#include <string.h>
+#include <ctype.h>
+
#include "pcrs.h"
-#include <stdio.h>
+
const char pcrs_h_rcs[] = PCRS_H_VERSION;
+/*
+ * Internal prototypes
+ */
+
+static int pcrs_parse_perl_options(const char *optstring, int *flags);
+static pcrs_substitute *pcrs_compile_replacement(const char *replacement, int trivialflag,
+ int capturecount, int *errptr);
/*********************************************************************
*
default: return "Unknown error";
}
}
- /* error > 0: No error */
+ /* error >= 0: No error */
return "(pcrs:) Everything's just fine. Thanks for asking.";
}
* Returns : option integer suitable for pcre
*
*********************************************************************/
-int pcrs_parse_perl_options(const char *optstring, int *flags)
+static int pcrs_parse_perl_options(const char *optstring, int *flags)
{
size_t i;
int rc = 0;
if (NULL == optstring) return 0;
- for (i=0; i < strlen(optstring); i++)
+ for (i = 0; i < strlen(optstring); i++)
{
switch(optstring[i])
{
case 's': rc |= PCRE_DOTALL; break;
case 'x': rc |= PCRE_EXTENDED; break;
case 'U': rc |= PCRE_UNGREEDY; break;
- case 'T': *flags |= PCRS_TRIVIAL; break;
+ case 'T': *flags |= PCRS_TRIVIAL; break;
default: break;
}
}
* the reason.
*
*********************************************************************/
-pcrs_substitute *pcrs_compile_replacement(const char *replacement, int trivialflag, int capturecount, int *errptr)
+static pcrs_substitute *pcrs_compile_replacement(const char *replacement, int trivialflag, int capturecount, int *errptr)
{
- int length, i, k, l, quoted;
+ int i, k, l, quoted;
+ size_t length;
char *text;
pcrs_substitute *r;
*errptr = PCRS_ERR_NOMEM;
return NULL;
}
- memset(r, '\0', length + 1);
+ memset(text, '\0', length + 1);
/*
*/
else
{
- while(i < length)
+ while (i < (int)length)
{
/* Quoting */
if (replacement[i] == '\\')
{
if (replacement[i+1] && strchr("tnrfae0", replacement[i+1]))
{
- switch(replacement[++i])
+ switch (replacement[++i])
{
case 't':
text[k++] = '\t';
text[k++] = '\f';
break;
case 'a':
- text[k++] = '\a';
+ text[k++] = 7;
break;
case 'e':
text[k++] = 27;
}
/* Backreferences */
- if (replacement[i] == '$' && !quoted && i < length - 1)
+ if (replacement[i] == '$' && !quoted && i < (int)(length - 1))
{
char *symbol, symbols[] = "'`+&";
r->block_length[l] = k - r->block_offset[l];
/* Numerical backreferences */
- if (isdigit(replacement[i + 1]))
+ if (isdigit((int)replacement[i + 1]))
{
- while (i < length && isdigit(replacement[++i]))
+ while (i < (int)length && isdigit((int)replacement[++i]))
{
r->backref[l] = r->backref[l] * 10 + replacement[i] - 48;
}
*********************************************************************/
pcrs_job *pcrs_compile_command(const char *command, int *errptr)
{
- int i, k, l, limit, quoted = FALSE;
+ int i, k, l, quoted = FALSE;
+ size_t limit;
char delimiter;
char *tokens[4];
pcrs_job *newjob;
tokens[l] = (char *) malloc(limit + 1);
- for (i=0; i <= limit; i++)
+ for (i = 0; i <= (int)limit; i++)
{
if (command[i] == delimiter && !quoted)
old = subject;
*result_length = subject_length;
- hits = 0;
+ hits = total_hits = 0;
for (job = joblist; job != NULL; job = job->next)
{
hits = pcrs_execute(job, old, *result_length, &new, result_length);
+
if (old != subject) free(old);
+
if (hits < 0)
{
- *result = NULL;
return(hits);
}
else
}
*result = new;
- return(hits);
+ return(total_hits);
+
}
int pcrs_execute(pcrs_job *job, char *subject, size_t subject_length, char **result, size_t *result_length)
{
int offsets[3 * PCRS_MAX_SUBMATCHES],
- offset, i, k,
+ offset,
+ i, k,
matches_found,
- newsize,
submatches,
max_matches = PCRS_MAX_MATCH_INIT;
+ size_t newsize;
pcrs_match *matches, *dummy;
char *result_offset;
* Find the pattern and calculate the space
* requirements for the result
*/
- newsize=subject_length;
+ newsize = subject_length;
- while ((submatches = pcre_exec(job->pattern, job->hints, subject, subject_length, offset, 0, offsets, 3 * PCRS_MAX_SUBMATCHES)) > 0)
+ while ((submatches = pcre_exec(job->pattern, job->hints, subject, (int)subject_length, offset, 0, offsets, 3 * PCRS_MAX_SUBMATCHES)) > 0)
{
job->flags |= PCRS_SUCCESS;
matches[i].submatches = submatches;
- for (k=0; k < submatches; k++)
+ for (k = 0; k < submatches; k++)
{
matches[i].submatch_offset[k] = offsets[2 * k];
/* Storage for matches exhausted? -> Extend! */
if (++i >= max_matches)
{
- max_matches *= PCRS_MAX_MATCH_GROW;
+ max_matches = (int)(max_matches * PCRS_MAX_MATCH_GROW);
if (NULL == (dummy = (pcrs_match *)realloc(matches, max_matches * sizeof(pcrs_match))))
{
free(matches);
/* Don't loop on empty matches */
if (offsets[1] == offset)
- if (offset < subject_length)
+ if ((size_t)offset < subject_length)
offset++;
else
break;
offset = 0;
result_offset = *result;
- for (i=0; i < matches_found; i++)
+ for (i = 0; i < matches_found; i++)
{
/* copy the chunk preceding the match */
- memcpy(result_offset, subject + offset, matches[i].submatch_offset[0] - offset);
+ memcpy(result_offset, subject + offset, (size_t)matches[i].submatch_offset[0] - offset);
result_offset += matches[i].submatch_offset[0] - offset;
/* For every segment of the substitute.. */
- for (k=0; k <= job->substitute->backrefs; k++)
+ for (k = 0; k <= job->substitute->backrefs; k++)
{
/* ...copy its text.. */
memcpy(result_offset, job->substitute->text + job->substitute->block_offset[k], job->substitute->block_length[k]);