Update the TODO list and bump the version to 0.7.12.

.gitignore

@@ -1,5 +1,6 @@
 Makefile
 config.h
+config.h.in
 config.log
 config.status
 src/*.o

INSTALL.md

@@ -5,11 +5,15 @@
 Autoconf and Automake are used, and the most basic compilation procedure consists of:
 
 ```
+autoreconf --install
 ./configure
 make
 make install # root rights probably needed here, prefix with sudo in such case
 ```
 
+(Invocation of `autoreconf` is only required if the source tree has been cloned from the
+git repository.)
+
 To see the list of flags accepted by the configure script, run:
 
 ```
@@ -18,5 +22,5 @@ To see the list of flags accepted by the configure script, run:
 
 Optionally, `mboxgrep` can be linked with the following libraries:
 
-- PCRE, to enable support for regular expressions compatible with Perl 5;
+- PCRE2, to enable support for regular expressions compatible with Perl 5;
 - zlib and bzlib, to enable support for compressed mbox folders.

Makefile.am

@@ -1 +1,2 @@
 SUBDIRS = doc src
+dist_data_DATA = contrib

NEWS.md

@@ -1,5 +1,11 @@
 # Changes of mboxgrep
 
+## Changes since 0.7.11
+
+- Port to the pcre2 library (pcre1 is no longer supported).
+- Check command-line options for conflicting matchers and actions.
+- Various minor code cleanups.
+
 ## Changes since 0.7.10
 
 - GNU Automake is now utilized instead of manually written Makefile.in files.

README.md

@@ -1,7 +1,7 @@
 # mboxgrep - selects e-mail messages matching a pattern
 
 `mboxgrep` is a `grep(1)`-like tool which scans mailboxes and selects
-e-mail messages matching pattern.
+e-mail messages matching a pattern.
 
 Full description of mboxgrep is contained in the documentation,
 which is provided both in manpage and texinfo format, to satisfy

TODO.md

@@ -3,7 +3,7 @@
 ## Behavior
 
 - [x] use cryptographic hashes for detecting duplicate messages
-- [ ] add checking for conflicting command-line options
+- [x] add checking for conflicting command-line options
 - [ ] support for deletion of messages after being matched and displayed
 - [x] ignore .overview when grepping Gnus folders
 - [x] inverted matching
@@ -18,7 +18,7 @@
 
 ## File formats, encodings and standards
 
-- [ ] migrate to pcre2, as pcre is obsolete
+- [x] migrate to pcre2, as pcre is obsolete
 - [ ] use a more modern hash function than MD5
 - [ ] MIME support
 - [ ] support for GnuPG

configure.ac

@@ -18,7 +18,7 @@
 # Process this file with autoconf to produce a configure script.
 
 # Yawn.
-AC_INIT([mboxgrep], [0.7.11], [dspiljar@datatipp.se], [mboxgrep], [https://www.mboxgrep.org/])
+AC_INIT([mboxgrep], [0.7.12], [dspiljar@datatipp.se], [mboxgrep], [https://www.mboxgrep.org/])
 #AM_INIT_AUTOMAKE
 AM_INIT_AUTOMAKE([foreign])
 AC_LANG([C])
@@ -34,34 +34,35 @@ AC_HEADER_DIRENT
 
 # Checks for libraries.
 
-# Check for PCRE library
-AC_ARG_WITH(pcre, [  --without-pcre          Compile without Perl regexp support],,
+# Check for PCRE2 library
+AC_ARG_WITH(pcre2, [  --without-pcre2          Compile without Perl regexp support],,
 [
-	AC_PATH_PROG(PCRE_CONFIG, pcre-config)
+	AC_PATH_PROG(PCRE2_CONFIG, pcre2-config)
 
-	if test "$PCRE_CONFIG"; then
-	   CFLAGS="$CFLAGS `$PCRE_CONFIG --cflags`"
-	   LIBS="$LIBS `$PCRE_CONFIG --libs`"
+	if test "$PCRE2_CONFIG"; then
+	   CFLAGS="$CFLAGS `$PCRE2_CONFIG --cflags`"
+	   LIBS="$LIBS `$PCRE2_CONFIG --libs32`"
 
 	   AC_LINK_IFELSE(
 	   [
-#include <pcre.h>
+#define PCRE2_CODE_UNIT_WIDTH 32
+#include <pcre2.h>
 
 int main ()
 {
   return 0;
 }
 	   ],
-	   AC_DEFINE(HAVE_LIBPCRE),
+	   AC_DEFINE(HAVE_LIBPCRE2),
 	   [
-		AC_MSG_NOTICE(found pcre-config but could not compile test program.)
-		AC_MSG_FAILURE(is PCRE properly installed?)
+		AC_MSG_NOTICE(found pcre2-config but could not compile test program.)
+		AC_MSG_FAILURE(is PCRE2 properly installed?)
 	   ]
 )
 	else
-	   AC_MSG_NOTICE(pcre-config not found)
-	   AC_MSG_NOTICE(trying to find PCRE anyway)
-	   AC_CHECK_LIB(pcre, main)
+	   AC_MSG_NOTICE(pcre2-config not found)
+	   AC_MSG_NOTICE(trying to find PCRE2 anyway)
+	   AC_CHECK_LIB(pcre2, main)
 	fi
 ]
 )

doc/mboxgrep.1

@@ -1,4 +1,4 @@
-.TH MBOXGREP 1 "24 Mar 2023"
+.TH MBOXGREP 1 "20 May 2023"
 .SH NAME
 mboxgrep \- displays email messages matching a pattern
 .SH SYNOPSIS
@@ -10,7 +10,7 @@ mboxgrep \- displays email messages matching a pattern
 This manual page refers to
 .B mboxgrep
 version
-.BR 0.7.11 .
+.BR 0.7.12 .
 .PP
 .B mboxgrep
 scans a

doc/mboxgrep.info

@@ -9,7 +9,7 @@ END-INFO-DIR-ENTRY
 
 File: mboxgrep.info,  Node: Top,  Up: (dir)
 
-This file documents 'mboxgrep' (version 0.7.11), a mailbox scanning
+This file documents 'mboxgrep' (version 0.7.12), a mailbox scanning
 utility.
 
    Copyright (C) 2000, 2001, 2002, 2003 Daniel Spiljar

doc/mboxgrep.texi

@@ -5,8 +5,8 @@
 @setchapternewpage odd
 
 @set EDITION 0.7
-@set VERSION 0.7.11
-@set UPDATED 24 Mar 2023
+@set VERSION 0.7.12
+@set UPDATED 20 May 2023
 
 @dircategory Mail
 @direntry

src/config.h.in

@@ -1,76 +0,0 @@
-/* src/config.h.in.  Generated from configure.ac by autoheader.  */
-
-/* Define to 1 if you have the <dirent.h> header file, and it defines `DIR'.
-   */
-#undef HAVE_DIRENT_H
-
-/* Define to 1 if you have the `fcntl' function. */
-#undef HAVE_FCNTL
-
-/* Define to 1 if you have the `flock' function. */
-#undef HAVE_FLOCK
-
-/* Define to 1 if you have the `fts_open' function. */
-#undef HAVE_FTS_OPEN
-
-/* Define to 1 if you have the `ftw' function. */
-#undef HAVE_FTW
-
-/* Define to 1 if you have the `bz2' library (-lbz2). */
-#undef HAVE_LIBBZ2
-
-/* Define to 1 if you have the `dmalloc' library (-ldmalloc). */
-#undef HAVE_LIBDMALLOC
-
-/* Define to 1 if you have the `garfield' library (-lgarfield). */
-#undef HAVE_LIBGARFIELD
-
-/* Define to 1 if you have the `pcre' library (-lpcre). */
-#undef HAVE_LIBPCRE
-
-/* Define to 1 if you have the `z' library (-lz). */
-#undef HAVE_LIBZ
-
-/* Define to 1 if you have the <ndir.h> header file, and it defines `DIR'. */
-#undef HAVE_NDIR_H
-
-/* Define to 1 if you have the `regcomp' function. */
-#undef HAVE_REGCOMP
-
-/* Define to 1 if you have the `strptime' function. */
-#undef HAVE_STRPTIME
-
-/* Define to 1 if you have the <sys/dir.h> header file, and it defines `DIR'.
-   */
-#undef HAVE_SYS_DIR_H
-
-/* Define to 1 if you have the <sys/ndir.h> header file, and it defines `DIR'.
-   */
-#undef HAVE_SYS_NDIR_H
-
-/* Name of package */
-#undef PACKAGE
-
-/* Define to the address where bug reports for this package should be sent. */
-#undef PACKAGE_BUGREPORT
-
-/* Define to the full name of this package. */
-#undef PACKAGE_NAME
-
-/* Define to the full name and version of this package. */
-#undef PACKAGE_STRING
-
-/* Define to the one symbol short name of this package. */
-#undef PACKAGE_TARNAME
-
-/* Define to the home page for this package. */
-#undef PACKAGE_URL
-
-/* Define to the version of this package. */
-#undef PACKAGE_VERSION
-
-/* Define to 1 if you have the ANSI C header files. */
-#undef STDC_HEADERS
-
-/* Version number of package */
-#undef VERSION

src/info.c

@@ -74,9 +74,9 @@ version (void)
 /*
   fprintf (stdout, "HAVE_LIBLOCKFILE ");
 */
-#ifdef HAVE_LIBPCRE
-  print_wrap ("HAVE_LIBPCRE");
-#endif /* HAVE_LIBPCRE */
+#ifdef HAVE_LIBPCRE2
+  print_wrap ("HAVE_LIBPCRE2");
+#endif /* HAVE_LIBPCRE2 */
 #ifdef HAVE_LIBZ
   print_wrap ("HAVE_LIBZ");
 #endif /* HAVE_LIBZ */
@@ -128,10 +128,10 @@ help (void)
            "Matching criteria:\n\n"
            "  -E,  --extended-regexp\tPATTERN is an extended regular expression\n"
            "  -G,  --basic-regexp\t\tPATTERN is a basic regular expression\n");
-#ifdef HAVE_LIBPCRE
+#ifdef HAVE_LIBPCRE2
   fprintf (stdout,
            "  -P,  --perl-regexp\t\tPATTERN is a Perl regular expression\n");
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
   fprintf (stdout,
            "  -e,  --regexp=PATTERN\t\tUse PATTERN as a regular expression\n"
            "  -i,  --ignore-case\t\tIgnore case distinctions\n"

src/main.c

@@ -99,11 +99,11 @@ main (int argc, char **argv)
 
   if (config.haveregex)
     {
-#ifdef HAVE_LIBPCRE
+#ifdef HAVE_LIBPCRE2
       if (config.regextype == REGEX_PERL)
         pcre_init ();
       else
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
         regex_init ();
     }
   else

src/mboxgrep.h

@@ -21,7 +21,7 @@
 #define MBOXGREP_H
 
 #define APPNAME "mboxgrep"
-#define VERSION "0.7.11"
+#define VERSION "0.7.12"
 #define BUGREPORT_ADDR "dspiljar AT datatipp.se"
 
 #define HOST_NAME_SIZE 256
@@ -120,7 +120,7 @@ typedef struct
   char hostname[HOST_NAME_SIZE];
   char *boxname, *outboxname, *pipecmd, *tmpfilename, *regex_s;
 
-  void *pcre_pattern, *pcre_hints, *posix_pattern;
+  void *pcre_pattern, *posix_pattern, *match_data;
   int res1, res2;
 
   action_t action;

src/misc.c

@@ -373,7 +373,7 @@ set_option_regextype (regextype_t regextype)
       exit (2);
     }
 
-#ifndef HAVE_LIBPCRE
+#ifndef HAVE_LIBPCRE2
   if (regextype == REGEX_PERL);
     {
       fprintf (stderr,
@@ -381,6 +381,6 @@ set_option_regextype (regextype_t regextype)
               APPNAME);
       exit (2);
     }
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
   config.regextype = regextype;
 }

src/re.c

@@ -21,52 +21,58 @@
 #include <stdio.h>
 #include <string.h>
 #include <regex.h>
-#ifdef HAVE_LIBPCRE
-#  include <pcre.h>
-#endif /* HAVE_LIBPCRE */
+#ifdef HAVE_LIBPCRE2
+#  define PCRE2_CODE_UNIT_WIDTH 32
+#  include <pcre2.h>
+#endif /* HAVE_LIBPCRE2 */
 #include "mboxgrep.h"
 #include "message.h"
 #include "wrap.h"               /* xcalloc() et cetera */
 
-#ifdef HAVE_LIBPCRE
+#ifdef HAVE_LIBPCRE2
 void
 pcre_init (void)
 {
-  int errptr;
-  const char *error;
+  int errornumber;
+  PCRE2_SIZE erroroffset;
 
   config.pcre_pattern =
-    (pcre *) pcre_compile (config.regex_s,
-                           (config.ignorecase ? PCRE_CASELESS : 0),
-                           &error, &errptr, NULL);
+    (pcre2_code *) pcre2_compile ((PCRE2_SPTR) config.regex_s, (PCRE2_SIZE) strlen (config.regex_s),
+                                  (config.ignorecase ? PCRE2_CASELESS : 0),
+                                  &errornumber, &erroroffset, NULL);
   if (config.pcre_pattern == NULL)
     {
       if (config.merr)
-        fprintf (stderr, "%s: %s: %s\n", APPNAME, config.regex_s, error);
+        {
+          PCRE2_UCHAR buffer[256];
+
+          pcre2_get_error_message (errornumber, buffer, sizeof(buffer));
+          fprintf (stderr, "%s: PCRE2 compilation failed at offset %d: %s\n",
+              APPNAME, (int) erroroffset, (char *) buffer);
+        }
       exit (2);
     }
+
+  config.match_data =
+    (pcre2_match_data* ) pcre2_match_data_create_from_pattern (config.pcre_pattern, NULL);
 }
 
 void
 pcre_match (message_t * msg)
 {
-  int of[BUFSIZ];
-
   if (config.headers)
     config.res1 =
-      pcre_exec ((pcre *) config.pcre_pattern,
-                 (pcre_extra *) config.pcre_hints,
-                 msg->headers, (int) strlen (msg->headers), 0, 0, of, BUFSIZ);
+      pcre2_match ((pcre2_code *) config.pcre_pattern,
+                   (PCRE2_SPTR) msg->headers, (int) strlen (msg->headers), 0, 0, config.match_data, NULL);
   if (config.body)
     config.res2 =
-      pcre_exec ((pcre *) config.pcre_pattern,
-                 (pcre_extra *) config.pcre_hints,
-                 msg->body, (int) strlen (msg->body), 0, 0, of, BUFSIZ);
+      pcre2_match ((pcre2_code *) config.pcre_pattern,
+                   (PCRE2_SPTR) msg->body, (int) strlen (msg->body), 0, 0, config.match_data, NULL);
 
   config.res1 = config.res1 ^ 1;
   config.res2 = config.res2 ^ 1;
 }
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
 
 void
 regex_init (void)

src/re.h

@@ -19,9 +19,9 @@
 
 #include "mboxgrep.h"
 
-#ifdef HAVE_LIBPCRE
+#ifdef HAVE_LIBPCRE2
 void pcre_init (void);
 void pcre_match (message_t * msg);
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
 void regex_init (void);
 void regex_match (message_t * msg);

src/scan.c

@@ -146,11 +146,11 @@ scan_mailbox (char path[])
       if (msg->from == NULL)
         msg->from = (char *) xstrdup ("nobody");
 
-#ifdef HAVE_LIBPCRE
+#ifdef HAVE_LIBPCRE2
       if (config.regextype == REGEX_PERL)
         pcre_match (msg);
       else
-#endif /* HAVE_LIBPCRE */
+#endif /* HAVE_LIBPCRE2 */
         regex_match (msg);
 
       if (config.dedup)