Fix #49: add support for -8 command line option to allow 8-bit data

This patch allows the user to disable the 8-bit data check in the log
message validator.  If you have experienced problems with logging any
unicode (utf-8) messages after v1.6, this option is for you.

The correct way to handle this is to add proper parser support for the
Unicode BOM, defined in RFC5424[1], as NetBSD syslogd does[2], search
for IS_BOM().

[1]: https://datatracker.ietf.org/doc/html/rfc5424#appendix-A.8
[2]: http://cvsweb.netbsd.org/bsdweb.cgi/src/usr.sbin/syslogd/syslogd.c?rev=1.138

Signed-off-by: Joachim Wiberg <troglobit@gmail.com>

man/syslogd.8

@@ -38,7 +38,7 @@
 .Nd log systems messages
 .Sh SYNOPSIS
 .Nm
-.Op Fl ?46AdFHKknsTtv
+.Op Fl ?468AdFHKknsTtv
 .Op Fl a Ar addr[/len][:port]
 .Op Fl a Ar name[:port]
 .Op Fl b Ar addr[:port]
@@ -129,6 +129,22 @@ to use IPv4 addresses only.
 Force
 .Nm
 to use IPv6 addresses only.
+.It Fl 8
+Tells
+.Nm
+not to interfere with 8-bit data.  Normally
+.Nm
+replaces C1 control characters
+.Pq ISO 8859 and Unicode characters
+with their
+.Dq M- Ns Em x
+equivalent.  Note, this option does not change the way
+.Nm
+alters control characters
+.Pq see Xr iscntrl 3 .
+They are always replaced with their
+.Dq ^ Ns Em x
+equivalent.
 .It Fl A
 Ordinarily,
 .Nm

src/syslogd.c

@@ -290,11 +290,12 @@ static void sys_seqno_save(void)
 int usage(int code)
 {
 	printf("Usage:\n"
-	       "  syslogd [-46AdFHKknsTtv?] [-a PEER] [-b NAME] [-f FILE] [-m INTERVAL]\n"
+	       "  syslogd [-468AdFHKknsTtv?] [-a PEER] [-b NAME] [-f FILE] [-m INTERVAL]\n"
-	       "                            [-P PID_FILE] [-p SOCK_PATH] [-r SIZE[:NUM]]\n"
+	       "                             [-P PID_FILE] [-p SOCK_PATH] [-r SIZE[:NUM]]\n"
 	       "Options:\n"
 	       "  -4        Force IPv4 only\n"
 	       "  -6        Force IPv6 only\n"
+	       "  -8        Allow all 8-bit data, e.g. unicode, does not affect control chars\n"
 	       "  -A        Send to all addresses in DNS A, or AAAA record\n"
 	       "  -a PEER   Allow PEER to use us as a remote syslog sink. Ignored when started\n"
 	       "            with -s. Multiple -a options may be specified:\n"
@@ -357,7 +358,7 @@ int main(int argc, char *argv[])
 	char *ptr;
 	int ch;
 
-	while ((ch = getopt(argc, argv, "46Aa:b:C:dHFf:Kkm:nP:p:r:sTtv?")) != EOF) {
+	while ((ch = getopt(argc, argv, "468Aa:b:C:dHFf:Kkm:nP:p:r:sTtv?")) != EOF) {
 		switch ((char)ch) {
 		case '4':
 			family = PF_INET;
@@ -367,6 +368,10 @@ int main(int argc, char *argv[])
 			family = PF_INET6;
 			break;
 
+		case '8':
+			mask_C1 = 0;
+			break;
+
 		case 'A':
 			send_to_all++;
 			break;

test/Makefile.am

@@ -1,5 +1,5 @@
 EXTRA_DIST       = lib.sh opts.sh
-EXTRA_DIST      += api.sh local.sh remote.sh fwd.sh mark.sh
+EXTRA_DIST      += api.sh local.sh unicode.sh remote.sh fwd.sh mark.sh
 CLEANFILES       = *~ *.trs *.log
 TEST_EXTENSIONS  = .sh
 TESTS_ENVIRONMENT= unshare -mrun
@@ -12,6 +12,7 @@ api_LDADD        = ../src/libsyslog.la
 
 TESTS            = opts.sh
 TESTS           += local.sh
+TESTS           += unicode.sh
 TESTS           += remote.sh
 TESTS           += api.sh
 TESTS           += fwd.sh

test/unicode.sh

@@ -0,0 +1,20 @@
+#!/bin/sh
+# Currently only same as local.sh but with unicode messages
+# From https://github.com/troglobit/sysklogd/issues/49
+# shellcheck disable=SC1090
+if [ x"${srcdir}" = x ]; then
+    srcdir=.
+fi
+. ${srcdir}/lib.sh
+setup -8
+
+MSG="öäüÖÄÜ߀¢§"
+MSG2="…‘’•"
+
+../src/logger -u "${SOCK}" ${MSG}
+grep ${MSG} "${LOG}" || FAIL "Cannot find: ${MSG}"
+
+../src/logger -u "${ALTSOCK}" ${MSG2}
+grep ${MSG2} "${LOG}" || FAIL "Cannot find: ${MSG2}"
+
+OK