I have fixed so that gnulib's 'make syntax-check' works again, and by running it I noticed these three complaints:
1) We use HAVE_CONFIG_H a lot: maint.mk: found use of HAVE_CONFIG_H; remove make: *** [sc_prohibit_have_config_h] Fel 1 Any objections to dropping all HAVE_CONFIG_H uses in InetUtils? GNU projects appears to be moving away from it, and there is no point for any project using gnulib (like InetUtils) to use HAVE_CONFIG_H since gnulib code already assumes there is a config.h. 2) Internationalization issues: ftp/main.c ftpd/ftpd.c ifconfig/ifconfig.c libinetutils/argcv.c ping/ping.c ping/ping6.c src/hostname.c src/inetd.c src/logger.c src/rcp.c src/rexec.c src/rexecd.c src/rlogin.c src/rlogind.c src/rsh.c src/rshd.c src/syslogd.c src/tftp.c src/tftpd.c src/traceroute.c src/uucpd.c talk/talk.c talkd/talkd.c telnet/main.c telnetd/telnetd.c tests/addrpeek.c tests/localhost.c maint.mk: the above files do not call bindtextdomain With the exception of the programs in tests/, it looks like this indicate a real bug: that there aren't any translations. InetUtils should be internationalized. We don't even have any po/ directory! I'm assuming this is an oversight? Marking strings has to be done manually and carefully to make sure we translate useful strings. It would quite an effort to resolve this in one go, however we can start on this and add strings as we move along. Any objections to creating a gettext domain for InetUtils and start marking strings for translation? 3) There is an error: prohibit_always_true_header_tests /bin/bash: line 0: cd: ./gnulib/lib: No such file or directory 0.03 prohibit_always_true_header_tests I'm not sure if this indicates a gnulib or Inetutils problem, but it seems this particular test is not ran properly. /Simon
