> I fail to see what the problem is, just do:
>
> Test if add_history is in libreadline, if it is not, test if it is in
> libhistory, else do not use add_history.
The patch already checks for add_history in libhistory, and if it does
not find it but the system has a suitable libreadline then the check
is repeated for libreadline.
AC_CHECK_LIB(readline, rl_bind_key,
[LIBREADLINE=-lreadline
AC_DEFINE(HAVE_LIBREADLINE, 1,
[Define to one if you have -lreadline])],
[LIBREADLINE=], "$LIBTERMCAP")
AC_SUBST(LIBREADLINE)
AC_SUBST(LIBTERMCAP)
AC_CHECK_LIB(history, add_history,
[LIBHISTORY=-lhistory
AC_DEFINE(HAVE_LIBHISTORY, 1,
[Define to one if you have -lhistory])],
[LIBHISTORY=])
AC_SUBST(LIBHISTORY)
# If libhistory does not provide add_history check if libreadline has it.
if test -z "$LIBHISTORY" && test -n "$LIBREADLINE"; then
AC_CHECK_LIB(readline, add_history,
[LIBHISTORY=-lreadline
AC_DEFINE(HAVE_LIBHISTORY, 1,
[Define to one if you have -lhistory])],
[LIBHISTORY=])
fi
Is this alright or you wanted something else?
Happy hacking,
Debarshi
--
GPG key ID: 63D4A5A7
Key server: pgp.mit.edu
_______________________________________________
bug-inetutils mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-inetutils
