[PATCH synaptics] Fail if no backends can be found

[Alan Coopersmith]

On 05/ 1/12 04:17 PM, Peter Hutterer wrote:
> Apparently the driver broke on Solaris when the backends were made optional
> in cf451f34e3bdd52a4eb072ce9f4b514eb83a1cfc. The driver would still build,
> but without backends there it has limited usefulness.
> 
> Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>
> ---
>  configure.ac |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/configure.ac b/configure.ac
> index dbf3fec..5c2c25d 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -120,7 +120,7 @@ case "${host}" in
>  	BUILD_PS2COMM="yes"
>  	;;
>  *)
> -	AC_MSG_RESULT([none])
> +	AC_MSG_ERROR("Cannot find suitable backends for this platform.")
>  	;;
>  esac
>  if test "x$BUILD_EVENTCOMM" = xyes; then

When applied to synaptics before Niveditha's recent patch to fix Solaris,
this does indeed catch the failure that broke Solaris:

checking which optional backends will be build... configure: error: "Cannot find
suitable backends for this platform."

Changing the "" to [] in the AC_MSG_ERROR seems to work as well and gets rid
of the extra quotes in the output:

checking which optional backends will be build... configure: error: Cannot find
suitable backends for this platform.

Since Linux, FreeBSD, NetBSD & Dragonfly were all already accounted for, and
Solaris is queued up, that seems to leave OpenBSD and Hurd that should be
alerted by this that something isn't quite right for them.

Reviewed-by: Alan Coopersmith <alan.coopersmith at oracle.com>

-- 
	-Alan Coopersmith-              alan.coopersmith at oracle.com
	 Oracle Solaris Engineering - http://blogs.oracle.com/alanc