[PATCH libinput 11/26] tablet: use libwacom to identify tablets for left-handedness

[sardemff7+wayland at sardemff7.net]

On 2015-02-24 07:21, Peter Hutterer wrote:
> A tablet hotplug event is rare and not a time-critical event, so we load the
> database on tablet init and throw it away again.
>
> Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>
> ---
>   configure.ac               | 14 ++++++++++++++
>   src/Makefile.am            |  2 ++
>   src/evdev-tablet.c         | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>   test/tablet.c              | 13 ++++++++++++-
>   test/valgrind.suppressions | 18 ++++++++++++++++++
>   5 files changed, 90 insertions(+), 2 deletions(-)
>
> diff --git a/configure.ac b/configure.ac
> index 1a4eb3a..95c822a 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -165,6 +165,19 @@ if test "x$build_tests" = "xyes"; then
>   	AC_PATH_PROG(VALGRIND, [valgrind])
>   fi
>
> +AC_ARG_ENABLE(libwacom,
> +	      AS_HELP_STRING([--enable-libwacom],
> +			     [Use libwacom for tablet identification (default=enabled)]),
> +	      [use_libwacom="$enableval"],
> +	      [use_libwacom="yes"])
> +if test "x$use_libwacom" = "xyes"; then
> +	PKG_CHECK_MODULES(LIBWACOM, [libwacom], [HAVE_LIBWACOM="yes"])
> +	AC_DEFINE(HAVE_LIBWACOM, 1, [Build with libwacom])
> +else
> +	HAVE_LIBWACOM="no"
> +fi
> +AC_SUBST(HAVE_LIBWACOM)

Do you use HAVE_LIBWACOM (as "yes"/"no") for real? I did not see any usage in this patch and it could be confusing. Using it in the configure summary does not require AC_SUBST.
I would just use ${use_libwacom}. Since you do not use automagic detection, it will always be set correctly for the summary. :-)


-- 

Quentin “Sardem FF7” Glidic