[Xcb] [PATCH 2/2] Fix strict-aliasing warning when getting generic event length.

[Peter Hutterer]

On Sun, Mar 28, 2010 at 10:58:17AM -0700, Jamey Sharp wrote:
> xcb_ge_event_t has its length field in the same place that
> xcb_generic_reply_t does, so there's no need to cast the generic reply.
> 
> Signed-off-by: Jamey Sharp <jamey at minilop.net>
> Cc: Peter Hutterer <peter.hutterer at who-t.net>
> ---
> 
> I don't know how to test generic events. Does this look right?

xinput --test-xi2 should do the job, but the patch looks good to me.

Cheers,
  Peter
 
>  src/xcb_in.c |    4 +---
>  1 files changed, 1 insertions(+), 3 deletions(-)
> 
> diff --git a/src/xcb_in.c b/src/xcb_in.c
> index 80f5523..a2f7312 100644
> --- a/src/xcb_in.c
> +++ b/src/xcb_in.c
> @@ -154,9 +154,7 @@ static int read_packet(xcb_connection_t *c)
>  
>      /* XGE events may have sizes > 32 */
>      if (genrep.response_type == XCB_XGE_EVENT)
> -    {
> -        eventlength = ((xcb_ge_event_t*)&genrep)->length * 4;
> -    }
> +        eventlength = genrep.length * 4;
>  
>      buf = malloc(length + eventlength +
>              (genrep.response_type == XCB_REPLY ? 0 : sizeof(uint32_t)));
> -- 
> 1.7.0
>