[PATCH] protocol: Extend wl_touch with touch point shape

[Dennis Kempin]

Thanks a lot for the comments guys!

I agree on all of the above and uploaded a v2 of the patch. I included
Peters description about the relationship to the frame event and
updated the documentation to better describe how that ellipse is
described. What major/minor means, how exactly the orientation is
specified, etc. Also added a comment about the center as Daniel
suggested.

Peter, I am not set on shape. How about 'contact area' as an alternative?


On Mon, Apr 4, 2016 at 5:11 AM, Daniel Stone <daniel at fooishbar.org> wrote:
> Hi,
>
> On 3 April 2016 at 23:35, Peter Hutterer <peter.hutterer at who-t.net> wrote:
>> Agreed. And please specify the exact details of what is considered the
>> major/minor axis (direction, orientation) and where the logical 0 is for the
>> orientation. Referencing the kernel isn't good enough here, we need to make
>> sure this is precise. Adding a few paragraphs to the documentation doesn't
>> hurt and saves headaches later. Other than that, it looks good imo, thanks.
>
> Yeah, AOL on all of the above and then it looks good to me as well.
> The only other thing I'd add: is the reported touchpoint always at the
> centre of the ellipse? If so, then it bears documenting, and if not
> then we may need to prepare for a future centrepoint-reporting event?
> The former probably makes the most sense, even if the kernel doesn't
> specify it.
>
> Cheers,
> Daniel