multi-touch-dev team mailing list archive

[Duncan McGreggor <duncan.mcgreggor@xxxxxxxxxxxxx>]

-------- Original Message --------
Subject: Re: [Private-sparsha-dev] tablets and auto-rotation
Date: Tue, 29 Jun 2010 12:38:52 +0200
From: Alberto Milone <alberto.milone@xxxxxxxxxxxxx>
To: Rafi Rubin <rafi@xxxxxxxxxxxxxx>
CC: Kees Cook <kees.cook@xxxxxxxxxxxxx>,        Rick Spencer
<rick.spencer@xxxxxxxxxxxxx>,        Bryce Harrington
<bryce@xxxxxxxxxxxxx>,        "Duncan M. McGreggor"
<duncan.mcgreggor@xxxxxxxxxxxxx>

On 29 June 2010 07:44, Rafi Rubin <rafi@xxxxxxxxxxxxxx> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 06/28/10 14:00, Kees Cook wrote:
>> Hi Bryce,
>>
>> On Mon, Jun 28, 2010 at 09:23:46AM -0700, Bryce Harrington wrote:
>>> On Mon, Jun 28, 2010 at 11:45:18AM -0400, Chase Douglas wrote:
>>>> On Mon, 2010-06-28 at 15:38 +0200, Alberto Milone wrote:
>>>>> I couldn't attend the UDS but I know that Federico Mena Quintero (at
>>>>> Novell) was working on this in a separate branch of the
>>>>> gnome-settings-daemon but he hasn't committed anything upstream yet (I
>>>>> guess he's focusing on different projects):
>>>>>
http://git.gnome.org/browse/gnome-settings-daemon/log/?h=randr-rotate-wacom-tablet
>>>>>
>>>>> As Rafi said, this solution also involves setting a gconf key to tell
>>>>> the daemon whether we want it to rotate input devices or not when
>>>>> screens are rotated.
>>>>>
>>>>> Currently Federico's code makes use of the xsetwacom tool instead of
>>>>> using Xinput (which is what we need). I'm very familiar with the
>>>>> gnome-settings-daemon code and, if there's enough interest at
>>>>> Canonical, I'm available to work on it but I don't own any Ntrig
>>>>> touchscreen and I can't use my Stantum tablet because it doesn't work
>>>>> with xinput Calibration.
>>>>
>>>> I'll second this approach. I believe this is the correct place to fix
>>>> the issue. I'll let Duncan handle how to pour resources into this, but
>>>> this looks promising.
>>>
>>> Kees Cook was also working on patch to g-s-d to solve this problem on
>>> his laptop at UDS.  IIRC, he'd also implemented it using xsetwacom, but
>>> we'd also talked about the changes to xinput; I don't know if he updated
>>> his patch for that, but I'm CCing him.
>>>
>>> Kees - mind pointing us to your tablet rotation patch?  Would at least
>>> be interesting to have it to compare with Federico's code.
>>
>> Sorry I didn't send this earlier, I kept getting distracted by other
>> things.  Yes, I wrote it to use XInput.  Also, it seems that g-s-d is not
>> actually the right place to put this, as it is just one of many tools
that
>> could be issuing a rotation request.  The patch is actually needed in
>> gnome-desktop.
>>
>> I've attached it to https://launchpad.net/bugs/599478
>>
>> -Kees
>
> Ok, good to see.  Is there a built version available in a ppa or other
repo?
>
> Also looks like it should be easy to add evdev stuff as well.
>
> Now for the tough part, for wacom devices in particular how should we
decide the
> defaults for which to auto rotate and which not too?  I'm sort of
wondering if
> it might just be prudent to brute force it with a list.
>
> Simply differentiating pen and touch is insufficient.  Most of the
wacom touch
> screens seem to have pen, and we would want to rotate those pens.  We
also see
> devices like the bamboo tablets which identify as touch, but aren't
attached to
> the screen and shouldn't auto rotate.
>
> And then there's messes like ntrig devices, where we're moving towards
using
> evdev for touch and the wacom x driver for pen :)
>
> Rafi

I've just forwarded you an email that Kees sent to Federico so that we
can continue the discussion there. You might want add the details of
your last email there, so that we can all access them (Federico works
for Novell).

Regards,

-- 
Alberto Milone
Sustaining Engineer (system)
Foundations Team
Canonical OEM Services