Profile

Cover photo
Peter Hutterer
Works at Red Hat
871,850 views
AboutPosts

Stream

Peter Hutterer

Shared publicly  - 
 
touchpad gestures will be coming to libinput with one of the upcoming releases.

http://who-t.blogspot.com.au/2015/06/libinput-touchpad-gestures.html
One of the bits we are currently finalising in libinput are touchpad gestures. Gestures on a normal touchscreens are left to the compositor and, in extension, to the client applications. Touchpad gestures are notably differen...
10
Add a comment...

Peter Hutterer

Shared publicly  - 
 
And only a few days after the previous release, libinput 0.17 is out. The tap timeout has been reduced to be more responsive and clickfinger now allows left-clicks with a thumb while the index finger is resting on the touchpad.

http://lists.freedesktop.org/archives/wayland-devel/2015-June/022448.html
[ANNOUNCE] libinput 0.17.0. Peter Hutterer peter.hutterer at who-t.net. Wed Jun 3 18:14:37 PDT 2015. Previous message: [PATCH weston v2 01/21] input: Pass axis events through pointer grab interfaces; Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] ...
8
Add a comment...

Peter Hutterer

Shared publicly  - 
 
Got a script that needs sudo somewhere during it's execution? Add this line:

sudo -n true 2>/dev/null || echo "password required" | festival --tts

And your box will tell you that/when you have to type in the password.
6
Add a comment...

Peter Hutterer

Shared publicly  - 
 
I wrote a short post about the new tap-and-drag features in libinput 0.16

http://who-t.blogspot.com.au/2015/06/extended-tap-and-drag-in-libinput.html
TLDR: as of libinput 0.16 you can end a touchpad tap-and-drag with a final additional tap libinput originally only supported single-tap and double-tap. With version 0.15 we now support multi-tap, so you can tap repeatedly to ...
7
Jim Basilio's profile photo
 
Can't tell you how excited I am to see this driver being worked/refined. The touchpad is such an integral part of the laptop experience and outside of OSX there's no OS that is any good at it. Windows is generally pretty terrible and spotty ... linux's X driver is about 85% solid but it's just impossible to improve b/c of coupling with X.

libinput's genesis and HOPEFULLY unification b/t Mir/Wayland will move touchpad's forward immensely.

thank you!!
Add a comment...

Peter Hutterer

Shared publicly  - 
 
$ wc -l libinput/src/*.(c|h)  
 15717 total
$  wc -l libinput/test/*.(c|h)
 16903 total

Probably the first project I've worked on where that happened.
22
Andrew Pinski's profile photoRay Strode's profile photo
2 comments
 
Doing it right
Add a comment...

Peter Hutterer

Shared publicly  - 
 
For the german speakers:
http://www.spiegel.de/static/happ/wissenschaft/2015/sprachatlas/v0/dist/#/questions
A quiz that tells you where you come from based on the words you use and how you pronounce them.

The town I grew up in was number two on the list of results, the top guess was the next large city ~30km away. Very impressive.
 ·  Translate
Können wir Sie aufgrund von Spracheigenheiten innerhalb des deutschsprachigen Europas verorten? Sagen Sie uns, wie Sie sprechen, und wir sagen Ihnen, woher Sie stammen.
4
2
Boudewijn Rempt's profile photoAlexander Neundorf's profile photo
Add a comment...

Peter Hutterer

Shared publicly  - 
 
I expect that at some Christmas party in Canberra, a couple of politicians will meet in a hidden corner and then go "ok, you said the stupidest dumbest shit this year and got away with it, here's your dollar".

I really hope they're having a bet running, because otherwise...
3
Christopher Halse Rogers's profile photoMatt Turner's profile photoPeter Hutterer's profile photoMartin Peres's profile photo
6 comments
 
WOW, that is a good one +Matt Turner !

I see, I did not know it was possible to be that bad!
Add a comment...

Peter Hutterer

Shared publicly  - 
 
Another day, another post. How we're using model-specific tags to apply custom configurations on some hardware:

http://who-t.blogspot.com.au/2015/06/libinput-and-model-specific.html
libinput provides a number of different out-of-the-box configurations, based on capabilities. For example: middle mouse button emulation is enabled by default if a device only has left and right buttons. On devices with a phy...
1
Jim Basilio's profile photoPeter Hutterer's profile photo
2 comments
 
F22 is already using libinput as default X driver. Look for xf86-input-libinput.

Though to get access to the config options you'll need support from your DE. Gnome, KDE, XFCE all have it already, no idea about unity.
Add a comment...

Peter Hutterer

Shared publicly  - 
 
libinput won't provide device type identifiers (e.g. "this is a touchpad"). Here's the reason why:

http://who-t.blogspot.com.au/2015/06/libinput-and-lack-of-device-types.html
libinput uses udev tags to determine what a device is. This is a significant difference to the X.Org stack which determines how to deal with a device based on an elaborate set of rules, rules grown over time, matured, but wit...
10
1
Igor Gnatenko's profile photo
Add a comment...

Peter Hutterer

Shared publicly  - 
 
libinput 0.16.0 is out, grab it while it's hot!

New features: tap-to-end-drag and disable-while-typing, the latter much to my disgust but palm detection is hard on the touchpads we have to deal with...

http://lists.freedesktop.org/archives/wayland-devel/2015-June/022400.html
[ANNOUNCE] libinput 0.16.0. Peter Hutterer peter.hutterer at who-t.net. Mon Jun 1 15:43:38 PDT 2015. Previous message: [PATCH libinput] evdev: use the udev ID_INPUT_POINTINGSTICK property; Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] ...
15
2
Rob Foehl's profile photoPeter Hutterer's profile photoSimos Xenitellis's profile photoJon Cruz's profile photo
9 comments
 
won't work, sorry. we don't know whether or how much the cursor moves in libinput, all we know is that we're sending relative events. but if we're e.g. at the edge of the screen then the cursor may not move at all. So this is in general a heuristic we can't rely on.

and I'd rather have simple heuristics that are predictable. users are better at dealing with bad but predictable interfaces than we are at writing good reliable ones :)
Add a comment...

Peter Hutterer

Shared publicly  - 
 
Is there a way to force the _typeof_ of an enum value so it corresponds to the actual enum type? i.e. get rid of the "signed vs unsigned comparison" warning (-Wsign-compare) in this bit:

enum myenum { FOO, BAR };

#define test(x, y) \
    do { \
        _typeof_(x) _a = x; \
        _typeof_(y) _b = y; \
        if (_a != _b) abort(); \
    } while(0)

int main(void) {
    enum myenum value = BAR;

    test(value, FOO);

    return 0;
}
1
Alison Chaiken's profile photoJonas Ådahl's profile photoThiago Macieira's profile photoPeter Hutterer's profile photo
6 comments
 
it wouldn't be used in C++ code, at least not in my case so I don't have to worry about this. meanwhile, the workaround was to assign the value to a temporary variable. Not as nice, but still readable and gets rid of the warning and is less prone to misc bugs :)
Add a comment...

Peter Hutterer

Shared publicly  - 
 
If you have a chromebook, please post the model number and the content of /sys/class/dmi/id/modalias
Trying to get some udev matching rules for chromebook touchpads and it'd be nice to get all chromebooks covered quickly.
Thanks.
1
4
Timo Paulssen's profile photoAdrian M Negreanu's profile photoSabine “Atari-Frosch” Engelhardt's profile photoRobert Schreiner's profile photo
4 comments
 
32 bit arm based Chromebooks won't have dmi as it does not exist for 32 bit arm.
Add a comment...
Story
Introduction
X.Org input maintainer
Links
Work
Employment
  • Red Hat
    Senior Software Engineer, present