Page MenuHomePhabricator

ecore-input : add Ecore_Device and device add/del/update events

Authored by zmike on Mar 15 2016, 3:38 AM.



There are requirements regarding source input device identification from input events
especially in Tizen. Evas input event callbacks are being used in applications and
ecore input events are being used by applications and some toolkits such as DALi, Volt,
Webkit and so on.

From the input device identification point of view, there is Evas_Device in evas common
and it can be useful to set/get the source device information associated with each evas input
event. However, in ecore input, there is no data structures and events which can be used
for helping input device identification with ecore input events.
Thus, I added Ecore_Device data structure and some events which can be used to store the source
device information to the ecore events.

Signed-off-by: Sung-Jin Park <>

Test Plan


Diff Detail

rEFL core/efl
No Linters Available
No Unit Test Coverage
Build Status
Buildable 1530
Build 1595: arc lint + arc unit
input.hacker retitled this revision from to ecore-input : add Ecore_Device and device add/del/update events.
input.hacker updated this object.
input.hacker edited the test plan for this revision. (Show Details)
input.hacker added reviewers: raster, devilhorns, zmike.
input.hacker added subscribers: ohduna, cedric, Unknown Object (User).

The following stuff was omitted. :)

#define ECORE_MAGIC_DEV           0x5e9e6510
zmike edited edge metadata.Mar 15 2016, 11:43 AM

I'd like to see some tests/usages of the api included in this patch. Otherwise it seems okay.

Hi, zmike. :)
This patch will be followed by other patch-sets including ecore-drm and ecore input evas side implementations.
Even the additional patches become added to this revision, it'll still be partial implementations and other patches need to be added more and more.
Would you really like to contain test cases/usage in this revision ? If you'd like to, then I'll do it. :)

Thanks and regards,
Sung-Jin Park

I believe it's a requirement that all new API has a corresponding test case?

devilhorns edited edge metadata.Mar 16 2016, 7:30 AM

Where applicable and possible to run the test, yes :)

Exactly. If there is already a testsuite for the part of the code you are adding more things you should add your tests there. If you add something completely new submitting it with a testsuite is what we want.

Exceptions apply for things that really hard to test. I would hear explanations if you want this exception.

jpeg requested changes to this revision.Jul 7 2016, 1:39 AM
jpeg added a reviewer: jpeg.

I believe this patch, along with D3858, should be put on hold for now, as we have no backend implementation (ecore wl2/drm2 or x11).

This revision now requires changes to proceed.Jul 7 2016, 1:39 AM
zmike commandeered this revision.May 1 2018, 10:04 AM
zmike abandoned this revision.
zmike edited reviewers, added: input.hacker; removed: zmike.

This should be handled by efl-input-device.