I found a way to keep eo_add() the way it was and gracefully degrade to
a portable (but not as fast) solution for compilers that don't support
the compound macros returning a value gnu extension: ({int a; a;}).
I'm reverting these changes now, and I'll introduce the fallback as soon
as I can.
This reverts commit b85bb37183.
This reverts commit 22eefc9626e15e78612396a9c7adca89ffc228e2.
ok - so equation changed. tizen 2.3 did ship with index starting at
1, so back to 1 (ugh - ugly). so apparently we have switch between 0
and 1 over history at various points (and no one in efl land uses this
api so we dont see it). but given that the largest userbase of efl atm
starts at 1 - we'll leave it there.
Summary:
This patch focuses on Combobox widget customization,
Multibuttonentry widget is used instead of entry for taking user input.
The idea is to make the widget look like {F28112} {F28115} when the multiple_selection is set.
To-DO:
1) Need to add scrollable interface to combobox when MBE is used (need some suggestions on it).
2) focus cycle is still buggy as genlist requires focus otherwise selected item will return NULL (sometimes)
Signed-off-by: divyesh purohit <div.purohit@samsung.com>
@feature
Test Plan: Please run combobox multiple selection example from elementart_test.
Reviewers: raster, shilpasingh, cedric
Subscribers: govi, rajeshps
Projects: #elementary
Differential Revision: https://phab.enlightenment.org/D3570
Summary:
Developer cannot notice that any description didn't applied due to missing description or typo.
This message will be helpful to make correct the application.
Reviewers: cedric, Hermet, raster
Subscribers: soohye.shin, minkyu, cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D3783
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
It has been decided that we would not use any namespace for interface
and they will sit in efl main namespace.
This patch doesn't correct the naming of the event has we don't have a
prefix for event. We do still have EFL_ANIMATOR_EVENT_ANIMATOR_TICK,
instead of a nicer EFL_EVENT_ANIMATOR_TICK.
EINVAL is bad, we can't go on. If we treat it like it's not a fatal
error we'll end up spinning on the fd and constantly retrying sends
on the dead wayland connection.
@fix
Coverity reports that 'ctx' may be NULL here and we should check it
before usage (as is done above).
Coverity CID1339785
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
This patch fixes a potential resource leak where we would previously
create a new eina_array and then possibly return from this function
(when checking validity of 's' parameter) without freeing the newly
created array.
Coverity CID1350291
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Summary:
The size of Colorclass is 20 bytes, but that of Elm_Color_Overlay 16 bytes.
Currently, there is no code to access last 4 bytes, but it can cause
seg fault by another patch.
Reviewers: cedric, zmike
Differential Revision: https://phab.enlightenment.org/D3784
so in feb 2015 seoz changed elm_genlist_item_index_get to start from 1
rather than 0. going back to elm code in 1.7 - it started at 0. this
is an api break that shouldn't have happend, but did. this fixes that.
yes - it's inconsistent with gengrid's index_get - but gengrid here is
wrong. nth_get starts at 0. this will get fixed with eo api's, and in
fact none of these index/nth api's should be in genlist's eo api.
legacy only. i can see why this was changed - it matches gengrid and
is more consistent, but we can't break things even if stupid.
@fix
eo.h and eo.c were added by mistake.
I couldn't quite test the build since the we still need ewekbit2
for configure to enable elm_web. @cedric has a bit of work left
here :)
The lcov tool offers the functionality to have an initial run over the code base
before the tests are executed to make sure it catches all files, even the ones
that are not loaded during the test run.
This is something we missed so far. The reports have only been in relation to
the files that actually have been loaded during the test. We missed quite a
few files which made our numbers inaccurate. Things like
modules/emotion/gstreamer1 have no tests yet and thus never showed up in the
coverage report. While it has 103 functions and over thousand lines which need
to get covered. With the baseline this is handled now. Thanks goes to the folks
at LibreOffice who described their lcov setup here:
https://wiki.documentfoundation.org/Development/Lcov
New numbers are lower now as we count in all the files never loaded which
decreases our percentages.
Overall coverage rate:
lines......: 30.2% (65119 of 215841 lines)
functions..: 34.0% (6361 of 18733 functions)
branches...: 23.6% (35627 of 151096 branches)
As ecore_drm_private.h already includes config.h header, we don't need
to include it here in these files also
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
As portions of this code have been derived from existing code in
Weston, we should also be including their copyright/licence text to
give credit.
NB: Fixes T3286
@fix
Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Reverting this at Felipe's request following my email. There are many
things I strongly object to in this commit. I've touched the surface of
those on the ML (which doesn't work at the moment), though we need to
better discuss it.
The gist:
1. dlsym is a really bad hack that is not even needed.
2. I don't see why eo should even be aware of promises. It's not aware
of list, hash and etc.
3. The eolian changes were done wrong.
This should have been discussed and consulted before done, even if only
because of the amount of hacks it includes and the cross-domain (ecore,
eo and eolian) nature of it.
This reverts commit f9ba80ab33.