Summary:
theme change was delayed untill some other change (i.e. mode change, clock update)
@fix
Test Plan:
1. create new style "test" for elm_clock
2. add this style to theme
3. compare results of following modifications of src/examples/clock_example.c:
```
/* am/pm */
ck = elm_clock_add(win);
elm_object_style_set(ck, "test"); //style is applied before ap_pm set
elm_clock_show_am_pm_set(ck, EINA_TRUE);
```
and
```
/* am/pm */
ck = elm_clock_add(win);
elm_clock_show_am_pm_set(ck, EINA_TRUE);
elm_object_style_set(ck, "test"); //style is applied after ap_pm set
```
ExR: style is applied in both cases
Result without this patch: style is applied only in first case
Reviewers: cedric, seoz, raster, reutskiy.v.v, Hermet
Reviewed By: Hermet
Differential Revision: https://phab.enlightenment.org/D2349
Summary:
Checkbox should support a third state "indeterminate" along with "Checked" and "Unchecked"
This third state is a state of checkbox which is shown when checkbox is neither Checked nor Unchecked.
- Added this new feature on the basis of a boolean variable's value.
- By default this boolean variable is disabled and checkbox will treat like old way.
- While adding this, I kept in mind, that applications which are already using checkbox, should not be affected, so I used 0=False=Unchecked, 1=True=Checked, and 2=Indeterminate
- Also added an example check_example_o2.c, which is using checkbox with both ways, using boolean, and using enum.
- Now also values can be set using boolean values, but it will give a type casting warning. As a boolean doen't support third state, so I used an enum int like.
- Added APIs to enable disable third state mode. elm_check_three_state_mode_set(check_obj, bool_val), and elm_check_three_state_mode_get(check_obj)
- Modified old APIs which were setting or getting states of checkbox.
- Added a state in theme of checkbox, with third state image.
Reviewers: seoz, raster, Sergeant_Whitespace, Hermet
Subscribers: Hermet, Sergeant_Whitespace, sachin.dev
Differential Revision: https://phab.enlightenment.org/D2249
Summary:
using theme of the object instead of default for subparts.
@fix
Reviewers: cedric, seoz, Hermet, raster, reutskiy.v.v
Differential Revision: https://phab.enlightenment.org/D2370
Summary: This fixed the elm_win_window_id_get function when running in
wayland to return the ecore_wl_window_id rather than the surface id.
This change is necessary so that anyone calling this function can use
the return value in ecore_wayland function calls.
NB: Thanks to Hermet for pointing this out ! :)
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
The surface and context should not be current when deleted.
Evas GL would work just fine but since it deletes the surface
and context immediately upon request, those should not be
current.
This is a slightly different behaviour from EGL.
Flipselector has some random whitespaces that do not belong.
Additionally there are a few hidden non-blocking spaces that prevent
edje_cc to compile the default.edc on arm. Appearantly on x86/amd64 this
works normally.
This patch works around T2285 while the issue should remain open as the
real issue, edje_cc treating nbsp as a whitespace on arm, is not fixed
through this patch.
ref T2285
Signed-off-by: Olliver Schinagl <o.schinagl@ultimaker.com>
Without any locking or thread-safe mechanism, the previous implementation
would have failed as soon as too many file_set() happened on the same
object. Indeed, file_set() can happen while the async open thread is
running. I shouldn't have blindly listened to Cedric :P
Summary:
Issue: Mirroring of elm_panel not working correctly.
Solution: Handle mirroring appropriately for animations.
@fix
Test Plan:
Case: 1
1. Open elementary_test
2. Mirror on
3. Open scrollable panel, click on toggle button.
Case: 2
1. Keep the scroller panel opened
2. Turn off Mirror
Case: 3
1. Turn on mirror
2. Check direction button in non-scrollable panel.
Reviewers: shilpasingh, raster, Hermet, Princekrdubey
Reviewed By: shilpasingh
Subscribers: poornima.srinivasan, subodh6129
Differential Revision: https://phab.enlightenment.org/D2237
Summary:
Some APIs need to return specific values for their usage.
For Examples, elm_xx_select_mode_get function need to return
ELM_OBJECT_SELECT_MODE_MAX when invalid input comes.
**Fixed API list :**
elm_list_mode_get
elm_list_multi_select_mode_get
elm_genlist_mode_get
elm_genlist_select_mode_get
elm_genlist_multi_select_mode_get
elm_genlist_item_type_get
elm_genlist_ttem_index_get
elm_genlist_item_select_mode_get
elm_gengrid_select_mode_get
elm_gengrid_multi_select_mode_get
elm_gengrid_item_index_get
Test Plan: Call API with NULL parameter and check return value.
Reviewers: Hermet, seoz, jaehwan, raster, eunue
Subscribers: id213sin
Differential Revision: https://phab.enlightenment.org/D2306
Since the latest elm colorclass changes we need eldbus-codegen to generate
some files. Not all setups have this in their normal $PATH so let allow these
setups to set the correct path during configure. This allows our Jenkins setup
as well as others to work again.
why have elm image repeat events always (and unable to stop it)? this
is neither useful nor convenient. it's just painful. it makes images
not behave like prety much every object. they already are set up to
capture events - then do it normally.
@fix
Without depth, glview / evas gl basically fall back to indirect rendering.
Also change bg color from yellowish to brown (and use PREMULTIPLIED
colors, not random values).
This will allow apps to set the proper viewport, even if the
actual surface geometry didn't change.
This fixes a rare situation where the mode is changed from indirect
to direct (eg. in elm_test glview).