* shadows should be uniformly moved to whatever thing they're supposed to be on (WIP)
* trans-clipper was a failure idea, this is now done in C
* signals renamed to be more consistent
* groups renamed to comp/frame/*
* comp/border/* groups preserved for compat
- Handled the deletion of image object correctly. Do not overuse/misuse
evas_object_data, it will make code less readable.
- Cleaned up button click callback. sharing one function for different
buttons is not a good idea in some cases. Just split the callbacks.
- Removed unnecessary evas_object_data_set().
Summary:
scroller is moved page by page if scroller's paging is enable
1. default is move to vertical.
2. move to horizontal with Shift key
Reviewers: seoz, Hermet, jaehwan, woohyun, raster, cedric
Reviewed By: seoz
Differential Revision: https://phab.enlightenment.org/D453
Summary: Minor change to make code a little simpler.
Test Plan: elementary_test -to accessibility
Reviewers: seoz, singh.amitesh
Reviewed By: seoz
Differential Revision: https://phab.enlightenment.org/D456
Panes smart callback depends on the signals from theme but it was broken
when dark theme was introduced.
"clicked" does not work yet, it should be fixed.
Thanks for the report, JackDanielZ.
eo_data_scope_get and eo_isa returns NULL with complaint when NULL is passed.
So, Removed unnecessary !obj check routine.
In addition, adopt EINA_UNLIKELY for these macros and fix indentation.
This API sets the attribute to show the input panel in case of only an user's explicit Mouse Up event.
It doesn't request to show the input panel even though it has focus.
EE_ENGINE_COMPARE reduces typing little bit, but it makes redundant function calls
because almost callers call it more than once.
And, there is no way to check whether ecore_evas_engine_name_get is null.
This patch removed EE_ENGINE_COMPARE with small refactoring.
In addition, removed unnecessary check routines from _elm_ee_xwin_get.
Summary:
Almost logic of both horizontal and vertical are same except different axis in
_els_box_layout.
This patch reduced code size and improved speed little bit by moving condition
out of the loop using some extra pointer variables.
In addition, removed unnecessary check routines.
There are already similar logic for them before and after.
Reviewers: raster, seoz, cedric
Reviewed By: raster
Differential Revision: https://phab.enlightenment.org/D423
Summary: These APIs allow to enable/disable a tooltip to move with respect to mouse pointer
Test Plan: elm_tooltip_movement_lock_set(), elm_tooltip_movement_lock_get(), elementary_test -to tooltip
Reviewers: singh.amitesh, seoz, raster
Reviewed By: raster
CC: raster
Differential Revision: https://phab.enlightenment.org/D369
Using the elm_scrollable_interface_content_viewport_resize_cb_set in
scrollable entry widget. this only happens in mobile profile when you
drag the content with your finger left/right when the content is
smaller than the gap between icons.
This accepts D187
Summary: called scroller bring in API in "scroll,anim,stop" callback, so occured infinite loop
Reviewers: seoz, jaehwan, raster, woohyun, cedric
Reviewed By: raster
Differential Revision: https://phab.enlightenment.org/D445
Summary: This file is is the attribution required by the license.
Reviewers: raster
Reviewed By: raster
Differential Revision: https://phab.enlightenment.org/D438
Since rELM166ca9e86a72, I moved EWebKit header into Elementary.h like other libraries.
But, it made possible build break while building test browser in webkit.
It's because EWebKit.h and EWebKit2.h have same symbols and test browser should
be compiled with EWebKit2.h although Elementary was built with ewebkit.
This patch reverts a part of changes in rELM166ca9e86a72, which includes EWebkit.h
in Elementary.h
now mirrored, scale and so on will be working fine.
but still wonder if the mirrored_set(), scale_set().. are called in the elm_widget_theme_object_set() inside.
then every widget won't be need to call it by theirselves...
this makes efl ignore certain env vars for thnigs and entirely removes
user modules (that no one ever used) etc. etc. to ensure that *IF* an
app is setuid, there isn't a priv escalation path that is easy.
and removed unnecessary local variable.
- Do not need to call the same function with same parameter many times.
Use local variable for them.
- Remove unnecessary local variable if it is not needed.
free/del func + free func get null as valid parameter.
Now ELM_SAFE_FREE is less useful than before. It reduced 5 lines to 1
but now it reduces 2 lines to 1.
ELM_WIDGET_DATA_GET was checking it was the correct class. This is
stupid and wrong (and makes tools rightfully complain) because these
functions are implementations of that class anyway, so that would never
be false.
Functions that outside facing (that is, not called from eo_do) should
use API_ENTRY.
When elm_widget_newest_focus_order_get is called with can_focus_only,
only focusable objects should be the candidates for newest focused object.
But there are cases that unfocusable object add +1 to newest_focus_order.
This API sets the attribute to show the input panel in case of only an user's explicit Mouse Up event.
It doesn't request to show the input panel even though it has focus.