Amitesh Singh [Fri, 14 Jul 2017 10:32:55 +0000 (19:32 +0900)]
efl.ui.image.zoomable: mark setter of img_region property as eo api
thanks to stefan for reporting.
jiin.moon [Fri, 14 Jul 2017 02:29:50 +0000 (11:29 +0900)]
edje_cc: Add copy item for proxy part
Summary:
Although user use "inherit" in a state of proxy part,
the inherit function does not copy the source_visible and
the source_clip value of parent to the state.
Test Plan: {
F1751132}sample
Reviewers: jpeg, cedric
Differential Revision: https://phab.enlightenment.org/D5017
@fix
Derek Foreman [Wed, 12 Jul 2017 21:31:19 +0000 (16:31 -0500)]
ecore_evas: Perform a tick before a manual render if we're not ticking
If we turned off the ticking with manual_render_set then perform a
manual render, we need animators to update.
ref T5462
Derek Foreman [Tue, 11 Jul 2017 21:43:33 +0000 (16:43 -0500)]
ecore_evas: Don't start custom tickers for manually rendered canvases
When a canvas is manually rendered the ticker is just a waste of cpu, and
worse - it can wake the drm back-end from dpms sleep, as the display needs
to be awake to generate vblanks.
We fire a DBG message when attempting to start an animator in this state
because it's frequently a bug that wastes battery life - (like E doing idle
cursor animations or clock updates while the display is off)
However, dpms off is not the only potential usage of manual render, so
another commit will follow shortly to fix the bug this commit introduces -
when using a backend with a custom ticker and doing manual render with
the display on, calling ecore_evas_manual_render() will not draw with
updated animator state.
Fix T5462
Again.
Really.
Derek Foreman [Wed, 12 Jul 2017 21:30:07 +0000 (16:30 -0500)]
ecore_evas_drm: Provide a last_tick_get callback
Allow the drm engine to provide a tick timestamp for the previous vblank
interval.
ref T5462
Derek Foreman [Wed, 12 Jul 2017 21:29:17 +0000 (16:29 -0500)]
ecore_evas: Add a function pointer for last_tick_get
Engines that provide their own tickers may need to be able to provide the
time of the last tick even if they weren't sending ticks to EFL at the
time.
This is a feature added during freeze as it's necessary to resolve a bug.
ref T5462
Derek Foreman [Wed, 12 Jul 2017 20:27:52 +0000 (15:27 -0500)]
ecore_drm2: Add a query for the next vblank time
I guess this is a feature, and we're deep in freeze, but:
a) this is critical for fixing T5462 properly without any side effects.
b) ecore_drm2 is all beta api
c) this should only affect wayland users
ref T5462
Derek Foreman [Tue, 11 Jul 2017 21:13:46 +0000 (16:13 -0500)]
ecore_evas: Refactor event catchers to call custom_tick_begin/end
Make the event catcher functions call the custom_tick_begin/end helpers
instead of directly calling the engine functions.
Derek Foreman [Wed, 12 Jul 2017 21:24:09 +0000 (16:24 -0500)]
ecore_evas engines: Add some nulls to the func arrays
Purely cosmetic, but gets the comments back in sync with reality.
Cedric BAIL [Thu, 13 Jul 2017 22:34:37 +0000 (15:34 -0700)]
ecore_evas: protect against application holding a reference on the canvas while destroying the ecore_evas.
Andy Williams [Thu, 13 Jul 2017 22:19:52 +0000 (23:19 +0100)]
doc: Include two missing properties
Stefan Schmidt [Thu, 13 Jul 2017 11:52:35 +0000 (13:52 +0200)]
elm: naviframe/prefs : do not include non-public header files
We are not shipping those (part of the private eo header files) and thus
we can not include those either. Simply removing them fixes the problem
I see with our ABI checker and it still passes al build tests I have.
Stefan Schmidt [Thu, 13 Jul 2017 11:04:44 +0000 (13:04 +0200)]
elm: build: fix indentation of file lists
I was looking over this file for other problems but this jumped into my
eye.
Stefan Schmidt [Thu, 13 Jul 2017 09:21:59 +0000 (11:21 +0200)]
eina: inarray: remove documentation pointer to non-existing functions
These do not longer exist in our current code base (did they ever?).
Thanks a lot to Vincent Torri for spotting this!
Marcel Hollerbach [Thu, 13 Jul 2017 08:37:34 +0000 (10:37 +0200)]
efl_ui_focus_manager: this whole thing is not just eo api its also beta
Marcel Hollerbach [Thu, 13 Jul 2017 08:23:56 +0000 (10:23 +0200)]
elementary: no need to provide the new legacy api
The api is beta AND only usable in eo, so no need to add those apis here
Amitesh Singh [Thu, 13 Jul 2017 02:16:48 +0000 (11:16 +0900)]
win: fix compiler warning
Stephen 'Okra' Houston [Wed, 12 Jul 2017 18:07:22 +0000 (13:07 -0500)]
Thermal theme: Remove text from theme and quit wasting space.
Derek Foreman [Wed, 12 Jul 2017 18:01:43 +0000 (13:01 -0500)]
ecore_anim: set the tick thread's fds CLOEXEC
If we don't set them CLOEXEC then they end up in the fd space of every
client Enlightenment launches, allowing any client to write to them
and mess with animator timings or freeze the compositor.
Mike Blumenkrantz [Wed, 12 Jul 2017 17:41:23 +0000 (13:41 -0400)]
efl-wl: initialize variable
CID 1377573
Stefan Schmidt [Wed, 12 Jul 2017 16:56:47 +0000 (18:56 +0200)]
build: efl_wl: make sure we have the two files for efl_wl included
These two files coming in via EXTRA_DIST have not been included becuase
they have been living inside an ifdef which would only have them
included in a dist when configured with wayland enabled. We always want
to have these files in the dist though. Building for wayland out of a
tarball release works now again.
Problem reported by William L. Thomson Jr.
Andrii Kroitor [Wed, 12 Jul 2017 16:21:15 +0000 (19:21 +0300)]
edje_edit: fix source generation
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: change x11 atom name var to normal char*
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: free atom name when sending selection
CID 1377542
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: free x11 pipe read buf on read fail
CID 1377520
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: return if pipe reading fails in x11 bridge
CID 1377526, 1377534
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: check returns in x11 selection request handler
CID 1377518
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: mmap byte for nul terminator in keymap string
CID 1377545
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl-wl: check return of ecore_main_fd_handler_fd_get
CID 1377541, 1377546, 1377519, 1377529, 1377543
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl_wl: check mkstemp return
CID 1377533
Mike Blumenkrantz [Wed, 12 Jul 2017 16:00:52 +0000 (12:00 -0400)]
efl_wl: error and cleanup on CLOEXEC fail
CID 1377530, 1377531
Chris Michael [Wed, 12 Jul 2017 15:24:37 +0000 (11:24 -0400)]
efl_wl: Fix resource leak
Coverity reports that 'name' is leaked here, so free the return
ecore_x_atom_name_get.
Fixes CID1377537
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 12 Jul 2017 15:02:16 +0000 (11:02 -0400)]
elput: Fix resource leak
Coverity CID1377539 says that we leak the eldbus reply here when we
error out, so be sure to cleanup 'reply'
Fixes CID1377539
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 12 Jul 2017 14:58:28 +0000 (10:58 -0400)]
ecore: Fix coverity resource leak
Coverity CID1377544 reports that poller is leaked here if we error out
and return null. As poller is created with calloc, we should free it
before returning.
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Youngbok Shin [Wed, 12 Jul 2017 09:37:50 +0000 (18:37 +0900)]
edje: give proper width to entry's cursor according to its theme
Summary:
evas_object_textblock_cursor_geometry_get() always return 0 as cursor's width
for BEFORE type cursor. It casued different cursor width when cursor_fg2 was shown.
The cursor_fg and cursor_fg2 must have same width according to its min width.
Also, the patch will enclose code for cursor_fg2 by cursor_fg code.
It is used when only there is cursor_fg.
@fix
Test Plan:
1. Give bigger width to entry's cursor by changing its theme.
2. See the width cursor_fg, cursor_fg2 when there is [LTR text + RTL text].
Reviewers: raster, herdsman, cedric, jpeg
Subscribers: stefan_schmidt
Differential Revision: https://phab.enlightenment.org/D4798
Jean-Philippe Andre [Wed, 12 Jul 2017 08:19:28 +0000 (17:19 +0900)]
evas_smart: Fix evas_smart_callback_description_find
bsearch was returnign a cb** not a cb*
Fixes T4540
Carsten Haitzler (Rasterman) [Wed, 12 Jul 2017 06:45:17 +0000 (15:45 +0900)]
po files - update lines/comments
Carsten Haitzler (Rasterman) [Wed, 12 Jul 2017 06:44:51 +0000 (15:44 +0900)]
ignore - add some build binary files to gitignore
Carsten Haitzler (Rasterman) [Wed, 12 Jul 2017 06:43:41 +0000 (15:43 +0900)]
evas mouse in/out/move event fix to remove "invalid move"
this should fix T5582 "properly" by emitting a move after an in only
if pointer x,y changed. this fixes rage and i hope still should handle
the original issue. let me know if it doesn't.
@fix
Stephen 'Okra' Houston [Wed, 12 Jul 2017 06:24:43 +0000 (01:24 -0500)]
Theme: Don't display freq on the meter for cpuclock.
Jean-Philippe Andre [Wed, 12 Jul 2017 01:55:44 +0000 (10:55 +0900)]
Revert "colorselector: Avoid API call on NULL object"
This reverts commit
b156f606e4fefb9b1d3ac0ca6d180e0d70e0140c.
The change should have been in EO itself. See the previous
patch.
Ref T5715
Jean-Philippe Andre [Wed, 12 Jul 2017 01:43:07 +0000 (10:43 +0900)]
eo: Don't resolve function calls for NULL objects
eo_function(NULL) always leads to a no-operation. A this
point it is basically considered the normal operation
with EO to just silently ignore NULL errors.
If the API function "eo_function" belongs to a class that
has not been loaded yet (eg. no object of this type has
been created yet), then the op associated to the API func
may not be defined yet.
Fixes T5715
Jean-Philippe Andre [Wed, 12 Jul 2017 01:24:11 +0000 (10:24 +0900)]
colorselector: Avoid API call on NULL object
Solves this error:
ERR<2321>:eo ../src/lib/efl/interfaces/efl_pack_layout.eo.c:1
efl_pack_layout_request() Unable to resolve op for api func
0x7fffed16dba0 for obj=(nil) ((null))
Fixes T5715
Derek Foreman [Tue, 11 Jul 2017 21:08:55 +0000 (16:08 -0500)]
ecore_evas_drm: Partially revert "Refuse to start a ticker for a manually rendered canvas"
BAILing here can break animators permanently, so best to just log the
issue and move on (and potentially allow the ticker to wake us from DPMS).
ref T5462
Partially reverts
a795629e8c9d7a7f444f8ce655eb9e24dce51429
Stefan Schmidt [Tue, 11 Jul 2017 21:00:00 +0000 (23:00 +0200)]
release: Update NEWS and bump version for 1.20.0-beta2 release
Stefan Schmidt [Tue, 11 Jul 2017 20:55:28 +0000 (22:55 +0200)]
tests: elm_cxx: add gitignore file for produced binary
Stefan Schmidt [Tue, 11 Jul 2017 19:58:15 +0000 (21:58 +0200)]
build: make sure we have the new fonts subdir in evas test for dist
This is one o the odds of the EXTRA_DIST2 workaround I did to avoid the
to long arguments error we started to get a while ago. Normally
autotools would handle this but as we copy the files in a hook we also
need to amke sure the folder is created.
Stefan Schmidt [Tue, 11 Jul 2017 19:54:25 +0000 (21:54 +0200)]
evas: update file path after move into fonts subdir
The file have been moved into a fonts subdir in commit
e8b35c779e7d7b.
Update the path here to get it included into the tarball to make sure
check passes during distcheck.
Cedric BAIL [Tue, 11 Jul 2017 18:02:47 +0000 (11:02 -0700)]
ecore: properly cleanup callbacks of future/promise.
Jean-Philippe Andre [Tue, 11 Jul 2017 09:58:51 +0000 (18:58 +0900)]
elm: Fix fileselector button "hidden files"
Fixes T5465
@fix
Romain Naour [Mon, 3 Jul 2017 21:26:16 +0000 (23:26 +0200)]
ecore_input: fix build without eeze
A dependency on HAVE_EEZE is missing in ecore_input_joystick_name_get():
lib/ecore_input/ecore_input_joystick.c: Dans la fonction « ecore_input_joystick_name_get »:
lib/ecore_input/ecore_input_joystick.c:632:4: erreur : unknown type name « Joystick_Info »
Joystick_Info *ji;
^~~~~~~~~~~~~
Fixes T5665
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Hosang Kim [Tue, 11 Jul 2017 07:25:44 +0000 (16:25 +0900)]
slider: Allow range limitations
Summary:
When user drags slider, slider value cannot be changed by API.
However the necessity of above behavior has emerged.
Because sometimes applications want limitation of slider value.
Test Plan: elementary_test -> slider -> Limited
Reviewers: woohyun, cedric, SanghyeonLee, singh.amitesh, jpeg
Reviewed By: jpeg
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D4883
Jean-Philippe Andre [Tue, 11 Jul 2017 07:02:59 +0000 (16:02 +0900)]
evas: Add a test case with a custom font
The font is **ugly** and was created with birdfont. Its license
shall be anything EFL requires (public domain in the ttf file).
Youngbok Shin [Tue, 11 Jul 2017 06:46:03 +0000 (15:46 +0900)]
evas font: append global font path when Evas initializes Fontconfig
Summary:
If there are appended font pathes before initializing Fontconfig,
Evas must care the font pathes after initializing Fontconfig.
@fix
Test Plan:
1. Call evas_font_path_global_append() with a private font path
before adding any Evas Object.
2. Add a Evas Textblock object which uses a font from the private font path.
3. See the font is not loaded properly without the patch.
Reviewers: raster, cedric, herdsman, jpeg, woohyun
Differential Revision: https://phab.enlightenment.org/D4867
JinYong Park [Tue, 11 Jul 2017 06:36:06 +0000 (15:36 +0900)]
popup: fix sizing eval logic when scrollable set TRUE
Summary:
When scrollable set TRUE,
scroller must be calculated its size except using item.
If content is removed, scroller doesn't calculate its size, so it has previous size.
So there is a empty space in this case.
@fix
Test Plan:
1. elementary_test -to popup
2. open "popup-center-title + list content + 2 button (with delete content)"
3. click "Delete" button
4. compare between scrollable set true and false
Reviewers: herb, singh.amitesh, jpeg, cedric, conr2d
Differential Revision: https://phab.enlightenment.org/D5010
Amitesh Singh [Tue, 11 Jul 2017 06:37:01 +0000 (15:37 +0900)]
elm test: label slide - use better color as background for label.
Amitesh Singh [Tue, 11 Jul 2017 06:12:45 +0000 (15:12 +0900)]
genlist: fix multiple focus edje signal issue
We need focus edje signal when item is focused or the already
focused item realizes. its wrong to call focus signal on
_elm_genlist_item_state_update()
fixes T4969
Bryce Harrington [Tue, 11 Jul 2017 02:35:25 +0000 (11:35 +0900)]
evas: Improve the documentation for evas_init() and evas_shutdown()
Summary:
Rephrase the documentation to read a bit more clearly, and restructure
the inclusions of stanzas from evas-buffer-simple.c to reflect the
recent updates to that example.
Reviewers: cedric
Subscribers: jpeg
Differential Revision: https://phab.enlightenment.org/D5015
Bryce Harrington [Tue, 11 Jul 2017 02:29:52 +0000 (11:29 +0900)]
evas: Fix stray periods
Summary: "to the Evas. @b buffer engine"
Reviewers: cedric
Subscribers: jpeg
Differential Revision: https://phab.enlightenment.org/D5014
Bryce Harrington [Tue, 11 Jul 2017 02:29:40 +0000 (11:29 +0900)]
gitignore: Ignore pc/efl-wl.pc
Reviewers: zmike
Subscribers: cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5013
Bryce Harrington [Tue, 11 Jul 2017 02:29:07 +0000 (11:29 +0900)]
examples: Move introduction to Ecore-Evas to evas-transparent.c
Summary:
The evas-transparent.c example is quite short, and so will allow for
more detailed explanation of setting up Ecore-Evas. Then
evas-object-manipulation.c can focus more on the explanation of objects
in Ecore-Evas.
Reviewers: cedric
Subscribers: jpeg
Differential Revision: https://phab.enlightenment.org/D5011
Derek Foreman [Mon, 10 Jul 2017 21:07:15 +0000 (16:07 -0500)]
ecore_evas_drm: Refuse to start a ticker for a manually rendered canvas
With atomic mode setting this has the hilarious side effect of waking
up the display from dpms.
Attempting to tick when manual render is set is likely a bug, so
log an ERR.
fix T5462
Derek Foreman [Mon, 10 Jul 2017 21:05:27 +0000 (16:05 -0500)]
ecore_evas: Don't start backend ticking for manually rendered canvases
Shortcut the code that starts up a back-end ticker if we're manually
rendered.
ref T5462
Marcel Hollerbach [Mon, 10 Jul 2017 17:57:56 +0000 (19:57 +0200)]
efl_ui_focus_manager: fix leaking safed_order
Marcel Hollerbach [Mon, 10 Jul 2017 14:22:32 +0000 (16:22 +0200)]
elementary: fix leaking manager objects
they are now automatically destroyed once the object is deleted.
Chris Michael [Mon, 10 Jul 2017 18:27:27 +0000 (14:27 -0400)]
elput: Update modifers for caps lock
Small patch to fix issue of not setting caps_lock modifier. As we
lookup this modifier index when creating keyboard info, we should be
testing for it and setting seat->modifiers accordingly.
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 10 Jul 2017 18:14:06 +0000 (14:14 -0400)]
elput: Fix support for setting keyboard led(s)
Small patch to update keyboard led(s) when caps, numlock, etc are
pressed. This patch adds some fields to internal (non-API exposed)
structures inside our private header (of a BETA-API library) and thus
should still be ok during freeze.
Fixes T5655
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 10 Jul 2017 17:13:59 +0000 (13:13 -0400)]
elput: Minor formatting fix
NB: No functional changes
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 10 Jul 2017 17:10:01 +0000 (13:10 -0400)]
ecore-wl2: Formatting fixes
NB: No functional changes
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 10 Jul 2017 17:02:43 +0000 (13:02 -0400)]
ecore-wl2: Add event handlers even if no data device manager exists
If we have no data device manager present, we would still like to
receive events when input devices get added or removed so these
handlers should be setup before we return from this function.
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 10 Jul 2017 17:02:02 +0000 (13:02 -0400)]
ecore-wl2: Rename internal function
As we are in the Ecore_Wl2 library, lets use a function name that
reflects this.
NB: No functional changes
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Jaehyun Cho [Mon, 10 Jul 2017 11:53:34 +0000 (20:53 +0900)]
Evas textblock: add null check in evas_textblock_cursor_free
Bryce Harrington [Mon, 10 Jul 2017 08:02:20 +0000 (17:02 +0900)]
examples: Flesh out the evas-object-manipulation example
Summary:
This is one of the basic examples that introduces Ecore-Evas, so needs
to be more detailed in commentary than other examples. It also points
out the use of the legacy API and directs the reader to the Eo example
where appropriate.
Reviewers: cedric
Subscribers: jpeg
Differential Revision: https://phab.enlightenment.org/D5008
Jean-Philippe Andre [Mon, 10 Jul 2017 05:21:13 +0000 (14:21 +0900)]
evas: Fix mouse event info in global events
When using the legacy API (and in fact also with the EO API) to
listen to mouse events (move, in, out...) on a window instead
of an actual evas object, some information was missing:
- buttons (bitmask of pressed buttons)
- prev.x/y (previous position)
This is because Evas had not handled the event yet at this
point, it was coming directly from ecore_evas with incomplete
information. This patch involves evas a little bit earlier, and
also fixes evas_events_legacy.c to have consistent values for
cur/prev canvas/ouput coordinates. See also
890a91785 and
484dae76e675318. Those commits were making the pointer coord
a seat-based property (instead of canvas-based) but the event
should already have those proper values before converting to
a legacy struct. This patch restores the meaning of the DUP
macros, as I observed 4 different coordinates from the app side
(instead of just 2: prev and cur).
Thanks to Andy for reporting the original issue on the ML!
Amitesh Singh [Mon, 10 Jul 2017 06:25:24 +0000 (15:25 +0900)]
Efl.Ui.Image: fix crash on passing NULL file in file_set()
img was crashing on elm_image_file_set(image, NULL, NULL).
fixes T5697
Carsten Haitzler (Rasterman) [Mon, 10 Jul 2017 02:30:35 +0000 (11:30 +0900)]
add missing license line for libefl to COPYING
LGPL will do - we never declared a license so take the most
restrictive one we have for this (LGPL).
Carsten Haitzler (Rasterman) [Sun, 9 Jul 2017 23:56:40 +0000 (08:56 +0900)]
default theme - add signals to pointer theme to fix glitch in x
in x11 the mouse pointer is separate to everything else on the screen,
and so when screensaver kicks in and we fade to/from black or we
suspend/resume and do the same... the mouse pointer stays annoyingly
visible and it just lookes like a bug. this allows that to be fixed by
allowing the pointer to be suspended or resumed... :)
@fix
Vincent Torri [Sun, 9 Jul 2017 09:08:08 +0000 (11:08 +0200)]
Ecore_Evas win32 module: fix memory leak
Vincent Torri [Fri, 7 Jul 2017 14:54:34 +0000 (16:54 +0200)]
Add ecore_win32_cursor_show() API and ecore_evas_win32_cursor_(un)set() functions
This fixes double mouse cursor in elm entries on Windows (for example)
@feature
@fix
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_cnp: fix selections for wl windows and FAKE wins
don't use hardcoded seat names and attempt to use wl protocol for
FAKE wins
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_cnp: use local cnp when elm_cnp owns the selection
no need to use protocol to send an app's own selection to itself
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_entry: don't use window_id to determine type of window
there's a real function for this...
ref
a173efd0a9892daca66bde76f7e90c66f39b9d23
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_entry: don't update PRIMARY selection under wayland
this doesn't exist yet, and elm_cnp forces PRIMARY updates to overwrite
CLIPBOARD which is not good
@fix
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elput: don't leak event pointers
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_cnp: mark x11 selection as lost when selection is lost in event
do not pretend that a widget maintains the selection when notified that
the selection has been lost
@fix
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_win: remove engine checks from window_id_get()
if these exist then just return the id
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_win: remove engine checks from wl_window_get
if there's a win, return it, otherwise this returns null anyway
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
elm_win: manually create a wl2 window for FAKE wins under wayland
this enables elm_cnp to function using protocol
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
ecore-wl2: create surfaces once compositor global is bound
it's expected that surfaces are created when a window exists
@fix
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
ecore-wl2: don't leak display event pointers
Mike Blumenkrantz [Fri, 7 Jul 2017 20:26:50 +0000 (16:26 -0400)]
ecore-wl2: add display object refs for windows
ensure the display lifetime exceeds the window lifetime
@fix
Amitesh Singh [Fri, 7 Jul 2017 06:52:51 +0000 (15:52 +0900)]
elm test: img zoomable - fix compiler warnings
efl/src/bin/elementary/test_photocam.c:779:19: note: initialize the variable '_zoom' to silence this warning
Jean-Philippe Andre [Fri, 7 Jul 2017 06:34:02 +0000 (15:34 +0900)]
win: Avoid CRI message before finalized
Test case was Bg EO API in elm_test
JinYong Park [Fri, 7 Jul 2017 05:11:55 +0000 (14:11 +0900)]
popup: fix wrong signal emit when remove all button
Summary:
In visuals_set function,
check action_area is NULL, and signal emit action_area,visible/hidden
So, before call visuals_set, action_area should be deleted.
Test Plan: Actually, action area has not min size, so it couldn't be observed.
Reviewers: herb, singh.amitesh, jpeg, cedric, conr2d
Reviewed By: jpeg
Differential Revision: https://phab.enlightenment.org/D5007
Amitesh Singh [Fri, 7 Jul 2017 06:39:37 +0000 (15:39 +0900)]
elm test: fake events - use efl_content_set instead
Amitesh Singh [Fri, 7 Jul 2017 06:24:15 +0000 (15:24 +0900)]
elm test: bg - use efl_content_set instead
there is no efl_packed on window. it won't work.
Amitesh Singh [Fri, 7 Jul 2017 05:54:35 +0000 (14:54 +0900)]
elm test: img zoomable - change anim code sample to use eo apis only
This sample was added to show the img anim usage.
lets make this sample to use eo apis only.
Uma Devika [Fri, 7 Jul 2017 02:48:25 +0000 (11:48 +0900)]
Evas and Eolian : Freed string buffer
Summary:
Summary : String buffer returned by eina_strbuf_new() is not freed in some cases
@Fix
Signed-off-by: Uma Devika <u.bodapati@samsung.com>
Reviewers: cedric, tasn, jpeg, raster, singh.amitesh
Subscribers: tanwar.umesh07, yashu21985, cedric, jpeg
Differential Revision: https://phab.enlightenment.org/D5000
Al Poole [Fri, 7 Jul 2017 02:19:31 +0000 (11:19 +0900)]
eina_file: fix unmap of unmapped region when Eina_File has copied data and not a map;
Summary:
This was causing problems on non-Linux architectures as eina_file_real_close unmapped not mapped data. Added a "copied" flag to Eina_File which is set on eina_file_virtualize (on copied data), and tested for when eina_file_real_close does the unmap. I'm surprised Linux allowed this. Certainly all of the BSDs crashed with the previous behaviour.
@fix T5479
Test Plan: Example inlcude Rage and Enlightenment Thumb on BSD systems which use eina_file_virtualize with emotion to obtain album artwork.
Reviewers: raster, cedric, jpeg
Reviewed By: jpeg
Maniphest Tasks: T5479
Differential Revision: https://phab.enlightenment.org/D5006
Jean-Philippe Andre [Thu, 6 Jul 2017 07:47:41 +0000 (16:47 +0900)]
evas,edje,elm: Mark all legacy objects as such