Subject: Re: [E-devel] [Patch][Ecore][Win32] Checking control character
I missed the updating WinCE. and..
Mr. Vincent Torri enlighten me about ChangeLog and NEWS also. Thanks!
SVN revision: 70219
Subject: [E-devel] [Patch][Ecore][Win32] Checking control character
The control characters are generated by holding down the Control key while
you strike another letter or symbol key.
Because of this reason, The Evas_Event_Key_Down in the
EVAS_CALLBACK_KEY_DOWN callback does not have proper keyname.
So I have shifted the control character to printing character. Please
review the patch and give any feedbacks. Thanks.
SVN revision: 70186
People, when you go and change ecore_event_add() to
_ecore_event_add(), please NOTICE THAT THE FREE FUNCTION IS NOT
AUTOMATICALLY SPECIFIED!
So specify these functions to _ecore_event_add() to stop leaking every
signal we receive from system.
SVN revision: 70177
Subject: [E-devel] [patch] ecore doxygen doc (3)
Date: Thu, 12 Apr 2012 15:39:16 +0900
Some leftovers, fix:
/tmp/ecore/src/lib/ecore_x/xlib/ecore_x_randr_12.c:1202: warning: The following parameters of ecore_x_randr_crtc_pos_relative_set(Ecore_X_Window root, Ecore_X_Randr_Crtc crtc_r1, Ecore_X_Randr_Crtc crtc_r2, Ecore_X_Randr_Output_Policy policy, Ecore_X_Randr_Relative_Alignment alignment) are not documented:
parameter 'root'
/tmp/ecore/src/lib/ecore_x/xcb/ecore_xcb_randr.c:921: warning: The following parameters of ecore_x_randr_mode_size_get(Ecore_X_Window root, Ecore_X_Randr_Mode mode, int *w, int *h) are not documented:
parameter 'root'
/tmp/ecore/doc/examples.dox:242: warning: unable to resolve reference to `ecore_event_example.c' for \ref command
SVN revision: 70119
Subject: [E-devel] [patch] ecore doxygen doc (2)
Date: Thu, 12 Apr 2012 12:46:04 +0900
Hi,
This is a big patch. It fixes:
- undef #EINA_{TRUE,FALSE} links
- @c for NULL and EINA_{TRUE,FALSE}
- some formatting/spello
- several missing return types
SVN revision: 70117
pointer grabs, shm_pool interface. Rename fields of the
wl_event_mouse_in/out structures to match other ecore event structs.
Add API functions for getting a list of outputs, for getting a windows
shell_surface, and for setting a windows parent.
SVN revision: 69997
documentation.
I didn't add the API declaration into the header yet because the API
name/parameter might be changed before release.
SVN revision: 69990
Subject: [E-devel] [patch] ecore doxygen doc
Here is a patch to correct some problems in ecore doxygen doc.
It fixes:
/tmp/ecore/doc/examples.dox:1173: warning: Unsupported xml/html tag
<some_num> found
/tmp/ecore/doc/examples.dox:1174: warning: Unsupported xml/html tag
<some_path> found
/tmp/ecore/doc/examples.dox:1176: warning: Unsupported xml/html tag
<some_num> found
/tmp/ecore/src/lib/ecore_con/ecore_con_ssl.c:714: warning: The
following parameters of
ecore_con_ssl_server_privkey_add(Ecore_Con_Server *svr, const char
*key_file) are not documented:
parameter 'svr'
/tmp/ecore/src/lib/ecore_con/Ecore_Con.h:1360: warning: The
following parameters of ecore_con_url_http_version_set(Ecore_Con_Url
*url_con, Ecore_Con_Url_Http_Version version) are not documented:
parameter 'url_con'
/tmp/ecore/src/lib/ecore_evas/Ecore_Evas.h:530: warning: argument
'demand_attention' of command @param is not found in the argument list
of ecore_evas_demand_attention_set(Ecore_Evas *ee, Eina_Bool demand)
/tmp/ecore/src/lib/ecore_evas/Ecore_Evas.h:530: warning: The following
parameters of ecore_evas_demand_attention_set(Ecore_Evas *ee,
Eina_Bool demand) are not documented:
parameter 'demand'
/tmp/ecore/src/lib/ecore_x/xcb/ecore_xcb_damage.c:129: warning:
Unsupported xml/html tag <empty> found
There are more things to fix due to API change but it's not so obvious
for me. I will first continue to check/correct trivial things in the
others efl doxygen doc.
SVN revision: 69820
I found a bug in ecore main loop while debuging cpu 100% issue on RSS
application.
1. [RSS] RSS application register two io watch callbacks(for G_IO_IN,
G_IO_OUT seperately) for a GIOChannel with g_io_add_watch().
2. [ecore] In _ecore_glib_context_query() function, g_main_context_query()
returns a fd list, and it has 2 fd items for the GIOChannel (channel
id = 20).
itr[0] (16, 1, 0)
itr[1] (15, 1, 0)
itr[2] (20, 1, 0) (G_IO_IN, 0)
itr[3] (20, 4, 0) (G_IO_OUT, 0)
itr[4] (18, 1, 0)
3. [ecore] In _ecore_glib_context_poll_from() function, create read, write,
exception fd list according to the events flags of each fd item.
[6 15 16 18 20], [20], []
4. [ecore] in _ecore_glib_select__locked() function, get active fd number from
select() call
select(21, [6 15 16 18 20], [20], [], NULL) = 1 (out [20])
5. [ecore] In _ecore_glib_context_poll_to() function, there is a bug on
setting revents flag.
(because of incorrect condition check - currently, the logic of the
function cannot handle this case)
itr[0] (16, 1, 0)
itr[1] (15, 1, 0)
itr[2] (20, 1, 4) (set incorrectly)
itr[3] (20, 4, 0) => this should be set as (20, 4, 4)!!!!
itr[4] (18, 1, 0)
6. [ecore] In _ecore_glib_select__locked(), g_main_context_check() function
returns false because of the above bug, so g_main_context_dispatch()
function will not be called.
>> After this, the 2~6 flow is executed repeatedly on ecore main loop
(because 20 out is still active fd) and this makes cpu 100% problem.
SVN revision: 69739
Fix ecore_wl_window surface attach of the buffer for shm and egl types.
NB: This ensures that both efl/elm clients work under the Wayland
Compositor AND the E17 Compositor.
NB: The issue here is that clients (when run under the Wayland
compositor) need to make sure 'iterate' has
been called before creating surfaces so that the wl_compositor is
setup. However, the actual E17 Compositor will hang if 'iterate' is called
to early.
SVN revision: 69707
Handle 'axis' events from multi-touch devices (also handles mouse
wheel).
Remove wl_buffer_damage calls (obsolete in newest wayland).
NB: This will require the 'latest' wayland code from their git repo as
of 2012-03-27.
SVN revision: 69672
- Comment out display_iterate during intialize (hangs compositor
initialize).
- Simplify ecore_wl_flush routine and avoid possible hangs in
wl_display_roundtrip.
- Add function (for trip ;) ) to be able to return the screen dpi.
- Implement functions for input grab/ungrab.
- Set root.x/y in mouse events.
- Remove freeing of mouse_move event (already handled generically).
- Handle setting output mm width/height (for calc of dpi).
- Implement setting surface input & opaque regions.
- Implement popup windows.
- Implement ecore_wl_window_transparent_set function.
- Implement function to allow setting Ecore_Wl_Window type.
- Implement setting the wayland mouse pointer to an image (via
ecore_wl_window_pointer_set).
- Comment out wl_display_iterate during initialize (can cause hangs in
compositor).
- Simplify ecore_wl_flush (to avoid compositor hangs).
SVN revision: 69644
This works in linux and windows, and should fix shm_detection on BSD (including Mac)
BSD, Mac and solaris users : please check that it compiles and shm_open is detected
SVN revision: 69616
Subject: [E-devel] [PATCH] ecore/ecore_fb: Tweaks in _ecore_fb_size_get
As the width and height are defined as static, no need to initialize it
again to zero. Also removed checks for 'valid pointers' as it is redundant.
SVN revision: 69556
Subject: [E-devel] [Patch][ecore_con] Simple bug fix
Date: Mon, 12 Mar 2012 13:45:02 +0900
Hello,
This is patch to prevent a segfault if curlmsg == NULL.
Thanks.
SVN revision: 69217
(interfaces_bound) to be raised when ecore_wayland has everything it
needs to function properly. (Used in the e17 wayland only code).
SVN revision: 69032
NOTE: I am now all for removing this lock. They trigger crash,
are not implementing the feature they are advertising, are not
maintained and impossible to fix.
SVN revision: 68987
Subject: [E-devel] [Patch] Ecore, Elementary: Supporting indicator
opacity mode
This is Myoungwoon Roy Kim.
This patches are for supporting the indicator's opacity mode and made by
Jeonhoon Park(jh1979.park@samsung.com) who is responsible for Indicator
application.
Requirements:
- In mobile device, Indicator area should be displayed as Opacity,
Transparency, or sometimes Translucency according to the UX
requirements.
This requirement can be applied in case of fullscreen based menu and
fullscreen applications like video player etc.
Functional requirements:
1. User can set indicator's opacity mode as opacity, transparency, and
translucency
2. User can get the current indicator's opacity mode.
Currently there are no APIs for supporting the upper functional
requirements.
Thus, he added support for indicator's opacity mode.
It is designed for EFL developers easily to set the indicator's opacity
like the existing indicator's mode.
He added a Elm_Win_Indicator_Opacity_Mode structure variable into
Elm_Win
structure.
Meanwhile, new two APIs are implemented independently from the existing
source code.
Anybody please review this and apply it to upstream code.
SVN revision: 68958
things. Implement alpha_set, transparent_set, and ignore_events for
ecore_evas_wayland_EGL.
NB: Big fat warning ... don't try to use alpha_set for EE_Wayland_Egl.
It segfaults for some unknown reason. If anyone knows EGL code, please
feel free to have a look. Any help is appreciated here.
SVN revision: 68574
EFL applications as Wayland Clients.
NB: This (along with the ecore_evas changes) fixes most (if not all)
outstanding issues what I am aware of, including the nasty resize bug
with wayland_egl.
SVN revision: 68515
* use AC_COMPILE_IFELSE after AC_PROG_CC has been called
* use EFL_CHECK_TESTS after pkg-config has been checked
* enable ecore_evas_extn only if its dependencies are found
SVN revision: 68312
Subject: [E-devel][Patch][ecore_con] Fix invalid curl handle removal by valgrind
Date: Wed, 22 Feb 2012 19:57:36 +0900
Hello,
discomfitor reports bugs by valigrind. I checked it with valgrid and
I found the clues
curl_multi_remove_handle() should not be called when multi handles
being performed. So I removed curl_multi_remove_handle() code from
_ecore_con_url_info_read()
Now, curl_multi_remove_handle() is only called for all easy handles
when a multi-handle ended.
Please review this simple patch.
SVN revision: 68287
NB: These may not be entirely correct, but since I am the only one using
xcb (apparently), and I don't ever use the RandR stuff, they are
sufficient for now. I'll debug them later when I have more time.
SVN revision: 68219
NOTES: It is now safer and faster. I doubt I will have more time before the release to finish
ecore_thread_message_run, nor to make the shutdown nicer.
SVN revision: 68164