Jean-Philippe Andre [Wed, 18 Mar 2015 01:04:27 +0000 (10:04 +0900)]
Evas: Really use GLES3 by default (fix logic)
I enabled gles3 by default in a previous commit but apparently my
logic was all wrong and it fell back to gles2.
Bryce Harrington [Wed, 18 Mar 2015 00:30:32 +0000 (20:30 -0400)]
ecore-wayland: Use calloc rather than malloc+memset
Summary:
In addition to being fewer lines of code,
malloc + memset to 0 is slower than calloc. See
http://stackoverflow.com/questions/2688466/why-mallocmemset-is-slower-than-calloc
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewers: zmike, cedric, devilhorns
Reviewed By: devilhorns
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2182
Bryce Harrington [Tue, 17 Mar 2015 23:52:43 +0000 (19:52 -0400)]
ecore-wayland: Doublecheck input validity
Summary:
Most of the other ecore-wayland API will return NULL in the (hopefully
theoretical) case of bad inputs; this API routine should so similarly
for consistency.
Reviewers: zmike, cedric
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2181
Bryce Harrington [Tue, 17 Mar 2015 23:23:48 +0000 (19:23 -0400)]
ecore_wayland: Improve API docs
Summary:
For calls that return pointers, specify that NULL is a
potential return value. Add API documentation for some misc. routines
missing it.
Reviewers: zmike, cedric, devilhorns
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2180
Chris Michael [Tue, 17 Mar 2015 19:53:38 +0000 (15:53 -0400)]
Merge branch 'devs/devilhorns/evas_wayland_shm'
This merge pulls in the Evas Wayland Shm engine refactor which
addressed resize issues causing crashes in E wl compositor.
Chris Michael [Tue, 17 Mar 2015 19:50:56 +0000 (15:50 -0400)]
ecore-evas-wayland: Remove frame callbacks from Ecore_Evas wayland
engine
Summary: Frame callbacks are now handled inside the engine itself and
are thus not needed here anymore
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 17 Mar 2015 19:49:09 +0000 (15:49 -0400)]
evas-wayland-shm: Refactor Evas Wayland Shm Engine (Fix T2201)
Summary: This patch set fixes the issue of Evas wayland shm engine
causing crashes when resizing efl apps in the E wayland compositor. It
refactors the evas engine to wait for release events on buffers, and
hooks into frame callbacks so that release events will get triggered
at the appropriate time.
@fix
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 17 Mar 2015 19:32:29 +0000 (15:32 -0400)]
evas-wayland-shm: Remove old wayland_shm engine swapper files
Summary: Part of engine refactor to address resize issue in E wayland
compositor
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Boram Park [Tue, 17 Mar 2015 13:30:59 +0000 (09:30 -0400)]
ecore-drm: add ecore_drm_devices_get to get the list of drm devices
Summary:
when enlightenment is working as wayland display server, enlightenment
changes KDSETMODE to KD_GRAPHICS in _ecore_drm_tty_setup(). However,
when enlightenment is killed by SIGSEGV, it doesn't changes KDSETMODE
to KD_TEXT because englightenment process doesn't call ecore_drm_tty_close().
To make possible enlightenment call ecore_drm_tty_close(), drm devices
should be exposed.
When enlightenment is killed by SIGSEGV, it will get drm devices with
ecore_drm_device_get_list(), and will call ecore_drm_launcher_disconnect(),
and ecore_drm_launcher_disconnect will call ecore_drm_tty_close() internally.
@feature
Change-Id: I1c594739ec96660a09cee77b823ace6548ee5282
Reviewers: zmike, cedric, raster, gwanglim, devilhorns
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D2159
Jean-Philippe Andre [Tue, 17 Mar 2015 10:19:28 +0000 (19:19 +0900)]
cmake: Add files for Emile, the serial killer
pierre lamot [Thu, 12 Mar 2015 11:13:08 +0000 (12:13 +0100)]
ecore_cocoa: add missing window state functions
implement missing window state functions:
* raise
* lower
* activate
* iconified_set
* withdrawn_set
* move
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Wed, 11 Mar 2015 14:31:36 +0000 (15:31 +0100)]
ecore_cocoa: add functions to allow min/max window size
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Tue, 3 Mar 2015 17:11:55 +0000 (18:11 +0100)]
ecore_cocoa: fix mouse scrollwheel direction
horizontal and vertical direction where inverted
@fix
pierre lamot [Tue, 3 Mar 2015 16:11:08 +0000 (17:11 +0100)]
ecore_cocoa: don't reject mouse events outside the window
mouse events ouside the window where rejected. that was creating
interactions where half an event was created, the application was
then not able to process correctly following events.
@fix
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Tue, 3 Mar 2015 16:01:03 +0000 (17:01 +0100)]
eio: don't restart a monitor on MacOSX when every paths have been removed.
This patch check that when a monitor is removed the FSEvent service, it is
not restarted if there is no more paths to monitor. This was generating an
error message from FSEvent.
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Tue, 24 Feb 2015 14:50:58 +0000 (15:50 +0100)]
evas: bypass lockfocus in NSView on OSX
this patch bypass calls to (un)lockFocus, avoid segv
when closing windows
This is certainly not the cleanest way to do it but it seems
to be working correctly at the moment. this may leads to
unexepected behavior which haven't been identified now
@fix
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Fri, 20 Feb 2015 13:35:16 +0000 (14:35 +0100)]
ecore_cocoa: add support for non latin keyboard events
this patch adds support for non latin keyboard layout such as
arabic, greek or hebrew, some language might not be functionnal
tough like korean or chineese
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Fri, 30 Jan 2015 17:21:14 +0000 (18:21 +0100)]
ecore_evas: rewrite render loop to conform with other engines.
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Fri, 30 Jan 2015 16:42:09 +0000 (17:42 +0100)]
evas: make current context on lockFocus for gl_cocoa backend.
make current context on lockFocus as suggested by (following best practice):
https://developer.apple.com/library/mac/documentation/GraphicsImaging/Conceptual/OpenGL-MacProgGuide
@fix
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
pierre lamot [Fri, 30 Jan 2015 16:15:55 +0000 (17:15 +0100)]
ecore_cocoa: release resources on window close event
@fix this patch:
catch the window close event from cocoa and send an ecore event
this event is catched by a handler in ecore_evas wich will
call the registered fn_delete_request (from elementary for instance)
/!\ this patch is currently incomplete and leads to a segv when
closing the last window
Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
Jean-Philippe Andre [Tue, 17 Mar 2015 03:21:00 +0000 (12:21 +0900)]
Evas GL: Fallback to safe GL configs with GLX as well
Jean-Philippe Andre [Tue, 17 Mar 2015 03:12:19 +0000 (12:12 +0900)]
Evas GL: Fallback to safe values of depth, stencil & msaa if not supported
Depth32, Stencil16 and MSAA are known to be unsupported on many platforms.
While applications should try not to request them, we can try to fallback
nicely and still render using depth24+stencil8 (which is often supported),
or reducing the number of MSAA samples (until 0 if not supported at all).
Cedric BAIL [Tue, 17 Mar 2015 08:59:16 +0000 (09:59 +0100)]
Merge branch 'devs/cedric/emile'
Here come a major code refactoring. There is still work to be done, especially on Ecore_Con_SSl
and also image compression, but it is already an improvement over current situation. Further
development should happen in master.
Emile does provide a library to put together serialization, compression and ciphering. It is a
low level library and can be use by anything above Eina.
Cedric BAIL [Tue, 17 Mar 2015 07:51:01 +0000 (08:51 +0100)]
emile: fix coding style with ecrustify.
Cedric BAIL [Tue, 17 Mar 2015 07:50:58 +0000 (08:50 +0100)]
emile: add a note regarding useless structure member.
Cedric BAIL [Tue, 17 Mar 2015 07:50:57 +0000 (08:50 +0100)]
ecore_con: fix build with -DISCOMFITOR and OpenSSL.
Cedric BAIL [Tue, 17 Mar 2015 07:50:56 +0000 (08:50 +0100)]
ecore_con: remove deprecated use of Eina_Binbuf functions.
Cedric BAIL [Tue, 17 Mar 2015 07:50:55 +0000 (08:50 +0100)]
edje: remove deprecated use of Eina_Binbuf functions.
Cedric BAIL [Tue, 17 Mar 2015 07:50:54 +0000 (08:50 +0100)]
eet: remove use of deprecated Eina_Binbuf functions.
Cedric BAIL [Tue, 17 Mar 2015 07:50:53 +0000 (08:50 +0100)]
emile: remove use of deprecated eina binbuf function.
Cedric BAIL [Tue, 17 Mar 2015 07:50:52 +0000 (08:50 +0100)]
eina: provide a generic eina_binbuf_manage_new.
eina_binbuf_manage_new provide the functionnality of both eina_binbuf_manage_new_length
and eina_binbuf_manage_read_only_new_length. Those function get deprecated.
Cedric BAIL [Tue, 17 Mar 2015 07:50:51 +0000 (08:50 +0100)]
emile: make sure given boolean are really boolean.
Cedric BAIL [Tue, 17 Mar 2015 07:50:50 +0000 (08:50 +0100)]
emile: remove binbuf from emile compression functions.
Cedric BAIL [Tue, 17 Mar 2015 07:50:49 +0000 (08:50 +0100)]
emile: use Eina_Log for Jpeg error message.
Cedric BAIL [Tue, 17 Mar 2015 07:50:49 +0000 (08:50 +0100)]
ecore_con: unbreak GNUTLS support.
Cedric BAIL [Tue, 17 Mar 2015 07:50:48 +0000 (08:50 +0100)]
emile: update to latest GNUTLS/GCRYPT API.
Vincent Torri [Tue, 17 Mar 2015 07:50:47 +0000 (08:50 +0100)]
emile: fix some warnings in gnutls component
Vincent Torri [Tue, 17 Mar 2015 07:50:46 +0000 (08:50 +0100)]
emile: ERROR is a macro already defined on Windows
Vincent Torri [Tue, 17 Mar 2015 07:50:45 +0000 (08:50 +0100)]
emile: fix emile_binbuf_sha1() signature when no cipher is used
Vincent Torri [Tue, 17 Mar 2015 07:50:44 +0000 (08:50 +0100)]
emile: fix array count
Cedric BAIL [Tue, 17 Mar 2015 07:50:44 +0000 (08:50 +0100)]
emile: cleanup comments and internal functions name.
Cedric BAIL [Tue, 17 Mar 2015 07:50:43 +0000 (08:50 +0100)]
emile: cleanup private header.
Cedric BAIL [Tue, 17 Mar 2015 07:50:42 +0000 (08:50 +0100)]
emile: cleanup header from useless cruft.
Cedric BAIL [Tue, 17 Mar 2015 07:50:42 +0000 (08:50 +0100)]
emile: follow convention for Emile_Compressor_Level.
Cedric BAIL [Tue, 17 Mar 2015 07:50:41 +0000 (08:50 +0100)]
emile: normalize compression header.
Cedric BAIL [Tue, 17 Mar 2015 07:50:40 +0000 (08:50 +0100)]
emile: normalize version number to 1.14.
Cedric BAIL [Tue, 17 Mar 2015 07:50:40 +0000 (08:50 +0100)]
emile: implement emile_cipher_module_get in a useful way.
Cedric BAIL [Tue, 17 Mar 2015 07:50:39 +0000 (08:50 +0100)]
emile: document the stable cipher function.
Cedric BAIL [Tue, 17 Mar 2015 07:50:38 +0000 (08:50 +0100)]
emile: actually this set of function are neat enough to be a public API.
Cedric BAIL [Tue, 17 Mar 2015 07:50:37 +0000 (08:50 +0100)]
emile: change the API to allow supporting other cipher in the future.
Cedric BAIL [Tue, 17 Mar 2015 07:50:36 +0000 (08:50 +0100)]
emile: mark ssl layer as a beta API for now.
Cedric BAIL [Tue, 17 Mar 2015 07:50:36 +0000 (08:50 +0100)]
eet: use Emile type for encoding.
Cedric BAIL [Tue, 17 Mar 2015 07:50:35 +0000 (08:50 +0100)]
emile: document Emile_Image.
Cedric BAIL [Tue, 17 Mar 2015 07:50:34 +0000 (08:50 +0100)]
emile: reduce code duplication between gnutls and openssl backend.
Add a visible emile_binbuf_sha1 implementation and deduplicate
emile_pbkdf2_sha1 (but does not explicitely expose it).
Cedric BAIL [Tue, 17 Mar 2015 07:50:33 +0000 (08:50 +0100)]
emile: remove ambiguity especially for Windows platform.
Thanks Vincent Torri for the review.
Cedric BAIL [Tue, 17 Mar 2015 07:50:33 +0000 (08:50 +0100)]
emile: add documentation for compression function.
Cedric BAIL [Tue, 17 Mar 2015 07:50:32 +0000 (08:50 +0100)]
emile: use an enum for level instead of just an int.
Cedric BAIL [Tue, 17 Mar 2015 07:50:31 +0000 (08:50 +0100)]
emile: update documentation version.
Cedric BAIL [Tue, 17 Mar 2015 07:50:30 +0000 (08:50 +0100)]
emile: split headers.
Cedric BAIL [Tue, 17 Mar 2015 07:50:29 +0000 (08:50 +0100)]
emile: fix crash when cspaces is left unchanged.
Cedric BAIL [Tue, 17 Mar 2015 07:50:29 +0000 (08:50 +0100)]
emile: build examples after adding emile.
Cedric BAIL [Tue, 17 Mar 2015 07:50:28 +0000 (08:50 +0100)]
eet: encode lossy image with AGRY88 or GRY8 whenever possible.
Cedric BAIL [Tue, 17 Mar 2015 07:50:27 +0000 (08:50 +0100)]
eet: refactoring jpeg encoding.
Cedric BAIL [Tue, 17 Mar 2015 07:50:26 +0000 (08:50 +0100)]
eet: handle AGRY88 encoding for JPEG encoded image.
The alpha and the grey being encoded in two separated grey JPEG.
Cedric BAIL [Tue, 17 Mar 2015 07:50:25 +0000 (08:50 +0100)]
emile: handle AGRY88 for JPEG image.
Cedric BAIL [Tue, 17 Mar 2015 07:50:24 +0000 (08:50 +0100)]
eet: fix error handling when decoding TGV file format.
Cedric BAIL [Tue, 17 Mar 2015 07:50:23 +0000 (08:50 +0100)]
eet: use Emile to decode jpeg.
Cedric BAIL [Tue, 17 Mar 2015 07:50:22 +0000 (08:50 +0100)]
emile: decode GRAY JPEG as GRY8.
Cedric BAIL [Tue, 17 Mar 2015 07:50:21 +0000 (08:50 +0100)]
emile: simplify error handling for jpeg data decoding.
Cedric BAIL [Tue, 17 Mar 2015 07:50:21 +0000 (08:50 +0100)]
evas: use emile to decompress JPEG files.
Cedric BAIL [Tue, 17 Mar 2015 07:50:20 +0000 (08:50 +0100)]
evas: simplify structure.
Cedric BAIL [Tue, 17 Mar 2015 07:50:19 +0000 (08:50 +0100)]
emile: add JPEG support.
Cedric BAIL [Tue, 17 Mar 2015 07:50:18 +0000 (08:50 +0100)]
eet: use Emile to decode TGV image section.
Cedric BAIL [Tue, 17 Mar 2015 07:50:17 +0000 (08:50 +0100)]
emile: remove use of custom structure and prefer Eina_Rectangle.
Get rid of warning inside of the jpeg loader that result of it. I do believe
this is not an ABI break on the loader API. If you disagree, please raise your
voice.
Cedric BAIL [Tue, 17 Mar 2015 07:50:16 +0000 (08:50 +0100)]
evas: use Emile to decode TGV.
Cedric BAIL [Tue, 17 Mar 2015 07:50:15 +0000 (08:50 +0100)]
evas: use Emile for defining all possible loader API.
Cedric BAIL [Tue, 17 Mar 2015 07:50:15 +0000 (08:50 +0100)]
emile: initial addition of emile image support.
Cedric BAIL [Tue, 17 Mar 2015 07:50:14 +0000 (08:50 +0100)]
eet: use eina_binbuf and emile to manipulate TGV compressed data.
Cedric BAIL [Tue, 17 Mar 2015 07:50:13 +0000 (08:50 +0100)]
emile: trying to fix security.
Cedric BAIL [Tue, 17 Mar 2015 07:50:13 +0000 (08:50 +0100)]
emile: Add SSL support.
Cedric BAIL [Tue, 17 Mar 2015 07:50:12 +0000 (08:50 +0100)]
build: display crypto system as a feature of Emile.
Cedric BAIL [Tue, 17 Mar 2015 07:50:10 +0000 (08:50 +0100)]
emile: remove left over #ifdef
Cedric BAIL [Tue, 17 Mar 2015 07:50:09 +0000 (08:50 +0100)]
emile: make the initialization part of backend cipher file to.
Cedric BAIL [Tue, 17 Mar 2015 07:50:08 +0000 (08:50 +0100)]
emile: with Emile and as long as you don't use any cipher, it will be just fine.
Cedric BAIL [Tue, 17 Mar 2015 07:50:07 +0000 (08:50 +0100)]
emile: split OpenSSL, GNUTLS and no cipher into separate file as a first step toward module.
Cedric BAIL [Tue, 17 Mar 2015 07:50:06 +0000 (08:50 +0100)]
emile: make it cross platform.
Cedric BAIL [Tue, 17 Mar 2015 07:50:05 +0000 (08:50 +0100)]
eet: initialize emile on demand.
Cedric BAIL [Tue, 17 Mar 2015 07:50:04 +0000 (08:50 +0100)]
ecore_con: initialize emile cipher support on demand.
Cedric BAIL [Tue, 17 Mar 2015 07:50:03 +0000 (08:50 +0100)]
emile: make cipher initialization optional.
Cedric BAIL [Tue, 17 Mar 2015 07:50:03 +0000 (08:50 +0100)]
eet: use Emile instead of Zlib and LZ4 directly.
Cedric BAIL [Tue, 17 Mar 2015 07:50:02 +0000 (08:50 +0100)]
emile: add compress/uncompress logic.
Cedric BAIL [Tue, 17 Mar 2015 07:50:01 +0000 (08:50 +0100)]
eet: move to use emile for ciphering logic.
Cedric BAIL [Tue, 17 Mar 2015 07:50:00 +0000 (08:50 +0100)]
emile: expose cipher/uncipher block logic.
Cedric BAIL [Tue, 17 Mar 2015 07:49:59 +0000 (08:49 +0100)]
ecore_con: depend on emile for initializing crypto library.
Cedric BAIL [Tue, 17 Mar 2015 07:49:58 +0000 (08:49 +0100)]
emile: move GNUTLS and OpenSSL initialization logic from Eet to Emile.
Cedric BAIL [Tue, 17 Mar 2015 07:49:57 +0000 (08:49 +0100)]
emile: initial introduction of Emile.
The intent of Emile is to be the common layer for serialisation, compression
and ciphering. It will expose the library we currently use internally to an
easier use from the outside (like gcrypt and lz4). It should improve portability.
Instead of pushing JSON, XML and what's not to Eina, I do think that they will
fit better in Emile.
As for the naming of Emile, you will need to be French and say :
"Un quoi ?" "Un serializer !"
Regarding why it is put there in the stack. Right now there is two users of
compression (eet and terminology), two users of cipher library (eet and ecore_con)
and a few handful of user for serialization (eina, eet, efreet, ecore_con, ...).
So the choice was quite simple, it needed to be below Eet. Now it could have been
on top of Eo or integrated into Eina.
One of the use case I am thinking of, is to compress Eo object when a canvas get
hidden/minized. For that it require Eo to use that library and it can't be a higher
level object. And with current implementation of Eo it is perfectly possible to
implement such idea. So not at Eo level.
As for Eina, I am starting to think it is getting to much things in its namespace.
I do believe that infact Eina_Simple_XML and Eina_File should after all have landed
in their own library. That's why I am putting the current logic in a new library.
It is going to expand, I want it to provide an few SAX like parser for JSON,
Eet_Data and protobuf with also an API like Eet_Data to directly feed those value
into a C structure without using a DOM at all. It would also be the right place
to experiment and benchmark for a new Eet_Data format that could be more efficient
to use.
So at the end, and due to how I see things going and being used, I do think it
is better of in its own library.
Cedric BAIL [Tue, 17 Mar 2015 07:49:56 +0000 (08:49 +0100)]
efl: autotools - move Eet section after Eo.
Cedric BAIL [Tue, 17 Mar 2015 07:49:54 +0000 (08:49 +0100)]
Revert "Add --disable-gui."
This reverts commit
3ed4f745907d3e842e64b341e2426c0c99cf3297.
Daniel Zaoui [Tue, 17 Mar 2015 07:32:54 +0000 (09:32 +0200)]
Eolian: add filename to error message
Carsten Haitzler (Rasterman) [Tue, 17 Mar 2015 03:04:10 +0000 (12:04 +0900)]
efl - bring back symlink complaint regardless of env vars
@fix