platform/upstream/efl.git
9 years agoefreetd - cache - fix handling a text index file with 0 length lines
Carsten Haitzler (Rasterman) [Wed, 1 Apr 2015 11:43:54 +0000 (20:43 +0900)]
efreetd - cache - fix handling a text index file with 0 length lines

this fixes an efreetd cache crash where files like icon.exts might
have lines with just a newline and no extension/content on the line.

@fix

9 years agoedje_calc: fix SEGFAULT on new parts
Andrii Kroitor [Wed, 1 Apr 2015 11:23:38 +0000 (07:23 -0400)]
edje_calc: fix SEGFAULT on new parts

Summary: new part added with edje_edit_part_add has NULL previous description

Reviewers: cedric, zmike, seoz, Hermet, raster

Subscribers: FurryMyad, cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D2272

9 years agoecore: Move documentation for ecore_timer_add at the correct place.
Stefan Schmidt [Wed, 1 Apr 2015 08:23:00 +0000 (10:23 +0200)]
ecore: Move documentation for ecore_timer_add at the correct place.

Many thanks to Vinicius dos Santos Oliveira for spotting this and
pointing it out on the mailing list.

9 years agoedje calc now more correctly forces recalc for child groups on state change
Mike Blumenkrantz [Wed, 1 Apr 2015 03:28:33 +0000 (23:28 -0400)]
edje calc now more correctly forces recalc for child groups on state change

in the event of fixed size -> non-fixed size (eg. previous commit optimization),
this calc would no longer occur, so we need to queue it. also if fixed.w or
fixed.h changes value for a group part, we must recalc the group to ensure correct
sizing occurs

9 years agoedje calc no longer performs calculations for group parts of fixed size > 0
Mike Blumenkrantz [Wed, 1 Apr 2015 03:27:01 +0000 (23:27 -0400)]
edje calc no longer performs calculations for group parts of fixed size > 0

when I said > 0 in the last commit message, I was thinking ahead to this commit
which I knew I would later have to make, but had not yet written because I had not
spent the requisite number of hours debugging the code to know that I needed to
have the check in both the code and the commit message

ref 3a451650d2f875077419e5777c49d04c618211f0

9 years agoEvas masking: Implement polygon masking (GL)
Jean-Philippe Andre [Wed, 1 Apr 2015 02:01:48 +0000 (11:01 +0900)]
Evas masking: Implement polygon masking (GL)

9 years agoEvas masking: Ensure texture is up-to-date before drawing (GL)
Jean-Philippe Andre [Wed, 1 Apr 2015 01:54:45 +0000 (10:54 +0900)]
Evas masking: Ensure texture is up-to-date before drawing (GL)

9 years agoEvas GL common: whitespace fixes
Jean-Philippe Andre [Wed, 1 Apr 2015 01:42:14 +0000 (10:42 +0900)]
Evas GL common: whitespace fixes

9 years agoEvas masking: Implement line masking in GL
Jean-Philippe Andre [Wed, 1 Apr 2015 01:39:36 +0000 (10:39 +0900)]
Evas masking: Implement line masking in GL

This commit also introduces a few whitespace changes...

9 years agoEvas GL generic: Simplify "scaled" images (used for masking)
Jean-Philippe Andre [Wed, 1 Apr 2015 00:58:05 +0000 (09:58 +0900)]
Evas GL generic: Simplify "scaled" images (used for masking)

Invert the meaning of scaled (w,h), so that im->(w,h) corresponds
to the final scaled size, and the original size is stored directly
in the texture itself.

This simplifies code a little bit.

Also, lift the limitation on the maximum texture size, as those
virtual textures are not limited by GPU texture size.

9 years agoEvas masking: Refactor GL code
Jean-Philippe Andre [Tue, 31 Mar 2015 09:58:51 +0000 (18:58 +0900)]
Evas masking: Refactor GL code

Use the same method as for map masking:
- Pass absolute geometry of the mask to the push functions,
- Compute absolute position in the shader (like gl_Position)
  and apply scaling factor + offset to sample the mask.

Masking now uses a single vec4 attribute instead of hi-jacking
other vertices.

This way, all masking shaders share the same code and are way
simpler. On the other hand, the vertex shaders have a little bit
more work to do.

9 years agoecore_file - fix nasty memory issues in ecore_file_app_exe_get()
Carsten Haitzler (Rasterman) [Wed, 1 Apr 2015 00:30:13 +0000 (09:30 +0900)]
ecore_file - fix nasty memory issues in ecore_file_app_exe_get()

valgrind was most unhappy with ecore_file_app_exe_get(). like:

==8331== Invalid write of size 1
==8331==    at 0x68DE90A: ecore_file_app_exe_get (ecore_file.c:994)
==8331==  Address 0x1348e58f is 0 bytes after a block of size 31 alloc'd
==8331==    at 0x4C28C20: malloc (vg_replace_malloc.c:296)
==8331== Invalid write of size 1
==8331==    at 0x68DE948: ecore_file_app_exe_get (ecore_file.c:1000)
==8331==  Address 0x1348e599 is 10 bytes after a block of size 31 alloc'd
==8331==    at 0x4C28C20: malloc (vg_replace_malloc.c:296)

etc. etc. - so i rewrote it cleanly using strbuf to save code and
effort. cleaner now and ACTUALLY works correctly... and no valgrind
complaints.

@fix

9 years agoedje object swallowing no longer forces a full recalc
Mike Blumenkrantz [Tue, 31 Mar 2015 23:15:17 +0000 (19:15 -0400)]
edje object swallowing no longer forces a full recalc

this (and/or related prior commits) seems to have hugely improved
edje performance across the board. particularly noticeable in the
settings pane animations for terminology

 #aprilfools

9 years agoedje object scale factor changes now trigger deferred recalcs
Mike Blumenkrantz [Tue, 31 Mar 2015 23:12:26 +0000 (19:12 -0400)]
edje object scale factor changes now trigger deferred recalcs

as with the previous commit, any user who wants geometry immediately
will be calling calc functions, so this is just going to be redundant
except in cases where calc is not wanted, in which case it is unnecessary
overhead

9 years agoedje group part init no longer forces synchronous hint updating during load
Mike Blumenkrantz [Tue, 31 Mar 2015 23:11:24 +0000 (19:11 -0400)]
edje group part init no longer forces synchronous hint updating during load

hints will be updated later and any user who wants these hints is going to
be calling edje calc functions anyway, causing this to be a redundant calc

9 years agoedje calc no longer forces proxy part source calcs for currently-hidden proxies
Mike Blumenkrantz [Tue, 31 Mar 2015 23:10:02 +0000 (19:10 -0400)]
edje calc no longer forces proxy part source calcs for currently-hidden proxies

this saves a bit of calc time when a proxy part is not currently visible

9 years agoedje calc no longer performs calculations for group parts of fixed size
Mike Blumenkrantz [Tue, 31 Mar 2015 23:07:46 +0000 (19:07 -0400)]
edje calc no longer performs calculations for group parts of fixed size

if the min/max of a part are identical and > 0, the part's min size is guaranteed
to be this size. there is no need to perform expensive recursive calcs here

9 years agoedje: add edje_file_iterator_new().
Cedric BAIL [Tue, 31 Mar 2015 21:23:32 +0000 (23:23 +0200)]
edje: add edje_file_iterator_new().

This list the actually opened Edje file by the current process.

9 years agoedje: fix build break.
Cedric BAIL [Tue, 31 Mar 2015 21:21:12 +0000 (23:21 +0200)]
edje: fix build break.

9 years agoedje: add edje_mmap_color_class_iterator_new().
Cedric BAIL [Tue, 31 Mar 2015 19:57:23 +0000 (21:57 +0200)]
edje: add edje_mmap_color_class_iterator_new().

This function make it possible to list the Color class of a specific Edje file.

9 years agoedje: add option to dump gnu style include dependencies in edje_cc.
Dinesh Dwivedi [Tue, 31 Mar 2015 15:40:30 +0000 (17:40 +0200)]
edje: add option to dump gnu style include dependencies in edje_cc.

Summary:
We were facing one problem in tizen sdk's build system as it does not trigger build for edc file
if only sub-edc files are changed. During analysis, we found that there is no option in edje_cc
for dumping include dependencies which other compiler (clang/ gcc etc) does have. We can do other
hack to solve this problem but it will be great if edje_cc can emit gnu style include dependency
target.

This patch will add support to generate gnu format include dependency file while compiling edc file.
similar to what gcc generates with option '-MMD -MF=<dep_file> -MT<dep_file>'
https://gcc.gnu.org/onlinedocs/gcc-4.9.2/gcc/Preprocessor-Options.html

Test Plan: no failure in existing test

Reviewers: raster, cedric

Reviewed By: cedric

Projects: #efl

Differential Revision: https://phab.enlightenment.org/D2263

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoevas/software_x11: Environment can influence the flow when the value is 1 not else.
Minkyoung Kim [Tue, 31 Mar 2015 05:13:06 +0000 (14:13 +0900)]
evas/software_x11: Environment can influence the flow when the value is 1 not else.

Summary: Do not try swapbuf init only if value of env 'EVAS_NO_DRI_SWAPBUF' is 1.

Reviewers: jpeg

Subscribers: spacegrapher, cedric, wonsik

Differential Revision: https://phab.enlightenment.org/D2259

9 years agoedje: get the color class value by iterating over all possible value provider.
Cedric BAIL [Tue, 31 Mar 2015 03:36:38 +0000 (05:36 +0200)]
edje: get the color class value by iterating over all possible value provider.

9 years agoedje: add edje_color_class_active_iterator_new()
Cedric BAIL [Tue, 31 Mar 2015 02:24:48 +0000 (04:24 +0200)]
edje: add edje_color_class_active_iterator_new()

This function make it possible to get a list of active Edje_Color_Class
in an application. Think about Enlightenment color class configuration,
but can now be done on any application dynamically.

@feature

9 years agoedje: make Edje_Color_Class a public structure.
Cedric BAIL [Tue, 31 Mar 2015 01:32:35 +0000 (03:32 +0200)]
edje: make Edje_Color_Class a public structure.

9 years agoEvas 3D: Fix shadow map shader
Oleksandr Shcherbina [Mon, 30 Mar 2015 23:34:32 +0000 (08:34 +0900)]
Evas 3D: Fix shadow map shader

Summary:
Silly mistake after this b9b5ced501f4642f03d8bcc8441fd3f7b55a7f52 commit.
Seems uninitialized components of vector gl_FragColor

Reviewers: cedric, Hermet, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2258

9 years agoedje: force NULL initialization of desc pointer.
Cedric BAIL [Mon, 30 Mar 2015 20:31:34 +0000 (22:31 +0200)]
edje: force NULL initialization of desc pointer.

9 years agoedje: this desc pointer is actually never set.
Cedric BAIL [Mon, 30 Mar 2015 20:31:10 +0000 (22:31 +0200)]
edje: this desc pointer is actually never set.

9 years agoecore_drm: use get_vt instead of parsing the tty attribute
Marcel Hollerbach [Mon, 30 Mar 2015 14:31:56 +0000 (10:31 -0400)]
ecore_drm: use get_vt instead of parsing the tty attribute

Summary: logind has a seperate function to get the vt number.

@fix

Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2252

9 years agoecore/idler - simpler checking.
ChunEon Park [Sat, 28 Mar 2015 06:43:50 +0000 (15:43 +0900)]
ecore/idler - simpler checking.

9 years agoeina: include blank lines in file iterator T2237
Andy Williams [Fri, 27 Mar 2015 21:36:45 +0000 (21:36 +0000)]
eina: include blank lines in file iterator T2237

any line that would not have been included before has length == 0

9 years agoecore_con: fix ftp upload function to follow documentation.
Srivardhan Hebbar [Fri, 27 Mar 2015 17:39:57 +0000 (18:39 +0100)]
ecore_con: fix ftp upload function to follow documentation.

Summary:
While creating a Ecore_con_url object, the url is given in this format "ftp://ftp.example.com". While uploading a file, this function was prefixing "ftp://" to this url which resulted in DNS failure, and upload fail. So corrected the issue.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2221

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoevas: fix gl_cocoa build after recent changes in evas_common_draw_context_font_ext_set.
Nicolas Aguirre [Thu, 26 Mar 2015 17:08:14 +0000 (18:08 +0100)]
evas: fix gl_cocoa build after recent changes in evas_common_draw_context_font_ext_set.

Now evas_common_draw_context_font_ext_set takes 3 more parameters, fix the build
by adding NULL for the function pointers. We really need to switch to gl_generic
here, it would avoid this kind of problems.

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeina: Change newline parsing to handle crlf better
Andy Williams [Fri, 27 Mar 2015 09:50:00 +0000 (09:50 +0000)]
eina: Change newline parsing to handle crlf better

Adjust test to demonstrate failure, existing code only worked for
windows newlines on blank line.
@fix T2236

9 years agoedje: edje_edit - recalculate object after chenge aspect preference
Vyacheslav Reutskiy [Fri, 27 Mar 2015 09:31:31 +0000 (11:31 +0200)]
edje: edje_edit - recalculate object after chenge aspect preference

9 years agoecore: trigger events on FD_READ, FD_CONNECT and FD_ACCEPT.
Nicolas Aguirre [Fri, 30 Jan 2015 13:25:04 +0000 (14:25 +0100)]
ecore: trigger events on FD_READ, FD_CONNECT and FD_ACCEPT.

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore_con: do not try to load /etc/resolv.conf when on windows.
Nicolas Aguirre [Fri, 30 Jan 2015 13:23:18 +0000 (14:23 +0100)]
ecore_con: do not try to load /etc/resolv.conf when on windows.

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years ago[Evas: Evas_3D] Improve shaders math
Dmytro Dadyka [Wed, 25 Mar 2015 23:07:49 +0000 (16:07 -0700)]
[Evas: Evas_3D] Improve shaders math

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2244

On many platforms the built-in mix(x, y, a) function faster
than its equivalent x * a + y * (1 - a)

9 years agoEcore: Use calloc() to initialize Ecore_Event_Key fully
Jean-Philippe ANDRE [Wed, 25 Mar 2015 22:29:49 +0000 (15:29 -0700)]
Ecore: Use calloc() to initialize Ecore_Event_Key fully

Some fields (eg. data) may not be properly initialized.
Use calloc() instead of malloc() will ensure proper initialization
even if we add more fields.

These fields (data) would even be passed down to Evas as
Evas_Event_Key_Down for instance.

9 years agoBuild: Fix horribly slow compilation times
Jean-Philippe ANDRE [Wed, 25 Mar 2015 18:19:53 +0000 (11:19 -0700)]
Build: Fix horribly slow compilation times

Cedric, our dear b0rker, introduced changes in the CFLAGS
generation when merging Emile. While the changes seem to make sense
at first sight (add the -I flags for the lib our new package depends on),
they were actually a terribly bad workaround.

The number of CFLAGS args would grow exponentially, slowing down libtool
a lot, which is known to be slow when it has a lot of arguments.

9 years agoedje: add of textures to Edje 3D node.
perepelits.m [Wed, 25 Mar 2015 18:31:25 +0000 (19:31 +0100)]
edje: add of textures to Edje 3D node.

Summary: Adding textures to edje_cc and some topblock keywords that I have forgotten to add in previous commits.

Reviewers: Hermet, raster, cedric

Reviewed By: cedric

Subscribers: cedric, artem.popov

Differential Revision: https://phab.enlightenment.org/D2129

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeina: add eina_list_data_idx().
Michal Jagiello [Wed, 25 Mar 2015 18:29:49 +0000 (19:29 +0100)]
eina: add eina_list_data_idx().

Summary:
Now the developer has to iterate the whole list to find the
index of the first occurence of the data.
I see that it is possible to get the index of the item for
the genlist widget, but for the eina_list not.

With these APIs it will be easier to implement *index_get
functions for the rest of widgets which contain items (Elm_List,
Elm_Ctxpopup etc.).

These functions returns the index of the given data or node in the eina_list.

Reviewers: Hermet, cedric

Subscribers: raster, cedric

Differential Revision: https://phab.enlightenment.org/D2189

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore_con: add example for ftp upload.
Srivardhan Hebbar [Wed, 25 Mar 2015 18:26:41 +0000 (19:26 +0100)]
ecore_con: add example for ftp upload.

Summary:
Added example for ftp upload. In the .gitignore only 2 files added which were missing. The differences it is showing is cos of reordering. I did ls and redirected the file to gitignore. So the files got reordered.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2223

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoevas: fix logic of 180 degree image rotation in software_generic backend.
kabeer khan [Wed, 25 Mar 2015 18:25:38 +0000 (19:25 +0100)]
evas: fix logic of 180 degree image rotation in software_generic backend.

Summary: Signed-off-by: kabeer khan <kabeer.khan@samsung.com>

Reviewers: cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2238

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeet: fix variable naming
Andrii Kroitor [Wed, 25 Mar 2015 17:55:40 +0000 (02:55 +0900)]
eet: fix variable naming

Summary: this file name is used for input, not for output.

Reviewers: cedric, seoz, Hermet, raster

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D2243

9 years agoWayland shm: Fix shadow variable warning
Jean-Philippe ANDRE [Wed, 25 Mar 2015 17:49:02 +0000 (10:49 -0700)]
Wayland shm: Fix shadow variable warning

9 years agoedje: add text_class_get() APIs
Sohyun Kim [Wed, 25 Mar 2015 17:45:59 +0000 (02:45 +0900)]
edje: add text_class_get() APIs

Summary: No APIs to get text_class for global hash and object hash

Reviewers: woohyun, cedric, raster

Reviewed By: cedric, raster

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2213

9 years ago[Evas: Evas_3D] Refactor shader system: refactored shadow map shade mode.
Dmytro Dadyka [Wed, 25 Mar 2015 17:37:16 +0000 (10:37 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored shadow map shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2240

9 years ago[Evas: Evas_3D] Refactor shader system: refactored parallax shade mode.
Dmytro Dadyka [Wed, 25 Mar 2015 17:35:42 +0000 (10:35 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored parallax shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2241

9 years agoEdje: Edje_Edit - code generation for TABLE part (description.table section)
Vitalii Vorobiov [Wed, 25 Mar 2015 15:56:47 +0000 (17:56 +0200)]
Edje: Edje_Edit - code generation for TABLE part (description.table section)

9 years agoEdje: Edje_Edit - fix typo in code generation of BOX part
Vitalii Vorobiov [Wed, 25 Mar 2015 15:17:49 +0000 (17:17 +0200)]
Edje: Edje_Edit - fix typo in code generation of BOX part

@fix

9 years agoecore_idler: + null check.
ChunEon Park [Wed, 25 Mar 2015 05:42:44 +0000 (14:42 +0900)]
ecore_idler: + null check.

like timer, animator we do check the idler validation
and prevert crash there just in user misusage case.

@fix

9 years agoevas: fix use after free.
Cedric BAIL [Tue, 24 Mar 2015 22:09:28 +0000 (23:09 +0100)]
evas: fix use after free.

Thanks to Thanatermesis for tracking this in T2135.

9 years agoecore-drm: Use better output names
Chris Michael [Tue, 24 Mar 2015 19:34:04 +0000 (15:34 -0400)]
ecore-drm: Use better output names

Summary: As there are various DVI and HDMI output names, update the
output connector types area to be more accurate so we get better
output names

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoEvas 3D: Print SHADERS instead of SHADERS_3D at compile time
Jean-Philippe ANDRE [Tue, 24 Mar 2015 18:47:29 +0000 (11:47 -0700)]
Evas 3D: Print SHADERS instead of SHADERS_3D at compile time

This is just a cosmetic change.

9 years agoEvas 3D: Add forgotten changes after generation of shaders
Jean-Philippe ANDRE [Tue, 24 Mar 2015 18:45:32 +0000 (11:45 -0700)]
Evas 3D: Add forgotten changes after generation of shaders

9 years agoEvas GL: Add missed GLES 3.0 functions.
mythri.venugopal [Tue, 24 Mar 2015 18:22:01 +0000 (11:22 -0700)]
Evas GL: Add missed GLES 3.0 functions.

Summary:
Initialise glVertexAttribDivisor & glVertexAttribIPointer
EvasGL function pointers.

Reviewers: wonsik, spacegrapher, jpeg

Reviewed By: jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2236

9 years ago[Evas: Evas_3D] Refactor shader system: refactored normal map shade mode.
Dmytro Dadyka [Tue, 24 Mar 2015 17:50:40 +0000 (10:50 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored normal map shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2232

9 years ago[Evas: Evas_3D] Refactor shader system: refactored color pick shade mode.
Dmytro Dadyka [Tue, 24 Mar 2015 17:50:36 +0000 (10:50 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored  color pick shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2230

9 years ago[Evas: Evas_3D] Refactor shader system: refactored diffuse shade mode.
Dmytro Dadyka [Tue, 24 Mar 2015 17:50:31 +0000 (10:50 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored diffuse shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2229

9 years ago[Evas: Evas_3D] Refactor shader system: refactored vertex color shade mode.
Dmytro Dadyka [Tue, 24 Mar 2015 17:50:25 +0000 (10:50 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored vertex color shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2228

9 years ago[Evas: Evas_3D] Refactor shader system: refactored flat shade mode.
Dmytro Dadyka [Tue, 24 Mar 2015 17:49:25 +0000 (10:49 -0700)]
[Evas: Evas_3D] Refactor shader system: refactored flat shade mode.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2224

9 years ago[Evas: Evas_3D] Fix shader code for using with GL_ES
Dmytro Dadyka [Tue, 24 Mar 2015 17:06:25 +0000 (10:06 -0700)]
[Evas: Evas_3D] Fix shader code for using with GL_ES

Summary: GL_ES variables precision added.

Reviewers: cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2234

9 years agoecore_drm: Add event of ECORE_DRM_EVENT_SEAT_ADD.
Seunghun Lee [Tue, 24 Mar 2015 14:45:11 +0000 (10:45 -0400)]
ecore_drm: Add event of ECORE_DRM_EVENT_SEAT_ADD.

Summary: To support multi-seat, ecore_drm need to issue the event to notify compositor of seat is added.

Test Plan: N/A

Reviewers: zmike, raster, gwanglim, devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2212

9 years agoecore-drm: Fixed null pointer free
Mun, Gwan-gyeong [Tue, 24 Mar 2015 13:11:07 +0000 (09:11 -0400)]
ecore-drm: Fixed null pointer free

Summary:
Fixed null pointer free
         when sd_session_get_seat() fails, it does not allocate string.

Reviewers: zmike, cedric, raster, gwanglim, devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2219

9 years agoecore_imf: Add NULL check before accessing evas device API
Jihoon Kim [Tue, 24 Mar 2015 11:33:45 +0000 (20:33 +0900)]
ecore_imf: Add NULL check before accessing evas device API

9 years agoevas: Add safety check for NULL pointer of clipper.
Youngbok Shin [Tue, 24 Mar 2015 03:53:34 +0000 (12:53 +0900)]
evas: Add safety check for NULL pointer of clipper.

Summary: It caused crash when clipper is NULL and it makes evas test-suite fail.

Test Plan: Run evas test-suite. (make check)

Reviewers: woohyun, tasn, herdsman, Hermet

Reviewed By: Hermet

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2215

9 years agoecore-imf: Added the device information in the Ecore_IMF_Event structs
Wonkeun Oh [Tue, 24 Mar 2015 01:26:20 +0000 (10:26 +0900)]
ecore-imf: Added the device information in the Ecore_IMF_Event structs

Summary:
Added the device information in the Ecore_IMF_Event structs.
Applications may require to know actual source of key event to handle it differently depending on device.
Even if key events are currently generated from different devices, application couldn't distinguish it.
This patch will support device information for the Ecore-imf key events.

Reviewers: jihoon

Reviewed By: jihoon

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2214

9 years agoevas/gl_x11: add env variables for disable partial rendering
Sunghyun Kim [Tue, 24 Mar 2015 00:30:59 +0000 (17:30 -0700)]
evas/gl_x11: add env variables for disable partial rendering

Summary: add 'EVAS_GL_PARTIAL_DISABLE' environment variables for disable partial rendering

Reviewers: cedric, jpeg

Reviewed By: jpeg

Subscribers: mer.kim, spacegrapher, cedric, wonsik

Differential Revision: https://phab.enlightenment.org/D2183

Signed-off-by: Jean-Philippe ANDRE <jpeg@videolan.org>
jpeg added atoi(s) and copied the code for gl_drm and wayland_egl.

9 years agoevas/gl_x11: Separate gl symbols check
Dongyeon Kim [Tue, 24 Mar 2015 00:17:37 +0000 (17:17 -0700)]
evas/gl_x11: Separate gl symbols check

Summary:
There is a restriction for some gpu drivers that
eglGetProcAddress must be called after eglMakeCurrent.
So separate egl/glx extensions check from gl_symbols
to be called inside eng_window_new.

Test Plan: egl and glx backend tests

Reviewers: cedric, jpeg

Subscribers: cedric, mer.kim, wonsik

Differential Revision: https://phab.enlightenment.org/D2193

9 years agoEvas: Fix invalid test case
Jean-Philippe ANDRE [Mon, 23 Mar 2015 22:55:23 +0000 (15:55 -0700)]
Evas: Fix invalid test case

No idea how it worked before. It shouldn't. It was invalid.
But somehow evas_suite decided to start crashing on this test.

9 years ago[Evas: Evas_3D] Refactor shaders: used m4 for removing duplication of code.
Dmytro Dadyka [Mon, 23 Mar 2015 21:59:19 +0000 (14:59 -0700)]
[Evas: Evas_3D] Refactor shaders: used m4 for removing duplication of code.

Reviewers: Hermet, raster, cedric, jpeg

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2076

9 years agoRevert "evas/gl_x11: Optimize out calls to make current"
Jean-Philippe ANDRE [Mon, 23 Mar 2015 20:58:26 +0000 (05:58 +0900)]
Revert "evas/gl_x11: Optimize out calls to make current"

This reverts commit 0585540bb3558b2ff834a1bb730e897c6301a224.

This broke Evas 3d examples. I also suspected some weird things and
wasn't 100% confident with this patch.

Closes T2215.
Thanks for the report.

9 years agoemile: meaningful initialization of some pointers.
Cedric BAIL [Mon, 23 Mar 2015 21:04:15 +0000 (22:04 +0100)]
emile: meaningful initialization of some pointers.

9 years agoecore_con: clean up, refactor and unpdate documentation.
Srivardhan Hebbar [Mon, 23 Mar 2015 18:44:03 +0000 (19:44 +0100)]
ecore_con: clean up, refactor and unpdate documentation.

Summary:
1. Have refactored code in ecore_con_local.c
2. Changed env variable from "TMP" to "TMPDIR".
3. Added check for negetive port number when type is ECORE_CON_LOCAL_USER.
4. Added check to check TMPDIR before assuming /tmp as temp directory.
5. Updated documentation in Ecore_Con.h explaining about local socket and port number.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2194

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore_exe: fix ecore_exe_pid_get on windows
Andrii Kroitor [Mon, 23 Mar 2015 18:37:21 +0000 (19:37 +0100)]
ecore_exe: fix ecore_exe_pid_get on windows

Summary:
ecore_exe_pid_get returns exe->pid value so it should be filled

@fix

Reviewers: raster, seoz, cedric, Hermet

Subscribers: NikaWhite, cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D2195

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoeina: add test cases for various eina xattr data type functions
vivek [Mon, 23 Mar 2015 18:29:55 +0000 (19:29 +0100)]
eina: add test cases for various eina xattr data type functions

Summary:
Added test cases for eina_xattr_string_set, eina_xattr_int_set and
eina_xattr_double_set functions

Signed-off-by: vivek <vivek.ellur@samsung.com>
Reviewers: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2218

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore-drm: Fix issue of not getting updated for output hotplug events
Chris Michael [Mon, 23 Mar 2015 16:17:11 +0000 (12:17 -0400)]
ecore-drm: Fix issue of not getting updated for output hotplug events

Summary: When setting up the Eeze_Udev_Watch for output hotplug
events, we should also be listening for the "change" event. As eeze
will now check the HOTPLUG property for drm events, we don't need to
check that here.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Fix issue of eeze_udev_watch_add not getting events for drm
Chris Michael [Mon, 23 Mar 2015 16:14:45 +0000 (12:14 -0400)]
eeze: Fix issue of eeze_udev_watch_add not getting events for drm

Summary: This fixes eeze_udev_watch_add for DRM events, Previously, we
were not getting a watch callback fired due to using the wrong
subsystem device type in
udev_monitor_filter_add_match_subsystem_devtype.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoevas: prevent crash.
ChunEon Park [Mon, 23 Mar 2015 13:00:27 +0000 (22:00 +0900)]
evas: prevent crash.

if user delete the interceptor in the interceptor callback,
the interceptors would be NULL.
So we need to check whether it's valid or not.

@fix

9 years agoemile: fix build on NEON enable system.
Cedric BAIL [Sun, 22 Mar 2015 16:41:36 +0000 (17:41 +0100)]
emile: fix build on NEON enable system.

T2214

9 years agoevil: fix access to a file mapping object
Vincent Torri [Fri, 20 Mar 2015 07:16:20 +0000 (08:16 +0100)]
evil: fix access to a file mapping object

Patch from Jorge Luis Zapatta (turran), a bit modified
by me

@fix

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore-drm: Fix build error caused by missing function declaration
Chris Michael [Sat, 21 Mar 2015 20:38:47 +0000 (16:38 -0400)]
ecore-drm: Fix build error caused by missing function declaration

Summary: This is mainly just a build fix. This issue occurred due to
the function declaration being After a call to use it while not having
a function prototype. This just adds the function prototype

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add more debug info when output is created or updated
Chris Michael [Sat, 21 Mar 2015 20:27:01 +0000 (16:27 -0400)]
ecore-drm: Add more debug info when output is created or updated

Summary: This patch adds more information so we can debug output
hotplug events better. It fixes an issue where any new outputs
from _ecore_drm_update_outputs where not getting the eeze udev watch
event created, and also addresses the issue of disconnected outputs
not sending the ecore_drm_event (both minor fixes)

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoevas-drm: Enable more information during flip error
Chris Michael [Sat, 21 Mar 2015 20:08:40 +0000 (16:08 -0400)]
evas-drm: Enable more information during flip error

Summary: This adds a bit more output during crtc flip errors so we can
debug multi-monitor setups

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeo: fix wrong check libaddr dependency.
ChunEon Park [Sat, 21 Mar 2015 04:35:14 +0000 (13:35 +0900)]
eo: fix wrong check libaddr dependency.

move checking libaddr to eo step to valid it.

9 years agoWayland: Fix compilation
Jean-Philippe ANDRE [Sat, 21 Mar 2015 03:13:28 +0000 (12:13 +0900)]
Wayland: Fix compilation

Bad merge?

9 years agoecore_wayland: Check for nulls before wayland calls
Bryce Harrington [Sat, 21 Mar 2015 02:07:27 +0000 (22:07 -0400)]
ecore_wayland:  Check for nulls before wayland calls

Summary:
Wayland API routines by policy do not check input parameters for invalid
values.  In particular, many calls are wrappers through
wl_proxy_marshal, which derefs its first argument without a check.

So, for all wayland calls, always null check the first argument.

Reviewers: zmike, cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2211

9 years agoecore_wayland: Check for null inputs
Bryce Harrington [Sat, 21 Mar 2015 02:06:58 +0000 (22:06 -0400)]
ecore_wayland: Check for null inputs

Summary:
The other functions in this module check inputs; this one should as
well.

Reviewers: zmike

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2210

9 years agoecore_wayland: Add some comments
Bryce Harrington [Sat, 21 Mar 2015 01:18:52 +0000 (21:18 -0400)]
ecore_wayland: Add some comments

Summary: Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>

Reviewers: zmike

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2209

9 years agoRevert "evas-3d: materials in .mtl format are deleted"
ChunEon Park [Fri, 20 Mar 2015 09:34:20 +0000 (18:34 +0900)]
Revert "evas-3d: materials in .mtl format are deleted"

This reverts commit 59271716ba5443bcba215f5dfbc190fc107ae4ab.

This patch submitted twice.

9 years agoevas render2 - restructure it to be an explicit api call - cleaner to do
Carsten Haitzler (Rasterman) [Fri, 20 Mar 2015 05:18:17 +0000 (14:18 +0900)]
evas render2 - restructure it to be an explicit api call - cleaner to do

9 years agoevas-3d: materials in .mtl format are deleted
Bogdan Devichev [Fri, 20 Mar 2015 05:33:23 +0000 (14:33 +0900)]
evas-3d: materials in .mtl format are deleted

Summary: Materials are not used in any loader and will not be used in nearest future.

Reviewers: raster, Hermet, cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2080

Conflicts:

src/examples/evas/Makefile.am

9 years agoevas: fix description about Evas_Device_Subclass
Jihoon Kim [Fri, 20 Mar 2015 05:42:46 +0000 (14:42 +0900)]
evas: fix description about Evas_Device_Subclass

9 years agoFix wrong drawing of shadows for some blending modes
Dmytro Dadyka [Fri, 20 Mar 2015 05:10:00 +0000 (14:10 +0900)]
Fix wrong drawing of shadows for some blending modes

Reviewers: cedric, Hermet

Reviewed By: Hermet

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2198

9 years agoemile: Remove unnecessary check in _emile_jpeg_head function
jiin.moon [Fri, 20 Mar 2015 05:02:58 +0000 (14:02 +0900)]
emile: Remove unnecessary check in _emile_jpeg_head function

Summary: Already checked it by "else if ((opts->w > 0) && (opts->h > 0))"

Reviewers: cedric, Hermet

Reviewed By: Hermet

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2199

9 years agoevas: fix build failure with old version freetype.
Youngbok Shin [Fri, 20 Mar 2015 04:24:31 +0000 (13:24 +0900)]
evas: fix build failure with old version freetype.

Summary:
Fix build failure with old version freetype.
It is caused for supporting colored font.

Reviewers: raster, jpeg, woohyun, Hermet

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D2200

9 years agoevas - font - new color bitmap support - rename funcs to match file
Carsten Haitzler (Rasterman) [Fri, 20 Mar 2015 01:49:49 +0000 (10:49 +0900)]
evas - font - new color bitmap support - rename funcs to match file

9 years agoevas-wayland-shm: Fix redrawing issues that were causing 'flashing' in latest engine...
Chris Michael [Thu, 19 Mar 2015 20:15:11 +0000 (16:15 -0400)]
evas-wayland-shm: Fix redrawing issues that were causing 'flashing' in latest engine code.

Summary: As reported by derek & zmike, the previous engine refactor
left some issues where the app would 'flash' during redraws due to
buffer handling. This commit fixes that issue along with flashing
during resize.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoevas-software-generic: Allow future rendering calls if evas has to skip a frame
Chris Michael [Thu, 19 Mar 2015 20:09:23 +0000 (16:09 -0400)]
evas-software-generic: Allow future rendering calls if evas has to skip a frame

Summary: If we do not get a new_region_for_update from an engine, it
could mean that the engine has skipped a frame. In that case, we still
want to try and render any future frames. Before this commit, evas
would never try to re-render anything again once it hit a failed
frame. To fix this, we test for a valid returned region, and if we
don't get one, mark this frame as failed so we can try again.

NB: Thanks for the help Cedric !! :)

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>