Mike Blumenkrantz [Mon, 24 Mar 2014 16:51:14 +0000 (12:51 -0400)]
edje_cc uses more setup macros to unify program statements
Mike Blumenkrantz [Mon, 24 Mar 2014 14:58:48 +0000 (10:58 -0400)]
edje_cc prevents sequence users from failing to sequence
Mike Blumenkrantz [Mon, 24 Mar 2014 14:58:09 +0000 (10:58 -0400)]
edje_cc should support sequences inside programs.program I guess
Anand [Mon, 24 Mar 2014 08:03:32 +0000 (17:03 +0900)]
evas: Fixed warning for uninitialized variables
Summary:
Warning fixed of evas
modules/evas/engines/gl_common/evas_gl_context.c: In function 'evas_gl_common_context_new':
modules/evas/engines/gl_common/evas_gl_context.c:392:32: warning: 'minor' may be used uninitialized in this function [-Wuninitialized]
modules/evas/engines/gl_common/evas_gl_context.c:314:8: note: 'minor' was declared here
modules/evas/engines/gl_common/evas_gl_context.c:392:16: warning: 'major' may be used uninitialized in this function [-Wuninitialized]
modules/evas/engines/gl_common/evas_gl_context.c:313:8: note: 'major' was declared here
@fix
Compilation Warning Fixed
Test Plan: Compile efl
Reviewers: singh.amitesh
CC: seoz, cedric
Differential Revision: https://phab.enlightenment.org/D656
Vorobiov Vitalii [Mon, 24 Mar 2014 07:05:25 +0000 (16:05 +0900)]
edje_edit: function that will return the source code of the loaded edje edit object.
Summary:
This function will return the source code of the edje edit object that is given to that function.
Also, that function will collect all resources required by this group.
That means all images, fonts, styles, data and color_classes that is mentioned
and set in descriptions of parts in this group (that is represented by given
Edje Edit object).
@feature
Added function for generating source code:
- edje_edit_source_generate(Evas_Object *obj);
Reviewers: cedric, seoz, raster
Reviewed By: raster
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D649
Yossi Kantor [Sun, 23 Mar 2014 14:20:37 +0000 (16:20 +0200)]
Eolian: Integration of Ecore Idler
Yossi Kantor [Sun, 23 Mar 2014 14:02:19 +0000 (16:02 +0200)]
Eolian: Integration of Ecore Job
Yossi Kantor [Sun, 23 Mar 2014 13:32:06 +0000 (15:32 +0200)]
Eolian: Integration of Ecore Poller
Daniel Zaoui [Mon, 24 Mar 2014 05:47:52 +0000 (07:47 +0200)]
Eolian: Integration of Ecore Timer
Daniel Zaoui [Sun, 23 Mar 2014 05:28:04 +0000 (07:28 +0200)]
Eolian/Generator: more stuff.
Better support of empty events/funcs lists
Better support of class constructor
Preparation for Eo2: separate Eo1 specific code and common code
Daniel Zaoui [Thu, 20 Mar 2014 15:33:25 +0000 (17:33 +0200)]
Eolian/Generator: detach Eo1 from Legacy generator.
This step is needed to clean the code and to prepare the integration of
Eo2 inside Eolian.
Except the eo_do invocation, there is no reason why legacy has to know
about Eo.
Yossi Kantor [Tue, 18 Mar 2014 14:00:14 +0000 (16:00 +0200)]
Eolian: Integration of Edje and Edje Edit
Cedric BAIL [Mon, 24 Mar 2014 02:41:20 +0000 (11:41 +0900)]
evas: cleaner fail handler.
Daniel Zaoui [Thu, 20 Mar 2014 15:52:19 +0000 (17:52 +0200)]
Eolian/Generator: generate Eo * instead of Evas_Object *
Jihoon Kim [Sun, 23 Mar 2014 04:36:00 +0000 (13:36 +0900)]
evas_gl: fix memory leak in case that version doesn't include dot
fix CID 1193479
ChunEon Park [Sat, 22 Mar 2014 12:09:58 +0000 (21:09 +0900)]
evas/gl - fill up missed blend mode.
Still there few mode are unsatisfied.
@fix
ChunEon Park [Sat, 22 Mar 2014 11:28:45 +0000 (20:28 +0900)]
evas/gl - don't set blend function in RENDER COPY mode.
it's enough that just disabling the blend mode.
Carsten Haitzler (Rasterman) [Sat, 22 Mar 2014 10:19:48 +0000 (19:19 +0900)]
warnings--
Carsten Haitzler (Rasterman) [Sat, 22 Mar 2014 10:09:01 +0000 (19:09 +0900)]
drm - shut up warnings (not actual bugs)
ChunEon Park [Sat, 22 Mar 2014 08:24:20 +0000 (17:24 +0900)]
evas/common - initialize only the available blend stuff that cpu is supporting.
Mike Blumenkrantz [Sat, 22 Mar 2014 06:18:54 +0000 (02:18 -0400)]
edje_cc can now use group.program_remove to remove inherited programs
@feature
Mike Blumenkrantz [Sat, 22 Mar 2014 04:51:05 +0000 (00:51 -0400)]
edje_cc no longer fails when program.name is specified after program.after
Mike Blumenkrantz [Sat, 22 Mar 2014 04:12:58 +0000 (00:12 -0400)]
edje_cc should probably disallow overriding of currently-sequencing program
Mike Blumenkrantz [Sat, 22 Mar 2014 03:33:28 +0000 (23:33 -0400)]
edje_cc now supports program.sequence for more easily chaining programs together
this is just some syntax shortening for program.after which causes program.action and program.script to create a new program and automatically chain it within the sequence{} block
recursive sequences not currently allowed/planned (don't be insane)
@feature
Mike Blumenkrantz [Sat, 22 Mar 2014 03:01:38 +0000 (23:01 -0400)]
edje_cc macros even more of its parser setup for consistency
Mike Blumenkrantz [Sat, 22 Mar 2014 02:36:22 +0000 (22:36 -0400)]
edje_cc typo which almost broke everything except nobody noticed
Mike Blumenkrantz [Sat, 22 Mar 2014 02:06:56 +0000 (22:06 -0400)]
edje_cc uses macros for some of its handler setup
this reduces code size, ensures consistency, and actually fixes bugs where the "filter" keyword was erroneously disallowed in some places
@fix
Mike Blumenkrantz [Sat, 22 Mar 2014 00:45:00 +0000 (20:45 -0400)]
edje_cc no longer reads every character of a script during parsing
Mike Blumenkrantz [Fri, 21 Mar 2014 20:49:23 +0000 (16:49 -0400)]
edje_cc group.remove should probably be group.part_remove
this allows things like group.program_remove to exist
Mike Blumenkrantz [Fri, 21 Mar 2014 20:37:21 +0000 (16:37 -0400)]
eeze_udev gets more helper functions
eeze_udev_syspath_get_parent_filtered, eeze_udev_syspath_check_property, eeze_udev_syspath_check_sysattr
@feature
Mike Blumenkrantz [Fri, 21 Mar 2014 19:45:17 +0000 (15:45 -0400)]
edje_cc now supports part.inherit for copying attributes of parts within a group more easily
@feature
Mike Blumenkrantz [Fri, 21 Mar 2014 19:35:08 +0000 (15:35 -0400)]
edje_cc error message for description with missing state is now useful
Mike Blumenkrantz [Fri, 21 Mar 2014 19:23:24 +0000 (15:23 -0400)]
edje_cc no longer errors when inheriting a description before/when state is not set
anonymous states are allowed in other places, so having this check is inconsistent and breaks inheriting. an error will be triggered at a later point.
Mike Blumenkrantz [Fri, 21 Mar 2014 18:28:29 +0000 (14:28 -0400)]
edje_cc now supports group.remove for removing parts from inherited groups
@feature
Mike Blumenkrantz [Fri, 21 Mar 2014 18:20:06 +0000 (14:20 -0400)]
eio monitors no longer trigger a CRI error during free
@fix
Jean-Philippe Andre [Fri, 21 Mar 2014 06:48:54 +0000 (15:48 +0900)]
configure.ac: Print eolian_gen if specified with --with-bin-eolian-gen
Carsten Haitzler (Rasterman) [Fri, 21 Mar 2014 06:49:36 +0000 (15:49 +0900)]
oops - remove debug echo
Jean-Philippe Andre [Mon, 17 Mar 2014 07:54:48 +0000 (16:54 +0900)]
Evas filters: fix potential memory leak
Unlikely use case trying to delete a source when there are no
sources yet.
@fix
Jean-Philippe Andre [Mon, 17 Mar 2014 07:53:29 +0000 (16:53 +0900)]
Evas filters: Fix Eo documentation of program.set
The generated header "looked" a bit broken and some bits of doc
were dropped during the transition. Not a very big deal. :)
Jean-Philippe Andre [Fri, 21 Mar 2014 03:04:35 +0000 (12:04 +0900)]
Evas filters: Allow BLUR and GROW with radius 0
Fallback to a standard blend instruction instead of failing
miserably just because the radius is 0.
Jean-Philippe Andre [Fri, 21 Mar 2014 02:52:52 +0000 (11:52 +0900)]
Evas filters: Add new instruction padding_set
This will allow forcing a specific value for the filter padding,
instead of relying on auto calculation.
Two advantages:
- Auto calculation can't be perfect, since it will add as much
padding as required for the full blur effect
- This prepares the path for animations with effects, where the
object size does not change over time
Jean-Philippe Andre [Thu, 20 Mar 2014 09:26:04 +0000 (18:26 +0900)]
Evas filters: Fix blur corner cases with small images
If the buffer size is smaller than the blurring kernel, then
special precautions must be taken to properly read the source
pixels. Also, fix the corner cases near the left & right edges
(or top & bottom).
Jean-Philippe Andre [Fri, 21 Mar 2014 03:20:11 +0000 (12:20 +0900)]
Evas filters: fix blur from rgba to output buffer
When blurring an RGBA buffer to the output buffer,
we don't need to convert the colorspace... but then we'll just
override what was already there.
Introduce a 'dirty' flag set to true whenever a command writes
to an output buffer.
Mike Blumenkrantz [Thu, 20 Mar 2014 23:36:00 +0000 (19:36 -0400)]
edje_cc now supports program.targets keyword for adding N targets in one line
@feature
Mike Blumenkrantz [Thu, 20 Mar 2014 23:26:49 +0000 (19:26 -0400)]
edje_cc also checks min args correctly for STATE_SET actions
@fix
Mike Blumenkrantz [Thu, 20 Mar 2014 23:19:50 +0000 (19:19 -0400)]
edje_cc now correctly handles lack of state int in STATE_SET action
@fix
Mike Blumenkrantz [Thu, 20 Mar 2014 18:58:57 +0000 (14:58 -0400)]
edje_cc no longer fails on collections.group{}
@fix
Daniel Zaoui [Thu, 20 Mar 2014 08:20:10 +0000 (10:20 +0200)]
Eolian/Lexer: fix parsing of parameters direction.
strncmp bytes number was not correct. A problem in the generation was
occurring when the parameter is @inout, as it is considered as @in and
the type was "out ...".
Carsten Haitzler (Rasterman) [Thu, 20 Mar 2014 08:20:53 +0000 (17:20 +0900)]
fix dbus service file complaint to work again.
Cedric BAIL [Thu, 20 Mar 2014 07:53:13 +0000 (16:53 +0900)]
evas: let's be more resistant even with things that should never happen.
Fix CID 1193212.
Cedric BAIL [Thu, 20 Mar 2014 07:51:48 +0000 (16:51 +0900)]
edje: let's be more careful with string.
Fix CID 1193213.
Cedric BAIL [Thu, 20 Mar 2014 07:50:48 +0000 (16:50 +0900)]
evas: let's be to safe than sorry.
Fix CID 1193214.
Cedric BAIL [Thu, 20 Mar 2014 07:48:38 +0000 (16:48 +0900)]
edje: let's be careful on the string content.
Fix CID 1193215.
Cedric BAIL [Thu, 20 Mar 2014 07:47:25 +0000 (16:47 +0900)]
eio: let's not divide by 0.
fix CID 1193208.
Jeff Grimshaw [Thu, 20 Mar 2014 07:35:41 +0000 (16:35 +0900)]
efl/eina: General cleanup of Eina documentation
Summary:
I've combed through the Eina source files and made enhancements to the
documentation, including:
- Document the undocumented
- Fixed some errors in Doxygen markup
- Moved some function documentation from implementation (.c or .x) to definition
(.h)
- Edited some of the entries to improve clarity
Test Plan: Reviewers
Reviewers: cedric
Reviewed By: cedric
CC: cedric
Differential Revision: https://phab.enlightenment.org/D639
Chris Michael [Thu, 20 Mar 2014 07:13:52 +0000 (07:13 +0000)]
ecore-drm: Fix absolute motion coordinate calculation
@bugfix: When calculating center point for absolute motion, use the
proper min_y value for calculating Y axis
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Daniel Zaoui [Wed, 19 Mar 2014 13:32:56 +0000 (15:32 +0200)]
Eolian: fix prototype for Eo_Base function.
The return type was defined as a out param. It creates issues in the
generation made by Eolian.
Tae-Hwan Kim [Thu, 20 Mar 2014 04:00:20 +0000 (13:00 +0900)]
Add new PLUGIN_RUN action type and new plugins.plugin handlers 1. If external library is commerical source and not opensource, we cannot include/build the library within edje. 2. If external library does not use general encodable sources, we...
Summary:
...cannot encode those things into edje.
In our case, we need vibration when longpressed. But those files are not
audio or image and cannot be encoded into edje. Also, this library is not
opensource so should not be linked directly with edje.
So we should call vibration API by using this plug-in.
Reviewers: raster, cedric, seoz, Hermet
CC: cedric
Differential Revision: https://phab.enlightenment.org/D588
Jean-Philippe Andre [Thu, 20 Mar 2014 02:13:32 +0000 (11:13 +0900)]
Evas image: Improve diagrams in comments
Purely non functional change, for readability :)
Jihoon Kim [Wed, 19 Mar 2014 23:44:56 +0000 (08:44 +0900)]
Fix dereference null return value
CID 1193210
CID 1193211
Stefan Schmidt [Wed, 19 Mar 2014 15:46:36 +0000 (16:46 +0100)]
eo: Make sure we have a va_end matching the va_start before returning.
Was fien on the normal path but missing on the error path. Also remove
the spurious break after the return. Would never be reached. Looks like
a copy and paste bug to me.
CID 1187638
Stefan Schmidt [Wed, 19 Mar 2014 15:40:52 +0000 (16:40 +0100)]
eolian/eo1_generator: Fix leak in error path
We allocated tmpstr so we should free it before we return in the error case.
CID 1191926
Stefan Schmidt [Wed, 19 Mar 2014 15:16:55 +0000 (16:16 +0100)]
evas/engine/drm: Avoid shadow of a function parameter
Having the same name for a parameter as well as a local var name could
cause quite some problems. Better avoid it.
Stefan Schmidt [Wed, 19 Mar 2014 10:48:46 +0000 (11:48 +0100)]
evas/gl_x11: Cast GLubyte to const char for strstr
The GLubyte we are getting back here are strings we want to search
in with strstr. Cast is only to avoid warnings.
Tom Hacohen [Wed, 19 Mar 2014 09:40:44 +0000 (09:40 +0000)]
Eolian helper: Depend on eolian_gen when building externally as well.
This means that eolian files will be regenerated when the generator is
updated.
Daniel Zaoui [Wed, 19 Mar 2014 05:52:04 +0000 (07:52 +0200)]
Eolian/Generator: Don't generate sub ids structure if not needed.
If no methods and properties are declared for a class, no base id and
sub_ids structure are generated.
@fix
Yossi Kantor [Tue, 18 Mar 2014 14:17:34 +0000 (16:17 +0200)]
Eolian: Added class constructor delegate to Evas Box
Daniel Zaoui [Tue, 18 Mar 2014 14:44:38 +0000 (16:44 +0200)]
Eolian/Lexer: increase buffer length for .eo files
@fix
Daniel Zaoui [Tue, 18 Mar 2014 14:01:24 +0000 (16:01 +0200)]
Eolian: Fix ctor/dtor generation
Daniel Zaoui [Tue, 18 Mar 2014 14:00:52 +0000 (16:00 +0200)]
Eolian/Generator: fix generation of events.
zmike [Tue, 18 Mar 2014 21:53:32 +0000 (17:53 -0400)]
ecore drm now semi-supports absolute motion (touchpads!)
this is a little bit sensitive. you have been warned
Mike Blumenkrantz [Tue, 18 Mar 2014 20:11:32 +0000 (16:11 -0400)]
ecore drm evdev probably shouldn't treat my touchpad as a keyboard (BTN_TOOL_FINGER)
this is pretty insane and likely needs more competent/reliable handling
Mike Blumenkrantz [Tue, 18 Mar 2014 18:24:06 +0000 (14:24 -0400)]
ecore-drm fails with appropriate error message on failing to open input
Tom Hacohen [Tue, 18 Mar 2014 14:56:57 +0000 (14:56 +0000)]
Evas text: Make tests more lax so they'll work on Ubuntu.
It seems that a different version of freetype is causing some different
values to be calculated for some glyphs. Also, we consider the whole
font list when calculating max ascent/descent, so there will always be
differences there.
This commit just laxes the tests, requiring the values to be at least
the values we expect from our font.
Fixes T1079
Tom Hacohen [Tue, 18 Mar 2014 12:07:57 +0000 (12:07 +0000)]
Evas box: Use the correct include for the generated eo file.
Chris Michael [Tue, 18 Mar 2014 11:23:49 +0000 (11:23 +0000)]
ecore-evas-drm: Add support for setting aspect of ecore_evas
@feature: This adds support for ecore_evas_aspect_set to the drm engine
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 11:21:02 +0000 (11:21 +0000)]
ecore-evas-drm: Add support for alpha & transparent setting
@feature: This adds (preliminary) support for ecore_evas_alpha_set and
ecore_evas_transparent_set
NB: This is just the function placeholders and does not actually
implement transparency yet
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 11:11:23 +0000 (11:11 +0000)]
ecore-evas-drm: Add support for maximized, fullscreen, withdrawn, and
ignore_events settings
@feature: This adds support for: ecore_evas_maximized_set,
ecore_evas_fullscreen_set, ecore_evas_withdrawn_set, and
ecore_evas_ignore_events_set functions to the drm engine
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 11:00:32 +0000 (11:00 +0000)]
ecore-evas-drm: Add support for ecore_evas_borderless_set
@feature: This adds support for setting the ecore_evas to borderless
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:52:10 +0000 (10:52 +0000)]
ecore-evas-drm: Add support for ecore_evas_iconified_set
@feature: This adds support for ecore_evas_iconified_set in ecore_evas
drm engine.
NB: This does not actually iconify the ecore_evas yet tho
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:50:26 +0000 (10:50 +0000)]
ecore-evas-drm: Add support for ecore_evas_layer_set
@feature: This adds support for ecore_evas_layer_set function used to
set the layer of the ecore_evas
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:48:16 +0000 (10:48 +0000)]
ecore-evas-drm: Add support for ecore_evas_object_cursor_set
@feature: This adds support for setting an Evas_Object as the cursor
to use on the ecore_evas
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:42:19 +0000 (10:42 +0000)]
ecore-evas-drm: Add support for setting size hints
@feature: This adds support for ecore_evas_size_min_set (and other
size hint functions).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:38:36 +0000 (10:38 +0000)]
ecore-evas-drm: Add support for ecore_evas_name_class_set
@feature: This adds support for setting the name & class of the
ecore_evas
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:33:38 +0000 (10:33 +0000)]
ecore-evas-drm: Add support for setting the title of the ecore_evas
@feature: This adds support for setting the title of the ecore_evas
with ecore_evas_title_set
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:30:44 +0000 (10:30 +0000)]
ecore-evas-drm: Add support for rotation set
@feature: This adds support for ecore_evas_rotation_set function and
updates the Evas rotation accordingly
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:27:02 +0000 (10:27 +0000)]
ecore-evas-drm: Add support for ecore_evas_move_resize
@feature: This adds support for the ecore_evas_move_resize function
and calls the appropriate internal functions for this ecore_evas
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:23:55 +0000 (10:23 +0000)]
ecore-evas-drm: Add support for ecore_evas_move function
@feature: This adds support for updating the ecore_evas location when
ecore_evas_move is called, and also calling any "move" function of the
ecore_evas.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:20:43 +0000 (10:20 +0000)]
ecore-evas-drm: Add support for setting the mouse in & mouse out
callbacks of the ecore_evas
@feature: Add support for setting the callback functions to call when
the ecore_evas gets mouse in or mouse out events
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:18:58 +0000 (10:18 +0000)]
ecore-evas-drm: Add support for setting focus_in & focus_out callbacks
@feature: Add support for setting the focus in & focus out callbacks
of the ecore_evas.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 10:15:05 +0000 (10:15 +0000)]
ecore-evas-drm: Add support for resize and move callbacks
@feature: Add support for setting the resize and move callbacks of the
ecore_evas
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 07:39:44 +0000 (07:39 +0000)]
ecore-wayland: Fix ecore_wl_init 'stalling' when used in a server
@fix: If we are using the ecore_wl library in a "server", we cannot
sit and 'sync' during the init process as that just leaves the server
in a stalled state waiting for ecore_wl_init to complete (which never
does because the server has not finished it's work).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 07:29:47 +0000 (07:29 +0000)]
evas-drm: Fix engine to use the output's framebuffer as window id
@fix: ecore_evas_window_get expects an Ecore_Window to be returned.
Because of this, we need to have a 'window' that can be returned via
ecore_evas. For this case, we will use the output's framebuffer id as
the 'window' so we need to set that into the engine info so that
ecore_evas can fetch it.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 07:27:33 +0000 (07:27 +0000)]
ecore-evas-drm: Fix setting of the ecore_evas property window and
register the prop.window for events.
@fix: ecore_evas_window_get expects an Ecore_Window to be returned so
we cannot use a void pointer in ecore_drm_device_window_get.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 18 Mar 2014 07:24:32 +0000 (07:24 +0000)]
ecore-drm: Fix ecore_drm_device_window_set to use an unsigned int
@fix: The majority of ecore_evas code expects that the 'window' returned
from ecore_evas_window_get be an Ecore_Window. Previously, this was
not possible as this function was using a void pointer. This change
fixes that issue.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Yossi Kantor [Mon, 17 Mar 2014 12:21:55 +0000 (14:21 +0200)]
Eolian: Integration of Evas Box
Jean-Philippe Andre [Tue, 18 Mar 2014 09:00:10 +0000 (18:00 +0900)]
Evas gl: Fix clip in image_draw if it's not set
In evas_gl_common_image_draw, if an image is drawn with a fresh context,
containing no clip and no cutouts, then it will be wrongly clipped to
the source image size instead of the destination surface size.
This case seems to never happen, ever, since the contexts are always
fully set by the render functions.
@fix
Jean-Philippe Andre [Tue, 18 Mar 2014 05:27:55 +0000 (14:27 +0900)]
Evas gl_x11: Do not premultiply pixels after reading FBO
The pixels are already premultiplied in such an FBO.
(cherry picked from commit
2b3e065ffeb9b587ca35909940beb59b25081e7f)
Jean-Philippe Andre [Tue, 18 Mar 2014 05:48:44 +0000 (14:48 +0900)]
Evas gl: Fix buffer dump filenames
evas_gl_common_buffer_dump can be used to dump all frames into
a series of PNG files. But the filename contained some garbage
characters (and potential segv, too).
(cherry picked from commit
a0f886138ed5a28d0d1596df3b805fca06d1ae31)
Jean-Philippe Andre [Tue, 18 Mar 2014 03:11:35 +0000 (12:11 +0900)]
Evas cserve2: Fix client crashes when a file changed
Clients must ignore changed file entries as they are not valid
anymore. The server will also mark the entry as invalid in case
of file update.
@fix