Jean-Philippe Andre [Fri, 18 Apr 2014 07:39:37 +0000 (16:39 +0900)]
Edje: Use LOSSY_ETC1 instead of LOSSY
Add a new flag in EDC files to specify ETC1 compression
should be enabled. It follows the same rules as the
current LOSSY flag for JPEG compression.
@feature
Jean-Philippe Andre [Fri, 18 Apr 2014 06:31:43 +0000 (15:31 +0900)]
rg_etc1: Fix encoding of solid color blocks
The ETC1 encoder is expected to write all 8 bytes of the
output data. But in case of a solid color block, it was writing
only 1 of the first 3 bytes (R, G, B). So lots of solid blocks
were containing invalid data (for instance: R + dR < 0 or > 255).
Jean-Philippe Andre [Fri, 18 Apr 2014 04:12:03 +0000 (13:12 +0900)]
Edje: Fix ETC1 encoding with EET and Edje
There were a few critical issues:
- Invalid pointer arithmetics on the input data (char vs. int)
- Invalid logic in the pixel duplication code
All of these due to bad copy and paste :(
Also, use LZ4HC instead of LZ4 when compression is enabled.
ETC1 encoding is so damn slow you won't see the difference between
LZ4 and LZ4HC compression times.
Jean-Philippe Andre [Thu, 17 Apr 2014 10:04:31 +0000 (19:04 +0900)]
Edje: Add support for ETC1 encoding
Edje will compress images as ETC1 provided that they don't
contain any alpha AND the command line includes -etc1.
Jean-Philippe Andre [Thu, 17 Apr 2014 10:03:19 +0000 (19:03 +0900)]
Eet: Use medium quality for ETC1 by default
And remove dithering.
This is the same as
07700a799ce759
Daniel Zaoui [Fri, 18 Apr 2014 06:21:35 +0000 (09:21 +0300)]
Eolian/Lexer: Fix coverity issues.
Fix CID 1203411
Fix CID 1203412
Daniel Zaoui [Fri, 18 Apr 2014 06:01:37 +0000 (09:01 +0300)]
Eolian/Generator: Fix coverity issues.
Fix CID 1203409
Fix CID 1203410
Daniel Zaoui [Thu, 17 Apr 2014 14:07:23 +0000 (17:07 +0300)]
Eolian/Lexer: fix when function name is composed of only one letter.
Cedric Bail [Thu, 17 Apr 2014 19:29:31 +0000 (21:29 +0200)]
edje: let's call this function before returning.
CID 1194718.
Cedric Bail [Thu, 17 Apr 2014 19:26:57 +0000 (21:26 +0200)]
rg_etc: check against meaningful size (pointer size is not)
CID 1195440.
Cedric Bail [Thu, 17 Apr 2014 19:21:54 +0000 (21:21 +0200)]
evas: let's check fwrite work as expected.
CID 1195441.
Cedric Bail [Thu, 17 Apr 2014 19:18:32 +0000 (21:18 +0200)]
evas: fix TGV loader to not leak ressource.
CID 1195443.
Cedric Bail [Thu, 17 Apr 2014 19:16:12 +0000 (21:16 +0200)]
ecore_evas: avoid using non initialized value for cursor position in X backend.
CID 1195445.
Cedric Bail [Thu, 17 Apr 2014 19:10:34 +0000 (21:10 +0200)]
eo: do not risk dereferencing NULL.
CID 1199563.
Felipe Magno de Almeida [Thu, 17 Apr 2014 19:08:43 +0000 (21:08 +0200)]
ecore: Fixed se of EINA_MAIN_LOOP_CHECK_RETURN macros with functions with return types
Summary:
Changed uses of EINA_MAIN_LOOP_CHECK_RETURN for EINA_MAIN_LOOP_CHECK_RETURN_VAL
for functions that doesn't have void return types.
These only error out when compiling with --with-profile=debug
@fixed
Reviewers: raster, cedric, smohanty
CC: cedric
Differential Revision: https://phab.enlightenment.org/D765
Andrii Kroitor [Thu, 17 Apr 2014 17:40:34 +0000 (19:40 +0200)]
edje_edit: image_del fixed
Summary: fixed segfault on broken refs in edje object
Reviewers: cedric, seoz, raster
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D759
Signed-off-by: Cedric Bail <cedric.bail@free.fr>
Lukasz Stanislawski [Thu, 17 Apr 2014 17:36:10 +0000 (19:36 +0200)]
eo: fix broken children iterator, remove redundant fields.
@fix
Summary: Tests added.
Reviewers: raster, JackDanielZ, tasn
CC: cedric
Differential Revision: https://phab.enlightenment.org/D763
Signed-off-by: Cedric Bail <cedric.bail@free.fr>
Wonguk Jeong [Thu, 17 Apr 2014 17:34:33 +0000 (19:34 +0200)]
efreet: check magic only for file which has size
Summary:
Due to unnecessary magic checking, there was freezing in /proc in efm.
Proc file's st_size is zero, but, it's readable. therfore, it takes unnecessary time in magic checking. And, there is no need to check magic in case of 0 sized regular files as well.
Therefore, skip magic check in case of st_size is zero.
Fixes T1173
Test Plan: enlightenment -> file browser (efm) -> get int /proc --> check whether efm freezes or not
Reviewers: raster, cedric, zmike
CC: seoz, cedric
Maniphest Tasks: T1173
Differential Revision: https://phab.enlightenment.org/D764
Signed-off-by: Cedric Bail <cedric.bail@free.fr>
Stefan Schmidt [Thu, 17 Apr 2014 15:01:45 +0000 (17:01 +0200)]
ecore_wl: Actually watch out for ECORE_FD_ERROR
Stefan Schmidt [Thu, 17 Apr 2014 14:16:57 +0000 (16:16 +0200)]
wcore_wl: Handle errors on wayland display fd
Cancel the callback if we receive an error on the wayland display
file descriptor.
Stefan Schmidt [Thu, 17 Apr 2014 14:37:44 +0000 (16:37 +0200)]
ecore_wl: Add FD_WRITE flag to handler as we monitor read and write
The _ecore_wl_cb_handle_data callback only has the flag for ECORE_FD_READ
set while the callback funtion also monitors and acts on writes. I wonder
if and how that worked before.
Daniel Zaoui [Thu, 17 Apr 2014 12:50:22 +0000 (15:50 +0300)]
Eolian/Generator: remove Eo1 generation
Daniel Zaoui [Wed, 16 Apr 2014 13:17:27 +0000 (16:17 +0300)]
Eolian/Lexer: support of complex types
Complex types are now supported in .eo files:
Eina_List * @own <Eo *>
The parser is in charge of creating a list of formatted basic types.
Carsten Haitzler (Rasterman) [Thu, 17 Apr 2014 07:35:43 +0000 (16:35 +0900)]
formatting fix (spacing)
Jean-Philippe Andre [Thu, 17 Apr 2014 05:59:52 +0000 (14:59 +0900)]
Evas gl: Fix texture border drawing
Since the introduction of color spaces other than RGBA8888 in
the GL engines, there was an issue with border images scaled in
GL. The left and right edges were simply not properly copied.
This would then show artifacts when scaling very thin images
(typically 2px wide).
Jean-Philippe Andre [Wed, 16 Apr 2014 09:28:56 +0000 (18:28 +0900)]
Evas TGV saver: Use medium quality by default
The default quality is 80 and high quality is horribly slow,
so default to medium quality instead (it's already very slow).
If you really want to shoot yourself in the foot and use high
quality, just set quality > 95 and go make some coffee.
Also, disable dithering, it creates horrible artifacts on real
life pictures (gradients and flat surfaces especially).
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 07:14:16 +0000 (16:14 +0900)]
fix overdraw issue in evas when clips change
the comments say it all. this stops adding massive full window updates
when clippers lose all children (and are visible) or gain a first
child while visible as all code hides these clip rects (or shows them)
in these cases and other usages just make no sense.
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 06:22:42 +0000 (15:22 +0900)]
remove unused code
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 06:22:08 +0000 (15:22 +0900)]
evas rects - missing update rect del if rect solid && not changed
small improvement
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 03:31:44 +0000 (12:31 +0900)]
eo - let's make NULL objects simply debug warnings, not errors.
due to recent changes a lot of objects are now NULL (correctly) and eo
complains on access of them. it's simply too noisy adding too many
if's all through code, so let's just make eo be sensible here.
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 03:27:40 +0000 (12:27 +0900)]
Revert "eo null object complaints - silence."
This reverts commit
1678d272af676a17e4608de655f231171c5a5af2.
... actually... on second thought... i think eo just needs to sensibly
handle null objects.
Carsten Haitzler (Rasterman) [Wed, 16 Apr 2014 03:18:06 +0000 (12:18 +0900)]
eo null object complaints - silence.
Stefan Schmidt [Tue, 15 Apr 2014 13:24:51 +0000 (15:24 +0200)]
Revert "build: Temporary disable debug log output during coverage"
This reverts commit
02a5a6bfea50e722bbb71b9cd813d32e453517ae.
Daniel fixed the underlaying problem in eolian debug statements which should
make this band aid unnecessary.
Matvey Konovalov [Tue, 15 Apr 2014 10:22:33 +0000 (19:22 +0900)]
evas - improve speed of unpremul with redundant pixel data
Jean-Philippe Andre [Tue, 15 Apr 2014 02:41:16 +0000 (11:41 +0900)]
Evas ETC1: Use LZ4HC for higher compression ratios
LZ4HC has a higher compression ratio than LZ4 but basically the
same decompression speed.
The performance cost during encoding is actually still pretty low
considering how expensive ETC1 compression can be (even at medium
quality).
Jean-Philippe Andre [Tue, 8 Apr 2014 08:19:38 +0000 (17:19 +0900)]
Evas: Use Evas_Colorspace to declare cspace
Why use int when we have a proper type?
All these APIs are internal.
Chris Michael [Tue, 15 Apr 2014 09:38:51 +0000 (10:38 +0100)]
ecore-wl: Check for existance of data device manager
@fix: Don't segfault if the running compositor never creates a data
device manager.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Tue, 15 Apr 2014 09:36:37 +0000 (10:36 +0100)]
ecore-wl: Don't create an input cursor surface unless the input has pointer capability.
@fix: There is no use for an input cursor surface if the seat does not
have the pointer ability (for kiosk cases, or touch-only cases where
a pointer is not created).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Fri, 11 Apr 2014 06:21:20 +0000 (07:21 +0100)]
ecore-evas-wl: Ooops, Fix opaque & input regions being set to incorrect values
@fix: As it turns out, we cannot just blindly set the regions here
during resize. Elementary apps will set the opaque region to account
for any edj frames, and having the region_set calls Here was
causing issues....
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Carsten Haitzler (Rasterman) [Tue, 15 Apr 2014 08:56:07 +0000 (17:56 +0900)]
fix recent png loader break with etc1 support that broke interlaced imgs
this fixes the png loader code to use png_read_row properly with the
number of passes needed to load aninterlaced image as well as handling
this right with scale ratio scaledown set.
Mike Blumenkrantz [Mon, 14 Apr 2014 18:39:21 +0000 (14:39 -0400)]
docs now have a slightly more english description and list libs alphabetically
Gwanglim Lee [Mon, 14 Apr 2014 13:38:58 +0000 (09:38 -0400)]
Fixed a problem with e wl server that sent invalid key value to wl client.
Summary:
This problem occurred due to xkb_keysym_t value of libxkbcommon by e wl server.
e wl server should pass keycode from evdev input device on to wl client.
In order that e wl server receives valid keycode Ecore_Event_Key should have
an extended data member. This patch should be applied with server side patch.
@fix
Test Plan: run e wl server -> create wl client -> type keys
Reviewers: raster, devilhorns, zmike
CC: cedric
Differential Revision: https://phab.enlightenment.org/D712
Daniel Zaoui [Mon, 14 Apr 2014 10:54:02 +0000 (13:54 +0300)]
Eolian: add support for cmake
Now with the file :-)
Stefan Schmidt [Mon, 14 Apr 2014 10:23:25 +0000 (12:23 +0200)]
build: Temporary disable debug log output during coverage
Due to the amount of generated debug messages from eolion generation
we end up with 3.8GB log files for elm builds when efl was build with
coverage enabled. Temporary disable this until eolion is fixed and
we can turn it on again.
Tom Hacohen [Mon, 14 Apr 2014 10:04:36 +0000 (11:04 +0100)]
Revert "Eolian: add support for cmake"
You forgot to add the file, you broke compilation for everyone.
This reverts commit
df6ef6aa78ddacfac8ccac162177ee42844b1611.
Tom Hacohen [Mon, 14 Apr 2014 09:36:09 +0000 (10:36 +0100)]
Eo: eo_do now returns called func's value + default ret fix.
It's now completely valid to do:
a = eo_do(obj, a_get());
or:
b = eo_do(obj, a_set(1), b_get());
Also, the default return value for eo2 functions is now also returned
when the object is invalid, not just when the object does not match
class.
It's a small refactor that fixed both issues at once.
@feature
@fix
Tom Hacohen [Mon, 14 Apr 2014 09:34:57 +0000 (10:34 +0100)]
Eo: Fix class checking.
Without this patch, NULL, and random garbage are detected as classes.
This fixes is.
Daniel Zaoui [Mon, 14 Apr 2014 09:42:21 +0000 (12:42 +0300)]
Eolian: add support for cmake
Daniel Zaoui [Mon, 14 Apr 2014 04:56:45 +0000 (07:56 +0300)]
Eolian/Lexer: disable INFO and DEBUG.
They are not needed on normal build. Two flags have been added to
enable/disable INF and DBG.
Two warnings have been removed since they are not printing interesting
and relevant information.
Yossi Kantor [Wed, 2 Apr 2014 07:46:53 +0000 (10:46 +0300)]
Eolian: Legacy integration of Evas Table
Yossi Kantor [Tue, 1 Apr 2014 07:10:29 +0000 (10:10 +0300)]
Eolian: Legacy Integration of Evas Text
Daniel Zaoui [Fri, 11 Apr 2014 07:38:53 +0000 (10:38 +0300)]
Eolian: fix generation of Eo2 functions definitions.
When keys are used in property, the Eo2 macro was not well chosen.
Mike Blumenkrantz [Fri, 11 Apr 2014 14:32:48 +0000 (10:32 -0400)]
eolian no longer generates return statements for void functions
Stefan Schmidt [Fri, 11 Apr 2014 09:50:47 +0000 (11:50 +0200)]
ecore_con: Force unsigned to unsigned comparison
num was changed to size_t which makes it unsigned while count is a normal
(signed) int. Comparing them does not always work as expected but in this
case we already checked if count is < 0 before and would have returned if
that would be true. Thus we can safely cast count to unsigned here as it
will be greater 0 anyway.
Stefan Schmidt [Fri, 11 Apr 2014 09:31:37 +0000 (11:31 +0200)]
engines/evas_drm: Remove some unused variables
Stefan Schmidt [Fri, 11 Apr 2014 09:26:18 +0000 (11:26 +0200)]
ecore_con: Remove superfluous >= 0 check for unsigned
Since num was changed from (signed) int to (unsigned) size_t in
16d7b981ebc22df86e4dc7719ae63677bff8b1e1 this check no longer
makes sense as it will always be true.
Yossi Kantor [Mon, 31 Mar 2014 11:54:02 +0000 (14:54 +0300)]
Eolian: Legacy integration of Evas Rectangle, Polygon, Line, Textgrid and Textblock
Yossi Kantor [Sun, 30 Mar 2014 11:08:46 +0000 (14:08 +0300)]
Eolian: Clean out Evas Eo header
Yossi Kantor [Sun, 30 Mar 2014 10:24:09 +0000 (13:24 +0300)]
Eolian: Clean out Ecore Audio Eo headers
Carsten Haitzler (Rasterman) [Fri, 11 Apr 2014 06:18:21 +0000 (15:18 +0900)]
set default help info for multisense to match new default on state
Daniel Zaoui [Thu, 10 Apr 2014 09:07:42 +0000 (12:07 +0300)]
Eo2: Fix other forgotten ret.
Mike Blumenkrantz [Thu, 10 Apr 2014 13:27:51 +0000 (09:27 -0400)]
"whether" has 2x 'h'
Tom Hacohen [Thu, 10 Apr 2014 09:04:34 +0000 (10:04 +0100)]
Edje multisense: Fixed compilation following change to eo2.
Carsten Haitzler (Rasterman) [Thu, 10 Apr 2014 09:04:09 +0000 (18:04 +0900)]
turn multisense on by default since ecore_audio is already
Tom Hacohen [Thu, 10 Apr 2014 08:46:04 +0000 (09:46 +0100)]
Move to Eo2 and adjust code accordingly
Eo2 is the new object system for the EFL, and the replacement for Eo. It
improves Eo1 in many ways that will be listed on the Wiki and discussed
on the ML before.
Another change this merge includes is using the slightly different API
eo2 offers in actual code.
For example:
eo_do(obj, elm_object_text_get(&text));
becomes:
eo_do(obj, text = elm_object_text_get());
@feature
Daniel Zaoui [Thu, 10 Apr 2014 08:42:54 +0000 (11:42 +0300)]
Eo: fix warning on printf
Daniel Zaoui [Sat, 5 Apr 2014 00:22:28 +0000 (03:22 +0300)]
Eolian: Fix generation of Eo APIs.
When return is forced to void, even if the function contains only
one parameter, the return type will be void.
Carsten Haitzler (Rasterman) [Thu, 10 Apr 2014 07:44:58 +0000 (16:44 +0900)]
fix invalid memory access where child cb frees table cache - new it seems
i havent seen this before so i would not cal lit an old bug that needs
a fix tag
Carsten Haitzler (Rasterman) [Thu, 10 Apr 2014 06:15:24 +0000 (15:15 +0900)]
formatting - imrove readability of ecore x event table decl
Tom Hacohen [Thu, 10 Apr 2014 02:11:29 +0000 (03:11 +0100)]
Eo2: Adjust test suite to recent changes.
Tom Hacohen [Thu, 3 Apr 2014 16:54:30 +0000 (17:54 +0100)]
Eo2: Make internal function static + improved debug output.
Tom Hacohen [Thu, 3 Apr 2014 09:27:43 +0000 (10:27 +0100)]
Eo: Removed redundant code.
Tom Hacohen [Wed, 2 Apr 2014 14:34:40 +0000 (15:34 +0100)]
Eo: Renamed eo files from eo2* to eo*.
Tom Hacohen [Wed, 2 Apr 2014 14:30:13 +0000 (15:30 +0100)]
Ecore audio tests: Updated to Eo2.
Tom Hacohen [Wed, 2 Apr 2014 13:21:57 +0000 (14:21 +0100)]
More adjustments for eo2 calls.
Tom Hacohen [Wed, 2 Apr 2014 12:16:04 +0000 (13:16 +0100)]
Eo: Reintroduce the events now that eo1 is gone.
Tom Hacohen [Wed, 2 Apr 2014 11:01:16 +0000 (12:01 +0100)]
Eo: EO_BASE_CLASS->EO_CLASS.
Tom Hacohen [Wed, 2 Apr 2014 10:42:00 +0000 (11:42 +0100)]
Eo: eo_base's data_*->key_data_*.
Tom Hacohen [Wed, 2 Apr 2014 10:41:25 +0000 (11:41 +0100)]
Eo: Changed eo_prefix for base class to 'eo'.
This is better than eo_base, as it's super common.
Tom Hacohen [Wed, 2 Apr 2014 10:16:09 +0000 (11:16 +0100)]
Adjust eo_do calls to work with the eo2 api.
Tom Hacohen [Wed, 2 Apr 2014 08:46:34 +0000 (09:46 +0100)]
Eo2: Updated naming Eo2->Eo.
Tom Hacohen [Wed, 2 Apr 2014 08:26:45 +0000 (09:26 +0100)]
Eo2: Removed more Eo1 code.
Tom Hacohen [Tue, 1 Apr 2014 15:21:13 +0000 (16:21 +0100)]
Eo2: Removed a lot of Eo1 code.
Tom Hacohen [Tue, 1 Apr 2014 13:07:53 +0000 (14:07 +0100)]
Eo2: Adjust to composite object changes.
Tom Hacohen [Thu, 10 Apr 2014 01:47:56 +0000 (02:47 +0100)]
Eolian: Generate Eo2.
Jérémy Zurcher [Wed, 22 Jan 2014 10:18:31 +0000 (11:18 +0100)]
eo2: call stack is now thread safe
create/destroy tls key (_eo2_call_stack_key) at eo_init()/eo_shutdown().
use _eo2_call_stack_get() to allocate the stack when required.
register _eo2_call_stack_free() as eina_tls_cb_new() delete callback.
Jérémy Zurcher [Tue, 21 Jan 2014 16:42:07 +0000 (17:42 +0100)]
eo2: add tests for thread safe call stack
Summary:
2 threads run 'eo2_do(o, a(), b());'
- A goes first, creates an object, enters 'eo2_do(o, a(), b());'
in a() call, it blocks, releases B and waits for it.
- B when released, creates an object, enters 'eo2_do(o, a(), b());'
in a() call, it joins and releases A, then blocks.
- A returns from a(); and enters b() using current call stack frame,
which is the one pushed by B! then pop the frame and releases B.
- B does as above using the stack pushed by A!
Jérémy Zurcher [Thu, 9 Jan 2014 13:43:52 +0000 (14:43 +0100)]
eo2: update class creation tests, follow new error policy
see previous commit
Jérémy Zurcher [Thu, 9 Jan 2014 13:43:26 +0000 (14:43 +0100)]
eo2: do not allow ill formed class creation
eo_class_new() returns NULL on error in _eo2_class_funcs_set().
it covers: NULL API func, API redefined, dich func override,
overriding non-existing fct.
Jérémy Zurcher [Thu, 9 Jan 2014 13:42:17 +0000 (14:42 +0100)]
eo2: if _eo2_class_funcs_set() fails eo_class_new() returns NULL
call to _eo2_class_funcs_set() is moved out of _eo_class_constructor()
into eo_class_new().
Jérémy Zurcher [Wed, 8 Jan 2014 13:48:36 +0000 (14:48 +0100)]
eo2: add tests for method call error msgs
Jérémy Zurcher [Wed, 8 Jan 2014 13:10:39 +0000 (14:10 +0100)]
eo2: tests: extract mgs checks into eo_error_msgs.c|h
Jérémy Zurcher [Wed, 8 Jan 2014 11:11:20 +0000 (12:11 +0100)]
eo2: _eo2_call_resolve() fails if op is EO_NOOP
Jérémy Zurcher [Wed, 8 Jan 2014 10:30:44 +0000 (11:30 +0100)]
eo2: add test cases for class construction errors
it covers: NULL API func, overriding non-existing fct,
API redefined, dich func override.
Jérémy Zurcher [Wed, 8 Jan 2014 10:13:34 +0000 (11:13 +0100)]
eo2: improve _eo2_class_funcs_set() errors support
as op descs are sorted, we can't output fct indexes in error msgs, but
as they are sorted using api_fct as key, we can detect multiple usage of
the same api_fct which leads to an unpredictable call to whatever is
returned by _eo2_api_op_id_get()->_eo2_api_desc_get().
it is still possible to instanciate an object of a not well defined class.
Jérémy Zurcher [Wed, 8 Jan 2014 10:10:48 +0000 (11:10 +0100)]
eo2: normalize ERR msgs
Jérémy Zurcher [Fri, 3 Jan 2014 15:02:00 +0000 (16:02 +0100)]
eo2: fix err msg in _dich_func_set()
Jérémy Zurcher [Fri, 3 Jan 2014 15:01:27 +0000 (16:01 +0100)]
eo2: improve _eo_op_id_name_get() to support eo1/2
Jérémy Zurcher [Fri, 3 Jan 2014 15:00:24 +0000 (16:00 +0100)]
eo2: add _eo2_op_id_desc_get()
Jérémy Zurcher [Fri, 3 Jan 2014 14:54:18 +0000 (15:54 +0100)]
eo2: improve err msg in _eo2_api_op_id_get() and _eo2_call_resolve()