Jean-Philippe Andre [Wed, 19 Feb 2014 06:57:49 +0000 (15:57 +0900)]
Evas filters: Fix curve filter with RGBA
The curve filter did not respect premultiplied values in
RGBA images. Unfortunately, the fastest solution is to unpremul
and re-premul everytime...
Jean-Philippe Andre [Tue, 18 Feb 2014 03:09:43 +0000 (12:09 +0900)]
Evas filters: Fix options parsing
Test case "blend() blend();" did not fail when it should have.
Also, return an error if the program was empty (all whitespaces).
Jean-Philippe Andre [Tue, 18 Feb 2014 08:01:54 +0000 (17:01 +0900)]
Evas filters: Fix padding in blur filter
The wrong value was used for the command's padding.
Also, remove mention of "motion" blur which is not
implemented.
Jean-Philippe Andre [Tue, 18 Feb 2014 03:02:45 +0000 (12:02 +0900)]
Evas filters: Mark parser functions as EAPI for tests
The filter unit test case will directly call the parser functions.
Cedric BAIL [Thu, 20 Feb 2014 03:21:55 +0000 (12:21 +0900)]
edje: Edje_Edit - make sure we don't use an not initialized value.
Should close CID 1181885 and CID 1181888.
Tom Hacohen [Wed, 19 Feb 2014 14:32:39 +0000 (14:32 +0000)]
Evas textblock: Fixed a few word start/end issues.
Fixed issues with valgrind complaints at the end of the textblock, and
Fixed wrong jumping with some special cases like "a a a a "."
This fixes T995.
Tom Hacohen [Wed, 19 Feb 2014 14:31:33 +0000 (14:31 +0000)]
Evas textblock tests: Improved textblock word start/end tests.
Removed a wrong test and added some more tests.
Carsten Haitzler (Rasterman) [Wed, 19 Feb 2014 13:45:30 +0000 (22:45 +0900)]
configure - add warning for non /usr prefixes!
Carsten Haitzler (Rasterman) [Wed, 19 Feb 2014 07:04:52 +0000 (16:04 +0900)]
ethumb - fix clients going right to disk and getting half written thumbs
client still do stats (not async) but at least they dont get
half-baked thumbs now
Daniel Kolesa [Wed, 19 Feb 2014 13:05:10 +0000 (13:05 +0000)]
edje: fix custom allocator with luajit
Tom Hacohen [Wed, 19 Feb 2014 12:01:08 +0000 (12:01 +0000)]
Evas text utils: Fixed walking compound clusters.
In some scripts, like Devanagari, clusters can be split across more
than just one glyph. This is now fixed.
Thanks to YoungBok Shin for reporting.
Tom Hacohen [Wed, 19 Feb 2014 12:03:56 +0000 (12:03 +0000)]
Evas textblock tests: Add a wrapping test for complex cluster wrapping.
This test uses some Devanagari text that should have more complex
clusters than what latin text can provide. This is a more complex
wrapping case that should be tested and haven't been tested until now.
Youngbok Shin [Wed, 19 Feb 2014 10:44:34 +0000 (10:44 +0000)]
evas - Added eina_stringshare_ref call for fallbacks, lang in font description.
Summary:
When the fdesc(Font Description) is duplicated,
ref of all of stringshare pointers should be increased.
But, in the evas_font_desc_dup API, we only increased ref for name string.
It can cause some of memory issues.
Reviewers: tasn, woohyun, seoz, Hermet
CC: cedric
Differential Revision: https://phab.enlightenment.org/D570
Jean-Philippe Andre [Wed, 19 Feb 2014 04:32:26 +0000 (13:32 +0900)]
Dox: Try to fix build for windows
Evil support was not included in the docs preview generator build.
Tested locally with Makefile & deps from the wiki (win64).
perepelits.m [Wed, 19 Feb 2014 03:18:42 +0000 (12:18 +0900)]
Edje_Edit: Add NULL-check
Summary:
Add NULL-check in:
edje_edit_state_add
_edje_edit_real_part_add
_edje_part_recalc (edje_calc.c)
Reviewers: cedric, seoz, raster
Reviewed By: raster
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D567
Jean-Philippe Andre [Wed, 19 Feb 2014 02:05:18 +0000 (11:05 +0900)]
Evas filters: Use strtok instead of strtok_r for mingw
The Windows build (mingw) does not know about strtok_r.
So, let's use the non-safe variant strtok instead.
Currently, this function is called from the main thread only,
so this should be fine :)
In the future it would be nice to not use strtok anymore,
but strtok_r everywhere, and add it to evil. Considering the
release coming soon, I'm not going to change something like that
now.
Cedric BAIL [Wed, 19 Feb 2014 02:02:33 +0000 (11:02 +0900)]
Revert "autotools: try improving thread detection."
This reverts commit
14c6378fc41906db14464e17aadc7d94a825fafd.
So either we get to fix T880 or T986. I do think the problem is more tricky than
that, so I am for the moment going to get T880 broken rather than T986 as I think
the later will show up in more case for our user base. Will try to find a proper
solution for everyone before the next release.
Stefan Schmidt [Tue, 18 Feb 2014 14:31:09 +0000 (15:31 +0100)]
release: Bump version to 1.9.0-beta1
Prepare tarballs for a first beta release.
Carsten Haitzler (Rasterman) [Tue, 18 Feb 2014 08:52:16 +0000 (17:52 +0900)]
ecore-evas - sdl module - limit to 1 window at a time
this should fix T693 and only allow 1 sdl window at any one time iwht
the sdl ecore-evas engine/module. it's simply invalid to use more than
one window with it.
Cedric BAIL [Tue, 18 Feb 2014 03:24:19 +0000 (12:24 +0900)]
autotools: try improving thread detection.
As already stated before reliably detecting CC, LIBS and CFLAGS parameters for
use with threads is still a pain in 2014. With this patch I am trying to solve
T880 and use ax_pthread.m4. Note that I do not trust this macro more than the
previous one, so if there is any build issue on any system regarding thread, I
may just rollback to the previous code. Still I hope it is an improvement and
that our build system detect thread more reliably.
Jean-Philippe Andre [Mon, 17 Feb 2014 11:23:07 +0000 (20:23 +0900)]
Evas filters: Fix documentation for displace
fillmode "none" is not supported, as it does not make much sense
(how do we displace pixels without an underlying displacement map?)
Jean-Philippe Andre [Mon, 17 Feb 2014 11:19:02 +0000 (20:19 +0900)]
Evas filters: Fix padding with "blend"
Test case was:
buffer : a (alpha);
blur (20, dst = a);
blend (src = a, ox = 30);
In that case, padding was 20, 30, 20, 20.
So the blurred buffer was clipped on screen.
Jean-Philippe Andre [Mon, 17 Feb 2014 10:54:47 +0000 (19:54 +0900)]
Evas filters: Implement "fillmode" for displace
The fillmode (stretch or repeat map) was present and documented
for displace, but not implemented. Easy copy & paste from the
mask filter.
Jean-Philippe Andre [Mon, 17 Feb 2014 10:49:48 +0000 (19:49 +0900)]
Evas filters doc: Add link to "fillmode"
Jean-Philippe Andre [Mon, 17 Feb 2014 10:37:32 +0000 (19:37 +0900)]
Evas filters: Fix crash in displace filter
Simplify and fix map traversal code
davemds [Mon, 17 Feb 2014 20:25:06 +0000 (21:25 +0100)]
Emotion generic: nothing to see here, just small formatting while reading the code.
ChunEon Park [Mon, 17 Feb 2014 14:05:36 +0000 (23:05 +0900)]
evas/image - update dirty region in the render time.
this fixes the proxy rendering that sub object of the source couldn't be dirty region set.
since the object could be invisible nor won't be pre-rendered neither.
Im supposing the mapped(proxy) object rendering sequence should be totally refactored
that should be separated with the normal rendering sequence.
Until that, this change will be alternative solution.
Carsten Haitzler (Rasterman) [Mon, 17 Feb 2014 12:19:28 +0000 (21:19 +0900)]
add new ignore files
Daniel Hirt [Mon, 17 Feb 2014 12:18:19 +0000 (21:18 +0900)]
Edje/entry: Add propagation handling in mouse_out for link anchors
Summary:
Link anchors are set to allow event propagation. In some cases, this
might lead to unwanted results in entry.
Adding a flag to event_flags allows to control this in Elementary.
One example that this fixes is having wrong mouse cursor handling when
having the mouse out of the link anchor, back to the entry parent
(in this case, additional handling in els_cursor.c is required so this
event is ignored if event_flag is set with EVAS_EVENT_FLAG_ON_HOLD.
Fixes T878.
Reviewers: tasn
CC: cedric
Maniphest Tasks: T878
Differential Revision: https://phab.enlightenment.org/D561
Jean-Philippe Andre [Mon, 17 Feb 2014 08:59:13 +0000 (17:59 +0900)]
Fix make distcheck
Makefile_Edje contained the EXTRA_DIST files for Emotion, which
doesn't make much sense. So move that to Makefile_Emotion.am.
But if we want to add the test files only when EFL_TESTS is
enabled, then make distcheck doesn't work.
So, this is a bit weird.
Yakov Goldberg [Mon, 17 Feb 2014 10:04:56 +0000 (12:04 +0200)]
Revert "ecore timer: change name of Eo constructor, to avoid clash in Eolian"
This reverts commit
f7808f1f22f1d149d61e2f5b086006e943a68b61.
Yakov Goldberg [Mon, 17 Feb 2014 10:04:40 +0000 (12:04 +0200)]
Revert "ecore animator: change name of Eo constructor, to avoid clash in Eolian"
This reverts commit
16c43796a6e888b858233e393c4b65c0ee6779a0.
Jean-Philippe Andre [Mon, 17 Feb 2014 07:56:28 +0000 (16:56 +0900)]
Gif: Fix animated gifs when used as proxy sources
This looks like a typo: if (animated > 1) when animated is a... Bool!
So, I am not entirely sure why this bug is visible in case of gif
proxies, all it seems that the load_data function may be called
multiple times when the object is visible. So gif close and reopen
happen properly, and the first frame can be decoded.
Jean-Philippe Andre [Mon, 17 Feb 2014 02:28:33 +0000 (11:28 +0900)]
Emotion: Add cmake definition files
Shameless copy & paste + sed from Evas stuff
Jean-Philippe Andre [Fri, 14 Feb 2014 07:50:31 +0000 (16:50 +0900)]
Evas filters: Fix alpha in displacement filter
If the displacement map has some alpha values (not 0xFF),
then the blending should take this alpha into account. This
part is fine.
BUT, since Evas relies on premultiplied colors... we have a
problem: R (dx) and G (dy) have already been scaled down.
Actually we would need to load the map in non premultiplied RGBA,
otherwise we'll lose precision on dx,dy as soon as A != 0xFF.
Well... I guess this will be a limitation of this filter, for now
at least. Most displacement maps shouldn't even have any alpha
anyways.
Jean-Philippe Andre [Fri, 14 Feb 2014 07:28:35 +0000 (16:28 +0900)]
Evas filters: Fix RGBA displacement filter
Well... it was just broken: typo, lack of normalization on alpha,
invalid formula for blending. Duh!
davemds [Sun, 16 Feb 2014 19:33:11 +0000 (20:33 +0100)]
Emotion test: more cleanups
* really free Frame_Data on EVAS_CALLBACK_FREE (data was NULL)
* place video windows with an incremental offset
davemds [Sun, 16 Feb 2014 18:53:17 +0000 (19:53 +0100)]
Emotion test: no need for an animator to update the time string.
It is yet handled in the frame decode callback, tested with vlc, gstreamer1, video and only audio.
davemds [Sun, 16 Feb 2014 18:37:55 +0000 (19:37 +0100)]
Emotion test: some small cleanups
* simpler and stronger way to search the theme file
* removed unused verbose option
* removed unused start_time var
davemds [Sun, 16 Feb 2014 18:04:13 +0000 (19:04 +0100)]
Emotion test: no need to search for a folder that no more exists
davemds [Sun, 16 Feb 2014 15:00:20 +0000 (16:00 +0100)]
Emotion test: add the volume slider also to the Reflex group.
NOTE: do we really need to maintain 2 different groups for the reflex effect?
The 2 groups differ only for the added reflection part.
Isn't enough to show/hide the reflection part when requested?
davemds [Sun, 16 Feb 2014 14:53:09 +0000 (15:53 +0100)]
Emotion test: Removed volume key bind. EINA_UNUSED--
Volume is now managed per-window, no more need for a global volume key.
davemds [Sun, 16 Feb 2014 14:43:42 +0000 (15:43 +0100)]
Emotion test: rename speed slider to alpha as it should be.
That slider was probaly used for speed years ago, it is used for alpha now, so call it with the right name.
davemds [Sun, 16 Feb 2014 14:32:59 +0000 (15:32 +0100)]
Emotion test: New volume slider. More readable alpha text.
Yakov Goldberg [Sun, 16 Feb 2014 13:42:57 +0000 (15:42 +0200)]
edje: fix Eo ids, to make Eolian keep silence
Yakov Goldberg [Sun, 16 Feb 2014 13:40:11 +0000 (15:40 +0200)]
edje: fix doxygen, to make Eolian keep silence
davemds [Sun, 16 Feb 2014 13:13:18 +0000 (14:13 +0100)]
Emotion generic interface: FILE_SET_DONE do not have any param.
Yakov Goldberg [Sun, 16 Feb 2014 09:50:13 +0000 (11:50 +0200)]
ecore timer: change name of Eo constructor, to avoid clash in Eolian
Yakov Goldberg [Sun, 16 Feb 2014 09:39:10 +0000 (11:39 +0200)]
ecore animator: change name of Eo constructor, to avoid clash in Eolian
davemds [Fri, 14 Feb 2014 20:49:08 +0000 (21:49 +0100)]
Emotion test: some small usability improvements.
* use different icons for play/pause/stop
* alpha slider moved to the right
* centered the progress text
davemds [Fri, 14 Feb 2014 19:51:25 +0000 (20:51 +0100)]
Emotion test theme: formatting
davemds [Fri, 14 Feb 2014 19:35:50 +0000 (20:35 +0100)]
Emotion test theme: tabs--
davemds [Fri, 14 Feb 2014 19:31:07 +0000 (20:31 +0100)]
Emotion test: tabs--
Daniel Kolesa [Fri, 14 Feb 2014 16:30:51 +0000 (16:30 +0000)]
evas/common: fix segfaults in evas_map_image_loop.c
Sung W. Park [Fri, 14 Feb 2014 17:33:47 +0000 (02:33 +0900)]
evas gl: added partial rendering support for qcom extension
When running in direct rendering mode, properly support partial
rendering if the extension is properly supported.
Also, fixed the SwapBufferwWithDamage rectangle coordinate bug.
It wasn't properly y-inverted before.
Jean-Philippe Andre [Fri, 14 Feb 2014 03:51:13 +0000 (12:51 +0900)]
cserve2: Fix minor clang warning
Jean-Philippe Andre [Fri, 14 Feb 2014 03:45:55 +0000 (12:45 +0900)]
Evas filters: Another clang warning fix
Prevent error case from crashing.
Jean-Philippe Andre [Fri, 14 Feb 2014 03:26:41 +0000 (12:26 +0900)]
Evas filters: More clang fixes
NULL ptr dereference.
Jean-Philippe Andre [Fri, 14 Feb 2014 03:20:51 +0000 (12:20 +0900)]
Doc: Fix compilation warning in preview_text_filter.c
Init variable properly.
Jean-Philippe Andre [Fri, 14 Feb 2014 03:18:58 +0000 (12:18 +0900)]
Evas filters: Fix more clang stuff and add safety checks
The prepare function should not fail, unless something is very
wrong. Also, return NULL instead of EINA_FALSE.
Jean-Philippe Andre [Fri, 14 Feb 2014 03:07:54 +0000 (12:07 +0900)]
Evas filters: Silence some clang warnings
Remove @hidden tag. It does not exist.
Initialize value of pow2_div. In theory not needed but better be safe :)
Jean-Philippe Andre [Fri, 14 Feb 2014 02:23:04 +0000 (11:23 +0900)]
Evas filters: Prevent division by zero
These can not happen, as all weights > 0.
Add a CRI message in case of div0.
Fixes CID 1174081, CID 1174080.
Jean-Philippe Andre [Fri, 14 Feb 2014 01:04:02 +0000 (10:04 +0900)]
Doc: Try to fix nightly builds
Nightly build failed with (make doc):
cd: ../../../src/lib/eo/.libs: No such file or directory
Looks like there's one too many ../
I guess the build dir is not the source dir on the build bots.
Rajeev Ranjan [Fri, 14 Feb 2014 01:01:30 +0000 (10:01 +0900)]
eina: fix in eina_rectangle_pool to make sorting effective
Summary: This patch fixes the bug related to sorting not happening in eina_rectangle_pool
Reviewers: cedric, raster, seoz, Hermet
Differential Revision: https://phab.enlightenment.org/D556
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 11:08:59 +0000 (20:08 +0900)]
eet - dump - if group type is not a handled type, assume unknown
fixes CID 1039307
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 11:04:29 +0000 (20:04 +0900)]
fix DSO complaint on build ... build break by jpeg.
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:44:40 +0000 (19:44 +0900)]
evas - table - fix theoretical leak that likely will never happen
this addresses CID 1039663
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:38:13 +0000 (19:38 +0900)]
evas - gl - core - checke for wrong context in error check
this fixes CID 1135266
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:35:17 +0000 (19:35 +0900)]
evas - gl x11 - fix glx texture from pixmap target rectangle mode check
even though we don't support rectangle bits in texture targets for
texture-from-pixmap the code checked and complained - problem is it
checked the wrong thing. fixes CID 1135267
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:31:29 +0000 (19:31 +0900)]
edje - edje_cc - remove logically dead code
fixes CID 1135268
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:27:11 +0000 (19:27 +0900)]
evas cserve2 - fix chekc of wrong return type
the wrong value was checked (not the returned value) - coverity caught
this. fixes CID 1135269
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:19:59 +0000 (19:19 +0900)]
evas - gl core - don't access null ptr on init
this fixes CID 1135270
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:11:52 +0000 (19:11 +0900)]
edje - fix bunch of possible null return errors with checks
fixes CID 1135274 CID 1135272 CID 1135275
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:05:43 +0000 (19:05 +0900)]
evas filter - fix uninitialized pointer in evas
fixes CID 1174077
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 10:02:08 +0000 (19:02 +0900)]
fix build break with preview_text_filter.c
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 09:59:16 +0000 (18:59 +0900)]
evas - filter - don't use longs when DATA32 is the right type
fixes CID 1174078
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 09:57:01 +0000 (18:57 +0900)]
evas - filter - fix memory leak if cmd not found by curve cmd
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 09:23:09 +0000 (18:23 +0900)]
evas - filters - remove logically dead code
this addresses CID 1174084 - fix to remove useless code.
Jean-Philippe Andre [Thu, 13 Feb 2014 08:26:33 +0000 (17:26 +0900)]
Evas filters: Improve doc automatization
Include example filter codes from src/examples in Doxygen.
Use these examples for the generation as well.
Jean-Philippe Andre [Thu, 13 Feb 2014 06:28:46 +0000 (15:28 +0900)]
Evas filters: Add image for the Syntax example
And change it a bit so the glow is nicer :)
Jean-Philippe Andre [Thu, 13 Feb 2014 06:10:57 +0000 (15:10 +0900)]
Evas filters: Add images to the html doc
Jean-Philippe Andre [Thu, 13 Feb 2014 06:06:17 +0000 (15:06 +0900)]
Doc: Add filter examples from the doc
The filter code is a copy and paste from the doc.
Maybe the code should be in an include file, so we're sure the
preview and the documentation code are the same?
Jean-Philippe Andre [Thu, 13 Feb 2014 03:38:37 +0000 (12:38 +0900)]
Doc: Add makefile for text filter previews
Build utility tool based on the local hidden .libs
Jean-Philippe Andre [Wed, 12 Feb 2014 12:10:34 +0000 (21:10 +0900)]
Evas filters: Add a small utility to take some screenshots
Let's add some real screenshots to the documentation!
Yay! Fancy effects inlined in the Doxygen! Well... soon :)
Carsten Haitzler (Rasterman) [Thu, 13 Feb 2014 06:40:29 +0000 (15:40 +0900)]
evas - proxy source visible check not done on update del
this fixes T842
Jean-Philippe Andre [Wed, 12 Feb 2014 09:22:22 +0000 (18:22 +0900)]
Doxygen: Reduce line gap in code fragments
Code should look compact, why use small fonts otherwise?
Reduce the gap between code lines (stored as separate <div>)
to 70%. This looks about right.
Jean-Philippe Andre [Tue, 11 Feb 2014 06:56:09 +0000 (15:56 +0900)]
Doxygen: Use larger fonts
Yes, I know. Someone hates large fonts.
But the ones currently used are so SMALL!
Also, this will just revert back to what's defined in e.css,
because body and list font size aren't even consistent.
Also, display the page titles like titles: big fat blue letters.
Jean-Philippe Andre [Wed, 12 Feb 2014 02:05:20 +0000 (11:05 +0900)]
Doxygen: Align param names vertically to the top
If a param description is long and spans on multiple lines,
then the param name should be aligned to the first description
line. Otherwise the layout becomes confusing.
Align [in] and [out] vertically like params.
Jean-Philippe Andre [Wed, 12 Feb 2014 08:52:01 +0000 (17:52 +0900)]
Evas filters: Remove undocumented files from Doxygen
Some files had Doxygen tags @file, which included them in the
Files list. But they are not documented, so remove them from
the index.
Jean-Philippe Andre [Wed, 12 Feb 2014 07:25:58 +0000 (16:25 +0900)]
Evas filters: Buffer names are case-sensitive
Or so claims the documentation, so let's use strcmp
instead of strcasecmp.
Jean-Philippe Andre [Tue, 11 Feb 2014 06:56:31 +0000 (15:56 +0900)]
Evas filters: Add the reference documentation (script language)
In Doxygen format, write the reference documentation for the filters.
It will contain a few examples only, should serve more as a reference
just like edcref.
This is for the script language itself, not for the Eo APIs or the
internal APIs (those are already documented).
Jean-Philippe Andre [Wed, 12 Feb 2014 06:58:06 +0000 (15:58 +0900)]
Evas filters: in transform, dst must be specified
Since the transform operation is (for now) a very simple tool,
it only works when src and dst have the same colorspace.
This commit forces users to specify dst, since "input" and "output"
have different colorspaces.
Jean-Philippe Andre [Wed, 12 Feb 2014 03:42:47 +0000 (12:42 +0900)]
Evas filters: Improve displace flags in scripts
These flags were a dumb integer instead of being a human
readable string.
They define how to handle cases were a pixel is read from
out of boundaries.
Cedric BAIL [Thu, 13 Feb 2014 01:13:46 +0000 (10:13 +0900)]
autotools: it is useless to destroy the cache everytime and request to use it.
This patch reenable configure cache and should make configure faster after a few
run.
Carsten Haitzler (Rasterman) [Wed, 12 Feb 2014 13:51:07 +0000 (22:51 +0900)]
edje segv on file update - try and fix this. not sure if this caused it
it serems as if maybe the edf->edjes can get an edje added twice...
Daniel Kolesa [Wed, 12 Feb 2014 13:27:01 +0000 (13:27 +0000)]
evas: reverted 9d9ea2e
Carsten Haitzler (Rasterman) [Wed, 12 Feb 2014 07:01:31 +0000 (16:01 +0900)]
update REAMDE version
Carsten Haitzler (Rasterman) [Wed, 12 Feb 2014 06:56:56 +0000 (15:56 +0900)]
distribution cleanup - remove os/arch from module dirs and simplify version
now that we don't load modules from user dirs anymore - just system,
we can nuke os/arch from module dir and just keep version. since abi
for a module should be stable for a 1.X release series, we dont need
micro, just 1.9 (or 1.10 etc.). this makes thnigs a bit cleaner and
simpler for emotion_generic_players
Carsten Haitzler (Rasterman) [Wed, 12 Feb 2014 04:07:30 +0000 (13:07 +0900)]
finish off long list of complaints for bad configure options
Albin Tonnerre [Wed, 12 Feb 2014 02:50:48 +0000 (11:50 +0900)]
eina: allow eina_time_get to fall back to other clocks if the first one fails
Summary:
eina_time_get tries to use only one clock which is defined at compile-time and
returns the result of that one. This causes problems on platforms where eg.
CLOCK_PROCESS_CPUTIME_ID is defined but the clock is actually not implemented
(ie. clock_gettime returns EINVAL), as we simply don't get any time at all.
Instead, make sure we include the code for all defined clocks and simply fall
back to other clocks if the previous ones aren't implemented.
Reviewers: cedric, raster
Reviewed By: cedric
CC: cedric
Differential Revision: https://phab.enlightenment.org/D547
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>