Chris Michael [Mon, 18 Nov 2013 12:52:38 +0000 (12:52 +0000)]
Remove read code.
Fixes Coverity CID1039498
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 18 Nov 2013 12:37:38 +0000 (12:37 +0000)]
ftell can return -1 to indicate error, so trap for that
Fixes Coverity CID1039358
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 18 Nov 2013 11:58:39 +0000 (11:58 +0000)]
ftell can return -1 to indicate error, so trap for that.
Fixes Coverity CID1039356
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 18 Nov 2013 11:48:00 +0000 (11:48 +0000)]
Malloc cannot accept a negative number so check 'num' parameter before
calling malloc.
Fixes Coverity CID1039352
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Cedric Bail [Mon, 18 Nov 2013 09:20:35 +0000 (18:20 +0900)]
evas: those if (re) are useless indeed.
Should fix CID 1129611 and CID 1129612 from this week coverity scan.
Cedric Bail [Mon, 18 Nov 2013 08:57:59 +0000 (17:57 +0900)]
eina: try to handle all possible 64bits pointers environment.
Cedric Bail [Mon, 18 Nov 2013 07:01:27 +0000 (16:01 +0900)]
evas: try to clarify licensing for people that are going to make their own Evas loader.
Chris Michael [Mon, 18 Nov 2013 10:15:54 +0000 (10:15 +0000)]
Remove dead assignments reported by nightly build
cw/cy are never used inside the "dosel" block, so no need to fetch
geometry and calculate difference here (already fetched below when
needed anyway).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 18 Nov 2013 08:04:07 +0000 (08:04 +0000)]
Partial revert of
eb84475503da9f40d6a79e
Actually, we don't need to free this data here.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Mon, 18 Nov 2013 07:56:13 +0000 (07:56 +0000)]
Check for valid drag type before trying to strdup.
When deleting the drag, free any dupped strings.
Hopefully this fixes Phab Ticket T471. Don't know for sure as I cannot
reproduce the crash here :/
Signed-off-by: Chris Michael <cp.michael@samsung.com>
ChunEon Park [Mon, 18 Nov 2013 07:10:55 +0000 (16:10 +0900)]
evas/render - optimize proxy rendering.
redraw proxy if the source is smart object and it's changed visually.
Daniel Juyung Seo [Mon, 18 Nov 2013 04:08:28 +0000 (13:08 +0900)]
evas_object_main.c: fixed formatting while reading the code.
Cedric BAIL [Sun, 17 Nov 2013 08:11:56 +0000 (09:11 +0100)]
eina: remove the value from the hash before changing it !
This should fix bug T439.
Carsten Haitzler (Rasterman) [Sun, 17 Nov 2013 05:03:36 +0000 (14:03 +0900)]
ecore-evas - example - fix lack of bg object in sub buffer
this fixes T229
Cedric BAIL [Sun, 17 Nov 2013 03:47:58 +0000 (04:47 +0100)]
evas: fix build without FontConfig.
Note that we still support building without FontConfig, so please
guard access to it.
Carsten Haitzler (Rasterman) [Sun, 17 Nov 2013 03:07:46 +0000 (12:07 +0900)]
ecore_fb - fix fb size get to use evas env vars for fb device
this makes evas ane ecore_fb agree on how to address the fb device.
this fixes terminology in fb on optimus (dual gpu) setups where you may
have to declare the exact fb to use (there are 2 of them, not just
fb0) and makes things consistent at least. this also should fix T253
Carsten Haitzler (Rasterman) [Sun, 17 Nov 2013 02:03:20 +0000 (11:03 +0900)]
ecore-audio - protect against more list-walking while list is modified
Carsten Haitzler (Rasterman) [Sun, 17 Nov 2013 01:24:59 +0000 (10:24 +0900)]
oops - remove printf debug.
Carsten Haitzler (Rasterman) [Sun, 17 Nov 2013 01:23:25 +0000 (10:23 +0900)]
evas - evasgl ... make osmesa sw render support work again
Carsten Haitzler (Rasterman) [Sat, 16 Nov 2013 09:46:36 +0000 (18:46 +0900)]
evas - gl - osmesa also try older osmesa major versions
Ryuan Choi [Sat, 16 Nov 2013 06:19:44 +0000 (15:19 +0900)]
cmake: Add EDJE_CC_EXECUTABLE variable in EdjeConfig.cmake
Ryuan Choi [Sat, 16 Nov 2013 06:17:58 +0000 (15:17 +0900)]
Add EezeConfig.cmake.in
Other components already install XXXConfig.cmake and XXConfigVersion.cmake,
but Eeze was missing.
Andrii Kroitor [Sat, 16 Nov 2013 00:55:42 +0000 (09:55 +0900)]
edje_edit: fixed SEGFAULT in _edje_real_part_free
Reviewers: cedric, seoz, raster
Reviewed By: raster
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D336
Carsten Haitzler (Rasterman) [Sat, 16 Nov 2013 00:46:19 +0000 (09:46 +0900)]
doc - update images.mk
Jean Guyomarc'h [Sat, 16 Nov 2013 00:43:23 +0000 (09:43 +0900)]
Fix build issues on Mac OS X due to sh differences.
Summary:
* efl.m4: add support for xterm-256color and fix display for the bsd echo. Fix autotools issue (present on Ubuntu also, but better handled).
* doc/Makefile.am: bsd echo may not handle -n option in sh
Reviewers: cedric
CC: cedric, seoz
Differential Revision: https://phab.enlightenment.org/D329
Sebastian Dransfeld [Fri, 15 Nov 2013 12:19:58 +0000 (13:19 +0100)]
m4: remove saving of CFLAGS
Since we don't modify cflags, no need to save them.
Carsten Haitzler (Rasterman) [Fri, 15 Nov 2013 10:16:03 +0000 (19:16 +0900)]
evas - fix neon blend code used for text rendering to not leave dirty end
Chris Michael [Fri, 15 Nov 2013 08:44:19 +0000 (08:44 +0000)]
Don't call input panel show/hide methods unless input_panel_enabled is
true.
Fixes Phab Ticket T501
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Fri, 15 Nov 2013 08:43:23 +0000 (08:43 +0000)]
Do not call the contexts' input_panel_show/hide methods unless
input_panel_enabled is True.
Part of Fix for Phab Ticket T501
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Fri, 15 Nov 2013 08:11:09 +0000 (08:11 +0000)]
Unify ecore_evas resize code across both shm and egl engines.
This reduces code duplication for ecore_evas resize routines inside
wayland engines by using a common _ecore_evas_wl_common_resize
function as the resize code for both engines was the same anyway.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Fri, 15 Nov 2013 08:10:15 +0000 (08:10 +0000)]
Add function prototype for _ecore_evas_wl_common_resize
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Cedric Bail [Fri, 15 Nov 2013 05:07:45 +0000 (14:07 +0900)]
evas: fix potential similar problem as with GL X11 backend with wrong context being used.
See commit
d2b647d2b378a88bee07610a7416d4257bbebbb6.
Wonsik Jung [Wed, 13 Nov 2013 07:58:01 +0000 (16:58 +0900)]
evas: fix a missing eglMakeCurrent in image_cache_flush/image_cache_set in the GL X11 backend.
[Problem] When glTextureDelete is called in image_cache_flush(), it sometimes doesn't work.
[Cause] glTextureDelete is called with the wrong eglContext.
[Solution] Call eng_window_use() in image_cache_flush() and image_cache_set() to use the correct eglContext.
Change-Id: Id7ab1aaeb456be6dbc5f09cb2731ace5399a5dce
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Gustavo Sverzut Barbieri [Thu, 14 Nov 2013 15:27:19 +0000 (13:27 -0200)]
allow ecore to not load system modules.
this won't go to changelog/news as it's mainly internal, I can't see
people wanting to use it except rare cases as edje tools.
Chris Michael [Thu, 14 Nov 2013 12:45:33 +0000 (12:45 +0000)]
Fix some formatting
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 14 Nov 2013 12:44:10 +0000 (12:44 +0000)]
Check for valid shell_surface, title, and class before trying to set
them else wayland segfaults (internal wayland code does no safety
checks). Also, Fix some formatting while I am here...
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 14 Nov 2013 11:26:43 +0000 (11:26 +0000)]
Remove blank line
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Sebastian Dransfeld [Thu, 14 Nov 2013 04:55:10 +0000 (05:55 +0100)]
m4: Save cflags before restoring them
If we do CFLAGS=${SAVE_CFLAGS}, it is important to do
SAVE_CFLAGS=${CFLAGS} first...
Sung W. Park [Wed, 13 Nov 2013 06:39:12 +0000 (15:39 +0900)]
evas_gl : Fixed macro substitution bug in evas_gl
In evas_gl_api_ext_def.h there're calls such as:
_EVASGL_EXT_DRVNAME(EGL_KHR_image_base)
The macro is defined in evas_gl_api_ext.c as:
(strstr(glexts, #name) != NULL || strstr(glueexts, #name) != NULL)
if (_EVASGL_EXT_CHECK_SUPPORT(name)) *ext_support = 1;
But EGL_KHR_image_base is itself a macro, which is defined
in EGL/eglext.h like this:
Thus, the _EVASGL_EXT_CHECK_SUPPORT macro will unwrap into:
(strstr(glexts, "1") != NULL || strstr(glueexts, "1") != NULL)
instead of intended:
(strstr(glexts, "EGL_KHR_image_base") != NULL ||
strstr(glueexts, "EGL_KHR_image_base") != NULL)
This patch fixes this by applying stringification earlier in
_EVASGL_EXT_DRVNAME
Bugfix reported by jinhyung.jo@samsung.com
Cedric Bail [Tue, 12 Nov 2013 11:12:54 +0000 (20:12 +0900)]
eina: adjust precision for all our targeted convertion.
Cedric Bail [Tue, 12 Nov 2013 10:29:45 +0000 (19:29 +0900)]
efl: let's try to force that thread things this way.
Cedric Bail [Tue, 12 Nov 2013 10:01:10 +0000 (19:01 +0900)]
eina: fix uninitialized data use.
Cedric Bail [Tue, 12 Nov 2013 07:06:22 +0000 (16:06 +0900)]
efl: let's try to detect in a portable way phtread around the world.
Linking to Pthread seems to be highly not portable. Look at lock.m4
macro if you want to understand the hell it is ! By following it
closely we should now have better portability than the 1.7.x release.
And of course than our alpha...
Vorobiov Vitalii [Tue, 12 Nov 2013 02:18:22 +0000 (11:18 +0900)]
edje: fixing the check whether the part is swallow.
Summary:
Some functions doesn't checking the part right.
For example:
edje_edit_part_clip_to_set
edje_edit_part_restack_below
edje_edit_part_restack_above
edje_edit_part_source_set
Reviewers: cedric, seoz
Reviewed By: cedric
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D325
Yakov Goldberg [Mon, 11 Nov 2013 11:01:52 +0000 (13:01 +0200)]
evas/textblock: add const for param in evas_object_textblock_style_set for consistency
Jihoon Kim [Mon, 11 Nov 2013 11:02:51 +0000 (20:02 +0900)]
Add @since in the doc of ecore_imf_input_panel_hide () API
Jihoon Kim [Mon, 11 Nov 2013 11:01:51 +0000 (20:01 +0900)]
ecore_imf: fix indentation
ChunEon Park [Mon, 11 Nov 2013 05:25:43 +0000 (14:25 +0900)]
evas/image - init default image state value explicitly.
maxerba [Sun, 10 Nov 2013 16:49:27 +0000 (17:49 +0100)]
Updating japanese translation
Cedric BAIL [Sun, 10 Nov 2013 11:07:59 +0000 (12:07 +0100)]
eina: avoid warning.
Cedric BAIL [Sun, 10 Nov 2013 08:43:57 +0000 (09:43 +0100)]
eina: let's copy va_list when passing it around to be more portable.
NOTE: EINA_VALUE_TYPE_DISPATCH_RETURN macro is a bit weird, it does
use external variable not passed to the macro and half of the parameter
given to it are just ignored...
Cedric BAIL [Sun, 10 Nov 2013 08:26:44 +0000 (09:26 +0100)]
ecore_evas: use portable infrastructure to detect page size.
Cedric BAIL [Sun, 10 Nov 2013 08:26:30 +0000 (09:26 +0100)]
evas: use portable infrastructure to detect page size.
Cedric BAIL [Sun, 10 Nov 2013 08:26:12 +0000 (09:26 +0100)]
eina: use portable infrastructure to detect page size.
Cedric BAIL [Sun, 10 Nov 2013 08:25:16 +0000 (09:25 +0100)]
eina: detect page size in a portable way.
Carsten Haitzler (Rasterman) [Sun, 10 Nov 2013 02:04:18 +0000 (11:04 +0900)]
ecore-audio - handle deletion of objects from outputs list in ctx fail cb
Carsten Haitzler (Rasterman) [Sun, 10 Nov 2013 01:45:33 +0000 (10:45 +0900)]
ecore-audio - remove the obj from the list before calling eo super destructor
Cedric BAIL [Sat, 9 Nov 2013 14:26:24 +0000 (15:26 +0100)]
eina: let's try to reduce the covered 64bits system.
Cedric Bail [Sat, 9 Nov 2013 11:43:20 +0000 (20:43 +0900)]
eina: let pass the right size.
Cedric Bail [Sat, 9 Nov 2013 11:07:27 +0000 (20:07 +0900)]
eina: fix eina_hash_int64 on Windows system.
Cedric Bail [Sat, 9 Nov 2013 11:06:59 +0000 (20:06 +0900)]
eina: eina_lock_debug is only on Eina_Lock.
Cedric Bail [Sat, 9 Nov 2013 11:06:36 +0000 (20:06 +0900)]
eina: fix warning of unsigned int being compared to an int.
Cedric Bail [Sat, 9 Nov 2013 10:50:15 +0000 (19:50 +0900)]
gitignore: ignore files generated by make check.
Cedric Bail [Sat, 9 Nov 2013 10:48:31 +0000 (19:48 +0900)]
efl: detect 64 bits correctly.
ChunEon Park [Sat, 9 Nov 2013 08:07:35 +0000 (17:07 +0900)]
evas/image - fixed typo.
Carsten Haitzler (Rasterman) [Sat, 9 Nov 2013 01:42:49 +0000 (10:42 +0900)]
warning--
Carsten Haitzler (Rasterman) [Sat, 9 Nov 2013 01:40:17 +0000 (10:40 +0900)]
evas - gl engine - this fixes the breakages in new nvidia drivers
nvidia > 319 broke alpha windows. this code that was left commented
out fixes it. it didn't used to have any use.
Carsten Haitzler (Rasterman) [Sat, 9 Nov 2013 01:15:35 +0000 (10:15 +0900)]
embryo - embryo_cc: fix buffer overrun warning
gcc warning was right. it's a pretty smart cookie too.
Carsten Haitzler (Rasterman) [Sat, 9 Nov 2013 01:11:12 +0000 (10:11 +0900)]
ecore_con - socks support - warning that shows invalid size calc
zmike - i think this was the problem. gcc warning caught it. size was
Tom Hacohen [Fri, 8 Nov 2013 12:10:50 +0000 (12:10 +0000)]
Eo: Fixed potential infinite loop with child deletion.
It looks like it should get into a loop for every object that has
children.
ChunEon Park [Fri, 8 Nov 2013 10:48:37 +0000 (19:48 +0900)]
evas/image - removed dead code.
ChunEon Park [Fri, 8 Nov 2013 10:22:12 +0000 (19:22 +0900)]
evas - removed dead code.
Carsten Haitzler (Rasterman) [Fri, 8 Nov 2013 08:42:34 +0000 (17:42 +0900)]
clean up previous commit with simple continues on invalid name
Youngbok Shin [Fri, 8 Nov 2013 08:40:57 +0000 (17:40 +0900)]
[edje_entry] Add null checking for name of anchor and item.
Summary:
Even if the anchor name is null, it will be appended to "anchors".
There are many null checking for name of anchor.
So it should be checked in geometry_get functions.
Test Plan:
Test with text "<a href =abc>We can't find name in this case</a>".
This text makes break the application. Because of a space between "href" and "=".
The name of anchor will be saved as null in "anchors".
Reviewers: woohyun, tasn, cedric
CC: cedric, raster
Differential Revision: https://phab.enlightenment.org/D318
Thiep Ha [Fri, 8 Nov 2013 08:30:44 +0000 (17:30 +0900)]
Remove selection handlers (and move to elm)
Summary:
Selection handlers are now implemented in elementary. So, we should remove them from edje.
The BLOCK_HANDLE mode should be removed also.
Reviewers: cedric, tasn
CC: cedric, raster
Differential Revision: https://phab.enlightenment.org/D312
Tom Hacohen [Thu, 7 Nov 2013 14:37:00 +0000 (14:37 +0000)]
Changed Eo class names to be consistent. #2
All the class names are now of the format: Elm_Type_Subtype_Extra
Thanks to Yakov for pointing out I forgot a few and for giving me a nice
list.
Chris Michael [Thu, 7 Nov 2013 14:30:49 +0000 (14:30 +0000)]
Fix getting the clipboard selection and add missing LOGFN macros
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 7 Nov 2013 14:29:49 +0000 (14:29 +0000)]
Add support for additional keyboard modifiers. Fix handling of Shift
key for capitals.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 7 Nov 2013 14:28:44 +0000 (14:28 +0000)]
Add some additional keyboard masks
This adds support for Windows logo, caps lock, num lock, scroll lock
and AltGr as keyboard modifiers
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Tom Hacohen [Thu, 7 Nov 2013 11:16:01 +0000 (11:16 +0000)]
Changed Eo class names to be consistent.
All the class names are now of the format: Lib_Type_Subtype_Extra.
Cedric Bail [Thu, 7 Nov 2013 10:56:04 +0000 (19:56 +0900)]
evas: Nice catch from Coverity, fix defect CID 1126097, 1126096 and 1126088.
Vyacheslav Reutskiy [Thu, 7 Nov 2013 09:01:52 +0000 (18:01 +0900)]
edje: add string "program_%p" (%p - edje_program pointer) to a default description of the program being build.
Fix deffect with a missing tokken 'name' in the program description block when
generated by edje_cc.
Reviewers: cedric, seoz
CC: cedric
Differential Revision: https://phab.enlightenment.org/D323
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Vorobiov Vitalii [Thu, 7 Nov 2013 08:58:42 +0000 (17:58 +0900)]
eet: Stopped SPANKing the naughty programmer while deleting dictionary.
Function edje_edit_save_all cause lots of SPANK SPANK, because
eet_dictionary_free is trying to delete string that is actually not a stringshare.
Reviewers: cedric, seoz, raster
Reviewed By: cedric
CC: reutskiy.v.v, cedric
Differential Revision: https://phab.enlightenment.org/D322
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Carsten Haitzler (Rasterman) [Thu, 7 Nov 2013 08:25:05 +0000 (17:25 +0900)]
fix readme to be 1.8.0
Sebastian Dransfeld [Tue, 5 Nov 2013 09:54:23 +0000 (10:54 +0100)]
ecore_x: accept -1 as valid mm size for no change
If we pass in w_mm = h_mm = -1 we want no change if w == cw and h == ch,
as we calculate w_mm and h_mm based on these later.
Sebastian Dransfeld [Tue, 5 Nov 2013 08:41:57 +0000 (09:41 +0100)]
ecore_x: store atoms internally
No need to request them every time.
Sebastian Dransfeld [Tue, 5 Nov 2013 08:39:51 +0000 (09:39 +0100)]
ecore_x: get correct property
Since the function is ecore_x_randr_output_connector_type_get, I guess
we want RR_PROPERTY_CONNECTOR_TYPE, not RR_PROPERTY_CONNECTOR_NUMBER.
Tom Hacohen [Wed, 6 Nov 2013 14:33:06 +0000 (14:33 +0000)]
Eo: remove ; after while (0) in Eo.h.
The whole point of having "do {} while (0);" is to force people to add a
; after the call to the macro. Therefore there should be no semicolon.
Guillaume Friloux [Wed, 6 Nov 2013 13:30:54 +0000 (14:30 +0100)]
This adds the test case for the DNS error patch.
Do not set the timeout value to something too low or you walk over
another bug i discovered and that needs a lil talk.
Guillaume Friloux [Wed, 6 Nov 2013 13:13:40 +0000 (14:13 +0100)]
Fixes a bug where when you try to connect to a server using an
inexistant DNS, or having a DNS resolution error, there isnt any event
generated for the application to know.
This is not a complete fix has you need to fix a timeout value for it to
happen, but still, it makes it possible to work around it.
More work is needed on this.
Cedric Bail [Wed, 6 Nov 2013 02:11:55 +0000 (11:11 +0900)]
evil: let's try a more modern declaration.
Cedric Bail [Tue, 5 Nov 2013 07:28:56 +0000 (16:28 +0900)]
epp: avoid to rely on undefined behavior for fopen.
In the case we didn't define the file to save to, fopen would have
been called with a NULL parameter. The result of that call is not
really defined and could have led to some crash at some point.
Ryuan Choi [Tue, 5 Nov 2013 06:45:52 +0000 (15:45 +0900)]
eina: Avoid increasing required alignment of target type warning on ARM in Eina_Hash.
Summary:
When building applications with -Wcast-align on ARM, there are following warnings.
increases required alignment of target type warnings. This impact eina_hash_murmur3.
Reviewers: cedric, seoz
Reviewed By: cedric
CC: cedric
Differential Revision: https://phab.enlightenment.org/D317
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Cedric Bail [Tue, 5 Nov 2013 02:12:20 +0000 (11:12 +0900)]
ecore: remove coroutine for now. Can easily be reverted for 1.9.
Rafael Antognolli [Mon, 4 Nov 2013 16:24:59 +0000 (14:24 -0200)]
ecore/wayland: Use eina_inlist instead of wl_list for inputs too.
Rafael Antognolli [Mon, 4 Nov 2013 16:18:39 +0000 (14:18 -0200)]
ecore/wayland: Use eina_inlist instead of wl_list for outputs too.
Rafael Antognolli [Mon, 4 Nov 2013 15:48:57 +0000 (13:48 -0200)]
ecore/wayland: Use eina_inlist instead of wl_list for globals.
Since we are allocating this list and returning it, let's make it more
EFL-like.
Chris Michael [Mon, 4 Nov 2013 13:29:31 +0000 (13:29 +0000)]
Require wayland libraries of at least 1.3.0
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Jerome Pinot [Mon, 4 Nov 2013 13:11:48 +0000 (22:11 +0900)]
Add "--with-id" configure option that can be used to specify a
string identifying the build (vendor, maintainer, etc). It defines
EFL_BUILD_ID with default to "none". It can be used by applications
or for bug reports by including Efl_Config.h
Differential Revision: https://phab.enlightenment.org/D288
Rafael Antognolli [Mon, 4 Nov 2013 13:06:49 +0000 (11:06 -0200)]
ecore/wayland: Move Ecore_Wl_Output to a private header.
This is just not being used outside of ecore_wayland library, so just
move it and nothing breaks.