Tom Hacohen [Fri, 12 Apr 2013 12:46:30 +0000 (13:46 +0100)]
Eo array: Add more testing.
Carsten Haitzler (Rasterman) [Fri, 12 Apr 2013 11:04:34 +0000 (20:04 +0900)]
Add subclasses for evas device api (new in 1.8) and document the evas
device api... :)
Tom Hacohen [Fri, 12 Apr 2013 10:06:30 +0000 (11:06 +0100)]
Eo: Added eo_event_callback_array_* functions.
These functions let you pass an array of callbacks instead of just one.
It's more memory efficient to use this if you just add a bulk of events
on the same object.
This commits breaks ABI, and breaks API of the EO_EV_CALLBACK_ADD/DEL
signals (the event info passed).
Stefan Schmidt [Fri, 12 Apr 2013 09:47:06 +0000 (10:47 +0100)]
eeze/sensor: Make sensor init and shutdown symbols private.
These are called only from eeze_init/shutdown. No need to have them public.
Daniel Willmann [Fri, 12 Apr 2013 09:10:25 +0000 (10:10 +0100)]
ecore_timer test: Pre-increment variable to prevent buffer overflow
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Sung W. Park [Fri, 12 Apr 2013 08:34:13 +0000 (17:34 +0900)]
[Evas] Updated eglMap/UnmapImageSEC extension definition
Recently Samsung specific eglMap/UnmapImageSEC has been updated to
support new features. The extension definition was updated in evas
accordingly.
Sung W. Park [Fri, 12 Apr 2013 08:02:43 +0000 (17:02 +0900)]
[Evas GL] Initialize evas gl engine after evas stuff is initialized first.
Cedric Bail [Fri, 12 Apr 2013 02:24:14 +0000 (11:24 +0900)]
evas: remove warning, no need for double ().
Cedric Bail [Fri, 12 Apr 2013 02:23:55 +0000 (11:23 +0900)]
edje: silence warning by doing explicit conversion.
Cedric Bail [Fri, 12 Apr 2013 02:19:39 +0000 (11:19 +0900)]
ecore: reduce risk of rounding error.
Valerii Kanunik [Fri, 12 Apr 2013 02:06:37 +0000 (11:06 +0900)]
ecore: add tests for ecore_poll module.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
ChunEon Park [Fri, 12 Apr 2013 02:06:03 +0000 (11:06 +0900)]
evas - removed unused
Dmytro Kaidalov [Fri, 12 Apr 2013 01:45:05 +0000 (10:45 +0900)]
ecore: add tests for Ecore_Evas.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Vitaliy Venzak [Fri, 12 Apr 2013 01:35:45 +0000 (10:35 +0900)]
ecore: add ecore timer test.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Valerii Kanunik [Fri, 12 Apr 2013 01:27:22 +0000 (10:27 +0900)]
ecore: Added tests for ecore_app_args procedures.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Tom Hacohen [Thu, 11 Apr 2013 14:10:42 +0000 (15:10 +0100)]
Eo base: Reduce object memory footprint.
Yakov Goldberg [Thu, 11 Apr 2013 15:00:24 +0000 (18:00 +0300)]
Evas textblock: Revert commit 3832ec5
with additional disabling of unused function
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Stefan Schmidt [Thu, 11 Apr 2013 14:35:53 +0000 (15:35 +0100)]
Revert "eeze/sensor: Expose less symbols for the initial API."
Sadly not ready yet as jenkins showed me. Need some more work on
the eeze_sensor_object_get front.
This reverts commit
081d55c51ca204e9f7decb45f5ff1d561a79c9e3.
Yakov Goldberg [Thu, 11 Apr 2013 11:00:15 +0000 (14:00 +0300)]
Evas textblock: fixing tests
Re-enabling tests, disabled in commit:
84fa062d6e958d4a8fd547c96ef571eab6106e4a
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Yakov Goldberg [Thu, 11 Apr 2013 13:50:20 +0000 (16:50 +0300)]
Evas text: fixing test
Adding conditional dependence on freebidi
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Tom Hacohen [Thu, 11 Apr 2013 14:06:59 +0000 (15:06 +0100)]
Eo callbacks: Improve memory footprint of callbacks.
We don't really need a double-linked-list with a pointer to the last item.
This saves us 2 pointers per callback, which is a decent improvement.
Stefan Schmidt [Thu, 11 Apr 2013 12:30:30 +0000 (13:30 +0100)]
eeze/sensor: Expose less symbols for the initial API.
Less is more. These symbols are only used in the internal modules and can
live in the private header without EAPI.
Carsten Haitzler (Rasterman) [Thu, 11 Apr 2013 12:23:07 +0000 (21:23 +0900)]
this is not a big change - not worth changelog, but allow gl partial
updates to be rounded up to specific tile sizes (and make it 16x16 by
default).
Stefan Schmidt [Thu, 11 Apr 2013 10:58:13 +0000 (11:58 +0100)]
eeze/sensor/tizen: Fix bug where we set the wrong sensor type to the object.
Eeze sensor and Tizen types can't be in sync as the order have been changed
several times in Tizen already. We need to keep the order stable for ABI
compliance though after 1.8.
While we did the translation in one direction we did not in the other. Fix
this bug that shows now.
Stefan Schmidt [Thu, 11 Apr 2013 10:21:07 +0000 (11:21 +0100)]
eeze/sensor: Use Eeze_Sensor_Type in sensor object.
We use the typedef everywhere else. This makes sure we also use it
in the struct to avoid any confusion. The API is new for upcoming
1.8 release so no problem in changing it now.
Cedric Bail [Thu, 11 Apr 2013 09:56:52 +0000 (18:56 +0900)]
evas: let's do less useless marshalling.
Cedric Bail [Thu, 11 Apr 2013 08:20:05 +0000 (17:20 +0900)]
evas: use a hash for the image scale cache.
Cedric Bail [Thu, 11 Apr 2013 07:25:36 +0000 (16:25 +0900)]
evas: don't call eo_data_get when it is really not necessary.
Cedric Bail [Thu, 11 Apr 2013 04:21:54 +0000 (13:21 +0900)]
evas: use eo_data_get less. Use clipees_has if you want to know the number of child now.
Cedric Bail [Thu, 11 Apr 2013 03:00:21 +0000 (12:00 +0900)]
evas: remove useless initialization.
Cedric Bail [Thu, 11 Apr 2013 02:58:53 +0000 (11:58 +0900)]
eo: let's cache this value.
Cedric Bail [Thu, 11 Apr 2013 02:58:22 +0000 (11:58 +0900)]
eina: make eina_mempool_alignof a little bit faster.
Cedric Bail [Thu, 11 Apr 2013 02:57:57 +0000 (11:57 +0900)]
eina: little speed improvement by reducing chance to call function.
Cedric Bail [Thu, 11 Apr 2013 00:58:09 +0000 (09:58 +0900)]
evas: reduce unecessary write.
Chris Michael [Thu, 11 Apr 2013 05:33:30 +0000 (06:33 +0100)]
Fix formatting.
Declare variables in the proper place.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Jihoon Kim [Thu, 11 Apr 2013 03:31:59 +0000 (12:31 +0900)]
edje_entry: fix typo in comment
Jihoon Kim [Thu, 11 Apr 2013 03:30:04 +0000 (12:30 +0900)]
edje_entry: fix formatting
Carsten Haitzler (Rasterman) [Thu, 11 Apr 2013 00:21:42 +0000 (09:21 +0900)]
and further work on buffer age support in evas... after more testing
on nvidia...
Rafael Antognolli [Wed, 10 Apr 2013 17:45:12 +0000 (14:45 -0300)]
ecore/wayland: Add support for double and triple click.
It's implemented exactly the same way as on the X11 backend. It does
handle multiple devices, though may need some adjustments when
multi-touch is added, since there's no support for multi-touch on
EFL-wayland yet.
Stefan Schmidt [Wed, 10 Apr 2013 13:48:29 +0000 (14:48 +0100)]
eeze/sensor: Check if sens is valid the same way we do in other cases.
Yakov Goldberg [Wed, 10 Apr 2013 13:59:33 +0000 (16:59 +0300)]
Evas textblock: disable some tests. FIXME
Some of tests fail on e5-jenkins-slave-x86_64-1.
Tests seem right and pass locally,
but not on the server.
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Yakov Goldberg [Wed, 10 Apr 2013 11:54:37 +0000 (14:54 +0300)]
Efl textblock/entry: additions to split BiDi cursor
- handling multiple runs, multiple lines, last char of line/par
- tests added
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Tom 'TAsn' Hacohen [Mon, 14 Jan 2013 16:18:07 +0000 (16:18 +0000)]
Efl textblock/entry: Added support for split BiDi cursor.
If logical cursor is between LTR/RTL text two cursors will be shown.
Upper cursor is shown for the text of the same direction as
paragraph, lower cursor - for opposite.
NOT DONE YET
Signed-off-by: Tom 'TAsn' Hacohen <tom@stosb.com>
Yakov Goldberg [Wed, 10 Apr 2013 06:48:55 +0000 (09:48 +0300)]
Evas textblock: fixing tests to work w/ or w/o harfbuzz
In this test harfbuzz creates ligatures,
so need to handle cases w/ or w/o harfbuzz
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Carsten Haitzler (Rasterman) [Wed, 10 Apr 2013 09:39:56 +0000 (18:39 +0900)]
add environemnt vars to switch bounding box to full region list calcs
and an env var to show debug for partial updates.
Carsten Haitzler (Rasterman) [Wed, 10 Apr 2013 06:38:53 +0000 (15:38 +0900)]
fix valgrind complaint in evas gl caching of surface caps... bad code
bad!
Carsten Haitzler (Rasterman) [Wed, 10 Apr 2013 00:40:37 +0000 (09:40 +0900)]
add mesa buffer release support to evas gl engine.
Carsten Haitzler (Rasterman) [Tue, 9 Apr 2013 23:36:33 +0000 (08:36 +0900)]
add gl buffer age support fully now for default automatic mode: query
extensions correctly in egl and glx too. need to hunt a partial swap
for glx now...
Cedric Bail [Wed, 10 Apr 2013 05:54:22 +0000 (14:54 +0900)]
eet: be picky and really use the correct format without shitty cast.
WooHyun Jung [Wed, 10 Apr 2013 04:53:37 +0000 (13:53 +0900)]
Modified ChangeLog and News for the latest modification.
WooHyun Jung [Wed, 10 Apr 2013 04:48:19 +0000 (13:48 +0900)]
edje_entry.c: When cursor is located to each edge, entry now doesn't grab events for cursor movement.
Tom Hacohen [Tue, 9 Apr 2013 14:19:40 +0000 (15:19 +0100)]
Eet: Pass the correct parameter to the correct format.
Daniel Willmann [Tue, 9 Apr 2013 13:13:47 +0000 (14:13 +0100)]
eo: Update tests to check for the correct error strings
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Daniel Willmann [Tue, 9 Apr 2013 12:41:44 +0000 (13:41 +0100)]
eo: Indicate that numbers in error messages are in hex - prefix with 0x
The message "Expected a and got b" is a bit misleading :-)
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Jérémy Zurcher [Tue, 9 Apr 2013 10:03:05 +0000 (12:03 +0200)]
eina: add eina_inlist_first and eina_inlist_last
- both as static inline functions
- test added in eina_inlist_simple
Carsten Haitzler (Rasterman) [Tue, 9 Apr 2013 07:52:49 +0000 (16:52 +0900)]
add a little code to alloc engine info in outs.. but we need a new
engine api for this... i guess we have no choice... it's time for a new
engine api anyway...
maxerba [Mon, 8 Apr 2013 16:50:12 +0000 (18:50 +0200)]
updating esperanto translation
Tom Hacohen [Mon, 8 Apr 2013 12:30:31 +0000 (13:30 +0100)]
Evas font: Fixed a bug with cluster sizes and ligatures.
This casused cursor position (among other things) to look wrong with texts
ending with ligatures.
Thanks to Yakov Goldberg for reporting.
Carsten Haitzler (Rasterman) [Mon, 8 Apr 2013 11:08:43 +0000 (20:08 +0900)]
evas device new/free -> add/del (new 1.8 api so no log).
Carsten Haitzler (Rasterman) [Mon, 8 Apr 2013 11:08:19 +0000 (20:08 +0900)]
clean up outputs list on evas free.
Carsten Haitzler (Rasterman) [Mon, 8 Apr 2013 11:02:58 +0000 (20:02 +0900)]
Add multiple output api support to evas... just "eo infra stubs" right
now.
Cedric Bail [Mon, 8 Apr 2013 11:03:43 +0000 (20:03 +0900)]
edje: always invalidate and recalc the correct Edje object.
Cedric Bail [Mon, 8 Apr 2013 06:05:58 +0000 (15:05 +0900)]
evas: correctly define bounding box in all case.
Cedric Bail [Mon, 8 Apr 2013 03:24:44 +0000 (12:24 +0900)]
edje: rename variable to match the rest of the library.
Cedric Bail [Mon, 8 Apr 2013 01:39:04 +0000 (10:39 +0900)]
edje: reuse existing index instead of a direct pointer (-30KB).
This one does account for the fact that we always read relative
information even when only computing a part of it.
Cedric Bail [Sun, 7 Apr 2013 15:14:50 +0000 (00:14 +0900)]
Revert "edje: reuse existing index instead of a direct pointer (-30KB)."
This reverts commit
6a6e6dda68bc91c031e893e62cfeaf8b84253829.
Cedric Bail [Sun, 7 Apr 2013 14:50:29 +0000 (23:50 +0900)]
edje: reuse existing index instead of a direct pointer (-30KB).
Cedric Bail [Sun, 7 Apr 2013 14:44:44 +0000 (23:44 +0900)]
evas: get the bounding box from the right object and always update it.
Cedric Bail [Sun, 7 Apr 2013 14:44:13 +0000 (23:44 +0900)]
evas: correctly initialize and get the bounding box at the right time.
Cedric Bail [Sun, 7 Apr 2013 11:35:09 +0000 (20:35 +0900)]
evas: revert useless change.
Ryuan Choi [Sun, 7 Apr 2013 11:35:53 +0000 (20:35 +0900)]
Updated .gitignore for tags.
Cedric Bail [Sun, 7 Apr 2013 11:18:50 +0000 (20:18 +0900)]
edje: fix little bug with table naming.
Cedric Bail [Sun, 7 Apr 2013 08:04:21 +0000 (17:04 +0900)]
edje: increase item number in the mempool to reduce its over head.
Cedric Bail [Sun, 7 Apr 2013 06:33:58 +0000 (15:33 +0900)]
evas: small reorder for a little bit of memory win.
Cedric Bail [Sun, 7 Apr 2013 06:11:56 +0000 (15:11 +0900)]
evas: bounding box is only relevant to smart object.
Cedric Bail [Sun, 7 Apr 2013 05:00:10 +0000 (14:00 +0900)]
evas: remove one useless pointer (-30KB).
Cedric Bail [Sun, 7 Apr 2013 02:39:20 +0000 (11:39 +0900)]
edje: don't use 30K of pointers for an indirection in some event path.
Cedric Bail [Sat, 6 Apr 2013 17:18:48 +0000 (02:18 +0900)]
edje: one less pointer per Edje_Real_Part.
Jérémy Zurcher [Fri, 5 Apr 2013 21:57:46 +0000 (23:57 +0200)]
eina_inlist: fix eina_inlist_sort doc
Rafael Antognolli [Fri, 5 Apr 2013 20:26:00 +0000 (17:26 -0300)]
ecore/wayland: Sync before assuming that there's no output.
This will make the mainloop block waiting for the output info, if it's
not there yet, but it's better than just returning with no screen size.
Rafael Antognolli [Fri, 5 Apr 2013 19:31:43 +0000 (16:31 -0300)]
ecore/wayland: Actually wait for the sync callback on ecore_wl_sync().
The wl_display_sync() request doesn't really wait for all the requests
to be processed, but instead sends a request for the "done" event.
Wayland relies on the fact that the requests are processed in order, so
when the "done" event is received, it means that all the other requests
requested prior to the respective "sync" have been processed already.
This commit makes the ecore_wl_sync() call actually wait for its "done"
event (thus blocking the ecore mainloop).
Cedric BAIL [Fri, 5 Apr 2013 06:18:23 +0000 (15:18 +0900)]
evas: stop over redraw of Evas_Map content with Smart Object.
Clip object would always stay in a changed state. Nobody was walking them
to reset. Now we do reset there state as soon as we do render them.
Jiyoun Park [Fri, 5 Apr 2013 04:05:25 +0000 (13:05 +0900)]
Evas font: fix font source have wrong current size.
Cedric BAIL [Fri, 5 Apr 2013 03:20:37 +0000 (12:20 +0900)]
eet: enable more debug information regarding memory allocation.
Cedric BAIL [Fri, 5 Apr 2013 03:14:50 +0000 (12:14 +0900)]
eet: remove use of prev to reduce memory usage (30KB on 32bits system).
Cedric BAIL [Fri, 5 Apr 2013 02:50:54 +0000 (11:50 +0900)]
eet: do the correct math for allocation.
U. Artie Eoff [Thu, 4 Apr 2013 18:00:53 +0000 (11:00 -0700)]
evas: fix documentation sample code format specifiers
%f causes 0.0000 to always be printed... changed to %d
to get the desired/correct output result.
Signed-off-by: U. Artie Eoff <ullysses.a.eoff@intel.com>
Tom Hacohen [Thu, 4 Apr 2013 15:05:27 +0000 (16:05 +0100)]
Evas font: Disable run-time emboldment in some cases.
If OS/2 table is available and the font is demi-bold, don't do runtime
emboldment.
Cedric Bail [Thu, 4 Apr 2013 14:43:23 +0000 (23:43 +0900)]
eina: add some more debug information.
Cedric Bail [Thu, 4 Apr 2013 13:54:12 +0000 (22:54 +0900)]
eina: align allocated memory to 4096.
Tom Hacohen [Thu, 4 Apr 2013 12:32:06 +0000 (13:32 +0100)]
Evas clouseau: Cleaned up and added more clouseau map info.
Yakov Goldberg [Thu, 4 Apr 2013 11:26:52 +0000 (14:26 +0300)]
evas/font: fixing tests
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Daniel Willmann [Thu, 4 Apr 2013 10:46:35 +0000 (11:46 +0100)]
edje_multisense: Remove unnecessary calls to eet_{open,close}
We have the Eet_File already open in ed->file->ef so use that.
Signed-off-by: Daniel Willmann <d.willmann@samsung.com>
Yakov Goldberg [Sun, 6 Jan 2013 07:53:17 +0000 (09:53 +0200)]
evas/font: fix handling querying char at coords (click on gap)
If query at x coord, which points to rigth half of LTR char,
next position will be returned. The same for left half of RTL char.
Signed-off-by: Yakov Goldberg <yakov.g@samsung.com>
Ryuan Choi [Thu, 4 Apr 2013 08:28:34 +0000 (09:28 +0100)]
Added EFLConfigVersion.cmake.in for all XXXConfigVersion.cmakeNeeds Review
Cedric BAIL [Thu, 4 Apr 2013 06:12:40 +0000 (15:12 +0900)]
eet: round above.
Cedric BAIL [Thu, 4 Apr 2013 03:36:03 +0000 (12:36 +0900)]
evas: let's not force an allocation and write for nothing.
Cedric BAIL [Thu, 4 Apr 2013 03:21:05 +0000 (12:21 +0900)]
eo: let's not duplicate our align code with a less efficient one.
Cedric BAIL [Thu, 4 Apr 2013 03:01:07 +0000 (12:01 +0900)]
eet: another big boolean removed.
By merging all boolean in their own array I avoid loosing 2.9 bytes on
32bits and 6.9 bytes on 64bits system.
Cedric BAIL [Thu, 4 Apr 2013 02:22:43 +0000 (11:22 +0900)]
eina: remove the most costly boolean ever.
We did use this 'begin' boolean to make eina_hash always allocate only
once per item it push in the hash. This boolean was alone at the end of
a structure. It would have costed us 4bytes on 32bits system and 8bytes
on 64bits. Removing it make elemines consume 100KB less on 32bits system.
We may have a speed impact on hash insertion here, but I don't think we
do use eina_hash_add and friends in any hot path, at the moment. If that
was the case there would be some way to mitigate this, just not worth it
at the moment.