Page MenuHomePhabricator
Feed Advanced Search

Today

bu5hm4n accepted D11906: ci: travis: enable efl-one in our all enabled build.

Damn, i thought this is already landed... sorry :)

Tue, Jun 2, 4:59 AM · efl

Yesterday

bu5hm4n accepted D11888: eo: Fix memory leak in efl_key_wref_set() api..

Seems to look fine. Lets see if it breaks for someone else ... (Its also kind of weird that there is a free func where you can tell to *not* free the node)

Mon, Jun 1, 11:06 PM · efl
bu5hm4n accepted D11907: efl_access_text: remove ptr usage in eo files.

Thx.

Mon, Jun 1, 11:05 PM · efl
bu5hm4n added a comment to D11888: eo: Fix memory leak in efl_key_wref_set() api..

I'll give it a look tomorrow :)

Mon, Jun 1, 3:04 AM · efl

Fri, May 29

bu5hm4n accepted D11903: ci: get rid of all extra asan handling.
Fri, May 29, 3:32 AM · efl
bu5hm4n added a comment to D11906: ci: travis: enable efl-one in our all enabled build.

Okay cool, thank you :)

Fri, May 29, 3:17 AM · efl
bu5hm4n accepted D11902: build: ensure we set all needed asan option for check run as well.

Yep, I did not see it until now :)

Fri, May 29, 3:17 AM · efl
bu5hm4n added a comment to D11906: ci: travis: enable efl-one in our all enabled build.

I really like this, just two things:

  • Do you think this is a good idea right now ? i don't know what changes, should we have it already on CI ?
  • Can we execute test-efl-one.py build at the end ? That will verify that nothing accidently drags in libeina or smth. like that when there was a build update.
Fri, May 29, 12:43 AM · efl

Thu, May 28

bu5hm4n added a comment to D11876: Support WebP Animation Image Files.

Only this one thing :)

Thu, May 28, 3:47 AM · efl
bu5hm4n added a comment to D11901: build: add efl-canvasl-wl to efl-one.

https://phab.enlightenment.org/D11892

Thu, May 28, 3:23 AM · efl
bu5hm4n added a comment to D11892: build: rely on automatic pkg file generation for eina.
Thu, May 28, 3:22 AM · efl
bu5hm4n requested review of D11901: build: add efl-canvasl-wl to efl-one.
Thu, May 28, 3:09 AM · efl
bu5hm4n added a child revision for D11900: Introduce a test to verify efl-one only links to the correct libs: D11901: build: add efl-canvasl-wl to efl-one.
Thu, May 28, 3:09 AM · efl
bu5hm4n requested review of D11900: Introduce a test to verify efl-one only links to the correct libs.
Thu, May 28, 2:50 AM · efl
bu5hm4n added a child revision for D11899: build: vg_common depends on evas internals: D11900: Introduce a test to verify efl-one only links to the correct libs.
Thu, May 28, 2:50 AM · efl
bu5hm4n added a comment to D11892: build: rely on automatic pkg file generation for eina.

Just checked, there is no header in efl/ that is required from eina ... maybe there was at some point and we got rid of it ?

Thu, May 28, 1:13 AM · efl
bu5hm4n requested review of D11899: build: vg_common depends on evas internals.
Thu, May 28, 1:00 AM · efl
bu5hm4n added a child revision for D11897: build: split off static parts from external parts in evas: D11899: build: vg_common depends on evas internals.
Thu, May 28, 1:00 AM · efl
bu5hm4n updated the diff for D11897: build: split off static parts from external parts in evas.

update

Thu, May 28, 1:00 AM · efl
bu5hm4n added a comment to D11897: build: split off static parts from external parts in evas.

https://travis-ci.org/github/Enlightenment/efl/builds/692050736

Thu, May 28, 12:34 AM · efl
bu5hm4n updated the diff for D11897: build: split off static parts from external parts in evas.

add a bit more eet

Thu, May 28, 12:11 AM · efl

Wed, May 27

bu5hm4n requested review of D11897: build: split off static parts from external parts in evas.
Wed, May 27, 11:32 PM · efl
bu5hm4n added a child revision for D11896: build: make eina drag in the -lm flag: D11897: build: split off static parts from external parts in evas.
Wed, May 27, 11:32 PM · efl
bu5hm4n requested review of D11896: build: make eina drag in the -lm flag.
Wed, May 27, 11:31 PM · efl
bu5hm4n added a child revision for D11895: build: add priv eo targets to efl_one_eo_deps: D11896: build: make eina drag in the -lm flag.
Wed, May 27, 11:31 PM · efl
bu5hm4n requested review of D11895: build: add priv eo targets to efl_one_eo_deps.
Wed, May 27, 11:31 PM · efl
bu5hm4n added a child revision for D11892: build: rely on automatic pkg file generation for eina: D11895: build: add priv eo targets to efl_one_eo_deps.
Wed, May 27, 11:31 PM · efl
bu5hm4n updated the diff for D11892: build: rely on automatic pkg file generation for eina.

rebase & update

Wed, May 27, 11:31 PM · efl
bu5hm4n added a comment to T8721: Additional link option needed.

mhm, give me time until tomorrow. I think eina should drag in -lm for everyone that uses it. as eina includes math.h. Will check and investigate tomorrow.

Wed, May 27, 9:11 AM · efl, E on FreeBSD
bu5hm4n added a comment to D11892: build: rely on automatic pkg file generation for eina.

That should not matter I think, eina headers do not include anything from there... I did not see issues in e or terminology ...

Wed, May 27, 7:07 AM · efl
bu5hm4n requested review of D11893: build: efl-one track all subdirs correctly.
Wed, May 27, 5:42 AM · efl
bu5hm4n added a child revision for D11892: build: rely on automatic pkg file generation for eina: D11893: build: efl-one track all subdirs correctly.
Wed, May 27, 5:42 AM · efl
bu5hm4n requested review of D11892: build: rely on automatic pkg file generation for eina.
Wed, May 27, 5:41 AM · efl
bu5hm4n removed a project from D11805: build: split up building of libs, bins, modules, benchmark: DO NOT MERGE.
Wed, May 27, 1:53 AM · efl
bu5hm4n removed a project from D11806: refactor build: DO NOT MERGE.
Wed, May 27, 1:53 AM · efl
bu5hm4n removed a project from D11808: build: time for efl-one: DO NOT MERGE.
Wed, May 27, 1:53 AM · efl
bu5hm4n added a comment to D11883: build: stop buildsystem from beeing a public dependency.

https://travis-ci.org/github/Enlightenment/efl/builds/691637373

Wed, May 27, 1:19 AM · efl
bu5hm4n updated the diff for D11883: build: stop buildsystem from beeing a public dependency.

Final update

Wed, May 27, 12:55 AM · efl
bu5hm4n updated the diff for D11808: build: time for efl-one.

Final update

Wed, May 27, 12:55 AM · efl
bu5hm4n updated the diff for D11806: refactor build.

Final update

Wed, May 27, 12:54 AM · efl
bu5hm4n updated the diff for D11805: build: split up building of libs, bins, modules, benchmark.

Final update

Wed, May 27, 12:54 AM · efl

Tue, May 26

bu5hm4n accepted D11838: elm_gesture_layer: Arrange the logic for delete the target object in gesture cb..

That looks indeed better, do you have some test code where i can reproduce this issue ?

Tue, May 26, 11:20 PM · efl
bu5hm4n added a comment to T8732: Memory Leak in efl_ui_focus_parent_provider_gen_content_item_map_set() api..

@smohanty please check the attached ref.

Tue, May 26, 2:41 AM
bu5hm4n requested review of D11883: build: stop buildsystem from beeing a public dependency.
Tue, May 26, 2:41 AM · efl
bu5hm4n added a child revision for D11808: build: time for efl-one: D11883: build: stop buildsystem from beeing a public dependency.
Tue, May 26, 2:41 AM · efl
bu5hm4n requested review of D11882: elm_gen****: free item_content map.
Tue, May 26, 2:35 AM · efl
bu5hm4n added a revision to T8732: Memory Leak in efl_ui_focus_parent_provider_gen_content_item_map_set() api.: D11882: elm_gen****: free item_content map.
Tue, May 26, 2:35 AM
bu5hm4n updated the diff for D11806: refactor build.

fix evas bug

Tue, May 26, 1:48 AM · efl
bu5hm4n updated the diff for D11808: build: time for efl-one.

rebase on master

Tue, May 26, 1:28 AM · efl
bu5hm4n updated the diff for D11806: refactor build.

rebase on master

Tue, May 26, 1:28 AM · efl
bu5hm4n updated the diff for D11805: build: split up building of libs, bins, modules, benchmark.

rebase on master

Tue, May 26, 1:27 AM · efl
bu5hm4n added a child revision for D11881: ecore_main_timechanges: fix include: D11805: build: split up building of libs, bins, modules, benchmark.
Tue, May 26, 1:27 AM · efl
bu5hm4n requested review of D11881: ecore_main_timechanges: fix include.
Tue, May 26, 1:27 AM · efl

Mon, May 25

bu5hm4n accepted D11854: build: lib: harmonize the use of package_c_args in all libs.
Mon, May 25, 7:08 AM · efl
bu5hm4n updated the diff for D11808: build: time for efl-one.

fix wrong package_c_args defines

Mon, May 25, 6:53 AM · efl
bu5hm4n updated the diff for D11806: refactor build.

fix comments

Mon, May 25, 6:53 AM · efl
bu5hm4n updated the diff for D11805: build: split up building of libs, bins, modules, benchmark.

rebase

Mon, May 25, 6:52 AM · efl
bu5hm4n requested changes to D11851: build: move EFL_BUILD to package_c_args used in all subprojects.

Wait, this is still missing the resetting to the state before, in line 407. (I am now sitting again at a real laptop, i can provide what i meant after 17:15)

Mon, May 25, 6:48 AM · efl
bu5hm4n accepted D11851: build: move EFL_BUILD to package_c_args used in all subprojects.

Yeah, thats how i meant it. Thank you.

Mon, May 25, 6:10 AM · efl
bu5hm4n added a comment to D11866: elm_genlist/elm_gengrid: fix memory leak caused by efl_wref_del..

Uhm, this here is kind of wrong. You can simply efl_del the object, if the object is still there after wards, you can simply delete the wref. Additionally, NULL'ing it out is also not really needed.
I think you can safe quite a few calls with fixing this, and make the code easier.

Mon, May 25, 2:12 AM · efl
bu5hm4n accepted D11846: ecore_evas_wayland_common: keep the backword compatibility for wayland dnd handler.

Seems good.

Mon, May 25, 1:59 AM · efl
bu5hm4n added a comment to T8719: are filters still in beta ?.

Our define guards for beta stuff are done via: #ifdef EFL_BETA_API_SUPPORT Anything that defines something else can simply be removed. The @beta state of a particular entity does not really matter to that.

Mon, May 25, 1:58 AM · efl
bu5hm4n updated subscribers of D11807: Merge efreet_mime efreet_trash into efreet.

Just something to drop into the round here:

Mon, May 25, 1:56 AM · efl
bu5hm4n updated subscribers of T8709: -Dnative-arch-optimization assumes all arm CPUs have neon instruction set.

I think we should add here @raster and @Hermet

Mon, May 25, 1:49 AM · Restricted Project
bu5hm4n requested review of D11877: elm_dnd: do not register widgets twice.
Mon, May 25, 1:46 AM · efl
bu5hm4n added a comment to D11855: build: modules: harmonize use of package_c_args.

I think we can just agree on a clang-format and let clang-format handle that ?

Mon, May 25, 12:22 AM · efl
bu5hm4n added a comment to D11854: build: lib: harmonize the use of package_c_args in all libs.

Yeah, exactly what @jptiz said.

Mon, May 25, 12:21 AM · efl
bu5hm4n requested changes to D11838: elm_gesture_layer: Arrange the logic for delete the target object in gesture cb..

Mike started to add test code for gesture detect

Mon, May 25, 12:18 AM · efl
bu5hm4n added a comment to T8661: efl_gfx_path: path property.

Hi, thank you for coming back to this.
As a general note:

Mon, May 25, 12:09 AM · efl: rendering, efl: api

Tue, May 19

bu5hm4n accepted D11855: build: modules: harmonize use of package_c_args.

Seems good.

Tue, May 19, 7:47 AM · efl
bu5hm4n requested changes to D11854: build: lib: harmonize the use of package_c_args in all libs.

Looks good beside the edje_external things.

Tue, May 19, 7:46 AM · efl
bu5hm4n added a comment to D11851: build: move EFL_BUILD to package_c_args used in all subprojects.

You need to set that variable before the subdir call. and you need to set that variable to the old value before the subdir call to binary, best case before line 408.

Tue, May 19, 7:38 AM · efl
bu5hm4n added a comment to T8706: CnP and Dnd Compatibility.

Dnd-4:(1) For the API user we only want to say when a element has Entered or Left a widget, the background why or how this happens is not a subject to the API user. The more important part is that the behaviour is consistent, and the way it is right now is IMO consistant, the alternative not.
Dnd-4(2): I just tested that here, when i am starting a drag by a external client or the client itself, _enter_state_change_cb gets called, flag is set correctly, enter is emitted, i am dropping in that widget, dropping event is called, _enter_state_change_cb gets called, left gets called, flag is resetted.

Tue, May 19, 6:59 AM
bu5hm4n added a comment to D11851: build: move EFL_BUILD to package_c_args used in all subprojects.

We can redefine package_c_args in line 408 to the state without -DEFL_BUILD=1. Otherwise, this looks exactly like the idea I had in mind when writing the mail. Thank you very much.

Tue, May 19, 4:46 AM · efl
bu5hm4n added a comment to T8706: CnP and Dnd Compatibility.

Dnd-4: You have to differenciate between a few things: 1. you have motion, enter & leave for the window. Which is driven directly by the protocol stuff from x / wl / win32 / cocoa. There motion, enter, leave are then via code broken down to the per widget position, enter & leave events. The per widget enter / leave events can only be called ONCE before the opposite happens. The position happens inbetween as often as possible. The fact that widget-enter and widget-leave are sometimes called from motion and sometimes from window-enter or window-leave cannot be changed, or we will miss enter and leave events in certain situations (dnd started, entered, not moved, and then dropped). So the behaviour you see there is correct, and still after writing all this, i dont get what the issue with this is. It is not opaque to the app from where the event is coming, and the app does not have access to window-motion or window-enter, or window-leave, so the app *cannot* know that.
To your second point: No that is wrong, dropping terminates the drop action, which will result in a call to _enter_state_change_cb which does that. Resetting the flags in _drop will not emit the needed DROP_LEFT event, which we probebly want to have for usecases like: Changing the theme based on that state etc. and the LEFT simple resets it to the default theme

Tue, May 19, 2:03 AM

Mon, May 18

bu5hm4n added a comment to T8706: CnP and Dnd Compatibility.

for Dnd-1: Yeah, issue. I have a fix here, will push when I find time.
for Dnd-4: Simply said: yes. _enter_state_change_cb is called when the drop somehow exists the window, or enters a window, that is not a motion event, and will not sent one (As we are not having the next position). All we know is, that the Drop has Entered or Left,. (The enter part is probebly only for consitancy). Due to the fact that these events are protected by target->currently_inside they are only emitted once. So this should not be any problem ?

Mon, May 18, 10:13 AM
bu5hm4n updated subscribers of T8706: CnP and Dnd Compatibility.

As I said in the original reply, that is nothing you should do or rely on, as the positioning could overlap, and then you have another issue.
Further more, I just checked with efl-1.19, and the code looks like that:

if (dropable) // leave last obj and enter new one
  {
     cnp_debug("leave %p\n", last_dropable->obj);
     cnp_debug("enter %p\n", dropable->obj);
     last_dropable->last.in = EINA_FALSE;
     last_dropable->last.type = NULL;
     dropable->last.in = EINA_TRUE;
     EINA_INLIST_FOREACH_SAFE(dropable->cbs_list, itr, cbs)
        if ((cbs->types & dropable->last.format) && cbs->entercb)
          cbs->entercb(cbs->enterdata, dropable->obj);
     EINA_INLIST_FOREACH_SAFE(last_dropable->cbs_list, itr, cbs)
        if ((cbs->types & last_dropable->last.format) && cbs->leavecb)
          cbs->leavecb(cbs->leavedata, last_dropable->obj);
  }
else // leave last obj
Mon, May 18, 3:55 AM

Fri, May 15

bu5hm4n added a comment to D11806: refactor build.

The files changes are needed as we are building the library now outside the library folder, which means we need a file object instead of a random string, interpreted as a path.

Fri, May 15, 3:45 AM · efl
bu5hm4n added a comment to D11808: build: time for efl-one.

Yeah that is expected. We need to build each and every shared object lib again as a static library, which doubles the amount of libraries we are building, that is sadly nothing we can get around right now.

Fri, May 15, 3:37 AM · efl

Tue, May 12

bu5hm4n added a comment to D11813: Efl.Ui.Radio_Legacy: Prevent unnecessary signal emit for contents.

Okay yeah, makes sense.

Tue, May 12, 7:56 AM · efl
bu5hm4n added a comment to D11813: Efl.Ui.Radio_Legacy: Prevent unnecessary signal emit for contents.

Can you elaborate why the icon signal is unnessesary ? If no icon is set, the theme will not receive the hidden signal, which was sent before ...

Tue, May 12, 7:14 AM · efl
bu5hm4n accepted D11815: Efl.Canvas.Object.Animation: Prevent crash when called .animation_pause_set..

Thank you :)

Tue, May 12, 6:12 AM · efl
bu5hm4n requested changes to D11815: Efl.Canvas.Object.Animation: Prevent crash when called .animation_pause_set..

I think you can just equip this with a if (pd->in), instead of moving it. The pause state will anyways be dropped when it is stopped, so this does not matter anymore. Additionally, I think it is usefull to have pause still beeing false during the first animation tick after the pausing stopped.

Tue, May 12, 5:49 AM · efl

Mon, May 11

bu5hm4n added a comment to T8705: FTBFS with -Dlua-interpreter=lua and lua 5.2.

I actually think that this is just a regression introduces in D10965. All that is needed config_h.set('ENABLE_LUA_OLD', '1') somewhere in the lua detection logic.

Mon, May 11, 7:33 AM · Restricted Project
bu5hm4n added a comment to T8706: CnP and Dnd Compatibility.
  1. I am not sure if i totally understand what you mean, the order of events here is defined by the order that the objects are added events to. I am not sure if it makes much sense to order the events in a way that you *first* get the entered and then the left. Depending on the gfx elements on the screen you could have two overlapping objects, which get entered called on, so if there is an assertion that the event flow will always be enter -> left -> enter -> left, then at least 2 overlapping objects do break this assertion. Can you explain to me for which usecase this would be important ? I am feeling like we would want to have a seperated API for that.
Mon, May 11, 6:22 AM

Sun, May 10

bu5hm4n added a project to D11806: refactor build: DO NOT MERGE.
Sun, May 10, 7:03 AM · efl
bu5hm4n added a project to D11808: build: time for efl-one: DO NOT MERGE.
Sun, May 10, 7:02 AM · efl
bu5hm4n requested review of D11808: build: time for efl-one.
Sun, May 10, 7:01 AM · efl
bu5hm4n added a child revision for D11806: refactor build: D11808: build: time for efl-one.
Sun, May 10, 7:01 AM · efl
bu5hm4n added a child revision for D11807: Merge efreet_mime efreet_trash into efreet: D11805: build: split up building of libs, bins, modules, benchmark.
Sun, May 10, 6:58 AM · efl
bu5hm4n added a parent revision for D11805: build: split up building of libs, bins, modules, benchmark: D11807: Merge efreet_mime efreet_trash into efreet.
Sun, May 10, 6:58 AM · efl
bu5hm4n requested review of D11807: Merge efreet_mime efreet_trash into efreet.
Sun, May 10, 6:57 AM · efl
bu5hm4n abandoned D11804: refactor build.
Sun, May 10, 6:56 AM · DO NOT MERGE, efl
bu5hm4n abandoned D11803: build: split up building of libs, bins, modules, benchmark.
Sun, May 10, 6:55 AM · efl
bu5hm4n abandoned D11802: Merge efreet_mime efreet_trash into efreet.
Sun, May 10, 6:55 AM · efl
bu5hm4n added a project to D11805: build: split up building of libs, bins, modules, benchmark: DO NOT MERGE.
Sun, May 10, 6:55 AM · efl
bu5hm4n requested review of D11806: refactor build.
Sun, May 10, 6:54 AM · efl
bu5hm4n added a child revision for D11805: build: split up building of libs, bins, modules, benchmark: D11806: refactor build.
Sun, May 10, 6:54 AM · efl
bu5hm4n requested review of D11805: build: split up building of libs, bins, modules, benchmark.
Sun, May 10, 6:53 AM · efl
bu5hm4n added a project to D11804: refactor build: DO NOT MERGE.
Sun, May 10, 6:49 AM · DO NOT MERGE, efl