Page MenuHomePhabricator
Feed Advanced Search

Yesterday

bu5hm4n requested changes to D10437: gesture_manager: Changed function param to reduce internal function calls..
Thu, Oct 17, 6:03 PM · efl

Wed, Oct 16

bu5hm4n created T8379: RFC: Replacing rb_tree in eina_hash with a array..
Wed, Oct 16, 9:18 AM · efl: data types

Thu, Oct 10

bu5hm4n placed T8334: efreet-mime.pc not correctly generated up for grabs.
Thu, Oct 10, 11:29 PM · Restricted Project
bu5hm4n added a comment to T8334: efreet-mime.pc not correctly generated.

There is no specification or anything saying that the Libs var in .pc files only has to carry one single Library flag. Hence the generated efreet_mime.pc file is completly valid, maybe there is an issue in the cmake config files - i don't know. However, this is not related to the generated .pc file, Ecore avahi, edje, [...] all of them have -lm in the Libs: *list* of flags.

Thu, Oct 10, 11:27 PM · Restricted Project
bu5hm4n added inline comments to D10353: evas/callbacks: emit hardcoded legacy smart callbacks always.
Thu, Oct 10, 9:17 AM · efl
bu5hm4n added inline comments to D10355: evas/smart: inhibit evas-internal smart callbacks when there are no subscribers.
Thu, Oct 10, 9:15 AM · efl
bu5hm4n accepted D10357: edje: block "recalc" event emission if it isn't being subscribed to.
Thu, Oct 10, 9:12 AM · efl
bu5hm4n accepted D10358: evas/smart: inhibit smart member add/del callbacks when no subscribers exist.
Thu, Oct 10, 9:12 AM · efl
bu5hm4n added a comment to D10359: ecore: inhibit "idle" event emission if no subscribers exist.

Phu, the problem with this revision is that it relies on pd->idlers, which is kind of weird. as it is done through the event catcher in efl_loop, which does not account events added in the constructor. Further more, this sort of event catcher stragety also bloats out event callback list, and should probebly be solved through inhieritance than a event. As it also is making the event adding *slow* (at least it takes twice the time.).

Thu, Oct 10, 9:11 AM · efl
bu5hm4n added inline comments to D10353: evas/callbacks: emit hardcoded legacy smart callbacks always.
Thu, Oct 10, 8:05 AM · efl
bu5hm4n accepted D10354: eo: block "invalidate" event emission when there are no subscribers.
Thu, Oct 10, 8:04 AM · efl
bu5hm4n updated the task description for T8288: Animation API.
Thu, Oct 10, 6:48 AM · efl: canvas
bu5hm4n requested review of D10350: efl_canvas_animation_rotate: adjust API.
Thu, Oct 10, 6:47 AM · efl
bu5hm4n added a revision to T8288: Animation API: D10350: efl_canvas_animation_rotate: adjust API.
Thu, Oct 10, 6:47 AM · efl: canvas
bu5hm4n added a comment to D10348: efl_ui_focus_object: do not emit focus_geometry_changed when not needed.

I did this here just like in evas, they are always just checking if something gets added, deletion is not interesting anymore.

Thu, Oct 10, 5:58 AM · efl
bu5hm4n requested review of D10348: efl_ui_focus_object: do not emit focus_geometry_changed when not needed.
Thu, Oct 10, 5:42 AM · efl
bu5hm4n added a revision to T8321: Slow event emission in eo: D10348: efl_ui_focus_object: do not emit focus_geometry_changed when not needed.
Thu, Oct 10, 5:42 AM · efl

Wed, Oct 9

bu5hm4n added a comment to D10321: tests/eina: use #ifdef guard for block which can only occur inside ifdef.

But what does it matter if this block can be executed or not ? There is code that sets things based on a value, that is affected by some build option. But the fact that this might be impacted by some build option does not really matter here, Its just code, that the compiler can remove, no need to do that by hand IMO.

Wed, Oct 9, 10:59 AM · efl
bu5hm4n added a comment to D10320: evas/language: #ifdef guard variable which is only used inside #ifdef.

different build options might result in different #defines, which might discover some oversight.

Wed, Oct 9, 10:58 AM · efl
bu5hm4n accepted D10319: tests/position_manager: initialize variable in declaration.
Wed, Oct 9, 10:56 AM · efl
bu5hm4n added a comment to D10319: tests/position_manager: initialize variable in declaration.

It is used ... it is ... returned, but yeah maybe better to init it with 0.

Wed, Oct 9, 10:56 AM · efl
bu5hm4n added a comment to D10320: evas/language: #ifdef guard variable which is only used inside #ifdef.

Does that one have a travis build ?

Wed, Oct 9, 10:55 AM · efl
bu5hm4n added a comment to D10321: tests/eina: use #ifdef guard for block which can only occur inside ifdef.

I have to admit, i dont like this. whatever coverity tells here, its easier to read if there is no ifdef arround it ?

Wed, Oct 9, 10:54 AM · efl
bu5hm4n accepted D10322: examples/emotion: remove dead code.
Wed, Oct 9, 10:53 AM · efl
bu5hm4n triaged T8321: Slow event emission in eo as High priority.
Wed, Oct 9, 10:52 AM · efl
bu5hm4n updated the diff for D10305: theme: rework a bit the tab_pager theme.

move boolean flag to single_selectable

Wed, Oct 9, 5:58 AM · efl
bu5hm4n placed T8310: pthread not found in applications using thread up for grabs.

This needs some solving in either your app or eina, but not in meson. eina.pc should not force you to -D_POSIX_C_SOURCE i guess. autotools also never did that.

Wed, Oct 9, 4:17 AM · efl: data types

Tue, Oct 8

bu5hm4n added a comment to D10306: efl_ui_item: add API that restricts a item from beeing unselected.

This is not solvable alone at the single_selection layer, as we would need something in the theme to ignore the clickable event of this beeing unpressed. We cannot just quickly change the selection state of the item again, as this would at least render one frame wrong.

Tue, Oct 8, 8:11 AM · efl
bu5hm4n requested review of D10307: efl_canvas_animation_player: fix double free.
Tue, Oct 8, 8:08 AM · efl
bu5hm4n added a child revision for D10306: efl_ui_item: add API that restricts a item from beeing unselected: D10307: efl_canvas_animation_player: fix double free.
Tue, Oct 8, 8:08 AM · efl
bu5hm4n requested review of D10306: efl_ui_item: add API that restricts a item from beeing unselected.
Tue, Oct 8, 7:40 AM · efl
bu5hm4n added a child revision for D10305: theme: rework a bit the tab_pager theme: D10306: efl_ui_item: add API that restricts a item from beeing unselected.
Tue, Oct 8, 7:40 AM · efl
bu5hm4n requested review of D10305: theme: rework a bit the tab_pager theme.
Tue, Oct 8, 7:16 AM · efl
bu5hm4n added a comment to T8310: pthread not found in applications using thread.

My eina.pc contains -lpthread ...

Tue, Oct 8, 3:23 AM · efl: data types
bu5hm4n triaged T8312: elm_list_select_mode_set -> ELM_OBJECT_SELECT_MODE_ALWAYS issue/question as High priority.
Tue, Oct 8, 1:36 AM · efl: widgets

Mon, Oct 7

bu5hm4n added a comment to T8288: Animation API.

@bu5hm4n Thank you for the description :) I replied as follows.

  • efl_canvas_animation:
    • I think duration is the same as Efl.Playable.length.
    • Since interpolator has its own factors, I think it should be an object. Can we make interpolator not ownable? Because I think there are some demands that app wants to implement animations with same interpolator. (e.g. Cubic Bezier with the same factors. If Interpolator is ownable, then app should create and set interpolator again and again.)
Mon, Oct 7, 11:57 AM · efl: canvas

Sun, Oct 6

bu5hm4n updated the diff for D9851: introduce a script that calulcates coverage of spec test suite.

update commit msg

Sun, Oct 6, 1:16 AM · efl
bu5hm4n updated the diff for D9850: introduce script that generates widget hirachy.

update commit msg

Sun, Oct 6, 1:15 AM · efl
bu5hm4n updated the diff for D9851: introduce a script that calulcates coverage of spec test suite.

repair from recent eolian breaks

Sun, Oct 6, 12:36 AM · efl
bu5hm4n updated the diff for D9850: introduce script that generates widget hirachy.

repair from recent eolian breaks

Sun, Oct 6, 12:36 AM · efl

Sat, Oct 5

bu5hm4n planned changes to D9850: introduce script that generates widget hirachy.

This is just broken now. Needs investigation.

Sat, Oct 5, 9:07 AM · efl
bu5hm4n planned changes to D9851: introduce a script that calulcates coverage of spec test suite.

This is just broken now. Needs investigation.

Sat, Oct 5, 9:07 AM · efl
bu5hm4n added a comment to D9851: introduce a script that calulcates coverage of spec test suite.

As said in the other revision, i do not want to have them in scripts/. Lets keep the discussion in D9851. (But i would prefer to move all statistic-related scripts to something that also tells that, in *one* go)

Sat, Oct 5, 9:01 AM · efl
bu5hm4n added a comment to D9850: introduce script that generates widget hirachy.

@zmike i want to leave it there for now, but i agree that this should be move. Lets get this in like this, and move this together with the "run all examples"-skript to some directory. I also do not want to have it in ./scripts/ as this is the place where the build related meson scripts are, and i do not want to mix those two types of scripts. (Maybe tree-helpers ? or statistics-scripts ?)

Sat, Oct 5, 9:00 AM · efl
bu5hm4n added a comment to D9836: efl_composite_model: make sure we are having a cmp method that works.

Ping ?

Sat, Oct 5, 8:56 AM · efl
bu5hm4n added a comment to D10289: eldbus: properly cleanup local variable during destruction..

Looks good, but since this is fixing (and maybe) causing new issues, can you sent this through travis ?

Sat, Oct 5, 8:40 AM · efl
bu5hm4n requested review of D10292: efl_ui_spec: introduce tests that ensure we are not leaking objects.
Sat, Oct 5, 8:34 AM · efl
bu5hm4n added a child revision for D10291: efl_ui_datepicker: no need to delete the manager here: D10292: efl_ui_spec: introduce tests that ensure we are not leaking objects.
Sat, Oct 5, 8:34 AM · efl
bu5hm4n requested review of D10291: efl_ui_datepicker: no need to delete the manager here.
Sat, Oct 5, 8:34 AM · efl
bu5hm4n added a comment to D10287: eldbus: Eldbus_Proxy require to handle self desctruction as refcounting doesn't fully protect..

Mhm, just a bunch of questions:

Sat, Oct 5, 1:53 AM · efl
bu5hm4n accepted D10286: eldbus: properly call parent finalize during creation of Eldbus.Model_Method..
Sat, Oct 5, 1:49 AM · efl
bu5hm4n added a comment to D10286: eldbus: properly call parent finalize during creation of Eldbus.Model_Method..

why did eo never catch this ? :/

Sat, Oct 5, 1:49 AM · efl
bu5hm4n accepted D10288: eldbus: connection are destroyed on invalidate, there is no point for any object to not cleanup on invalidate..
Sat, Oct 5, 1:44 AM · efl
bu5hm4n added inline comments to D10289: eldbus: properly cleanup local variable during destruction..
Sat, Oct 5, 1:43 AM · efl

Fri, Oct 4

bu5hm4n added a comment to T5810: EFL "new" theme for EO objects.

Now that we have released objects with the theme beeing in the default theme, we cannot split files anymore i guess ?

Fri, Oct 4, 3:39 AM · Restricted Project, efl: widgets
bu5hm4n added a comment to T5332: Edje & evas: prevent some API calls when an object is swallowed.

I do not think this is realistically possible tbh.
There is *no* reliable way of resolving that. The ways that would be possible would also be hackable, as you can always just push a different owner. Additionally, this would clutter the API a lot. For me this is honestly a won't fix.

Fri, Oct 4, 3:20 AM · efl: layout engine, Restricted Project
bu5hm4n removed subtasks for T5354: Provide infrastructure for MVVM: T7382: [MVVM] Efl.Ui.View_List Homogenous Support, T5365: Use efl.part for efl.ui.property_bind.
Fri, Oct 4, 3:16 AM · efl (efl-1.24), efl: mvvm, efl: data types
bu5hm4n removed a parent task for T5365: Use efl.part for efl.ui.property_bind: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:16 AM · efl (efl-1.23), efl: mvvm, efl: widgets
bu5hm4n removed a parent task for T7382: [MVVM] Efl.Ui.View_List Homogenous Support: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:16 AM · efl: mvvm
bu5hm4n removed a parent task for T7659: Fix efl_part issues in C#: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:15 AM · efl: mvvm
bu5hm4n removed subtasks for T5354: Provide infrastructure for MVVM: T7659: Fix efl_part issues in C#, T7485: [MVVM] Add a Factory class that help implementing cached item factory.
Fri, Oct 4, 3:15 AM · efl (efl-1.24), efl: mvvm, efl: data types
bu5hm4n removed a parent task for T7485: [MVVM] Add a Factory class that help implementing cached item factory: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:15 AM
bu5hm4n removed subtasks for T5354: Provide infrastructure for MVVM: T8153: Efl.Ui.Factory_Bind work on string, T7670: Remove Efl.Model_Container_Item, T7668: Autogenerate property_bind View , T7528: Improve lifecycle of Child model, T7458: Composite Model renaming, T7405: [MVVM] Multistyle Factory, T7404: [MVVM] Factory text property handles, T7380: [MVVM] Efl.Ui.Viewlist renaming., T5355: Replacement of genlist for MVVM.
Fri, Oct 4, 3:14 AM · efl (efl-1.24), efl: mvvm, efl: data types
bu5hm4n removed a parent task for T5355: Replacement of genlist for MVVM: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl (efl-1.23), efl: mvvm, efl: widgets
bu5hm4n removed a parent task for T7405: [MVVM] Multistyle Factory: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl: mvvm
bu5hm4n removed a parent task for T7404: [MVVM] Factory text property handles: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM
bu5hm4n removed a parent task for T7380: [MVVM] Efl.Ui.Viewlist renaming.: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM
bu5hm4n removed a parent task for T7458: Composite Model renaming: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl (efl-1.22), efl: mvvm
bu5hm4n removed a parent task for T7528: Improve lifecycle of Child model: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl: mvvm, efl (efl-1.22), efl: data types
bu5hm4n removed a parent task for T7668: Autogenerate property_bind View : T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl: mvvm
bu5hm4n removed a parent task for T7670: Remove Efl.Model_Container_Item: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl (efl-1.22), efl: mvvm, efl: data types, Restricted Project
bu5hm4n removed a parent task for T8153: Efl.Ui.Factory_Bind work on string: T5354: Provide infrastructure for MVVM.
Fri, Oct 4, 3:14 AM · efl (efl-1.23), efl: mvvm, efl: data types
bu5hm4n removed subtasks for T5333: Cleanup View interface aka finish MVVM: T7832: Provide helper logic for text binded property, T7473: Add a create event on Efl.Ui.Factory, T7472: Asynchronous create function in Efl.Ui.Factory, T7471: Add an efl.model.property_ready_get and efl.model.properties_ready_get., T7470: [MVVM] Create a ViewModel helper class, T7469: Find a better name for Efl.Factory.model_connect.
Fri, Oct 4, 3:13 AM · efl (efl-1.23), efl: mvvm, efl: data types
bu5hm4n removed a parent task for T7469: Find a better name for Efl.Factory.model_connect: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · efl: mvvm, Restricted Project
bu5hm4n removed a parent task for T7472: Asynchronous create function in Efl.Ui.Factory: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · efl: data types, Restricted Project
bu5hm4n removed a parent task for T7470: [MVVM] Create a ViewModel helper class: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · efl: mvvm
bu5hm4n removed a parent task for T7471: Add an efl.model.property_ready_get and efl.model.properties_ready_get.: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · Restricted Project, efl: data types, Restricted Project
bu5hm4n removed a parent task for T7473: Add a create event on Efl.Ui.Factory: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · Restricted Project, efl: data types, Restricted Project
bu5hm4n removed a parent task for T7832: Provide helper logic for text binded property: T5333: Cleanup View interface aka finish MVVM.
Fri, Oct 4, 3:13 AM · efl (efl-1.23), efl: mvvm, efl: data types
bu5hm4n added a comment to T7956: Defining the behavior of iterator<struct> or iterator<uint>.

Ping ping ping ?

Fri, Oct 4, 3:11 AM · Restricted Project, efl, Restricted Project
bu5hm4n removed a subtask for T5326: Refactor ctxpopup, popup and notify: T5328: Refactor/rename container, pack, ... interface.
Fri, Oct 4, 3:07 AM · efl: widgets, Restricted Project
bu5hm4n removed parent tasks for T5328: Refactor/rename container, pack, ... interface: T5326: Refactor ctxpopup, popup and notify, T5327: Refactor toolbar.
Fri, Oct 4, 3:07 AM · efl, Restricted Project
bu5hm4n removed a subtask for T5327: Refactor toolbar: T5328: Refactor/rename container, pack, ... interface.
Fri, Oct 4, 3:07 AM · efl: widgets, Restricted Project
bu5hm4n removed parent tasks for T5327: Refactor toolbar: T6360: EO: Item and Item_List API's, T5326: Refactor ctxpopup, popup and notify.
Fri, Oct 4, 3:07 AM · efl: widgets, Restricted Project
bu5hm4n removed a subtask for T6360: EO: Item and Item_List API's: T5327: Refactor toolbar.
Fri, Oct 4, 3:07 AM · efl (efl-1.23), efl: mvvm, efl: widgets
bu5hm4n removed a subtask for T5326: Refactor ctxpopup, popup and notify: T5327: Refactor toolbar.
Fri, Oct 4, 3:07 AM · efl: widgets, Restricted Project
bu5hm4n added a comment to T5716: Base background object.

I think we do not want to have state passing for disabled things etc. on the background, as the background is just a passiv static thing floating arround in the widget ? So i think this is done ?

Fri, Oct 4, 3:06 AM · efl: widgets, Restricted Project
bu5hm4n added a comment to T5317: Add a tab/pageview widget.

I think this is done ?

Fri, Oct 4, 2:44 AM · efl: widgets, Restricted Project
bu5hm4n added a comment to T5320: Create Gesture Framework in Elementary.

I think this is done ?

Fri, Oct 4, 2:44 AM · efl: widgets, Restricted Project

Thu, Oct 3

bu5hm4n added a comment to T8305: RFC: Items.

The reason for Context here is that you can have the item in context mode, or normal mode. What other names would you like ? (I am bad at nameing ... we all know that ... :P)

Thu, Oct 3, 3:52 AM · efl: api
bu5hm4n requested review of D10277: efl_test_multi_selectable: ensure consistant state across mode switch.
Thu, Oct 3, 3:42 AM · efl

Wed, Oct 2

bu5hm4n requested review of D10275: meson: pass correct PACKAGE_DATA_DIR.
Wed, Oct 2, 10:00 AM · efl
bu5hm4n added a revision to T8304: Some widget previews are broken: D10275: meson: pass correct PACKAGE_DATA_DIR.
Wed, Oct 2, 10:00 AM · efl
bu5hm4n updated subscribers of T8305: RFC: Items.
Wed, Oct 2, 9:17 AM · efl: api
bu5hm4n created T8305: RFC: Items.
Wed, Oct 2, 9:16 AM · efl: api
bu5hm4n updated the diff for D10088: elm: add basics test to create a destroy *every* widget with errors.

remove gl ... AND STAY GONE.

Wed, Oct 2, 8:04 AM · efl
bu5hm4n added a comment to T5327: Refactor toolbar.

Here a little RFC:

Wed, Oct 2, 7:51 AM · efl: widgets, Restricted Project
bu5hm4n updated the diff for D10088: elm: add basics test to create a destroy *every* widget with errors.

use ck_assert_ptr_ne

Wed, Oct 2, 6:14 AM · efl
bu5hm4n added a comment to T8288: Animation API.

@Jaehyun_Cho what do you think about those things ?

Wed, Oct 2, 4:50 AM · efl: canvas
bu5hm4n planned changes to D10088: elm: add basics test to create a destroy *every* widget with errors.

That check.h feature is available since 2! TWO! years, come on, why is your distro still running arround with that old stuff...

Wed, Oct 2, 4:39 AM · efl