In D10675#199808, @segfaultxavi wrote:

Efl.Ui.Multi_Selectable_Index_Range and Efl.Ui.Multi_Selectable_Object_Range?

efl_ui_list_view_example_3 still has some issue with focus, but I am not sure where it is coming from. Maybe it is a different task.

In D10598#199062, @ali.alzyod wrote:

In D10598#198963, @cedric wrote:

As I said the patch looks good and should go on, but not without addressing the past behavior in a way that application that might rely on it would be broken. I have proposed two possible path. If you have other idea to fix it, I will be fine. As long as the application can still have that feature in some way.

I think we will add new functions to change override classes at file level (get/set). I think classes are only (size, color, and text)

Great.

rebase

rebase

rebase

rebase and fix assert.

rebase

rebase

rebase

rebase

rebase

rebase

As I said the patch looks good and should go on, but not without addressing the past behavior in a way that application that might rely on it would be broken. I have proposed two possible path. If you have other idea to fix it, I will be fine. As long as the application can still have that feature in some way.

In D10598#198702, @a.srour wrote:

In D10598#198373, @cedric wrote:

This patch is a good idea, I think, but it is most likely breaking some existing application as before when you applied a style it would propagate everywhere the same file was loaded. If an application does rely on this it would be broken I would think. I am not expert on this part of the code, so I might be wrong, but if I am right, I think that we should most likely continue to update the edf->styles hash on every call to edje_object_text_class_set and handle the fallback case when a style is not found on the object to look on the file.

From function description of edje_object_text_class_set() (Sorry can't find link due to broken links in documentation, but you will find it in code)

...
@brief Sets Edje text class.
This function sets the text class for the edje
...

So this function should change text_class at edje object level not file level.

In D10634#198768, @bu5hm4n wrote:

The assert is not found ...

This is the test you're looking for.

There is a test in phab somewhere waiting for you to review it. I forgot the number will try to find later.

Because their wasn't a test for it until now 😓

In T8126#146909, @bu5hm4n wrote:

In T8126#146903, @cedric wrote:

Before all that kind of optimization, I think we will need to revise the API of the position manager and its internal a bit. Main issue I am seeing:

• The collection needs to throttle position manager sizing information request and make sure batch request are not overwhelming the main loop.

There have been a nice way to distinguish between caching sizing calls, and real sizing infos. That was not enough ?

rebase

rebase and improve logic to reduce over fetching model/entity.

rebase

Before all that kind of optimization, I think we will need to revise the API of the position manager and its internal a bit. Main issue I am seeing:

• The collection needs to throttle position manager sizing information request and make sure batch request are not overwhelming the main loop.
• Any change on the item size is currently throwing all cache out from the position manager (and in general the position manager give up on its cache way to easily).
• There is no way for the collection view to inform the position manager of information like item average size or total view port size which would avoid batching size request unnecessarily.

This isn't an ideal patch, but I don't really have a better idea on how to make this happen on theme apply otherwise as scrollable is an interface :-(

After some more back and forth with @bu5hm4n discussion on IRC, it seems very hard to actually merge those two multiselect interface as they do have different constraint.