87cf5188b466d3fa1e8c8ff7ad6d5039b28e298a should sort this.

return true if we added a callback from an always handler

Make sure clean up is called after all the adds

stronger comments

slightly better "documentation"

What is "size"? is that the physical size of the screen or the resolution of it?

Well, anything with ecore_wl2 in it should be considered right out, because the whole point of this is to surrender that namespace, and sort of continuing to use it with a prefix will be confusing...

fix typo in commit log

@vtorri, not for you, @raster commented that he changed the patch when landing.

Just out of curiosity, did you try compiling after you made changes and before landing?

I've tried to understand this but I get lost when trying to figure out what happens in evas_object_change() after this patch.

Nice! Looks like a sizeable reduction in complexity at the cost of a feature that seems dubious at best - and as cedric says, can be provided manually when it's actually necessary. I feel sometimes like EFL carries too much technical debt for features that very few devs actually need, so it's nice to see this go before eo stabilizes. 😄

Reorder the whitelist, add a couple of new events, and list where it's from

Not hugely familiar with this code, have asked one question.

Cool - I recently stumbled across that define and wondered what it was all about.

I like this a lot. :)

rebase after removing temp var

Remove temp variable

Just wanting to know if it's safe to assume pd is passed

Dear reviewers, please pay close attention to the final hunk - I think this should be fine, but my EOAPI knowledge is thin on the ground.

Should be orders of magnitude faster now, however, this mere 228k file is actually a pathologically nasty test case as it has 3 lines that are over 33000 chars long (check it with wc -L)

Thanks, that fflog2 is excellent.

Can you do that with WAYLAND_DEBUG=1 so we can see what the client/compositor conversation looks like to get there?