- User Since
- Dec 16 2017, 8:25 AM (75 w, 2 d)
I'm not sure about inclusion because:
Fri, May 24
The proposed script contains a hack that is not suitable for upstream.
(not sure about removing that property since it's somewhat public API isn't it?)
Thu, May 23
Hmm, looks like arc diff forgot to update commit message. In either case, this change is in master.
Reword a comment.
- Fix coding style: align references to right
I'm targeting 5.16.X.
The solution would be to store the plugins configuration encrypted (or in general to allow for encrypted config sections)
Wed, May 22
Tue, May 21
remove no longer actual comment
XInput is not a mandatory dependency. :'(
Mon, May 20
Fri, May 17
Wed, May 15
Before merging this feature I urge you to look through the support information attached to our bug reports and check how many users actually change anything in the effects.
So let's rewrite it
Tue, May 14
Mon, May 13
The rabbit hole goes a bit deeper. On X11, all screens are rendered in a single pass, which means that per screen color correction is a huge pita.
Naming nitpick: I suggest to rename m_at.at to m_at.toplevel in order to avoid repetition.
Fri, May 10
@since 5.XX becomes @since 5.59
Make some parts of code more visually appealing.
Move PointerAxisSource enum to InputRedirection class.
On one hand, we have broken clients. On the other hand, we have our promise to keep KWin/X11 feature frozen.
Adjust doxygen comment style.
Seems a shame for this to get blocked over a tiny comment over about where an enum should be.
Wed, May 8
The difference of opinion is do we write code patching rendering and input or write code patching the window management.
As I said previously, I'd prefer a solution where Toplevel::geometry() is logical geometry of the client. This way we don't have to adjust lots of code and if we implement this in more or less "generic" way then it should be quite easy to add support for a bit controversial _GTK_FRAME_EXTENTS (distros could patch kwin for example). xdg-surface::set_window_geometry and _GTK_FRAME_EXTENTS look quite the same, except that the former provides us absolute values.
Tue, May 7
Sat, May 4
Fri, May 3
Leave a todo comment.
Don't handle unmapped clients for now.
Thu, May 2
Tue, Apr 30
Arguably it won't help us with all cases because the clients would have to voluntarily participate in the protocol.
Mon, Apr 29
Apr 26 2019
Wrote relevant matching kwin patch to close effects
Hmm, in general this is quite delicate case because KWin/X11 is feature frozen...
Could you please address my inline comments?
Please land together with other revisions.