Bug 467426

It's caused by the new triangle filter introduced in Plasma 5.27.

(In reply to Fushan Wen from comment #1)
> It's caused by the new triangle filter introduced in Plasma 5.27.

Thanks. That certainly matches my memory of roughly how long I've seen this issue.

*** Bug 469556 has been marked as a duplicate of this bug. ***

*** Bug 470155 has been marked as a duplicate of this bug. ***

*** Bug 470012 has been marked as a duplicate of this bug. ***

A possibly relevant merge request was started @ https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/3027

A possibly relevant merge request was started @ https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/3079

Git commit b5ee178bc7d6bd5958f433592fb2983228a2453e by Fushan Wen.
Committed on 22/07/2023 at 06:14.
Pushed by fusionfuture into branch 'master'.

TriangleMouseFilter: fix incorrect handling of hover events and inactive state

1. Even if inactive, we still need to record the current item so when active becomes
   true after the child item is hovered, the filter can still work correctly.
2. When resetTimer is still active (waiting), the saved item may
   already have changed, so don't eat the event when the timer is
   still running. This fixes containsMouse not updating correctly.

Explanation:

When the previous HoverEnter is still blocked by the timer, a new HoverEnter
arrives but Qt doesn't send a HoverLeave if the previous item hasn't received
a HoverEnter, so sometimes m_interceptedHoverItem is not reset before the new
item's HoverEnter arrives.

And the new item's HoverEnter can arrive after the timer is triggered, so the
old item also receives a HoverEnter and thus the old item wrongly thinks it's
hovered.

This also moves all relevant properties to a unified struct, so when
resetting items, relevant properties will also be reset.
FIXED-IN: 5.27.7

M  +34   -29   components/trianglemousefilter/trianglemousefilter.cpp
M  +28   -6    components/trianglemousefilter/trianglemousefilter.h

https://invent.kde.org/plasma/plasma-workspace/-/commit/b5ee178bc7d6bd5958f433592fb2983228a2453e

Git commit 959404ce00e156142ef4e5e5e6087f2b4b850661 by Fushan Wen, on behalf of Derek Christ.
Committed on 22/07/2023 at 07:45.
Pushed by fusionfuture into branch 'Plasma/5.27'.

TriangularMouseFilter: Filter event everytime when there was no movement at all

(cherry picked from commit 9b4cd3b04e8696f9bc7a6be247d45e34ced080bc)

M  +23   -12   components/trianglemousefilter/trianglemousefilter.cpp
M  +1    -1    components/trianglemousefilter/trianglemousefilter.h

https://invent.kde.org/plasma/plasma-workspace/-/commit/959404ce00e156142ef4e5e5e6087f2b4b850661

Git commit 2186fa259708ded8ce8e9441da23ecad28824593 by Fushan Wen, on behalf of Derek Christ.
Committed on 22/07/2023 at 07:45.
Pushed by fusionfuture into branch 'Plasma/5.27'.

Improve responsiveness of triangular filter

The responsiveness of the filter is improved by adding two new
metrics to the filter logic:     - Firstly, the direction of the cursor
movement is considered. If (in the default configuration) the filter is
currently active and the user moves the cursor to the left (away from the
content area), the filter stops intercepting the movement events. This
results in a noticeable improvement in responsiveness without compromise
the objective of the filter.     - Secondly, a velocity metric (as
already brought up in BUG 438950) is added. This ensures that a very slow
movement of the cursor results in disabling the triangular filter. This,
also, improves the the feeling of responsiveness significantly. Still,
there is a need for the timer in the filter, as when the cursor has
stopped its movement completely, it will no longer send HoverMove events
to the filter, thus it wouldn't be possible to detect a complete stop
of the cursor.

(cherry picked from commit 6c18efd13cabde92f82fa881a7ebb29beda7344b)

M  +84   -45   components/trianglemousefilter/trianglemousefilter.cpp
M  +8    -1    components/trianglemousefilter/trianglemousefilter.h

https://invent.kde.org/plasma/plasma-workspace/-/commit/2186fa259708ded8ce8e9441da23ecad28824593

Git commit 9bdc101051d878addad8f59521a0b0d72be9e61f by Fushan Wen.
Committed on 22/07/2023 at 07:45.
Pushed by fusionfuture into branch 'Plasma/5.27'.

TriangleMouseFilter: fix incorrect handling of hover events and inactive state

1. Even if inactive, we still need to record the current item so when active becomes
   true after the child item is hovered, the filter can still work correctly.
2. When resetTimer is still active (waiting), the saved item may
   already have changed, so don't eat the event when the timer is
   still running. This fixes containsMouse not updating correctly.

Explanation:

When the previous HoverEnter is still blocked by the timer, a new HoverEnter
arrives but Qt doesn't send a HoverLeave if the previous item hasn't received
a HoverEnter, so sometimes m_interceptedHoverItem is not reset before the new
item's HoverEnter arrives.

And the new item's HoverEnter can arrive after the timer is triggered, so the
old item also receives a HoverEnter and thus the old item wrongly thinks it's
hovered.

This also moves all relevant properties to a unified struct, so when
resetting items, relevant properties will also be reset.
FIXED-IN: 5.27.7
(cherry picked from commit b5ee178bc7d6bd5958f433592fb2983228a2453e)

M  +34   -29   components/trianglemousefilter/trianglemousefilter.cpp
M  +28   -6    components/trianglemousefilter/trianglemousefilter.h

https://invent.kde.org/plasma/plasma-workspace/-/commit/9bdc101051d878addad8f59521a0b0d72be9e61f

*** Bug 472658 has been marked as a duplicate of this bug. ***

Very many thanks for fixing this. It now feels much more comfortable :)