[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-29 Thread David Edmundson
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #12 from David Edmundson  ---
Edit, it's not that commit, but it is that area of code.

For some reason the tooltip skips all the paths that set region

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-29 Thread David Edmundson
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #11 from David Edmundson  ---
Did some looking. 

I'm pretty sure it's 309fbf1837863c483b04491eda8af8bc5235260a in kwin

(which means it's not frameworks and we have a whil)

adding some debug:


@@ -411,8 +411,10 @@ QRegion BlurEffect::blurRegion(const EffectWindow *w)
const
+

 if (value.isValid()) {
+qDebug() << "start" << w->shape() << w->rect();
.
+qDebug() << "end" <<  region;
 } else if (w->decorationHasAlpha() && 

results in



22:17:59.165 KWin::BlurEffect::blurRegion start QRegion(0,0 364x239) QRect(0,0
364x239)
22:17:59.165 KWin::BlurEffect::blurRegion end QRegion(null)

not sure why it makes a difference &'ing these two rects should result in the
same thing.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-29 Thread Nate Graham
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #10 from Nate Graham  ---
Only affects X11, not the Wayland session.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-27 Thread Nate Graham
https://bugs.kde.org/show_bug.cgi?id=423405

Nate Graham  changed:

   What|Removed |Added

   Priority|NOR |VHI

--- Comment #9 from Nate Graham  ---
Marking VHI as we can't ship this regression in 5.20.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-27 Thread Nate Graham
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #8 from Nate Graham  ---
That didn't help sadly.

Do you have any idea what could have been responsible for this, or where I
should look?

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-24 Thread Vlad Zahorodnii
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #7 from Vlad Zahorodnii  ---
> but was unsuccessful as it did not build past a certain point in time. :(

iirc there was a breaking change in kscreenlocker 5.19. so you need sometimes
to switch between kscreenlocker 5.18 and kscreenlocker 5.19. :/

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-23 Thread Nate Graham
https://bugs.kde.org/show_bug.cgi?id=423405

--- Comment #6 from Nate Graham  ---
Hmm, I just bisected plasma-workspace, plasma-desktop, and plasma-framework
back to June 1st and I didn't manage to find the guilty commit.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 423405] Certain Plasma dialogs are sometimes or always super transparent with no blur

2020-06-23 Thread Nate Graham
https://bugs.kde.org/show_bug.cgi?id=423405

Nate Graham  changed:

   What|Removed |Added

 CC||plasma-b...@kde.org
Version|git master  |master
  Component|compositing |general
   Target Milestone|--- |1.0
   Assignee|kwin-bugs-n...@kde.org  |k...@davidedmundson.co.uk
Product|kwin|plasmashell

--- Comment #5 from Nate Graham  ---
Cool, moving to plasmashell then. Glad I'm not insane!

-- 
You are receiving this mail because:
You are watching all bug changes.