Here are the notes from the design discussion we had today about how to fix "two kioslaves accessing the same MTP device at the same time cannot work":

(I'd normally add spaces after commas in QT_VERSION_CHECK)

+1, looks good to me (but I'm no QJSEngine expert)

Excellent, thanks for the contribution.

This class was doing it with a timer, I wonder if that should be part of the recommendation ("use QLineEdit with a timer"), or if we don't recommend delaying the searching as a general solution.
(e.g. another typical fix for the "I type one letter and the result is huge" issue is to require 3 letters).

Much better, thanks ;)

Depends whether you want people to #include <Zodiac> (and zodiac.h) (the KArchive solution) or to #include <KHolidays/Zodiac> (and kholidays/zodiac.h) (the KParts solution, i.e. with namespace)

Yes, Pino is right. Trust Pino ;)

The check is trying to verify that possibleMountPoints() works, by picking a mount point that we know has to exist, the root mount point. Anything else is hard to rely on, in a unittest ;)
At least that was the logic in the old days, before all this magic that circumvents having entries in fstab...

Ah, hmm, my suggestion would allow to display a warning messagebox, but not to offer "continue/cancel" type of interaction.

You can just emit the warning signal, from the job.

Great job, thanks for the unittest.

I'm confused by the removal of PREFIX.

• forwarded message27 KBDownload

+1, looks good to me (I just wonder how well KDE applications actually behave with KWallet disabled...)

Fix pushed as https://commits.kde.org/kdelibs4support/c3fe9344a8e7be1347e97523da19b5395ce9b3db

OK maybe I misunderstood the comment "Welcome to the wondeful world of / on zfs with boot-environments :)."

OK, now we're getting somewhere.
If ssl-dev isn't installed, this code is compiled without SSL support, and fails

Yes it does, and that's where I get the "zfs" information from.

Yep, exactly.

That doesn't mean it's the right solution :-)

The reason I don't like it, is that X509 isn't a class ;-)

Ah, I see. This fails indeed:

CMake Error at CMakeLists.txt:4 (include):
  include could not find load file:

In D14591#304636, @ervin wrote:

Depends on the query part. It doesn't have to list the whole collection, it could just fetch the single item + ancestors and then return a list containing just the top most ancestor. Your predicate could then just return true if said parent was indeed a project. That fetch should not be expensive since in the current design it'll go through the cache anyway.

This commit seems to break CI with Qt 5.9, please check