Even if not split up, it would be neat if possible to build the core and non-core part independently (see current KIOCORE_ONLY variable)

It probably makes no sense to be part of Release Service any longer then.

In D29514#665912, @ngraham wrote:

In D29514#665911, @asturmlechner wrote:

In D29514#665869, @ngraham wrote:

Hmm, on my system, the file is located at /usr/src/xserver/include/xserver-properties.h, and this doesn't seen to find it:

Where is synaptics-properties.h located on your system, if you even have been building touchpad kcm so far?

It's in /usr/include/xorg.

I must confess I've never actually been able to build it, so I appreciate your efforts here to fix that.

In that case yes it must have failed so far during build, now cmake will warn you already. It is now the same as with Evdev, Libinput and Synaptics, if you don't have them in system paths you would have to somehow feed the alternative paths to cmake.

In D29514#665869, @ngraham wrote:

Hmm, on my system, the file is located at /usr/src/xserver/include/xserver-properties.h, and this doesn't seen to find it:

In D29119#657938, @elvisangelaccio wrote:

Does PackageKit work on Windows/MacOS? I don't think so. We should probably run this code only on Unix, no?

This commit is missing in release/20.04 branch.

Without ifdefs this is effectively raising minimum KF5 dependency to >=5.69:

This should also go to 1.7.

This should be in release/20.04 branch as well *imo*.

In downstream I've simply removed that menu item for now.

Thanks for keeping pipewire optional, this is in line with how we have been packaging xdg-desktop-portal-kde in Gentoo so far, we had just been conditionally disabling the cmake finding of these 3 libraries to do it, and will switch over to using the new option then.

re-add accidentally removed required includes to FindMySQLe.cmake

updated for the case when mysqle is shipped with mariadb

Works for me!

scrobble it does 👍

Waiting for @wbauer to confirm it works as well - btw, does openSUSE still provide mysqle to test -DWITH_MYSQL_EMBEDDED=on?

In D28165#634067, @vmatare wrote:

Please test again, and thanks a lot for the review so far.

In D28165#633145, @wbauer wrote:

The content of my /usr/include/mysql/mysql_version.h (which is just a symlink to mariadb_version.h):

In D24805#633417, @hindenburg wrote:

Is there really any reason to fully revert? Why not just make it not the default per the above comment.

Same check should imo also be added for phononsettings. This is shipped with phonon package, but kmix only indirectly pulls in that via knotifications, where it is optional.

Confirming successful build here as well.

That diff now looks like a diff in top of your previous diff.

reverting after the original change did not go through review anyway.

And here we have a sandbox error: https://forums.gentoo.org/viewtopic-t-1110252.html

This change should have raised minimum KACCOUNTS_MIN_VERSION as it breaks with 19.12.3 and below.

In D26306#585587, @asturmlechner wrote:

Thanks for that, works for me!

What consequence is there for using minuet with fluidsynth-2 without this patch? So far it had built and linked to it without issues.

Thanks for that, works for me!

In D26050#583007, @ognarb wrote:

In D26050#582981, @asturmlechner wrote:

In D26050#578862, @anthonyfieroni wrote:

Maybe we can drop 62, but not 72.

If this is for master I don't see why we could keep anything pre-0.79. Distros already have to patch the hell out of 3.1.0 which has led to diverging code because several changes are conflicting. That one needs a point release to settle that.

And meanwhile master needs a fix for 0.83 too.

What about releasing Calligra as a part of the Release Service (a release every 4 months with some Bugfix release between)? The latest release was almost 2 years ago, and we have more than 500 commits in master.

In D26050#578862, @anthonyfieroni wrote:

Maybe we can drop 62, but not 72.

If this is acceptable I would not limit it to !LINUX; minimum dependencies may as well be desirable for those users.

What do we do here then going further, am I supposed to apply a downstream patch for konsole 19.12.x?

Is it really necessary to split KF5 dependencies in two parts here? I don't see this being done anywhere else, KF5_MIN_VERSION is always global. Likewise, the Qt5 line change is simply unnecessary.

That fix is already in master: https://phabricator.kde.org/R375:90ebe8e37b8699ef26c10fb2928cf673cad04a06

In D24805#550880, @hiangel wrote:

Do you have an example where having ccache on by default would cause a problem? When I asked what makes Konsole different from Mesa, libinput, etc, I was expecting to see an example.

ccache is nothing new, you can imagine it will be a hot topic on a source-based distribution, so we have years of experience with it. Issues are mostly sandbox violations as are typically used for packaging builds, but also build errors, so much that on bug reports it is regularly demanded to rebuild without ccache.

In D24805#550829, @hiangel wrote:

E.g. recently I created this task.

In D24805#550829, @hiangel wrote:

And packaging is disabling it.

Ok, so, I think there's some confusion between us. You're talking about "packaging" however here you want to make a commit into Konsole sources, i.e. to have influence on both packaging and building.

In D24805#550813, @hiangel wrote:

Basically, any Meson-based project has ccache autodetection.

In D24805#550794, @hiangel wrote:

ccache should never be enabled by default for packaging builds.

Why?

In D22210#547637, @davidedmundson wrote:

1. A period where both X and wayland say "Plasma".
   
   1 is super super bad. Which means we need to patch one of the Plasma entries first. I would block this happening.

It is the status quo for everyone using a DM other than SDDM on <unaware-distro>, so any decision here is going to be an improvement.

Switch back to Plasma (Wayland)

In D22210#547621, @ngraham wrote:

"Plasma on Xorg" is a yucky string to display to users.

well, it is what Gnome apparently does, so at least it would have lead to some uniformity. But no strong opinion here.

In D22210#547621, @ngraham wrote:

IMO it would be preferable to display "Plasma on Wayland (Wayland) for a limited period of time while the Plasma and SDDM versions aren't yet aligned.

...except I wasn't sure if you could call Plasma running "on" Wayland in the same sense as it is running "on" Xorg.

copy that for Gentoo.