https://roy.marples.name/blog/dhcpcd-8-0-0-released
https://roy.marples.name/blog/dhcpcd-8-0-1-released

https://roy.marples.name/blog/dhcpcd-8-0-2-released

> I don't anticipate any more releases for a while as this is looking
really good now!

ssh-keysign is used for host-based authentication, and is designed to be used
as SUID-root program. OpenSSH defaults to referencing it from libexec, which
cannot be made SUID in Nix.

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/cmake-cursesui-qt5ui/versions

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/glib/versions

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/jsoncpp/versions

jsoncpp: 1.9.0 -> 1.9.1

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/phonon-qt5/versions

Because we're using plain buildtype these have to be
passed manually.

See: https://gitlab.gnome.org/GNOME/gtk/blob/3.24.10/meson.build#L59

With autotools the mapping for the the options to the defines was:

yes:     G_ENABLE_DEBUG G_ENABLE_CONSISTENCY_CHECKS
minimum: G_ENABLE_DEBUG G_DISABLE_CAST_CHECKS
no:      G_DISABLE_CAST_CHECKS G_DISABLE_ASSERT G_DISABLE_CHECKS

So we're passing the exact ones that would've been used for minimum.

Additionally it isn't a good idea to pass the equivalents used for "no" 
as it eliminates G_ENABLE_DEBUG which disables pre-condition checks and 
assertions. The actual option only existed to serve people who needed a 
specific build of GTK for very specific environments. And now they are
much better served with meson's plain buildtype and figuring out what to
pass themselves.

This is what would have been passed before with the release
buildtype.

See: https://gitlab.gnome.org/GNOME/glib/blob/2.60.4/meson.build#L208

This is what would have been passed before with the release
buildtype.

See: https://gitlab.gnome.org/GNOME/gnome-settings-daemon/blob/GNOME_SETTINGS_DAEMON_3_32_1/meson.build#L73

Mirror c7c727f.

I fail to see where or for what it is useful for.

libsForQt5.phonon: 4.10.2 -> 4.10.3

glib,gtk: correct CLAGS for plain buildtype

glib: 2.60.4 -> 2.60.6

Clean up the expression and prepare for tests. Unfortunately, I could not get them pass.

Semi-automatic update generated by
https://github.com/ryantm/nixpkgs-update tools. This update was made
based on information from
https://repology.org/metapackage/vala/versions

enchant: 2.2.3 -> 2.2.4

https://gitlab.gnome.org/GNOME/vala/raw/0.40.16/NEWS

cmakeWithGui: 3.14.5 -> 3.15.1

This reverts commit bf39fc1.

This reverts commit c220e00.

vala_0_40: 0.40.15 -> 0.40.16

gnome3.vala: 0.44.5 -> 0.44.6

Replay Qt 5.12.3 version bump

Broke the build as the pname variable was used to create the URL
Introduced by 46420bb

kde: multiple staging fixes

There ver very many conflicts, basically all due to
name -> pname+version.  Fortunately, almost everything was auto-resolved
by kdiff3, and for now I just fixed up a couple evaluation problems,
as verified by the tarball job.  There might be some fallback to these
conflicts, but I believe it should be minimal.

Hydra nixpkgs: ?compare=1538299

Fortunately the issue was caught as a hash mismatch.

Thanks to Symphorien: #67426

Attributes: firefox-esr-60, tor-browser(-8-5), icecat
Thanks to Symphorien: #67426

There were several more conflicts from name -> pname+version;
all auto-solved by kdiff3, hopefully OK.

Hydra nixpkgs: ?compare=1538611

* upstream/master:
  vimPlugins: update (#67823)
  mattermost-desktop: 4.2.0 -> 4.2.3 (#67717)
  lightspark: init at 0.8.1 (#67425)
  anki: 2.1.11 -> 2.1.14 (#67738)
  nixos/{namecoind,bitcoind}: removing the altcoin prefix
  rust-cbindgen: 0.8.7 -> 0.9.0