Update math/libqalculate to 2.2.1

[jhale]

Summary

Update math/libqalculate to 2.2.1 and math/qalculate to 2.2.1.
Several kde@ ports needed CMake module adjustments to find the newer library.
I don't want to step on any toes in regards to https://reviews.freebsd.org/D14413,
so if this needs to wait until after that has landed, that's fine.

Test Plan

Standard poudriere build tests completed.
10.4-i386 - all pass
11.1-amd64 - all pass
12.0-CURRENT-amd64 - all pass

[tcberner]

Hi there

Thanks, for the offer to wait. I'll gladly take it :)

Mfg Tobias

[tcberner]

@jhale D14413 has landed.

[jhale]

Update against current ports tree

[tcberner]

math/libqalculate/Makefile

7 ↗

(On Diff #39615)

^DISTVERSIONPREFIX=v + USE_GITHUB

math/qalculate/Makefile

7 ↗

(On Diff #39615)

^ instead of setting MASTER_SITES and DISTNAME this should be

DISTVERSIONPREFIX=v
[...]
USE_GITHUB=yes

[jhale]

math/qalculate/Makefile

7 ↗

(On Diff #39615)

This pulls in the full release tarball, so we don't have to run autoreconf and a bunch of other stuff.

[mat]

math/libqalculate/Makefile

7 ↗

(On Diff #39615)

No, this is NOT the same thing. What you are proposing is using an extract from the git repository, which, in this case, is wrong, as there is a released tarball.

Releases tarballs are hand made, usually they have all the configure things taken care of, and you do not need to use autoreconf, also, they often have the documentation already generated so that it is not required to depend on doxygen and stuff.

We should not use git extracts when releases are available.

[tcberner]

math/libqalculate/Makefile

7 ↗

(On Diff #39615)

Oh, sorry, I was under the wrong impression that there was magic involved :(