databases/influxdb: Update to 1.6.2

[fernape]

Summary

Via PR 231232

Waiting for maintainer approval/timeout.

I also reordered some of the variable blocks and simplify the man pages installation

Test Plan

• portlint -AC OK (some false positives show up)
• poudriere builds for {10.4,11.1}{amd64,i386}, 12i386 OK
• run test in 10.4amd64 OK (basic testing like firing up the db server, connecting, listing the available databases and such)

[tcberner]

Looks good from my side.

[mat]

databases/influxdb/Makefile

92 ↗

(On Diff #47836)

This is already in USE_RC_SUBR, remove SUB_FILES.

[fernape]

Remove redundant SUB_FILES

[fernape]

Does this look good?

[tcberner]

Still looks good to me, yes :)

[mat]

head/databases/influxdb/Makefile

14

I wonder if someone is keeping track of those "needs go" ONLY_FOR_ARCHS, for when go, for instances, grows aarch64 support.

15

these architectures

[fernape]

head/databases/influxdb/Makefile

14

I suppose we'll need to grep the tree for ONLY_FOR_ARCHS.*go.* when that time comes

15

As it is now:

make
===>  influxdb-1.6.2 is only for armv6 armv7 i386, while you are running amd64
(reason: depends on go which is not supported in this architecture).

It makes sense in singular since we are referring to the arch the user is building on, right?

[mat]

head/databases/influxdb/Makefile

14

For ports where there is a reason about go, yeah, sure :-)

If it were me, I'd remove those as go is the limiting factor, technically, this works everywhere. In poudriere the ports will be skipped, and for other tools, or if building manually, the build will fail early because go won't build.

15

Ah, yes, I missed the IGNORE+= while you are running ${ARCH} while reading the code ^^