resurect lua-language-server

[dave_freedave.net]

Summary

I guess technically it was building but it would SIGBUS. The fix for that in bee.lua has been merged upstream. Far enough for both the luamake and lua-language-server which are not the same versions.

Test Plan

There is a known flaky test for lua-language-server which I have hit, fairly often actually. But more than half the time make test is going to pass all unit tests.

[adamw]

So, what you have at the end there (the .if ${OSVERSION} < 15 block) feels like an antipattern, but I can't think of any better way to do it. I've tried out a couple alternatives and they just made it more complicated. At the end of the day, it's a valid approach and it works, and also it only needs to be in there until 14 goes EOL (now that 15 is out, 14 has a defined death day).

I say, fix those small things, do a final sweep to make sure that the port is where you want it, and then let's get it committed!

devel/lua-language-server/Makefile

15

This block of comments should be removed; they are helpful locally, but not in the ports tree. Plus, 13 is going away in a matter of months.

22

These two (GH_PROJECT / GH_TAGNAME) are defaults and should be removed.

[dave_freedave.net]

adopt review comments

[dave_freedave.net]

Not that I doubted you, but I did rebuild after removing GH_PROJECT and GH_TAGNAME.

[dave_freedave.net]

devel/Makefile

1630 ↗

(On Diff #167499)

I see I didn't sort this right, should I update diff or is that fixable on merge?

[adamw]

devel/Makefile

1630 ↗

(On Diff #167499)

So first off, I loved seeing it there, as it shows you've put a lot of work into understanding how the ports build system works. Great job!

But, it's actually not helpful to have in a patch. Category Makefiles get out-of-date pretty quickly, and I'd much rather yyp it in neovim than have it fail in patch(1) and need to recover it by hand anyway.

So, while you absolutely did the right thing including it here, I'd actually appreciate if you could remove it from the patch entirely 😄

[dave_freedave.net]

removed change to ports/devel/Makefile

[dave_freedave.net]

devel/lua-language-server/Makefile

68–72

it looks like deve/glib20 relied upon the existence of the header file. That also doesn't appear to need bsd.port.[pre|post].mk and can use bsd.port.mk like normal.

Not looking for ways to slow down commit, but consistency helps everybody. Thoughts?

[dave_freedave.net]

Oh relying upon the header file would also help in case inotify gets backported to stable/14 in the future.