Looks good only except I think I have missed some things in the actual Mk/Uses/ruby.mk. We have both <none> and none that is no arguments and none as an argument. I will add these in an upcoming commit after you wordsmith what I have tried to describe here.

However I have added the relevant lines inline. Can you just make sure the wordings are correct?

documentation/content/en/books/porters-handbook/uses/_index.adoc
1720	We have to add `none` as an argument too. As it's used.
1722	(none) states that only `RUN_DEPENDS` is added for the ports and relevant variables are loaded.
1725	I think we should also mention that this is the default.
1727	`none` states that no dependencies are added to the port but loads the required variables. This is required for the package:lang/ruby[] ports itself only.

Add bofh@'s suggestions.

Harbormaster completed remote builds in B49314: Diff 116030.Jan 29 2023, 6:17 PM

In D38121#869097, @bofh wrote:

Looks good only except I think I have missed some things in the actual Mk/Uses/ruby.mk. We have both <none> and none that is no arguments and none as an argument. I will add these in an upcoming commit after you wordsmith what I have tried to describe here.

Excellent, thanks for the review.

If we are not too late it could be nice to change none (as parameter) for something else. I can't come up with a name right now, but nodep might be one, or maybe noselfdep.

However I have added the relevant lines inline. Can you just make sure the wordings are correct?

documentation/content/en/books/porters-handbook/uses/_index.adoc
1727	I omitted the part for the variables since it is assumed that every other option (run, build, etc) will load the necessary variables if needed too.

In D38121#869472, @fernape wrote:

In D38121#869097, @bofh wrote:

Looks good only except I think I have missed some things in the actual Mk/Uses/ruby.mk. We have both <none> and none that is no arguments and none as an argument. I will add these in an upcoming commit after you wordsmith what I have tried to describe here.

Excellent, thanks for the review.

If we are not too late it could be nice to change none (as parameter) for something else. I can't come up with a name right now, but nodep might be one, or maybe noselfdep.

I had a discussion with the portmgr@ and based on our discussion and further searching for needles in the haystack please do not mention none as an argument here. Currently there is no consumer which is using USES=ruby:none so we will get rid of this in my next commit.

Other than that it's perfectly fine to ship it.

However I have added the relevant lines inline. Can you just make sure the wordings are correct?

In D38121#869988, @bofh wrote:

In D38121#869472, @fernape wrote:

In D38121#869097, @bofh wrote:

Looks good only except I think I have missed some things in the actual Mk/Uses/ruby.mk. We have both <none> and none that is no arguments and none as an argument. I will add these in an upcoming commit after you wordsmith what I have tried to describe here.

Excellent, thanks for the review.

If we are not too late it could be nice to change none (as parameter) for something else. I can't come up with a name right now, but nodep might be one, or maybe noselfdep.

I had a discussion with the portmgr@ and based on our discussion and further searching for needles in the haystack please do not mention none as an argument here. Currently there is no consumer which is using USES=ruby:none so we will get rid of this in my next commit.

Other than that it's perfectly fine to ship it.