[MNG-7727] Remove DefaultMavenProfilesBuilder

[gnodet]

JIRA issue: https://issues.apache.org/jira/browse/MNG-7727

[michael-o]

Does this make sense to fiddle with compat although we want to get rid of it?

[elharo]

I'm not sure I understand the purpose of this. This also is singificant enough to deserve a JIRA issue.

[gnodet]

I'm not sure I understand the purpose of this. This also is singificant enough to deserve a JIRA issue.

The purpose is to:

• remove references to plexus-utils
• fix encoding problems with BOM parsing

But given, the code is not used afaik, I'm not sure it's worth fixing it, so removing also solves the problem.

[gnodet]

Does this make sense to fiddle with compat although we want to get rid of it?

Well, we want to get rid of it, but I think that won't be tomorrow. So in the mean time....

[michael-o]

Does this make sense to fiddle with compat although we want to get rid of it?

Well, we want to get rid of it, but I think that won't be tomorrow. So in the mean time....

I see, then a JIRA issue is imperative

[gnodet]

@elharo would you prefer removing the DefaultMavenProfilesBuilder implementation class completely ?

[elharo]

When did we deprecate it? If we can simply remove it in 4.0.0, that solves all problems. I will raise a question on the dev mailing list about whether we should start ripping out the deprecated methods and classes in 4.0.0.

I do not think we should replace it with a broken version that simply returns null. Compile-time errors are better than runtime errors.

[rmannibucau]

@elharo it already returns null in 99% of the cases so guess this is not a behavior change

[slachiewicz]

maybe backport it to 3.9?

[gnodet]

maybe backport it to 3.9?

For what benefit ?

[slachiewicz]

maybe backport it to 3.9?

For what benefit ?

to receive faster feedback from end users

[elharo]

LGTM. I'm happy for these classes to be deleted.