Guillaume Nodet

> [@sebastian-bayerl-enmacc](https://github.com/sebastian-bayerl-enmacc) can you try with `-Denforcer.ignoreCache` ? Also try with `-X` as somewhere some hint may be logged, like this one https://github.com/apache/maven-enforcer/blob/d4be5532fa112460a30c462d2ebb94fc64278075/maven-enforcer-plugin/src/main/java/org/apache/maven/plugins/enforcer/internal/EnforcerRuleCache.java#L62 > > [@gnodet](https://github.com/gnodet) This looks sus: https://github.com/apache/maven-enforcer/blob/d4be5532fa112460a30c462d2ebb94fc64278075/maven-enforcer-plugin/src/main/java/org/apache/maven/plugins/enforcer/internal/EnforcerRuleCache.java#L72...

I think we're missing a `mvnsh` integration. This can be done by adding a command to the builtin command registry: https://github.com/apache/maven/blob/bcf98382a62f2445a3c65fec5a9c680424414e18/impl/maven-cli/src/main/java/org/apache/maven/cling/invoker/mvnsh/builtin/BuiltinShellCommandRegistryFactory.java#L94

It looks like a duplicate of #11389 ?

> > @gnodet ping > > > > > > @tballison did you use `mvnup` to "upgrade"? > > > > Y. I tried to follow directions. :rofl: > >...

> Should we maybe at the same time archive/EOL plexus-classworlds? > Also, is this a fork of it? Or rewrite? As if some third-party may use plexus-classworlds, is this new...

> I've read at maven.apache.org/whatsnewinmaven4.html, that the consumer POM only contains information about `compile` and `runtime` scoped dependencies. This need to be update for rc-5 which does not flatten the...

> Having `directDependencyStrategy` at dependency level looks strange. > > WDYT of something like: > > * `dependencyVersionType`: `strict` (select exactly this version), `require` (accept this or later), `reject` >...

> [@gnodet](https://github.com/gnodet) , I have read the proposal again, and I do not see user story in it. In other words, I do not see what would really change for...

Actually, I think this comes down to just a different VersionSelector, something like `flexible-highest` instead of `highest`. That may be sufficient.

@vlsi I'm thinking that we just need a new strategy in the resolver. We recently introduced `highest` in the resolver, but that one does not take direct dependencies into account,...