cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864547699
ITs look good.
And thinking more about this, this should be maybe just a boolean flag?
`classic(transitive)`? As I would leave out completely the "transitive" and the
"default" as t
cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864470425
I see no difference to what happens if you use two libs that declare
different version on some artifact, really. And depMgt (_unlike BOM imports_)
let it all to you to sort it out in "maven
rmannibucau commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864468358
@cstamas right, now consider 2 lib authors for your project (3rd maven build
to make it clear) and also maven 3 consumers (let's ignore 2, agree), it
becomes a mess for us :(
--
This
cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864465206
IMO, this change -- if goes in -- will just make POMs _simpler_ and not
_more complex_. Essentially what this makes possible is that lib author (whos
lib is used somewhere deep in your proj
rmannibucau commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864461670
@cstamas I understand that and ultimately I hope we reach your proposal I do
like, but I don't want to be at the cost of making builds more complex and user
most lost, this is how I end
cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864458580
And as regarding "blowing", lets see what ITs has to say. Also we could
maybe add IT for https://issues.apache.org/jira/browse/MRESOLVER-235
--
This is an automated message from the Apach
cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864456808
This setting cannot be per dependency, at least resolver does not support
that. The depMgr is set per session. And what current PR changes is really only
this: are depMgt sections obeyed on
rmannibucau commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864449457
What I worry about it is the fact we are moving the current issues a step
further but I don't understand yet how it does fix anything, this kind of
configuration looks per dependency, s
gnodet commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864442381
> Hi @cstamas , what is the plan to identify mvn 4 consumed artifacts to
switch the impl, using the mvn4 specific pom metadata file? in current state it
breaks probably too easily IMHO and I
cstamas commented on PR #1357:
URL: https://github.com/apache/maven/pull/1357#issuecomment-1864436787
@rmannibucau this PR is draft on purpose. Actually, I'd say "consider all
resolver related Maven PRs draft" I do, as am just throwing ideas or missing
Maven features, to improve use of (so
10 matches
Mail list logo
