You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@maven.apache.org by Romain Manni-Bucau <rm...@gmail.com> on 2021/10/12 06:11:45 UTC
Follow up of MNG-6354 (cdi-api)
Hi all,
Michael wonders if we drop cdi-api from maven 3.8.x in
https://issues.apache.org/jira/browse/MNG-6354.
The topic is worth it since it creates several issues and some quite hidden
by mojo.
The only expected impact is to not provide @Typed anymore and have to
replace it by an alternative. If not done the plugin will run but an
injection can become ambiguous but not sure how much real case we can get
out there - none in maven itself but maybe in a few mojo.
So questio is: do we backport it or not with this very low risk or not.
I'm +0.99 to do it due to the risk but the quite common benefit.
Romain Manni-Bucau
@rmannibucau <https://twitter.com/rmannibucau> | Blog
<https://rmannibucau.metawerx.net/> | Old Blog
<http://rmannibucau.wordpress.com> | Github <https://github.com/rmannibucau> |
LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book
<https://www.packtpub.com/application-development/java-ee-8-high-performance>
Re: Follow up of MNG-6354 (cdi-api)
Posted by Romain Manni-Bucau <rm...@gmail.com>.
Le dim. 17 oct. 2021 à 13:36, Tamás Cservenák <ta...@cservenak.net> a
écrit :
> Agreee, maven stuff should use sisu @Typed annotation as core does.
>
> T
>
> On Sun, Oct 17, 2021, 12:41 Michael Osipov <mi...@apache.org> wrote:
>
> > Am 2021-10-12 um 08:11 schrieb Romain Manni-Bucau:
> > > Hi all,
> > >
> > > Michael wonders if we drop cdi-api from maven 3.8.x in
> > > https://issues.apache.org/jira/browse/MNG-6354.
> > > The topic is worth it since it creates several issues and some quite
> > hidden
> > > by mojo.
> > > The only expected impact is to not provide @Typed anymore and have to
> > > replace it by an alternative. If not done the plugin will run but an
> > > injection can become ambiguous but not sure how much real case we can
> get
> > > out there - none in maven itself but maybe in a few mojo.
> > >
> > > So questio is: do we backport it or not with this very low risk or not.
> > >
> > > I'm +0.99 to do it due to the risk but the quite common benefit.
> >
> > As far as I understand the Sisu/Guice version of @Typed must be used
> > right? I think we can make this one for 4.0.0. For 3.8.x, am not sure,
> > for 3.9.0 very likely. The most important thing is to document this in
> > the release notes what people need to do.
> >
> > Questions:
> > * Is the fix in Maven trivial?
>
Yes, same as for v4.
> * Is the fix in components and plugins trivial and backwards compatible?
>
Trivial yes, backward compat no until we do more and complicated stuff (for
a few usage).
This is why for 3.8 it is not obvious if we include it IMHO.
Think delaying for 3.9 can be a wise proposal so I join you on this one.
>
> > M
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> > For additional commands, e-mail: dev-help@maven.apache.org
> >
> >
>
Re: Follow up of MNG-6354 (cdi-api)
Posted by Tamás Cservenák <ta...@cservenak.net>.
Agreee, maven stuff should use sisu @Typed annotation as core does.
T
On Sun, Oct 17, 2021, 12:41 Michael Osipov <mi...@apache.org> wrote:
> Am 2021-10-12 um 08:11 schrieb Romain Manni-Bucau:
> > Hi all,
> >
> > Michael wonders if we drop cdi-api from maven 3.8.x in
> > https://issues.apache.org/jira/browse/MNG-6354.
> > The topic is worth it since it creates several issues and some quite
> hidden
> > by mojo.
> > The only expected impact is to not provide @Typed anymore and have to
> > replace it by an alternative. If not done the plugin will run but an
> > injection can become ambiguous but not sure how much real case we can get
> > out there - none in maven itself but maybe in a few mojo.
> >
> > So questio is: do we backport it or not with this very low risk or not.
> >
> > I'm +0.99 to do it due to the risk but the quite common benefit.
>
> As far as I understand the Sisu/Guice version of @Typed must be used
> right? I think we can make this one for 4.0.0. For 3.8.x, am not sure,
> for 3.9.0 very likely. The most important thing is to document this in
> the release notes what people need to do.
>
> Questions:
> * Is the fix in Maven trivial?
> * Is the fix in components and plugins trivial and backwards compatible?
>
> M
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
>
Re: Follow up of MNG-6354 (cdi-api)
Posted by Michael Osipov <mi...@apache.org>.
Am 2021-10-12 um 08:11 schrieb Romain Manni-Bucau:
> Hi all,
>
> Michael wonders if we drop cdi-api from maven 3.8.x in
> https://issues.apache.org/jira/browse/MNG-6354.
> The topic is worth it since it creates several issues and some quite hidden
> by mojo.
> The only expected impact is to not provide @Typed anymore and have to
> replace it by an alternative. If not done the plugin will run but an
> injection can become ambiguous but not sure how much real case we can get
> out there - none in maven itself but maybe in a few mojo.
>
> So questio is: do we backport it or not with this very low risk or not.
>
> I'm +0.99 to do it due to the risk but the quite common benefit.
As far as I understand the Sisu/Guice version of @Typed must be used
right? I think we can make this one for 4.0.0. For 3.8.x, am not sure,
for 3.9.0 very likely. The most important thing is to document this in
the release notes what people need to do.
Questions:
* Is the fix in Maven trivial?
* Is the fix in components and plugins trivial and backwards compatible?
M
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org