See inline comments

Robert

On Sun, 21 Oct 2018 20:11:14 +0200, Karl Heinz Marbaise <khmarba...@gmx.de> wrote:

Hi to all Devs,

* https://issues.apache.org/jira/browse/MNG-6492 review ok

   To be honstest I see only a question about the details of it
   by Michael Osipov. Nor do I see any implemented
   code changes etc.

   From my point of view no reason to postpone the release.

   In the end: -1 postpone to next release

IIRC some addAll-methods have been replaced with single add-methods, because they could cause some exception (I guess ConcurrentModificationException) when elements were added while some other process was trying to read. Assuming we all know about the addAll I guess there was a good reason to write it like this, little bit of shame there's no comment...


* https://issues.apache.org/jira/browse/MNG-6490 Karl, can You review?

   I would like to have the opinion of other devs as well.

   From my point view Ok to merge: +1 from me.

It is not circular, but awkward as the dependency is part of the build of this project. If it used to work, well, I guess we should accept it. It doesn't really hurt, but I think there's a better solution possible.


* https://issues.apache.org/jira/browse/MNG-6069 also for review

   Unfortunately the IT's tell us there are issues also
   on the fix/MNG-6096 branch which tells me it is not
   that simple as expected.

   From my point of view: -1 postpone to next release

Agree


* https://issues.apache.org/jira/browse/MNG-5693 code for review
   + few ITs failed to changed output [2] ITs needs to be adjusted.

   @Sylwester: Can you can create an appropriate branch in IT's
               so we check if eveything works as expected.

Would have been a nice to have, we can do this later.


* https://issues.apache.org/jira/browse/MNG-6481

   From my point of view -1 cause for the release not critical.

   This means to postpone it to the next release.

   More important if core is working fine with JDK11 which is
   the case.


Agree, support Java 11 runtime is much more important

* quote from Sylwester: I also verified release with Synk.io
   and we have only one report to upgrade Guava [4] to
   version 24.1.1 or above (now with Guice 4.2 we use 23.6)

   I have created MNG-6497 for this.
   See what IT's etc. will tell us.

   If all IT's are Ok I will VOTE: +1 for that.

   If I read the description I would say it is not really
   important for Maven Core cause as far as I know we don't
   do any serialization etc.


Kind regards
Karl Heinz Marbaise

[4]: https://app.snyk.io/vuln/SNYK-JAVA-COMGOOGLEGUAVA-32236

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org

Reply via email to