Replies: 5 comments 3 replies
-
Overall I agree with those goals. 2.0 would be a good time to get rid of deprecations. |
Beta Was this translation helpful? Give feedback.
-
I am totally OK on this, one major feature of Virtual thread support is done now. Release 1.2 and do some clean up can help further faster iterations |
Beta Was this translation helpful? Give feedback.
-
This sounds great.
I like this goal. We might need something to make sbt happy, though: with the following test project:
sbt now gives:
|
Beta Was this translation helpful? Give feedback.
-
There are some binary breaking changes that I have planned for Pekko core 2.0.x which would mean the Pekko ecosystem would also have to be rebuilt. Some are changes that would have ordinarily been done in the early days (i.e. Scala 3 inlining) which had to be reverted because it broke binary compatibility. There are other changes such as planned hooks for tracing/instrumentation which may require API changes (also propagate to Pekko-http) I would personally be wary of trying to signal and hence avoid rebuilding for 2.0.x for Pekko modules as it gives the wrong impression to users. It may also be impossible as some libraries in Pekko connectors need Java 17 and at that point you may as rebuild all of Pekko for 2.0.x. Building Pekko libraries for 2.0.x shouldn't be problem as we can just do milestone releases, it's just tedious. |
Beta Was this translation helpful? Give feedback.
-
I still think we are on track for doing a Pekko 1.2.0 release in a few weeks - maybe late in the month. With the main branch, we are developing 2.0.0. I think it could be 6-12 months before a full release but we could do a 2.0.0-M1 in a month or 2. So far, Java 17 has become the min and about half of the old Akka-era deprecations have been removed. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
We're in the process of releasing 1.2.0-M2. There are no major issues blocking us from basically closing 1.2 development and aim for a 1.2.0 release. I would suggest waiting a month to allow people to try out the milestone release.
If we agree with this plan, we could press on with creating a 1.2.x branch and backfitting any bug fixes to this branch so that they can appear in the 1.2.0 release or a subsequent 1.2.x patch release.
This would free up the main branch and allow us to start making some bigger changes with an eventual 2.0.0 release in mind.
The 2.0.0 changes could include:
We will continue to fix important bugs for Pekko 1 - so I would prefer if we don't entertain too many requests to not make changes based on some users having very conservative requirements for support for old versions of Java, Scala or other 3rd party libs.
I'd be interested in everyone's opinion on these topics.
Beta Was this translation helpful? Give feedback.
All reactions