Akka 2.7.0 is safe to port as it's Apache 2.0 too; we should not diverge much, many changes in Akka are actually fine. I will try to port the testkit-related things this weekend. I will spend all my weekend working on the stream things, because I want the stream to be great, In case someone is building an MCP Server/Client with Pekko Stream.
And ZSTD is really needed and great, as the Pekko 1.3.0 is the last one we support Java 8 , so I really suggest we contain all the features in Akka 2.7.0 何品 PJ Fanning <[email protected]> 于2025年11月7日周五 02:29写道： > The Akka BSL license releases the changes as Apache license 3 years > after a release. Akka 2.7.0 was released a little over 3 years ago. > Its changes are now available to be added to Pekko. > > On Thu, 6 Nov 2025 at 19:26, Brendan Doyle <[email protected]> wrote: > > > > Sorry if this is a dumb question, but isn’t porting anything from Akka > >2.6 > > a license / legal issue once it moved onto BSL? I thought the expectation > > was that anyone working on Pekko shouldn’t be looking at Akka to not > derive > > any intellectual work from the project post 2.6. > > > > On Thu, Nov 6, 2025 at 10:22 AM Matthew de Detrich <[email protected] > > > > wrote: > > > > > In regards to the zstd-compression work, I was waiting for a response > to > > > https://github.com/apache/pekko/pull/2409#issuecomment-3469014680 but > > > since > > > none came through I will try and implement the backpressure friendly > buffer > > > accumulation version that Arnout suggested with a time box. If I don't > > > manage to complete it within that time box then I will just do the dumb > > > simple version, as in any case the most important thing is to > > > unblock/create an API for pekko-http. > > > > > > I understand the concerns regarding having zstd-jni as part of the main > > > pekko-streams artifact but doing it this way would add a lot of > complexity, > > > time and effort to the solution for what I see not that much gain (it > might > > > be a different story if there was multiple production ready competing > > > suitable implementations of zstd on JVM but there isn't). Hopefully at > some > > > point zstd will actually get added to JVM in the future, at which > point we > > > can use it without even having to mess around with any artifacts/API > with a > > > future Pekko release. > > > > > > On Thu, Nov 6, 2025 at 1:37 PM PJ Fanning <[email protected]> > wrote: > > > > > > > I don't think that we should port all the Akka changes. I'm happy > > > > enough with what we have already. If someone really wants a > particular > > > > change, they can volunteer to add it. > > > > > > > > On Thu, 6 Nov 2025 at 13:28, kerr <[email protected]> wrote: > > > > > > > > > > I think we should port all features of Akka 2.7.0 to Pekko 1.3.0, > but > > > > I'm > > > > > quite busy at work right now... > > > > > The stream operators I want to add is already. > > > > > it was tracked here: > > > > > > > > > > https://github.com/apache/pekko/issues/2390 > > > > > > > > > > 何品 > > > > > > > > > > > > > > > PJ Fanning <[email protected]> 于2025年11月6日周四 02:33写道： > > > > > > > > > > > Hi everyone, > > > > > > > > > > > > We have a few changes awaiting release. > > > > > > * https://github.com/apache/pekko/milestone/21?closed=1 > > > > > > > > > > > > There are a couple of PRs under discussion which we should > hopefully > > > > > > get into the 1.3.x branch shortly. Does anyone have any other > items > > > > > > that they want to talk about getting into 1.3.0? > > > > > > > > > > > > If there are no objections to doing a release, I can act as > release > > > > > > manager. > > > > > > If anyone else wants to take on the role, please let us know. > > > > > > > > > > > > PJ > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > To unsubscribe, e-mail: [email protected] > > > > > > For additional commands, e-mail: [email protected] > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > To unsubscribe, e-mail: [email protected] > > > > For additional commands, e-mail: [email protected] > > > > > > > > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
