releases 2025-11-17

https://github.com/polyfy/polylith/releases/tag/v0.3.31: A tool used to develop Polylith based architectures in Clojure and ClojureScript: • Fix Java 21 compatibility by upgrading core.rrb-vector to 0.2.0, issue https://github.com/polyfy/polylith/issues/579 • Handle invalid keywords correctly when exporting workspaces to disk, issue https://github.com/polyfy/polylith/issues/581 • Replace http://badgen.net with http://shields.io (on-demand generated icons), to fix the icons on the start page (readme.md), issue https://github.com/polyfy/polylith/issues/583 • New video: Gerben Dekker https://www.youtube.com/watch?v=k49496sH9aw how you can ship faster and collaborate better, using Polylith and other tools (in Python). He starts talking about Polylith at 9:20.

3

What's the issue with clojure.rrb vector with Java 21? babashka is on 0.1.2 and I've never noticed any issue

I will upgrade of course, but would be nice to understand the issue

I couldn't reproduce the problem myself, so asked Cursor, and it said this was a know problem (of course, I can't trust that, but I had no other idea). The reporter hasn't responded for two weeks, so I don't really know if it solves his problem, but I decided to ship it anyway (he may come back later)!

I can't find anything in the changelogs of rrb vector about JVM 21

It could be that it has no effect at all, so maybe just wait to see if the guy that reported it will come back. There is a stack trace in issue https://github.com/polyfy/polylith/issues/579, you can take a quick look at it (if you feel like it) and you might immediately see what the problem is (as I said, I haven't been able to reproduce it). But I'm not asking you to spend time on this of course!

SequencedCollection isn't used in rrb vector

so I don't understand how he could have gotten that stacktrace

it could be that he's running a fork? maybe ask about this dependencies

I'm on my phone so I have not checked that issue but that SequencedCollection issue only occurs if you compile on jdk22+ and try to run that compiled code on an earlier jdk, yes?

(And therefore nothing to do with rrb vector per se)

I think you're right, core.rrb-vector probably has nothing to do with the error, it just happened to be the first library to use a sequence collection.