Fork me on GitHub
#cider
<
2020-02-29
>
benedek11:02:25

thanks for the clj-refactor release @bozhidar

bozhidar12:02:44

@benedek Youโ€™re welcome! ๐Ÿ™‚ I had a couple of hours to kill today and I thought it was high time we released something, as the previous release was from summer 2018.

vemv19:02:19

Hi ๐Ÿ‘‹ I make some developer tooling, which justifies placing cider/cider-nrepl in project.clj in my CI, even if I don't directly use it. I noticed that bumping cider/cider-nrepl "0.21.1" -> cider/cider-nrepl "0.24.0" breaks some clojure.test tests, only on JDK 11. Reason being indiscernible to me (at least at first) Does this sound familiar?

vemv19:02:19

Hi ๐Ÿ‘‹ I make some developer tooling, which justifies placing cider/cider-nrepl in project.clj in my CI, even if I don't directly use it. I noticed that bumping cider/cider-nrepl "0.21.1" -> cider/cider-nrepl "0.24.0" breaks some clojure.test tests, only on JDK 11. Reason being indiscernible to me (at least at first) Does this sound familiar?

bozhidar09:03:41

I saw your ticket, but didnโ€™t have time to comment on it yet. Iโ€™m guessing itโ€™s related to use moving away from c.j.classpath to orchard.java.classpath.

vemv09:03:42

Yes I took a look at it. I guess a custom classloader is set to the current thread, which interferes with c.j.p's assumptions Might be considered 33% a bug in c.j.c, 33% in orchard, and 33% in applications like mine which invoke (classpath-directories) and take its result at face value