Le mardi 27 février 2024, 18:33:43 CET sebb a écrit : > On Tue, 27 Feb 2024 at 17:01, Benjamin Marwell <bmarw...@apache.org> wrote: > > > Compiling for Java 8 with > > > Java 17 -release 8 is not the same as compiling with javac from JDK 8. > > > They do not produce the same byte code. +1 that's a fact
> > > There is a need to compile *with* JDK 8, not just compile *for* JDK 8. I'd also add: do you also expect to run the unit and integration tests with JDK 8? could running be decoupled from compiling? > > > > And when would that be needed? I'd say it's a question of taste/aversion to risk: I personally fully trust JDK 17 --release 8 to not produce anything wrong against JDK 8, but some may not trust That's why I personally would accept building with JDK 17 --release 8 but would like to run UT/ITs with JDK 8 > > Isn't that needed for reproducible builds? simply no one concrete example from so many, just picking a random one: plexus-utils, that targets Java 8 bytecode https://github.com/jvm-repo-rebuild/reproducible-central/blob/master/content/org/codehaus/plexus/plexus-utils/README.md as you can see, some releases were done with JDK 11, some with JDK 17, some on *nix and others on Windows Each is reproducible, even if the required environment to get the same binaries as the reference published to Maven Central for each is different while at it, I'll add: this does not mean that for each release, any environment different from the reference one gives a bad output; it just gives a different binary that is expected to be functionnally equivalent = what we blindly trusted before we checked binaries --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org For additional commands, e-mail: dev-h...@maven.apache.org
