Conversation
|
|
||
| ThisBuild / githubWorkflowJavaVersions := Seq( | ||
| JavaSpec.temurin("8"), | ||
| // Java 17 first: publish job uses the head of this list when downloading staged artifacts; sbt 2 (Scala 3 axis) needs 17+. |
There was a problem hiding this comment.
FWIW I'm proposing to make Java 17 the minimum in #205 but haven't dared pull the trigger yet.
build.sbt
Outdated
| ThisBuild / scmInfo := Project.scmInfo | ||
| ThisBuild / description := Project.description | ||
|
|
||
| def sbtVersionForPlugin(scalaBinary: String): String = |
There was a problem hiding this comment.
maybe name this clearer to reflect that this is the version we run the scripted tests with?
There was a problem hiding this comment.
This one is not only for the scripted tests but also for the plugin itself - it is used twice in this file.
There was a problem hiding this comment.
ah, right, I didn't notice. In that case are we sure we want these to be the same? I'd say we'd want to build with the "latest" sbt but test against the "minimum supported".
There was a problem hiding this comment.
Ok, fixed this. Testing against minimal sbt 1.6.0 as per README, and 2.0.0-RC11 for sbt 2. I don't think it's reasonable to specify a minimal supported version for sbt 2 to be an RC version; we should wait until 2.0.0 and set the minimal to it.
.github/workflows/ci.yml
Outdated
| tar xf targets.tar | ||
| rm targets.tar | ||
|
|
||
| - name: Download target directories (3.8.2) |
There was a problem hiding this comment.
using the cache is an attack vector and this job is also used for releases - does it really make a big difference? Does this come in via githubWorkflowGenerate? Could we disable it?
There was a problem hiding this comment.
This behaviour was already in the build before this PR: see line https://github.com/sbt/sbt-sbom/pull/219/changes#diff-b803fcb7f17ed9235f1e5cb1fcd2f5d3b2838429d4368ae4c57ce4436577f03fR158 which does the same for Scala 2.12.
We could disable this behavior with ThisBuild / githubWorkflowArtifactUpload := false which would make the publish job run slower but will remove caching. WDYT?
There was a problem hiding this comment.
Ah, gotcha. I think that would be good - we're not so active that it matters a lot for us anyway
There was a problem hiding this comment.
Fixed, now the caching is disabled.
2 participants
No description provided.