hboutemy commented on PR #206:
URL: 
https://github.com/apache/maven-artifact-plugin/pull/206#issuecomment-4124369111

   I'm thinking at this very concrete example: 
https://github.com/jvm-repo-rebuild/reproducible-central/blob/master/content/io/cucumber/gherkin/README.md
   
   Where they use ranges, and:
   - sometimes I was lucky to rebuild early enough to get the same resolved 
version, like 39.0.0
   - sometimes I rebuilt too late, and I can't force Maven resolution to NOT 
use newer versions, like 37.0.1 and 38.0.0 (I added a manual comment in 
diffoscope output)
   
   in this case, when another version is resolved that the initial build, it 
has an impact on `.class` content
   But often it does not impact `.class`
   
   the more I write, the more I think we're not describing the same 
"Reproducible Builds" semantics
   It could sometimes impact CycloneDX files, because they list dependencies. 
Or it will impact `.war` or `shade` output or anything that really copies the 
dependencies.
   
   But in general, version ranges that give "un-stable over time" resolution is 
not impacting "Reproducible Builds" in terms of getting the same output
   And even if it impacts output, I'm sure that by configuring an extension 
that locks a little bit resolution, we could force the build to get the same 
output = what Reproducible Builds is about = checking that nothing has been 
cheated in the output
   
   
   that being said, what to put in `.buildcompare`, I confess I don't know: in 
fact, I don't know how to describe the issue in plain text even in this PR
   
   what would describe the gherkin case?
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to