Github user srowen commented on the issue:
https://github.com/apache/spark/pull/21610
Certainly. I'm saying that these are nearly the same thing, because the
binary release is, mostly, a collection of compiled Spark code and all its
transitive dependencies. This is why it's correct to start from that list. It's
not 100% what makes it into the release: some bundled code doesn't show up this
way (shaded stuff). It also bears some manual sense checking. But it appeared
to give the correct list of dependencies.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
