Christopher Tubbs

We could set up our own standalone cluster for a lot of tests, at pre-integration-test phase of the build lifecycle, and shut it down at post-integration-test. There was an attempt...

The standalone stuff that's there now is newer than that history. What was already removed was definitely a case of fix it or remove it. But I think we can...

When this is done, the key Range object could also be a improved with some similar API changes. Although, we might want to wait until we have a better ByteSequence...

Haven't re-reviewed yet, but just thought of a suggestion (I don't know if it's workable). Internally, this could use Guava Range classes. As long as they're not exposed to this...

It occurs to me that this entire class is basically just Guava's Range with Hadoop's Text as the generic parameter, `Range`. I don't think we should force users to use...

`fs.defaultFS` is only used for qualifying unqualified paths (paths without a filesystem specified). It is probably bad practice to rely on this mechanism when writing Hadoop client code (it's too...

> I'm wondering if we should just close this PR and punt on Java 21 source compatibility for now. Can just leave it open until we work through the issues....

First, I just want to mention I'm not sure this is the best venue for a discussion. This is probably better suited to the developer mailing list. But I'll respond...

I could probably start helping by providing some examples of alternatives, if you could highlight some specific pain points with the alternatives you've tried so far.

(this response was originally discussed with @ivakegg outside of this ticket; I'm sharing here, to summarize that conversation and to share my views with the wider community) > to be...