Steve Loughran

In #6686 I'm creating a new utils class for reflection access, nothing else. And proposing that all tests of the API use reflection to be really confident it works and...

ok, i'm going to add one more change, and this time we do go near that production code. how about logging a warning in the FileOutputCommitter constructor at Line 160?...

Like I said on the jira, I don't want this. It has the same scale issues encountered on abfs as #[6399](https://github.com/apache/hadoop/pull/6399) and #[6378](https://github.com/apache/hadoop/pull/6378), the same correctness problems on GCS as...

you can use env var resolution within a hadoop core-site file; which lets you at the values with defaults when unset. on locked down config loading (oozie etc) then only...

ok. if you look into _SUCCESS json from an s3a or the manifest committer, then the job id is one of the root attributes, as is the source there's a...

@shangxinli looking forward to your comments -anything you can do to test will be wonderful too!

@danielcweeks that's a good point about pluggability. 1. an interface/implementation split in parquet would line you up later to choose the back end, maybe? 2. I've done an initial pass...

@gszadovszky thanks for your comments, will update the PR

Thanks for the numbers; I am deep in the aws v2 sdk migration right now and haven't had a chance to work on this.

...back on this. @parthchandra you know that hadoop trunk is built on the v2 sdk now?