I think that with a few reviews, we could have a minimally working DSv2 SQL API ready in a couple weeks. We have PRs ready for the last of the major items, like ALTER TABLE, INSERT INTO, REPLACE TABLE, and a v2 catalog interface for the session catalog. Mainly, we're just waiting for final reviews on those and then I would be comfortable moving to QA for 3.0.
The built-in file sources unfortunately have some work to do to work with SQL, so I'm not including those in that statement. That would also not include all of the non-essential SQL statements, like SHOW TABLES or CREATE TABLE LIKE. Most of the non-essential SQL could be done fairly quickly, if anyone has time! On Fri, Jun 28, 2019 at 4:53 PM Sean Owen <sro...@gmail.com> wrote: > That's a good question. Although we had penciled in 'middle of the > year' I don't think we're in sight of a QA phase just yet, as I > believe some key items are still in progress. I'm thinking of the Hive > update, and DS v2 work (?). > > I'm also curious to hear what broad TODOs people see for 3.0? we > probably need to start focusing them down and/or deferring some parts, > in order to get a 3.0 release out anytime soon. Is that feasible? I > have a guess that it won't end up being released this quarter, but, > just a guess. > > On Fri, Jun 28, 2019 at 5:14 PM Long, Andrew > <loand...@amazon.com.invalid> wrote: > > > > Hey Friends, > > > > > > > > Is there a timeline for spark 3.0 in terms of the first RC and final > release? > > > > > > > > Cheers Andrew > > --------------------------------------------------------------------- > To unsubscribe e-mail: dev-unsubscr...@spark.apache.org > > -- Ryan Blue Software Engineer Netflix
