David Li
David Li
See https://github.com/Bit-Quill/arrow/pull/1 to get the build (though not the tests) passing
Looks like this still doesn't build, I'll take another look soon
Hmm, well, it builds locally, though it fails with a different error…and I can't replicate the CI. Seeing what happens with my fork's CI here: https://github.com/lidavidm/arrow/tree/flight-sql-jdbc Otherwise I might ask...
Ok, I'm dumb - it's because the version number got updated again. I'll put up another PR to fix that and the compilation failures. I also think there'll be some...
@jduo see https://github.com/Bit-Quill/arrow/pull/5
Great, CI looks good now (minus known extant issues). I'll give this a final review soon
CI failures are unrelated/addressed elsewhere
We do have to decide what to do about the Parquet tests, though…
So the Parquet tests are now failing, example: https://github.com/apache/arrow/blob/623a15e7f7a45578733956714c8dddcc9f66f015/cpp/src/parquet/arrow/arrow_reader_writer_test.cc#L2816-L2830 It seems we (understandably) ignore the child array slots corresponding to top-level nulls, either on writing or on reading Parquet, but...
See https://github.com/kaoutherab/arrow/pull/1 for an attempt at this, though I think we're going to want to go and test each type specifically for this case.