hi Micah -- I realized that the PR [1] actually tests for _no_ record
batches. We actually already have a working integration test for
length-0 batches [2]

[1]: https://github.com/apache/arrow/pull/3871
[2]: 
https://github.com/apache/arrow/blob/master/integration/integration_test.py#L1034

On Wed, May 22, 2019 at 1:40 PM Micah Kornfield <emkornfi...@gmail.com> wrote:
>
> If no one can get to it sooner, I should be able to get a fix for Java
> sometime next week.
>
> On Tue, May 21, 2019 at 11:45 AM Wes McKinney <wesmck...@gmail.com> wrote:
>
> > https://github.com/apache/arrow/pull/3871
> >
> > On Tue, May 21, 2019 at 1:26 PM Paul Taylor <ptay...@apache.org> wrote:
> > >
> > > I'd be happy to PR a fix for JS today if someone can link me to Wes's PR.
> > >
> > > On 5/21/19 11:02 AM, Wes McKinney wrote:
> > > > I agree also. As a practical use case, the results of a request made
> > > > with Arrow Flight might yield an empty result set. I'm not sure if
> > > > this needs to be formally noted in the specification documents but it
> > > > might not hurt.
> > > >
> > > > If someone can fix the Java implementation we could enable the
> > > > integration test (minus JavaScript for now) in my PR
> > > >
> > > > On Tue, May 21, 2019 at 12:47 AM Ravindra Pindikura <
> > ravin...@dremio.com> wrote:
> > > >> On Tue, May 21, 2019 at 10:35 AM Micah Kornfield <
> > emkornfi...@gmail.com>
> > > >> wrote:
> > > >>
> > > >>> Today, the format docs are ambiguous on whether zero sized batches
> > are
> > > >>> supported.  Wes opened a PR [1] for empty record batches that shows
> > C++
> > > >>> handles them but Java and javascript fail to handle them.
> > > >>>
> > > >>>
> > > >>> I'd like to propose:
> > > >>> 1.  Make it explicit in the format docs, that 0 size record batches
> > are
> > > >>> supported
> > > >>> 2.  Update Java and javascript implementations to work with them (I
> > can put
> > > >>> the Java work on my backlog, but would need a volunteer for JS).
> > And any
> > > >>> other implementations that don't currently handle them.
> > > >>>
> > > >>> Thoughts?
> > > >>>
> > > >> Will need to add a test case for gandiva also - and fix if it shows
> > up any
> > > >> bugs. but, I agree we should support zero sized batches.
> > > >>
> > > >>
> > > >>
> > > >>> Thanks,
> > > >>> Micah
> > > >>>
> > > >>
> > > >> --
> > > >> Thanks and regards,
> > > >> Ravindra.
> >

Reply via email to