Hi all!

I am the author of Tonic, I'd love to see the rust flight implementation done 
with Tonic. David, it looks like what you implemented with tower-grpc should 
work just fine with tonic as well. I am also interested in helping out. Since, 
I guess most of my experience up to now is with Tonic itself, I wanted to make 
myself available to you all for help or special implementations coming from the 
Tonic side. I am also willing to help with the flight rust implementation. So 
please let me know around that what I can do!
Thanks,
Lucio

On Oct 17 2019, at 7:26 pm, Neville Dipale <nevilled...@gmail.com> wrote:
> Thanks Andy,
>
> Please see https://github.com/apache/arrow/pull/4167#issuecomment-543381089
> for the status of the PR. We have a few missing data types (fixed list,
> timezone to timestamp, etc.) that are currently stopping me from testing
> the reading of files.
>
> I'm trying out creating a fixed size list, and I'll open a PR for that if
> my attempt works.
>
> On Fri, 18 Oct 2019 at 01:10, Andy Grove <andygrov...@gmail.com> wrote:
> > Thanks for all the updates. I'd like to get involved and help out with this
> > effort as well. I don't have any major work planned for DataFusion for
> > 1.0.0 now other than maybe moving to the new parquet ArrowReader, if it is
> > ready in time.
> >
> > I have been chatting with the author of the Rust Flatbuffer project about
> > some of the issues and I can take an action to follow up with that.
> >
> > I have also been talking with one of the authors of Tonic, and I believe
> > they might be interested in helping here too.
> >
> > Let me know how else I can help out with this effort.
> > Andy.
> >
> >
> > On Thu, Oct 17, 2019 at 2:42 PM Neville Dipale <nevilled...@gmail.com>
> > wrote:
> >
> > > Good evening
> > > With support for testing against integration files now done, I've resumed
> > > work on the IPC reader. If I don't encounter trouble reading the existing
> > > files, I expect to be done with this work by the end of the weekend. I
> >
> > had
> > > taken the approach of one large PR to include all Rust-supported Arrow
> > > types.
> > >
> > > I'm not sure of how long the writer would take, but I remain committed to
> > > having this work completed by 1.0.
> > >
> > > I have to catch up on null-type roundtrip and the padding alignment work
> > as
> > > I haven't been able to keep abreast with development these last few
> >
> > months.
> > > Also, the Rust flatbuffer issues that we've had haven't progressed in the
> > > relevant repo, so it still makes ergonomics not great, but at least our
> > > users don't have to worry about that.
> > >
> > > @Andy I have good experience with gRPC in Rust, and also want to see
> > Flight
> > > support landing soon.
> > >
> > > On Thu, 17 Oct 2019, 17:06 David Li, <li.david...@gmail.com> wrote:
> > > > Just for reference, it's possible once the basic IPC support is
> > merged; I
> > > > had a proof of concept, though it needs to be updated to use Tonic over
> > > > tower-grpc, actually implement the zero-copy optimizations, provide a
> > >
> > > real
> > > > API, etc.
> > > >
> > > > https://github.com/apache/arrow/pull/4167#issuecomment-529695811
> > > > On Thu, Oct 17, 2019, 10:51 Wes McKinney <wesmck...@gmail.com> wrote:
> > > > > I hope to see Flight in all the reference implementations eventually.
> > > > > Having hardened IPC support is a pre-requisite, it would be ideal to
> > > > > have Rust as a participant in the integration tests
> > > > >
> > > > > On Thu, Oct 17, 2019 at 9:41 AM Andy Grove <andygrov...@gmail.com>
> > > > wrote:
> > > > > >
> > > > > > I was approached directly about adding Flight support to the Rust
> > > > > > implementation, and said I would start a discussion here on the
> > > > >
> > > >
> > >
> > > mailing
> > > > > > list.
> > > > > >
> > > > > > There is ongoing work with IPC and integration and I believe that
> > it
> > > > > would
> > > > > > make sense to start looking at adding Flight support.
> > > > > >
> > > > > > I'd like to hear what others think though.
> > > > > > Thanks,
> > > > > > Andy.

Reply via email to