Re: [Rust] Column names in FFI_ArrowSchema

No reason in particular: please go ahead! tips: anything that is to be deallocated when exporting must go to "private data" and deallocated by the extern "C" release bit, as we do in the DataType string. Think of that function as the "Drop" equivalent for ffi. Best, Jorge On Sat, Feb 20, 2021

[Rust] Column names in FFI_ArrowSchema

When I export to the C data interface structs with array.to_raw(), I'm seeing that FFI_ArrowSchema.name is always null. And looking at ArrowArray::try_new, it appears that FFI_ArrowSchema is only ever created with a format argument; name and metadata are never set to anything. Is there any

Re: Intermittent (but frequent) flight integration failures on master

@mrkn submitted a PR to add backtraces which was merged recently: https://github.com/apache/arrow/pull/9524 However I think the abort is a red herring - the C++ integration client aborts when anything fails. The UNAVAILABLE makes me think that the C++ client is starting and running before the

Re: [Python] A user friendly way to filter parquet partitions

We could add a section to the docs that points out some ecosystem add-on projects On Thu, Feb 18, 2021 at 10:03 PM Bill Zhao wrote: > > Hi Micah, > > Thank you for looking into this matter. > > I understand your goal of having minimal dependency and also solve > the problem from C++ for

Re: [C++] Conventions around C++ shared_ptr in the code base?

On Fri, Feb 19, 2021 at 10:30 AM Micah Kornfield wrote: > > > > > Agreed on Array::data(), in fact I made this change in #9490 (ARROW-9196) > > Thank you! > > > > One convention I'd like to append: we mostly avoid convenience typedefs, > > but an > > exception is the common case of `vector>`, for

Re: [C++] Conventions around C++ shared_ptr in the code base?

> > Agreed on Array::data(), in fact I made this change in #9490 (ARROW-9196) Thank you! > One convention I'd like to append: we mostly avoid convenience typedefs, > but an > exception is the common case of `vector>`, for > which we > allow and encourage the use of `{class name}Vector`

Intermittent (but frequent) flight integration failures on master

I noticed that the Rust/CPP integration tests are failing seemingly intermittently on master (and on Rust PRs). Is anyone else was looking into this? I will have time tomorrow, but likely not today. There didn't seem to be any obviously open PRs or JIRA There are several commits that the little

Re: [C++] Conventions around C++ shared_ptr in the code base?

I don't think there's any benefit in keeping RecordBatch abstract. Making it concrete would probably reduce overhead slightly as well. People who need abstract tabular data capabilities can still implement Table (though I wonder if that ability has ever been used productively). Regards

Re: [C++] Conventions around C++ shared_ptr in the code base?

Thanks for looking into this, Micah. One convention I'd like to append: we mostly avoid convenience typedefs, but an exception is the common case of `vector>`, for which we allow and encourage the use of `{class name}Vector` typedefs. (Conversely, nothing should be named /^\w+Vector$/ which isn't

Re: [C++] The best method to pass null from struct to its children & visitors

Hi Ying, Le 19/02/2021 à 07:06, Ying Zhou a écrit : > > Now I’m working on fixing the last concerns on my ORC writer > https://github.com/apache/arrow/pull/8648 > and have two questions. > > I have a need to standardize an Arrow Array so that it

[NIGHTLY] Arrow Build Report for Job nightly-2021-02-19-0

Arrow Build Report for Job nightly-2021-02-19-0 All tasks: https://github.com/ursacomputing/crossbow/branches/all?query=nightly-2021-02-19-0 Failed Tasks: - conda-linux-gcc-py36-aarch64: URL: