Re: [DISCUSS] Improving Contributor Guidelines

2021-03-29 Thread Micah Kornfield
I opened PR for "MINOR" changes https://github.com/apache/arrow/pull/9763 On Sun, Mar 21, 2021 at 8:59 AM Antoine Pitrou wrote: > > Ok, then it sounds ok to me. > > Regards > > Antoine. > > > Le 20/03/2021 à 20:05, Wes McKinney a écrit : > > FTR "[MINOR]" is what Apache Spark uses, e.g. > > > >

Re: [R][Rust][IPC] Attempting to pass RecordBatch from R to Rust via C ABI

2021-03-29 Thread Eric Burden
(1) Zero-copy would be awesome. (2) I'm absolute crap at C. (3) I'll be honest, at this point, I'm just happy to have a way to get it working. That said, the return trip back from Rust using the 'coerce RecordBatch to a buffer/byte vector' strategy seems overly complicated. I managed it by extracti

Bintray deprecation

2021-03-29 Thread Antoine Pitrou
Hello, The Apache INFRA team issued a statement about Bintray deprecation, which can be read here (text pasted below): https://mail-archives.apache.org/mod_mbox/www-builds/202103.mbox/%3CCAN0Gg1dSbHnzO%2BQYsq4qAOy94a2Mhwz57JHqx7vjUyj1qt%2BwdA%40mail.gmail.com%3E """ We have secured a replace

Re: [R][Rust][IPC] Attempting to pass RecordBatch from R to Rust via C ABI

2021-03-29 Thread Neal Richardson
Nothing in https://github.com/apache/arrow/blob/master/r/src/py-to-r.cpp requires Python or reticulate--perhaps we should rename the file if we're getting non-Python applications of the C data interface in R. The reticulate methods in https://github.com/apache/arrow/blob/master/r/R/python.R consume

Re: [R][Rust][IPC] Attempting to pass RecordBatch from R to Rust via C ABI

2021-03-29 Thread Wes McKinney
If you are looking for true zero-copy R/Rust interop, then using the C interface is the way to go. You shouldn't need to depend on Python to have this, so we could need to refactor some things on the R side to compartmentalize anything relating to Python specifically. On Sun, Mar 28, 2021 at 10:04

[NIGHTLY] Arrow Build Report for Job nightly-2021-03-29-0

2021-03-29 Thread Crossbow
Arrow Build Report for Job nightly-2021-03-29-0 All tasks: https://github.com/ursacomputing/crossbow/branches/all?query=nightly-2021-03-29-0 Failed Tasks: - gandiva-jar-ubuntu: URL: https://github.com/ursacomputing/crossbow/branches/all?query=nightly-2021-03-29-0-github-gandiva-jar-ubuntu -

Re: [Gandiva] Active maintainers?

2021-03-29 Thread Kirill Lykov
Thank you Micah for the effort. I think [python][gandiva] was added by Anyscale. Earlier, I've managed to invite them for the commit review yet they are currently not active with it. But these particular PRs are more about documentation and fixing some minor problems and do not require in-depth kno