[Java] Large Memory Allocators (Taking a dependency on JNA?)

2020-01-17 Thread Micah Kornfield
With the recently merged changes to the underlying ArrowBuf APIs to allow 64-bit memory address spaces there is some follow-up work to actually confirm it works. I opened a JIRA [1] to track this work. The main question is how to provide an allocator that supports larger memory chunks. It

[jira] [Created] (ARROW-7610) [Java] Finish support for 64 bit int allocations

2020-01-17 Thread Micah Kornfield (Jira)
Micah Kornfield created ARROW-7610: -- Summary: [Java] Finish support for 64 bit int allocations Key: ARROW-7610 URL: https://issues.apache.org/jira/browse/ARROW-7610 Project: Apache Arrow

Re: 0.16 release preparation

2020-01-17 Thread Neal Richardson
Hi folks, Looks like we're still a couple of days away from closing out 0.16. The wiki shows 24 issues still open/in-progress, 4 of which are "blockers". These blockers are: * ARROW-6801 [Rust] Arrow source release tarball is missing benchmarks (breaks the release script) * ARROW-7596 [Python]

[jira] [Created] (ARROW-7609) [C++] Build error by passing -DARROW_HIVESERVER2=ON

2020-01-17 Thread Yosuke Shiro (Jira)
Yosuke Shiro created ARROW-7609: --- Summary: [C++] Build error by passing -DARROW_HIVESERVER2=ON Key: ARROW-7609 URL: https://issues.apache.org/jira/browse/ARROW-7609 Project: Apache Arrow Issue

Re: Python 2.7 support: dropping it after Arrow 0.16 ?

2020-01-17 Thread Wes McKinney
hi Joris, Personally I am OK with dropping 2.7 after the next major release (0.16.0). According to pypistats.org there are still a lot of 2.7 installs but we aren't seeing much incoming maintenance support for this, so it doesn't seem reasonable to burden ourselves with this. Rather than hacking

Re: [Discuss][Rust] Policy regarding "unsafe"

2020-01-17 Thread Andy Grove
I agree that we need to audit use of unsafe and think carefully about how we use it. This blog post is somewhat tangential since it is primarily about the actix-web situation with the author quitting open source, but "unsafe" played a large role in it, and I think this is worth reading.

[jira] [Created] (ARROW-7608) [C++][Dataset] Expose more informational properties

2020-01-17 Thread Neal Richardson (Jira)
Neal Richardson created ARROW-7608: -- Summary: [C++][Dataset] Expose more informational properties Key: ARROW-7608 URL: https://issues.apache.org/jira/browse/ARROW-7608 Project: Apache Arrow

Re: [Discuss][Rust] Policy regarding "unsafe"

2020-01-17 Thread Neville Dipale
Hi Paddy, Arrow Developers, I've given this some thought, and I preliminarily think that perhaps we can audit our use of unsafe and evaluate where we can remove it, propagate it upwards (and provide safe alternatives) or provide some safety to callers. Looking at the 3 options that Paul Kernfeld

Re: [Discuss][Rust] Policy regarding "unsafe"

2020-01-17 Thread Neville Dipale
Hi Paddy, On Fri, 10 Jan 2020 at 04:13, paddy horan wrote: > Hi All, > > This time last year there was a brief discussion on the usage of unsafe in > Rust (a user on github raised the issue and I created the JIRA). [1] > > So far we mostly avoid unsafe in the public API's. The thinking here is

[jira] [Created] (ARROW-7607) [C++] Add to cpp/examples minimal examples of using Arrow as a dependency of another CMake project

2020-01-17 Thread Wes McKinney (Jira)
Wes McKinney created ARROW-7607: --- Summary: [C++] Add to cpp/examples minimal examples of using Arrow as a dependency of another CMake project Key: ARROW-7607 URL: https://issues.apache.org/jira/browse/ARROW-7607

[jira] [Created] (ARROW-7606) [Java]Update checkstyle plugin to latest version

2020-01-17 Thread Jiajia Li (Jira)
Jiajia Li created ARROW-7606: Summary: [Java]Update checkstyle plugin to latest version Key: ARROW-7606 URL: https://issues.apache.org/jira/browse/ARROW-7606 Project: Apache Arrow Issue Type:

[jira] [Created] (ARROW-7605) [C++] Merge private je_arrow symbols into produced libarrow.a

2020-01-17 Thread Wes McKinney (Jira)
Wes McKinney created ARROW-7605: --- Summary: [C++] Merge private je_arrow symbols into produced libarrow.a Key: ARROW-7605 URL: https://issues.apache.org/jira/browse/ARROW-7605 Project: Apache Arrow

[jira] [Created] (ARROW-7604) [C++] Avoid duplicate object file names

2020-01-17 Thread Wes McKinney (Jira)
Wes McKinney created ARROW-7604: --- Summary: [C++] Avoid duplicate object file names Key: ARROW-7604 URL: https://issues.apache.org/jira/browse/ARROW-7604 Project: Apache Arrow Issue Type:

[jira] [Created] (ARROW-7603) [CI][Crossbow] Nightly centos 8 job fails

2020-01-17 Thread Neal Richardson (Jira)
Neal Richardson created ARROW-7603: -- Summary: [CI][Crossbow] Nightly centos 8 job fails Key: ARROW-7603 URL: https://issues.apache.org/jira/browse/ARROW-7603 Project: Apache Arrow Issue

[jira] [Created] (ARROW-7602) [Archery] Add more build options

2020-01-17 Thread Francois Saint-Jacques (Jira)
Francois Saint-Jacques created ARROW-7602: - Summary: [Archery] Add more build options Key: ARROW-7602 URL: https://issues.apache.org/jira/browse/ARROW-7602 Project: Apache Arrow

Python 2.7 support: dropping it after Arrow 0.16 ?

2020-01-17 Thread Joris Van den Bossche
Hi all, The support for Python 2.7 has been discussed before. At some point, it was decided to support it until the end of 2019 (ARROW-331 , the docs state that pyarrow "pledges to maintain compatibility with Python 2.7 until the end of 2019").

[NIGHTLY] Arrow Build Report for Job nightly-2020-01-17-0

2020-01-17 Thread Crossbow
Arrow Build Report for Job nightly-2020-01-17-0 All tasks: https://github.com/ursa-labs/crossbow/branches/all?query=nightly-2020-01-17-0 Failed Tasks: - centos-8: URL: https://github.com/ursa-labs/crossbow/branches/all?query=nightly-2020-01-17-0-azure-centos-8 - gandiva-jar-osx: URL:

[jira] [Created] (ARROW-7601) [Doc] [C++] Create a doc page about fuzzing

2020-01-17 Thread Antoine Pitrou (Jira)
Antoine Pitrou created ARROW-7601: - Summary: [Doc] [C++] Create a doc page about fuzzing Key: ARROW-7601 URL: https://issues.apache.org/jira/browse/ARROW-7601 Project: Apache Arrow Issue

Re: [C++] Arrow added to OSS-Fuzz

2020-01-17 Thread Antoine Pitrou
Le 17/01/2020 à 07:33, Marco Neumann a écrit : > Hey Antoine,  > > Thanks a lot also from my side.  > > The build is likely currently succeeding due to the Fuzzing work done by > fuzzit. Actually, around ~30 crashes were found on the first run or so. I think one of the important things here