Hi,
I feel that '.stream' is too generic. How about '.arrows'?
JSON Lines uses 'l' suffix for extension: '.jsonl'
https://jsonlines.org/#conventions
Thanks,
--
kou
In
"Re: Please Review: Application for a Media Type" on Thu, 22 Apr 2021
06:44:51 +0200,
Jorge Cardoso Leitão wrote:
> Tha
Hi I am seeing a similar problem when serializing tables with lists of
dictionary encoded elements: each resulting chunk is pointing to the first
chunk’s original dictionary.
Is this a known issue/limitation.
I can follow with a repro otherwise.
Thank you
Radu
> On Sep 28, 2020, at 1:26 PM, Wes
Matt,
I just created the JIRA: https://issues.apache.org/jira/browse/ARROW-12517
. I'd be happy to pick it up as well but since it sounds like you're
already in the code at the moment and it's a small change, I'll leave it to
you. Thanks for your work on the Go library already!
For anyone else
This is a verification script problem:
https://github.com/apache/arrow/pull/10135
It should not run Gandiva related Ruby tests with
ARROW_GANDIVA=0.
In <55363261-b056-440b-dbfe-222f8f13f...@python.org>
"Re: [VOTE] Release Apache Arrow 4.0.0 - RC3" on Thu, 22 Apr 2021 13:44:54
+0200,
Antoi
+1
I've been using the development version for assignments and I had 0 problems
On Thu, Apr 22, 2021 at 4:12 PM Ian Cook wrote:
> +1 (non-binding)
>
> Verified C++ source on Windows 10 with only known failures[1] that
> seem to be caused by using Visual Studio 2019.
>
> [1] https://issues.apac
+1 (non-binding)
Verified C++ source on Windows 10 with only known failures[1] that
seem to be caused by using Visual Studio 2019.
[1] https://issues.apache.org/jira/browse/ARROW-11675
On Wed, Apr 21, 2021 at 5:30 PM Krisztián Szűcs wrote:
>
> Hi,
>
> I would like to propose the following rele
Thanks for the clarifications, much appreciated.
Looking closer, I realize `arrjson` is anyway separating out the values, which
isn't what I'd want to ship a table from one service to another.
What's a good way to embed the table as a byte stream that can be "read back
out" the other end?
I se
+1 (non-binding)
Verified wheels, sources, and binaries on macOS 11.2 using the verification
script (except for Java Integration, Glib, and Ruby). Like Antoine I ran
into the same issue with Ruby.
I also installed Arrow and the R package locally + ran some adhoc tests
using some of our benchmarks
+1 (binding)
Verified the source release, binaries and wheels on macOS Big Sur.
The automatized verifications scripts have passed as well [1]
[1]: https://github.com/apache/arrow/pull/10126
On Thu, Apr 22, 2021 at 3:56 PM David Li wrote:
>
> +1 (non-binding)
>
> Verified wheels, sources, and ap
You're absolutely correct and not missing anything, this is definitely an
opportunity to make the flight record reader a bit more useful. I like the idea
of the getLatestMetadata that you mentioned is on the java side. Can you file a
JIRA issue for this? Given that I'm doing a lot of updates to
Micah is correct, the arrjson package is used for the internal integration
testing using the specific JSON format for that integration testing which is
not likely what Users would want when converting Arrow to JSON.
There is not currently a recommended way to serialize an instance of
arrow.Tabl
+1 (non-binding)
Verified wheels, sources, and apt binaries on Ubuntu 18.04.
Best,
David
On 2021/04/21 21:30:33, Krisztián Szűcs wrote:
> Hi,
>
> I would like to propose the following release candidate (RC3) of Apache
> Arrow version 4.0.0. This is a release consisting of 719
> resolved JIRA
I tried to verify the source release on Ubuntu 20.04 with
ARROW_GANDIVA=0 TEST_JAVA=0 TEST_INTEGRATION=0 TEST_CSHARP=0.
It succeeded until the Ruby bindings:
+ bundle exec ruby test/run-test.rb
Traceback (most recent call last):
9: from test/run-test.rb:48:in `'
8: from test/ru
Arrow Build Report for Job nightly-2021-04-22-0
All tasks:
https://github.com/ursacomputing/crossbow/branches/all?query=nightly-2021-04-22-0
Failed Tasks:
- conda-linux-gcc-py36-arm64:
URL:
https://github.com/ursacomputing/crossbow/branches/all?query=nightly-2021-04-22-0-drone-conda-linux-g
Yes, these soft-float math (in libm.so) makes Arm binary extremely slow.
-Original Message-
From: Antoine Pitrou
Sent: Thursday, April 22, 2021 17:20
To: dev@arrow.apache.org
Subject: Re: [C++] Indeterminate poor performance of random number generator
Le 22/04/2021 à 03:38, Yibo Cai a é
On 4/22/21 9:38 AM, Yibo Cai wrote:
On 4/21/21 6:07 PM, Antoine Pitrou wrote:
Le 21/04/2021 à 11:41, Yibo Cai a écrit :
On 4/21/21 5:17 PM, Antoine Pitrou wrote:
Le 21/04/2021 à 11:14, Yibo Cai a écrit :
When running benchmarks on Arm64 servers, I find some benchmarks are extremely
slow w
Le 22/04/2021 à 03:38, Yibo Cai a écrit :
Both using same libstdc++.
But std::bernoulli_distribution is inlined, so they are indeed different for
clang and gcc.
https://godbolt.org/z/aT84x5Yec
Looks a pure compiler thing.
It looks like clang generates calls to logl() and __divtf3() (soft-fl
+1
I got only known Python problems.
I ran the followings on Debian GNU/Linux sid:
* LANG=C \
TZ=UTC \
ARROW_CMAKE_OPTIONS="-DBoost_NO_BOOST_CMAKE=ON" \
CUDA_TOOLKIT_ROOT=/usr \
dev/release/verify-release-candidate.sh source 4.0.0 3
* dev/release/verify-release-candid
Hi,
I've re-uploaded binaries.
Thanks,
--
kou
In
"Re: [VOTE] Release Apache Arrow 4.0.0 - RC3" on Thu, 22 Apr 2021 03:11:09
-1000,
Weston Pace wrote:
> I'm getting a failure during the download files check...
>
> Traceback (most recent call last):
> File "/home/centos/arrow/dev/releas
19 matches
Mail list logo
