I’ve done some testing using pyarrow.jvm as part of jdbc optimization.

https://arrow.apache.org/docs/python/integration/python_java.html

Pull a dataset using a JVM environment with JDBC.

Map JDBC arrow result set to a python pyarrow result set.

ra = jpype.JPackage("org").apache.arrow.memory.RootAllocator(sys.maxsize)

batch = jpype.JPackage("org").apache.arrow.adapter.jdbc.JdbcToArrow.sqlToArrow(
    jdbc_connection,
    sql,
    ra
)

arrow_table = pyarrow.jvm.record_batch(batch)


From: Chang She <[email protected]>
Sent: Monday, October 24, 2022 11:44 AM
To: [email protected]
Subject: Re: Arrow in jupyter


External Email: Use caution with links and attachments
Awesome, I will continue on that JIRA.

Thanks!

On Mon, Oct 24, 2022 at 6:23 AM Neal Richardson 
<[email protected]<mailto:[email protected]>> wrote:
Hi Chang,
I don't recall seeing an Arrow issue about this specifically, or any examples 
of doing that, but that sounds nice. 
https://issues.apache.org/jira/browse/ARROW-17608<https://urldefense.com/v3/__https:/issues.apache.org/jira/browse/ARROW-17608__;!!KSjYCgUGsB4!fr_nJ6s_jfTijlw-O2aBexeefqDuRbFByff-p8Zi9Y8VGZrzsqSy-tCzdikV_lCdBGz9Yl-nhYFCIrA$>
 (C interface support in JS) is probably a prerequisite.

Neal

On Sun, Oct 23, 2022 at 2:45 PM Chang She <[email protected]<mailto:[email protected]>> 
wrote:
couldn’t*

On Sun, Oct 23, 2022 at 11:42 AM Chang She <[email protected]<mailto:[email protected]>> 
wrote:
Hi, just curious if anyone’s played with using Arrow to ship data between 
languages in jupyter/ipython, specifically python and js (that doesn’t require 
a flight server or writing the data to disk).

Eg, can I call do something like IPython.kernel.execute(“table.filter(…)”) and 
read the results directly into arrowjs instead of converting from plain text / 
json?

Could find an existing issue and figure I’d ask before I go chase this down.

Thanks!

Chang

This message may contain information that is confidential or privileged. If you 
are not the intended recipient, please advise the sender immediately and delete 
this message. See 
http://www.blackrock.com/corporate/compliance/email-disclaimers for further 
information.  Please refer to 
http://www.blackrock.com/corporate/compliance/privacy-policy for more 
information about BlackRock’s Privacy Policy.


For a list of BlackRock's office addresses worldwide, see 
http://www.blackrock.com/corporate/about-us/contacts-locations.

© 2022 BlackRock, Inc. All rights reserved.

Reply via email to