-
Notifications
You must be signed in to change notification settings - Fork 28.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[SPARK-38563][PYTHON] Upgrade to Py4J 0.10.9.5 #35907
Conversation
cc @dongjoon-hyun @wangyum FYI |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Actually, we cannot use the same JIRA ID because branch-3.2 has Py4J 0.10.9.4
already with SPARK-38563. Could you use a new JIRA ID for Py4J 0.10.9.5
? You can still land it to branch-3.2 too.
SPARK-38563 solved the resource leakage in |
Oh no. That's not released yet. i reverted it from branch-3.2 too. |
BTW, Python 3.10 already works with Spark 3.2 too - Py4J upgrade broke that (unofficial) support. |
Technically what you said is correct because |
Oh, got it. If you reverted cleanly from all branches, there is no problem. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
+1, LGTM.
Merged to master, branch-3.3 and branch-3.2. |
### What changes were proposed in this pull request? This PR is a retry of #35871 with bumping up the version to 0.10.9.5. It was reverted because of Python 3.10 is broken, and Python 3.10 was not officially supported in Py4J. In Py4J 0.10.9.5, the issue was fixed (py4j/py4j#475), and it added Python 3.10 support officially with CI set up (py4j/py4j#477). ### Why are the changes needed? See #35871 ### Does this PR introduce _any_ user-facing change? See #35871 ### How was this patch tested? Py4J sets up Python 3.10 CI now, and I manually tested PySpark with Python 3.10 with this patch: ```bash ./bin/pyspark ``` ``` import py4j py4j.__version__ spark.range(10).show() ``` ``` Using Python version 3.10.0 (default, Mar 3 2022 03:57:21) Spark context Web UI available at http://172.30.5.50:4040 Spark context available as 'sc' (master = local[*], app id = local-1647571387534). SparkSession available as 'spark'. >>> import py4j >>> py4j.__version__ '0.10.9.5' >>> spark.range(10).show() +---+ | id| +---+ ... ``` Closes #35907 from HyukjinKwon/SPARK-38563-followup. Authored-by: Hyukjin Kwon <[email protected]> Signed-off-by: Hyukjin Kwon <[email protected]> (cherry picked from commit 97335ea) Signed-off-by: Hyukjin Kwon <[email protected]>
This PR is a retry of #35871 with bumping up the version to 0.10.9.5. It was reverted because of Python 3.10 is broken, and Python 3.10 was not officially supported in Py4J. In Py4J 0.10.9.5, the issue was fixed (py4j/py4j#475), and it added Python 3.10 support officially with CI set up (py4j/py4j#477). See #35871 See #35871 Py4J sets up Python 3.10 CI now, and I manually tested PySpark with Python 3.10 with this patch: ```bash ./bin/pyspark ``` ``` import py4j py4j.__version__ spark.range(10).show() ``` ``` Using Python version 3.10.0 (default, Mar 3 2022 03:57:21) Spark context Web UI available at http://172.30.5.50:4040 Spark context available as 'sc' (master = local[*], app id = local-1647571387534). SparkSession available as 'spark'. >>> import py4j >>> py4j.__version__ '0.10.9.5' >>> spark.range(10).show() +---+ | id| +---+ ... ``` Closes #35907 from HyukjinKwon/SPARK-38563-followup. Authored-by: Hyukjin Kwon <[email protected]> Signed-off-by: Hyukjin Kwon <[email protected]> (cherry picked from commit 97335ea) Signed-off-by: Hyukjin Kwon <[email protected]>
This PR is a retry of apache#35871 with bumping up the version to 0.10.9.5. It was reverted because of Python 3.10 is broken, and Python 3.10 was not officially supported in Py4J. In Py4J 0.10.9.5, the issue was fixed (py4j/py4j#475), and it added Python 3.10 support officially with CI set up (py4j/py4j#477). See apache#35871 See apache#35871 Py4J sets up Python 3.10 CI now, and I manually tested PySpark with Python 3.10 with this patch: ```bash ./bin/pyspark ``` ``` import py4j py4j.__version__ spark.range(10).show() ``` ``` Using Python version 3.10.0 (default, Mar 3 2022 03:57:21) Spark context Web UI available at http://172.30.5.50:4040 Spark context available as 'sc' (master = local[*], app id = local-1647571387534). SparkSession available as 'spark'. >>> import py4j >>> py4j.__version__ '0.10.9.5' >>> spark.range(10).show() +---+ | id| +---+ ... ``` Closes apache#35907 from HyukjinKwon/SPARK-38563-followup. Authored-by: Hyukjin Kwon <[email protected]> Signed-off-by: Hyukjin Kwon <[email protected]> (cherry picked from commit 97335ea) Signed-off-by: Hyukjin Kwon <[email protected]>
What changes were proposed in this pull request?
This PR is a retry of #35871 with bumping up the version to 0.10.9.5.
It was reverted because of Python 3.10 is broken, and Python 3.10 was not officially supported in Py4J.
In Py4J 0.10.9.5, the issue was fixed (py4j/py4j#475), and it added Python 3.10 support officially with CI set up (py4j/py4j#477).
Why are the changes needed?
See #35871
Does this PR introduce any user-facing change?
See #35871
How was this patch tested?
Py4J sets up Python 3.10 CI now, and I manually tested PySpark with Python 3.10 with this patch: