dmsuehir commented on PR #9969:
URL: 
https://github.com/apache/incubator-gluten/pull/9969#issuecomment-2986141172

   @PHILO-HE Thanks, I've tried that out and ran into a couple of issues:
   * The install directory for `install_spark_resources.sh` is hardcoded to 
`/opt`. When trying to run this without sudo, I get a permission denied error
   * After running `install_spark_resources.sh`, the directory has the Spark 
test resource files, but it does not have the Spark binaries. This works for 
running unit tests, but without binaries the doc generation script fails. You 
can see from this 
[PR](https://github.com/apache/incubator-gluten/pull/9499/files#diff-cc9b23a1706d7aac3ffaa94c96692d477c2cfb375b994d23289190d0f581e3c4R125),
 which is trying to add a GHA weekly run for the doc generation script, that 
it's needing to download the Spark binaries and then combine that with the test 
resource files from the `install_spark_resources.sh` directory.
   
   I can mitigate these issues with a couple of changes to the 
`install_spark_resource.sh` script:
   * Adding an optional second positional arg to the script for the spark dir 
(if it's not provided, it can still default to `/opt`)
   * Extracting the bin/python dir from the spark .tgz file (right now the 
script only grabs `jars`)


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to