Known issue - PySpark import statements fail for .jar files installed through environment
You can upload a custom library file with a .jar format into a Fabric environment. Although uploading and publishing of the environment succeeds, the import command fails in the PySpark session.
Status: Open
Product Experience: Data Engineering
Symptoms
In a PySpark session, the import library statement fails with a Class not found error. The custom library with a .jar format was installed through an environment.
Solutions and workarounds
To work around this issue, you can use the following command to install the library in Notebook session: %%configure -f { "conf": { "spark.jars": <>", } }
Alternatively, you can switch from a PySpark to a Scala session if applicable.
Related content
Feedback
https://aka.ms/ContentUserFeedback.
Coming soon: Throughout 2024 we will be phasing out GitHub Issues as the feedback mechanism for content and replacing it with a new feedback system. For more information see:Submit and view feedback for