Databricks + Sedona TypeError JavaPackage object is not callable

[MrPowers]

I created a Databricks 16.4 LTS ML cluster with Spark 3.5.2 and Scala 2.12.

Here's the init script:

cp /Workspace/Shared/sedona/1.7.1/geotools-wrapper-1.7.1-28.5.jar /databricks/jars
cp /Workspace/Shared/sedona/1.7.1/sedona-spark-shaded-3.5_2.12-1.7.1.jar /databricks/jars

/databricks/python3/bin/pip install apache-sedona==1.7.1
/databricks/python3/bin/pip install geopandas==1.0.1
/databricks/python3/bin/pip install keplergl==0.3.7
/databricks/python3/bin/pip install pydeck==0.9.1

Here is the error:

[jiayuasu]

This means the sedona jars were not found. Are you sure they were loaded into /databricks/jars folder? Can you check via your cluster runtime terminal?

[jiayuasu]

OK. You will need to run the following code (basically explicitly call SedonaContext.create).

from pyspark.sql import functions as f

from sedona.spark import *

sedona = SedonaContext.create(spark)

df = sedona.sql("SELECT array(0.0, 1.0, 2.0) AS values")

min_value = f.array_min("values")
max_value = f.array_max("values")

df = df.select(ST_Point(min_value, max_value))

df.show()

I don’t know why the create is needed since Sedona on Databricks alreay did the same via the SesonaSqlExtension in config. But anyway, this fixes the issue