Spark JVM & Spark Connect compatibility library.
Here is code that works on Spark JVM:
data = [("Java", "20000"), ("Python", "100000"), ("Scala", "3000")]
spark.sparkContext.parallelize(data).toDF()
This code doesn't work with Spark Connect. The congruity library rearranges the code under the hood, so the old syntax works on Spark Connect clusters as well:
from congruity import *
data = [("Java", "20000"), ("Python", "100000"), ("Scala", "3000")]
spark.sparkContext.parallelize(data).toDF()