Hello! I am using the delta-spark (v.

2.4.0

) and pyspark (v.

3.4.1

) packages to attempt to unit test our Databricks notebooks locally. I ran into a problem where the

DeltaTable.forName()

method throws a

pyspark.errors.exceptions.captured.ParseException

when using a 3-layer-namespace for the

tableOrViewName

parameter. Below is an example error:

pyspark.errors.exceptions.captured.ParseException: 
[PARSE_SYNTAX_ERROR] Syntax error at or near '.'.(line 1, pos 21)

== SQL ==
catalog.database.test_table
---------------------^^^

The method works correctly when naming a table using a 2-layer-namespace, i.e.

database.test_table

. Given that 3-layer-namespace support was added to Spark 3.4 and delta-spark 2.4.x should be compatible with Spark 3.4, I expected the

DeltaTable.forName()

method to support this. This thread on the Databricks Community forums also discusses this issue, and the poster mentions that there may be something missing in the delta-core libraries. Is this indeed a bug/missing feature or have I misunderstood something here?