allisonwang-db commented on code in PR #42422:
URL: https://github.com/apache/spark/pull/42422#discussion_r1289395623
##########
python/pyspark/sql/tests/test_udtf.py:
##########
@@ -1795,6 +1796,65 @@ def terminate(self):
assertSchemaEqual(df.schema, StructType().add("col1",
IntegerType()))
assertDataFrameEqual(df, [Row(col1=10), Row(col1=100)])
+ def test_udtf_with_named_arguments(self):
+ @udtf(returnType="a: int")
+ class TestUDTF:
+ def eval(self, a, b):
+ yield a,
+
+ self.spark.udtf.register("test_udtf", TestUDTF)
+
+ for i, df in enumerate(
+ [
+ self.spark.sql("SELECT * FROM test_udtf(a=>10, b=>'x')"),
+ self.spark.sql("SELECT * FROM test_udtf(b=>'x', a=>10)"),
+ ]
+ ):
+ with self.subTest(query_no=i):
+ assertDataFrameEqual(df, [Row(a=10)])
+
+ def test_udtf_with_kwargs(self):
+ @udtf(returnType="a: int, b: string")
+ class TestUDTF:
+ def eval(self, **kwargs):
+ yield kwargs["a"], kwargs["b"]
+
+ self.spark.udtf.register("test_udtf", TestUDTF)
+
+ for i, df in enumerate(
+ [
+ self.spark.sql("SELECT * FROM test_udtf(a=>10, b=>'x')"),
+ self.spark.sql("SELECT * FROM test_udtf(b=>'x', a=>10)"),
Review Comment:
What would be the error message if the named argument is used incorrectly?
For example
1. duplicated input argument names: a => 10, a => 10
2. non-existing argument name: c => 10
3. incorrect combination of positional and named arguments: test_udtf(a =>
10, 'x')
I am afraid that if we directly leverage Python's kwargs, the error messages
wouldn't be as user-friendly as the SQL function ones.
--
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]
