zhengruifeng commented on code in PR #39925:
URL: https://github.com/apache/spark/pull/39925#discussion_r1103956971
##########
python/pyspark/sql/tests/connect/test_connect_basic.py:
##########
@@ -356,6 +356,64 @@ def test_join_condition_column_list_columns(self):
)
self.assert_eq(joined_plan3.toPandas(), joined_plan4.toPandas())
+ def test_join_ambiguous_cols(self):
+ # SPARK-41812: test join with ambiguous columns
+ data1 = [Row(id=1, value="foo"), Row(id=2, value=None)]
+ cdf1 = self.connect.createDataFrame(data1)
+ sdf1 = self.spark.createDataFrame(data1)
+
+ data2 = [Row(value="bar"), Row(value=None), Row(value="foo")]
+ cdf2 = self.connect.createDataFrame(data2)
+ sdf2 = self.spark.createDataFrame(data2)
+
+ cdf3 = cdf1.join(cdf2, cdf1["value"] == cdf2["value"])
+ sdf3 = sdf1.join(sdf2, sdf1["value"] == sdf2["value"])
+
+ self.assertEqual(cdf3.schema, sdf3.schema)
+ self.assertEqual(cdf3.collect(), sdf3.collect())
+
+ cdf4 = cdf1.join(cdf2, cdf1["value"].eqNullSafe(cdf2["value"]))
+ sdf4 = sdf1.join(sdf2, sdf1["value"].eqNullSafe(sdf2["value"]))
+
+ self.assertEqual(cdf4.schema, sdf4.schema)
+ self.assertEqual(cdf4.collect(), sdf4.collect())
+
+ cdf5 = cdf1.join(
+ cdf2, (cdf1["value"] == cdf2["value"]) &
(cdf1["value"].eqNullSafe(cdf2["value"]))
+ )
+ sdf5 = sdf1.join(
+ sdf2, (sdf1["value"] == sdf2["value"]) &
(sdf1["value"].eqNullSafe(sdf2["value"]))
+ )
+
+ self.assertEqual(cdf5.schema, sdf5.schema)
+ self.assertEqual(cdf5.collect(), sdf5.collect())
+
+ cdf6 = cdf1.join(cdf2, cdf1["value"] ==
cdf2["value"]).select(cdf1.value)
+ sdf6 = sdf1.join(sdf2, sdf1["value"] ==
sdf2["value"]).select(sdf1.value)
+
+ self.assertEqual(cdf6.schema, sdf6.schema)
+ self.assertEqual(cdf6.collect(), sdf6.collect())
+
+ cdf7 = cdf1.join(cdf2, cdf1["value"] ==
cdf2["value"]).select(cdf2.value)
+ sdf7 = sdf1.join(sdf2, sdf1["value"] ==
sdf2["value"]).select(sdf2.value)
+
+ self.assertEqual(cdf7.schema, sdf7.schema)
+ self.assertEqual(cdf7.collect(), sdf7.collect())
+
+ def test_invalid_column(self):
Review Comment:
add tests for `df1.select(df2.col)` cc @cloud-fan
--
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]
