amaliujia commented on code in PR #38653:
URL: https://github.com/apache/spark/pull/38653#discussion_r1022273256
##########
python/pyspark/sql/tests/connect/test_connect_plan_only.py:
##########
@@ -70,6 +70,37 @@ def test_filter(self):
self.assertEqual(plan.root.filter.condition.unresolved_function.parts,
[">"])
self.assertEqual(len(plan.root.filter.condition.unresolved_function.arguments),
2)
+ def test_fill_na(self):
+ df = self.connect.readTable(table_name=self.tbl_name)
+
+ plan = df.fillna(value=1)._plan.to_proto(self.connect)
+ self.assertEqual(len(plan.root.fill_na.values), 1)
+ self.assertEqual(plan.root.fill_na.values[0].long_value, 1)
+ self.assertEqual(plan.root.fill_na.cols, [])
+
+ plan = df.na.fill(value="xyz")._plan.to_proto(self.connect)
+ self.assertEqual(len(plan.root.fill_na.values), 1)
+ self.assertEqual(plan.root.fill_na.values[0].string_value, "xyz")
+ self.assertEqual(plan.root.fill_na.cols, [])
+
+ plan = df.na.fill(value="xyz", subset=["col_a",
"col_b"])._plan.to_proto(self.connect)
+ self.assertEqual(len(plan.root.fill_na.values), 1)
+ self.assertEqual(plan.root.fill_na.values[0].string_value, "xyz")
+ self.assertEqual(plan.root.fill_na.cols, ["col_a", "col_b"])
+
+ plan = df.na.fill(value=True, subset=("col_a", "col_b",
"col_c"))._plan.to_proto(
+ self.connect
+ )
+ self.assertEqual(len(plan.root.fill_na.values), 1)
+ self.assertEqual(plan.root.fill_na.values[0].bool_value, True)
+ self.assertEqual(plan.root.fill_na.cols, ["col_a", "col_b", "col_c"])
+
+ plan = df.fillna({"col_a": 1.5, "col_b":
"abc"})._plan.to_proto(self.connect)
+ self.assertEqual(len(plan.root.fill_na.values), 2)
+ self.assertEqual(plan.root.fill_na.values[0].double_value, 1.5)
+ self.assertEqual(plan.root.fill_na.values[1].string_value, "abc")
+ self.assertEqual(plan.root.fill_na.cols, ["col_a", "col_b"])
Review Comment:
Yeah I am actually thinking we maybe needs at least one for each API to make
sure they pass through. Then use such plan_only test case to improve the
coverage for different usage/cases.
--
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]
