You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2022/12/09 11:25:27 UTC
[GitHub] [spark] grundprinzip commented on a diff in pull request #38946: [SPARK-41414][CONNECT][PYTHON] Implement date/timestamp functions
grundprinzip commented on code in PR #38946:
URL: https://github.com/apache/spark/pull/38946#discussion_r1044356233
##########
python/pyspark/sql/tests/connect/test_connect_function.py:
##########
@@ -645,6 +645,153 @@ def test_string_functions(self):
sdf.select(SF.encode("c", "UTF-8")).toPandas(),
)
+ # TODO(SPARK-41283): To compare toPandas for test cases with dtypes marked
+ def test_date_ts_functions(self):
+ from pyspark.sql import functions as SF
+ from pyspark.sql.connect import functions as CF
+
+ query = """
+ SELECT * FROM VALUES
+ ('1997/02/28 10:30:00', '2023/03/01 06:00:00', 'JST', 1428476400, 2020, 12, 6),
+ ('2000/01/01 04:30:05', '2020/05/01 12:15:00', 'PST', 1403892395, 2022, 12, 6)
+ AS tab(ts1, ts2, tz, seconds, Y, M, D)
+ """
+ # +-------------------+-------------------+---+----------+----+---+---+
+ # | ts1| ts2| tz| seconds| Y| M| D|
+ # +-------------------+-------------------+---+----------+----+---+---+
+ # |1997/02/28 10:30:00|2023/03/01 06:00:00|JST|1428476400|2020| 12| 6|
+ # |2000/01/01 04:30:05|2020/05/01 12:15:00|PST|1403892395|2022| 12| 6|
+ # +-------------------+-------------------+---+----------+----+---+---+
+
+ cdf = self.connect.sql(query)
+ sdf = self.spark.sql(query)
+
+ # With no parameters
+ for cfunc, sfunc in [
+ (CF.current_date, SF.current_date),
+ ]:
+ self.assert_eq(
+ cdf.select(cfunc()).toPandas(),
+ sdf.select(sfunc()).toPandas(),
+ )
+
+ # current_timestamp
+ # [left]: datetime64[ns, America/Los_Angeles]
+ # [right]: datetime64[ns]
+ plan = cdf.select(CF.current_timestamp())._plan.to_proto(self.connect)
+ self.assertEqual(
+ plan.root.project.expressions.unresolved_function.function_name, "current_timestamp"
Review Comment:
you could do something along the lines that you call first spark, then connect then spark again and make sure that connect is in the middle between all values and that the second spark one is after the first one.
##########
python/pyspark/sql/tests/connect/test_connect_function.py:
##########
@@ -645,6 +645,153 @@ def test_string_functions(self):
sdf.select(SF.encode("c", "UTF-8")).toPandas(),
)
+ # TODO(SPARK-41283): To compare toPandas for test cases with dtypes marked
+ def test_date_ts_functions(self):
+ from pyspark.sql import functions as SF
+ from pyspark.sql.connect import functions as CF
+
+ query = """
+ SELECT * FROM VALUES
+ ('1997/02/28 10:30:00', '2023/03/01 06:00:00', 'JST', 1428476400, 2020, 12, 6),
+ ('2000/01/01 04:30:05', '2020/05/01 12:15:00', 'PST', 1403892395, 2022, 12, 6)
+ AS tab(ts1, ts2, tz, seconds, Y, M, D)
+ """
+ # +-------------------+-------------------+---+----------+----+---+---+
+ # | ts1| ts2| tz| seconds| Y| M| D|
+ # +-------------------+-------------------+---+----------+----+---+---+
+ # |1997/02/28 10:30:00|2023/03/01 06:00:00|JST|1428476400|2020| 12| 6|
+ # |2000/01/01 04:30:05|2020/05/01 12:15:00|PST|1403892395|2022| 12| 6|
+ # +-------------------+-------------------+---+----------+----+---+---+
+
+ cdf = self.connect.sql(query)
+ sdf = self.spark.sql(query)
+
+ # With no parameters
+ for cfunc, sfunc in [
+ (CF.current_date, SF.current_date),
+ ]:
+ self.assert_eq(
+ cdf.select(cfunc()).toPandas(),
+ sdf.select(sfunc()).toPandas(),
+ )
+
+ # current_timestamp
+ # [left]: datetime64[ns, America/Los_Angeles]
+ # [right]: datetime64[ns]
+ plan = cdf.select(CF.current_timestamp())._plan.to_proto(self.connect)
+ self.assertEqual(
+ plan.root.project.expressions.unresolved_function.function_name, "current_timestamp"
+ )
+
+ # localtimestamp
+ plan = cdf.select(CF.localtimestamp())._plan.to_proto(self.connect)
+ self.assertEqual(
+ plan.root.project.expressions.unresolved_function.function_name, "localtimestamp"
Review Comment:
see above
--
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: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org