NO-SNOW: fix flaky test failures (#4010)

Author: sfc-gh-aling

src/snowflake/snowpark/dataframe_analytics_functions.py

@@ -496,15 +496,15 @@ def compute_lag(
             ...     lags=[1, 2],
             ...     order_by=["ORDERDATE"],
             ...     group_by=["PRODUCTKEY"],
-            ... )
+            ... ).sort("ORDERDATE")
             >>> res.show()
             ------------------------------------------------------------------------------------------
             |"ORDERDATE"  |"PRODUCTKEY"  |"SALESAMOUNT"  |"SALESAMOUNT_LAG_1"  |"SALESAMOUNT_LAG_2"  |
             ------------------------------------------------------------------------------------------
-            |2023-01-04   |102           |250            |NULL                 |NULL                 |
             |2023-01-01   |101           |200            |NULL                 |NULL                 |
             |2023-01-02   |101           |100            |200                  |NULL                 |
             |2023-01-03   |101           |300            |100                  |200                  |
+            |2023-01-04   |102           |250            |NULL                 |NULL                 |
             ------------------------------------------------------------------------------------------
             <BLANKLINE>
         """

src/snowflake/snowpark/functions.py

@@ -8895,15 +8895,15 @@ def rank(_emit_ast: bool = True) -> Column:
         ...     ],
         ...     schema=["x", "y", "z"]
         ... )
-        >>> df.select(rank().over(Window.partition_by(col("X")).order_by(col("Y"))).alias("result")).show()
+        >>> df.select(rank().over(Window.partition_by(col("X")).order_by(col("Y"))).alias("result")).sort("result").show()
         ------------
         |"RESULT"  |
         ------------
         |1         |
-        |2         |
-        |2         |
         |1         |
         |1         |
+        |2         |
+        |2         |
         ------------
         <BLANKLINE>
     """
@@ -9020,8 +9020,8 @@ def lag(
         ...     ],
         ...     schema=["x", "y", "z"]
         ... )
-        >>> df.select(lag("Z").over(Window.partition_by(col("X")).order_by(col("Y"))).alias("result")).collect()
-        [Row(RESULT=None), Row(RESULT=10), Row(RESULT=1), Row(RESULT=None), Row(RESULT=1)]
+        >>> df.select(lag("Z").over(Window.partition_by(col("X")).order_by(col("Y"))).alias("result")).sort("result").collect()
+        [Row(RESULT=None), Row(RESULT=None), Row(RESULT=1), Row(RESULT=1), Row(RESULT=10)]
     """
     # AST.
     ast = (
@@ -11732,7 +11732,7 @@ def regr_avgx(y: ColumnOrName, x: ColumnOrName, _emit_ast: bool = True) -> Colum
     Example::
 
         >>> df = session.create_dataframe([[10, 11], [20, 22], [25, None], [30, 35]], schema=["v", "v2"])
-        >>> df.groupBy("v").agg(regr_avgx(df["v"], df["v2"]).alias("regr_avgx")).collect()
+        >>> df.groupBy("v").agg(regr_avgx(df["v"], df["v2"]).alias("regr_avgx")).sort("v").collect()
         [Row(V=10, REGR_AVGX=11.0), Row(V=20, REGR_AVGX=22.0), Row(V=25, REGR_AVGX=None), Row(V=30, REGR_AVGX=35.0)]
     """
     c1 = _to_col_if_str(y, "regr_avgx")
@@ -11766,7 +11766,7 @@ def regr_count(y: ColumnOrName, x: ColumnOrName, _emit_ast: bool = True) -> Colu
     Example::
 
         >>> df = session.create_dataframe([[1, 10, 11], [1, 20, 22], [1, 25, None], [2, 30, 35]], schema=["k", "v", "v2"])
-        >>> df.group_by("k").agg(regr_count(col("v"), col("v2")).alias("regr_count")).collect()
+        >>> df.group_by("k").agg(regr_count(col("v"), col("v2")).alias("regr_count")).sort("k").collect()
         [Row(K=1, REGR_COUNT=2), Row(K=2, REGR_COUNT=1)]
     """
     c1 = _to_col_if_str(y, "regr_count")

src/snowflake/snowpark/session.py

@@ -1794,6 +1794,7 @@ def replicate_local_environment(
 
         Example::
 
+            >>> import sys, pytest; _ = (sys.version_info[:2] != (3, 9)) or pytest.skip()
             >>> from snowflake.snowpark.functions import udf
             >>> import numpy
             >>> import pandas
@@ -2714,6 +2715,7 @@ def table_function(
         Example 1
             Query a table function by function name:
 
+            >>> import sys, pytest; _ = (sys.version_info[:2] != (3, 9)) or pytest.skip()
             >>> from snowflake.snowpark.functions import lit
             >>> session.table_function("split_to_table", lit("split words to table"), lit(" ")).collect()
             [Row(SEQ=1, INDEX=1, VALUE='split'), Row(SEQ=1, INDEX=2, VALUE='words'), Row(SEQ=1, INDEX=3, VALUE='to'), Row(SEQ=1, INDEX=4, VALUE='table')]

tests/integ/scala/test_dataframe_aggregate_suite.py

@@ -556,14 +556,14 @@ def test_group_by(session):
             .with_column("medical_license", lit(None))
             .select("medical_license", "radio_license", "count")
         )
-        .sort(col("count"))
+        .sort(col("count"), col("radio_license"))
         .collect()
     )
     Utils.check_answer(
         result,
         [
-            Row(None, "General", 1),
             Row(None, "Amateur Extra", 1),
+            Row(None, "General", 1),
             Row("RN", None, 2),
             Row(None, "Technician", 2),
             Row(None, None, 3),