sqlframe 3.31.4__tar.gz → 3.32.0__tar.gz

{sqlframe-3.31.4 → sqlframe-3.32.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.31.4
+Version: 3.32.0
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.31.4 → sqlframe-3.32.0}/sqlframe/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '3.31.4'
-__version_tuple__ = version_tuple = (3, 31, 4)
+__version__ = version = '3.32.0'
+__version_tuple__ = version_tuple = (3, 32, 0)

{sqlframe-3.31.4 → sqlframe-3.32.0}/sqlframe/base/dataframe.py

@@ -260,6 +260,13 @@ class BaseDataFrame(t.Generic[SESSION, WRITER, NA, STAT, GROUP_DATA]):
     def __copy__(self):
         return self.copy()
 
+    def __repr__(self) -> str:
+        fields = [f"{field.name}: {field.dataType}" for field in self.schema]
+        return "DataFrame[" + ", ".join(fields) + "]"
+
+    def _display_(self) -> str:
+        return self.__repr__()
+
     @property
     def _typed_columns(self) -> t.List[CatalogColumn]:
         raise NotImplementedError

{sqlframe-3.31.4 → sqlframe-3.32.0}/sqlframe/base/functions.py

@@ -494,21 +494,21 @@ def skewness(col: ColumnOrName) -> Column:
         func_name = "SKEW"
 
     if session._is_duckdb or session._is_snowflake:
+        col = Column.ensure_col(col)
         when_func = get_func_from_session("when")
         count_func = get_func_from_session("count")
-        count_star = count_func("*")
+        count_col = count_func(col)
         lit_func = get_func_from_session("lit")
         sqrt_func = get_func_from_session("sqrt")
-        col = Column.ensure_col(col)
         full_calc = (
             Column.invoke_anonymous_function(col, func_name)
-            * (count_star - lit_func(2))
-            / (sqrt_func(count_star * (count_star - lit_func(1))))
+            * (count_col - lit_func(2))
+            / (sqrt_func(count_col * (count_col - lit_func(1))))
         )
         return (
-            when_func(count_star == lit_func(0), lit_func(None))
-            .when(count_star == lit_func(1), lit_func(float("nan")))
-            .when(count_star == lit_func(2), lit_func(None))
+            when_func(count_col == lit_func(0), lit_func(None))
+            .when(count_col == lit_func(1), lit_func(None))
+            .when(count_col == lit_func(2), lit_func(0.0))
             .otherwise(full_calc)
         )
 

{sqlframe-3.31.4 → sqlframe-3.32.0}/sqlframe.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sqlframe
-Version: 3.31.4
+Version: 3.32.0
 Summary: Turning PySpark Into a Universal DataFrame API
 Home-page: https://github.com/eakmanrq/sqlframe
 Author: Ryan Eakman

{sqlframe-3.31.4 → sqlframe-3.32.0}/tests/integration/engines/test_int_functions.py

@@ -694,6 +694,8 @@ def test_skewness(get_session_and_func):
     assert math.isclose(df.select(skewness(df.c)).first()[0], 0.7071067811865475, rel_tol=1e-5)
     df = session.createDataFrame([{"a": 2.0}, {"a": None}])
     assert df.agg(skewness(df.a)).collect() == [Row(value=None)]
+    df = session.createDataFrame([{"a": 1}, {"a": 2}])
+    assert df.agg(skewness(df.a)).collect() == [Row(value=0.0)]
 
 
 def test_kurtosis(get_session_and_func):