[FIX] _builtin_table import in groupby apply (changed in pandas>=1.3.…

…0) (#2184)

I have recently had an issue while upgrading pandas to the latest version in my Databricks environment :

`AttributeError: type object 'SelectionMixin' has no attribute '_builtin_table'`


Pandas has recently refactored the way we import the _builtin_table and is now part of the `pandas.core.common` module instead of being an attribute of the `pandas.core.base.SelectionMixin` class.

PR that was merged in the 1.3.0 : pandas-dev/pandas#40857

This suggestion should solve the issue !

databricks/koalas/groupby.py

@@ -1105,7 +1105,12 @@ def apply(self, func, *args, **kwargs) -> Union[DataFrame, Series]:
         1    52
         Name: B, dtype: int64
         """
-        from pandas.core.base import SelectionMixin
+        if LooseVersion(pd.__version__) >= LooseVersion("1.3.0"):
+            from pandas.core.common import _builtin_table
+        else:
+            from pandas.core.base import SelectionMixin
+
+            _builtin_table = SelectionMixin._builtin_table
 
         if not isinstance(func, Callable):  # type: ignore
             raise TypeError("%s object is not callable" % type(func).__name__)
@@ -1133,9 +1138,9 @@ def apply(self, func, *args, **kwargs) -> Union[DataFrame, Series]:
 
         if is_series_groupby:
             name = kdf.columns[-1]
-            pandas_apply = SelectionMixin._builtin_table.get(func, func)
+            pandas_apply = _builtin_table.get(func, func)
         else:
-            f = SelectionMixin._builtin_table.get(func, func)
+            f = _builtin_table.get(func, func)
 
             def pandas_apply(pdf, *a, **k):
                 return f(pdf.drop(groupkey_names, axis=1), *a, **k)
@@ -1292,7 +1297,12 @@ def filter(self, func) -> Union[DataFrame, Series]:
         5    6
         Name: B, dtype: int64
         """
-        from pandas.core.base import SelectionMixin
+        if LooseVersion(pd.__version__) >= LooseVersion("1.3.0"):
+            from pandas.core.common import _builtin_table
+        else:
+            from pandas.core.base import SelectionMixin
+
+            _builtin_table = SelectionMixin._builtin_table
 
         if not isinstance(func, Callable):  # type: ignore
             raise TypeError("%s object is not callable" % type(func).__name__)
@@ -1324,7 +1334,7 @@ def pandas_filter(pdf):
                 return pd.DataFrame(pdf.groupby(groupkey_names)[pdf.columns[-1]].filter(func))
 
         else:
-            f = SelectionMixin._builtin_table.get(func, func)
+            f = _builtin_table.get(func, func)
 
             def wrapped_func(pdf):
                 return f(pdf.drop(groupkey_names, axis=1))