add docstring, use ABCSeries

Author: ResidentMario

Diff for: pandas/core/missing.py

@@ -15,6 +15,7 @@
                                  needs_i8_conversion, is_integer,
                                  is_period_arraylike)
 from pandas.types.missing import isnull, valid_fill_value
+from pandas.types.generic import ABCSeries
 
 
 def mask_missing(arr, values_to_mask):
@@ -622,10 +623,38 @@ def fill_zeros(result, x, y, name, fill):
 
 def maybe_fill(obj, value, errors):
     """
+
+    Fillna error coercion routine.
+
+    Construct a piecewise polynomial in the Bernstein basis, compatible
+    with the specified values and derivatives at breakpoints.
+
+    Parameters
+    ----------
+    obj : Series of DataFrame
+        The Series or DataFrame for which a fill value is being evaluated.
+        If obj is a DataFrame this method simply returns True (e.g. the fillna
+        operation is allowed to continue) because it will be broken up and
+        parsed as a sequence of sub-Series later on.
+    value : object
+        The value to be used as a fill for the object.
+    errors : {'raise', 'coerce', or 'ignore'}
+        How invalid fill values will be handled. If 'raise', a TypeError will
+        be raised. If 'coerce', the column need to be coerced to a more
+        general dtype; the fillna op simply continues. If 'ignore', the fill
+        value will not be applied to the column: this method will pass fillna
+        a flag that tells it to terminate immediately.
+
+    Returns
+    -------
+    continue : bool
+        Whether or not, based on the values and the error mode, the fill
+        operation ought to continue.
+    """
+    """
     fillna error coercion routine, returns whether or not to continue.
     """
-    from pandas import Series
-    if isinstance(obj, Series):
+    if isinstance(obj, ABCSeries):
         if errors is None or errors == 'coerce':
             return True
         else: