ENH: try to preserve the dtype on combine_first for the case where the two DataFrame objects have the same columns

Author: danielhrisca

pandas/core/frame.py

@@ -6482,6 +6482,37 @@ def combiner(x, y):
 
             return expressions.where(mask, y_values, x_values)
 
+        if self.columns.equals(other.columns):
+
+            if self.empty and len(other) == other_idxlen:
+                return other.copy()
+            if self.index.equals(other.index):
+                return self.copy()
+            elif self.empty:
+                return other.copy()
+            elif other.empty:
+                return self.copy()
+            else:
+                new_index = self.index.union(other.index)
+                self_index = self.index
+                diff_index = other.index.difference(self.index)
+
+                columns = {}
+
+                for col in self.columns:
+                    this_vals = self[col]
+                    other_vals = other[col][diff_index]
+                    new_dtype = find_common_type([this_vals.dtype, other_vals.dtype])
+
+                    series = Series([this_vals.values[0]] * len(new_index), dtype=new_dtype, index=new_index)
+
+                    series[self_index] = this_vals
+                    series[diff_index] = other_vals
+
+                    columns[col] = series
+
+                return DataFrame.from_dict(columns)
+
         return self.combine(other, combiner, overwrite=False)
 
     def update(