Add "force_suffixes" flag to pd.merge

Author: kopytjuk

pandas/core/frame.py

@@ -11046,6 +11046,7 @@ def merge(
         right_index: bool = False,
         sort: bool = False,
         suffixes: Suffixes = ("_x", "_y"),
+        force_suffixes: bool = False,
         copy: bool | lib.NoDefault = lib.no_default,
         indicator: str | bool = False,
         validate: MergeValidate | None = None,
@@ -11065,6 +11066,7 @@ def merge(
             right_index=right_index,
             sort=sort,
             suffixes=suffixes,
+            force_suffixes=force_suffixes,
             indicator=indicator,
             validate=validate,
         )

pandas/core/reshape/merge.py

@@ -153,6 +153,7 @@ def merge(
     right_index: bool = False,
     sort: bool = False,
     suffixes: Suffixes = ("_x", "_y"),
+    force_suffixes: bool = False,
     copy: bool | lib.NoDefault = lib.no_default,
     indicator: str | bool = False,
     validate: str | None = None,
@@ -395,6 +396,7 @@ def merge(
             right_index=right_index,
             sort=sort,
             suffixes=suffixes,
+            force_suffixes=force_suffixes,
             indicator=indicator,
             validate=validate,
         )
@@ -411,6 +413,7 @@ def _cross_merge(
     right_index: bool = False,
     sort: bool = False,
     suffixes: Suffixes = ("_x", "_y"),
+    force_suffixes: bool = False,
     indicator: str | bool = False,
     validate: str | None = None,
 ) -> DataFrame:
@@ -447,6 +450,7 @@ def _cross_merge(
         right_index=right_index,
         sort=sort,
         suffixes=suffixes,
+        force_suffixes=force_suffixes,
         indicator=indicator,
         validate=validate,
     )
@@ -966,6 +970,7 @@ def __init__(
         right_index: bool = False,
         sort: bool = True,
         suffixes: Suffixes = ("_x", "_y"),
+        force_suffixes: bool = False,
         indicator: str | bool = False,
         validate: str | None = None,
     ) -> None:
@@ -978,6 +983,8 @@ def __init__(
         self.on = com.maybe_make_list(on)
 
         self.suffixes = suffixes
+        self.force_suffixes = force_suffixes
+
         self.sort = sort or how == "outer"
 
         self.left_index = left_index
@@ -1088,7 +1095,7 @@ def _reindex_and_concat(
         right = self.right[:]
 
         llabels, rlabels = _items_overlap_with_suffix(
-            self.left._info_axis, self.right._info_axis, self.suffixes
+            self.left._info_axis, self.right._info_axis, self.suffixes, self.force_suffixes
         )
 
         if left_indexer is not None and not is_range_indexer(left_indexer, len(left)):
@@ -3007,7 +3014,7 @@ def _validate_operand(obj: DataFrame | Series) -> DataFrame:
 
 
 def _items_overlap_with_suffix(
-    left: Index, right: Index, suffixes: Suffixes
+    left: Index, right: Index, suffixes: Suffixes, force_suffixes: bool = False
 ) -> tuple[Index, Index]:
     """
     Suffixes type validation.
@@ -3023,7 +3030,11 @@ def _items_overlap_with_suffix(
             "Provide 'suffixes' as a tuple instead."
         )
 
-    to_rename = left.intersection(right)
+    if not force_suffixes:
+        to_rename = left.intersection(right)
+    else:
+        to_rename = left.union(right)
+
     if len(to_rename) == 0:
         return left, right
 

pandas/tests/reshape/merge/test_merge.py

@@ -2369,6 +2369,52 @@ def test_merge_suffix(col1, col2, kwargs, expected_cols):
     tm.assert_frame_equal(result, expected)
 
 
+@pytest.mark.parametrize("force_suffixes", [False, True])
+def test_merge_suffix_with_force_simple(force_suffixes):
+    a = DataFrame({"A": [1, 2, 3, 98], "B": [4, 5, 6, 99], "ALPHABET": ["A", "B", "C", "Z"]})
+    b = DataFrame({"a": [1, 2, 3], "b": [4, 5, 6], "alphabet": ["a", "b", "c"]})
+
+    if force_suffixes:
+        expected = DataFrame([[1, 4, "A", 1, 4, "a"], [2, 5, "B", 2, 5, "b"], [3, 6, "C", 3, 6, "c"]], columns=["A_x", "B_x", "ALPHABET_x", "a_y", "b_y", "alphabet_y"])
+    else:
+        expected = DataFrame([[1, 4, "A", 1, 4, "a"], [2, 5, "B", 2, 5, "b"], [3, 6, "C", 3, 6, "c"]], columns=["A", "B", "ALPHABET", "a", "b", "alphabet"])
+
+    result = merge(a, b, left_on=["A", "B"], right_on=["a", "b"],
+                   force_suffixes=force_suffixes)
+    tm.assert_frame_equal(result, expected)
+
+@pytest.mark.parametrize(
+    "col1, col2, kwargs, expected_cols",
+    [
+        (0, 0, {"suffixes": ("", "_dup")}, ["0", "0_dup"]),
+        (0, 0, {"suffixes": (None, "_dup")}, [0, "0_dup"]),
+        (0, 0, {"suffixes": ("_x", "_y")}, ["0_x", "0_y"]),
+        (0, 0, {"suffixes": ["_x", "_y"]}, ["0_x", "0_y"]),
+        ("a", 0, {"suffixes": (None, "_y")}, ["a", "0_y"]),
+        (0.0, 0.0, {"suffixes": ("_x", None)}, ["0.0_x", 0.0]),
+        ("b", "b", {"suffixes": (None, "_y")}, ["b", "b_y"]),
+        ("a", "a", {"suffixes": ("_x", None)}, ["a_x", "a"]),
+        ("a", "b", {"suffixes": ("_x", None)}, ["a_x", "b"]),
+        ("a", "a", {"suffixes": (None, "_x")}, ["a", "a_x"]),
+        (0, 0, {"suffixes": ("_a", None)}, ["0_a", 0]),
+        ("a", "a", {}, ["a_x", "a_y"]),
+        (0, 0, {}, ["0_x", "0_y"]),
+    ],
+)
+def test_merge_suffix_with_force(col1, col2, kwargs, expected_cols):
+    # issue: 24782
+    a = DataFrame({col1: [1, 2, 3]})
+    b = DataFrame({col2: [4, 5, 6]})
+
+    expected = DataFrame([[1, 4], [2, 5], [3, 6]], columns=expected_cols)
+
+    result = a.merge(b, left_index=True, right_index=True, force_suffixes=True, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+    result = merge(a, b, left_index=True, right_index=True, force_suffixes=True, **kwargs)
+    tm.assert_frame_equal(result, expected)
+
+
 @pytest.mark.parametrize(
     "how,expected",
     [
@@ -2577,6 +2623,7 @@ def test_categorical_non_unique_monotonic(n_categories):
     tm.assert_frame_equal(expected, result)
 
 
+
 def test_merge_join_categorical_multiindex():
     # From issue 16627
     a = {