BUG: Raise MergeError when suffixes result in duplicate column names (#61422)

Author: Farsidetfs

doc/source/whatsnew/v3.0.0.rst

@@ -851,6 +851,7 @@ Reshaping
 - Bug in :meth:`DataFrame.unstack` producing incorrect results when manipulating empty :class:`DataFrame` with an :class:`ExtentionDtype` (:issue:`59123`)
 - Bug in :meth:`concat` where concatenating DataFrame and Series with ``ignore_index = True`` drops the series name (:issue:`60723`, :issue:`56257`)
 - Bug in :func:`melt` where calling with duplicate column names in ``id_vars`` raised a misleading ``AttributeError`` (:issue:`61475`)
+- Bug in :meth:`DataFrame.merge` where user-provided suffixes could result in duplicate column names if the resulting names matched existing columns. Now raises a :class:`MergeError` in such cases. (:issue:`61402`)
 
 Sparse
 ^^^^^^

pandas/core/reshape/merge.py

@@ -3062,13 +3062,16 @@ def renamer(x, suffix: str | None):
     if not llabels.is_unique:
         # Only warn when duplicates are caused because of suffixes, already duplicated
         # columns in origin should not warn
-        dups = llabels[(llabels.duplicated()) & (~left.duplicated())].tolist()
+        dups.extend(llabels[(llabels.duplicated()) & (~left.duplicated())].tolist())
     if not rlabels.is_unique:
         dups.extend(rlabels[(rlabels.duplicated()) & (~right.duplicated())].tolist())
+    # Suffix addition creates duplicate to pre-existing column name
+    dups.extend(llabels.intersection(right.difference(to_rename)).tolist())
+    dups.extend(rlabels.intersection(left.difference(to_rename)).tolist())
     if dups:
         raise MergeError(
             f"Passing 'suffixes' which cause duplicate columns {set(dups)} is "
-            f"not allowed.",
+            "not allowed.",
         )
 
     return llabels, rlabels

pandas/tests/reshape/merge/test_merge.py

@@ -3060,3 +3060,12 @@ def test_merge_on_all_nan_column():
         {"x": [1, 2, 3], "y": [np.nan, np.nan, np.nan], "z": [4, 5, 6], "zz": [4, 5, 6]}
     )
     tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("suffixes", [("_dup", ""), ("", "_dup")])
+def test_merge_for_suffix_collisions(suffixes):
+    # GH#61402
+    df1 = DataFrame({"col1": [1], "col2": [2]})
+    df2 = DataFrame({"col1": [1], "col2": [2], "col2_dup": [3]})
+    with pytest.raises(MergeError, match="duplicate columns"):
+        merge(df1, df2, on="col1", suffixes=suffixes)