Backport PR #40604: REGR: replace with multivalued regex raising (#40648)

Co-authored-by: Matthew Zeitlin <[email protected]>

Author: simonjayhawkins

doc/source/whatsnew/v1.2.4.rst

@@ -18,6 +18,7 @@ Fixed regressions
 - Fixed regression in :meth:`DataFrame.sum` when ``min_count`` greater than the :class:`DataFrame` shape was passed resulted in a ``ValueError`` (:issue:`39738`)
 - Fixed regression in :meth:`DataFrame.to_json` raising ``AttributeError`` when run on PyPy (:issue:`39837`)
 - Fixed regression in :meth:`DataFrame.where` not returning a copy in the case of an all True condition (:issue:`39595`)
+- Fixed regression in :meth:`DataFrame.replace` raising ``IndexError`` when ``regex`` was a multi-key dictionary (:issue:`39338`)
 -
 
 .. ---------------------------------------------------------------------------

pandas/core/internals/blocks.py

@@ -894,10 +894,20 @@ def comp(s: Scalar, mask: np.ndarray, regex: bool = False) -> np.ndarray:
 
         rb = [self if inplace else self.copy()]
         for i, (src, dest) in enumerate(pairs):
+            convert = i == src_len  # only convert once at the end
             new_rb: List["Block"] = []
-            for blk in rb:
-                m = masks[i]
-                convert = i == src_len  # only convert once at the end
+
+            # GH-39338: _replace_coerce can split a block into
+            # single-column blocks, so track the index so we know
+            # where to index into the mask
+            for blk_num, blk in enumerate(rb):
+                if len(rb) == 1:
+                    m = masks[i]
+                else:
+                    mib = masks[i]
+                    assert not isinstance(mib, bool)
+                    m = mib[blk_num : blk_num + 1]
+
                 result = blk._replace_coerce(
                     to_replace=src,
                     value=dest,

pandas/tests/frame/methods/test_replace.py

@@ -644,6 +644,28 @@ def test_regex_replace_numeric_to_object_conversion(self, mix_abc):
         tm.assert_frame_equal(res, expec)
         assert res.a.dtype == np.object_
 
+    @pytest.mark.parametrize(
+        "to_replace", [{"": np.nan, ",": ""}, {",": "", "": np.nan}]
+    )
+    def test_joint_simple_replace_and_regex_replace(self, to_replace):
+        # GH-39338
+        df = DataFrame(
+            {
+                "col1": ["1,000", "a", "3"],
+                "col2": ["a", "", "b"],
+                "col3": ["a", "b", "c"],
+            }
+        )
+        result = df.replace(regex=to_replace)
+        expected = DataFrame(
+            {
+                "col1": ["1000", "a", "3"],
+                "col2": ["a", np.nan, "b"],
+                "col3": ["a", "b", "c"],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
     @pytest.mark.parametrize("metachar", ["[]", "()", r"\d", r"\w", r"\s"])
     def test_replace_regex_metachar(self, metachar):
         df = DataFrame({"a": [metachar, "else"]})