ENH: Allow dictionaries to be passed to pandas.Series.str.replace

doc/source/whatsnew/v2.2.0.rst

@@ -223,6 +223,7 @@ Other enhancements
 - :func:`tseries.api.guess_datetime_format` is now part of the public API (:issue:`54727`)
 - :meth:`ExtensionArray._explode` interface method added to allow extension type implementations of the ``explode`` method (:issue:`54833`)
 - :meth:`ExtensionArray.duplicated` added to allow extension type implementations of the ``duplicated`` method (:issue:`55255`)
+- Allow dictionaries to be passed to :meth:`pandas.Series.str.replace` via ``pat`` parameter (:issue:`51748`)
 - Allow passing ``read_only``, ``data_only`` and ``keep_links`` arguments to openpyxl using ``engine_kwargs`` of :func:`read_excel` (:issue:`55027`)
 - DataFrame.apply now allows the usage of numba (via ``engine="numba"``) to JIT compile the passed function, allowing for potential speedups (:issue:`54666`)
 - Implement masked algorithms for :meth:`Series.value_counts` (:issue:`54984`)

pandas/core/strings/accessor.py

@@ -1395,8 +1395,8 @@ def fullmatch(self, pat, case: bool = True, flags: int = 0, na=None):
     @forbid_nonstring_types(["bytes"])
     def replace(
         self,
-        pat: str | re.Pattern,
-        repl: str | Callable,
+        pat: str | re.Pattern | dict | None = None,
+        repl: str | Callable | None = None,
         n: int = -1,
         case: bool | None = None,
         flags: int = 0,
@@ -1410,11 +1410,14 @@ def replace(
 
         Parameters
         ----------
-        pat : str or compiled regex
+        pat : str, compiled regex, or a dict
             String can be a character sequence or regular expression.
+            Dictionary contains <key : value> pairs of strings to be replaced
+            along with the updated value.
         repl : str or callable
             Replacement string or a callable. The callable is passed the regex
             match object and must return a replacement string to be used.
+            Must have a value of None if `pat` is a dict
             See :func:`re.sub`.
         n : int, default -1 (all)
             Number of replacements to make from start.
@@ -1448,6 +1451,7 @@ def replace(
             * if `regex` is False and `repl` is a callable or `pat` is a compiled
               regex
             * if `pat` is a compiled regex and `case` or `flags` is set
+            * if `pat` is a dictionary and `repl` is not None.
 
         Notes
         -----
@@ -1457,6 +1461,15 @@ def replace(
 
         Examples
         --------
+        When `pat` is a dictionary, every key in `pat` is replaced
+        with its corresponding value:
+
+        >>> pd.Series(['A', 'B', np.nan]).str.replace(pat={'A': 'a', 'B': 'b'})
+        0    a
+        1    b
+        2    NaN
+        dtype: object
+
         When `pat` is a string and `regex` is True, the given `pat`
         is compiled as a regex. When `repl` is a string, it replaces matching
         regex patterns as with :meth:`re.sub`. NaN value(s) in the Series are
@@ -1519,8 +1532,11 @@ def replace(
         2    NaN
         dtype: object
         """
+        if isinstance(pat, dict) and repl is not None:
+            raise ValueError("repl cannot be used when pat is a dictionary")
+
         # Check whether repl is valid (GH 13438, GH 15055)
-        if not (isinstance(repl, str) or callable(repl)):
+        if not isinstance(pat, dict) and not (isinstance(repl, str) or callable(repl)):
             raise TypeError("repl must be a string or callable")
 
         is_compiled_re = is_re(pat)
@@ -1540,10 +1556,21 @@ def replace(
         if case is None:
             case = True
 
-        result = self._data.array._str_replace(
-            pat, repl, n=n, case=case, flags=flags, regex=regex
-        )
-        return self._wrap_result(result)
+        if isinstance(pat, dict):
+            res_output = self._data
+            for key, value in pat.items():
+                result = res_output.array._str_replace(
+                    key, value, n=n, case=case, flags=flags, regex=regex
+                )
+                res_output = self._wrap_result(result)
+
+        else:
+            result = self._data.array._str_replace(
+                pat, repl, n=n, case=case, flags=flags, regex=regex
+            )
+            res_output = self._wrap_result(result)
+
+        return res_output
 
     @forbid_nonstring_types(["bytes"])
     def repeat(self, repeats):

pandas/tests/strings/test_find_replace.py

@@ -355,6 +355,21 @@ def test_endswith_nullable_string_dtype(nullable_string_dtype, na):
 # --------------------------------------------------------------------------------------
 # str.replace
 # --------------------------------------------------------------------------------------
+def test_replace_dict_invalid(any_string_dtype):
+    # GH 51914
+    series = Series(data=["A", "B_junk", "C_gunk"], name="my_messy_col")
+    msg = "repl cannot be used when pat is a dictionary"
+
+    with pytest.raises(ValueError, match=msg):
+        series.str.replace(pat={"A": "a", "B": "b"}, repl="A")
+
+
+def test_replace_dict(any_string_dtype):
+    # GH 51914
+    series = Series(data=["A", "B", "C"], name="my_messy_col")
+    new_series = series.str.replace(pat={"A": "a", "B": "b"})
+    expected = Series(data=["a", "b", "C"], name="my_messy_col")
+    tm.assert_series_equal(new_series, expected)
 
 
 def test_replace(any_string_dtype):