[ENH] Add optional removal of accents on functions.clean_names, enabled by default. (#506)

Author: mralbu

AUTHORS.rst

@@ -76,3 +76,4 @@ Contributors
 - `@puruckertom <https://github.com/puruckertom>`_ | `contributions <https://github.com/ericmjl/pyjanitor/pulls?utf8=%E2%9C%93&q=is%3Apr+author%3Apuruckertom>`_
 - `@thomasjpfan <https://github.com/thomasjpfan>`_ | `contributions <https://github.com/ericmjl/pyjanitor/issues?q=is%3Aclosed+mentions%3Athomasjpfan>`_
 - `@jiafengkevinchen <https://github.com/jiafengkevinchen>`_ | `contributions <https://github.com/ericmjl/pyjanitor/pull/480#issue-298730562>`_
+- `@mralbu <https://github.com/mralbu>`_ | `contributions <https://github.com/ericmjl/pyjanitor/issues/502>`_

CHANGELOG.rst

@@ -3,7 +3,7 @@ v0.18.1 (on deck)
 - [ENH] add preserve_position kwarg to deconcatenate_column with tests by @shandou and @ericmjl
 - [DOC] add contributions that did not leave ``git`` traces by @ericmjl
 - [ENH] add inflation adjustment in finance submodule by @rahosbach
-
+- [ENH] add optional removal of accents on functions.clean_names, enabled by default by @mralbu
 
 For changes that happened prior to v0.18.1,
 please consult the closed PRs,

janitor/functions.py

@@ -2,6 +2,7 @@
 
 import datetime as dt
 import re
+import unicodedata
 import warnings
 from fnmatch import translate
 from functools import partial, reduce
@@ -121,6 +122,7 @@ def clean_names(
     strip_underscores: str = None,
     case_type: str = "lower",
     remove_special: bool = False,
+    strip_accents: bool = True,
     preserve_original_columns: bool = True,
 ) -> pd.DataFrame:
     """
@@ -189,6 +191,9 @@ def _remove_special(col):
     if remove_special:
         df = df.rename(columns=_remove_special)
 
+    if strip_accents:
+        df = df.rename(columns=_strip_accents)
+
     df = df.rename(columns=lambda x: re.sub("_+", "_", x))
     df = _strip_underscores(df, strip_underscores)
 
@@ -208,6 +213,18 @@ def _normalize_1(col_name: str) -> str:
     return result
 
 
+def _strip_accents(col_name: str) -> str:
+    """
+    Removes accents from a DataFrame column name.
+    .. _StackOverflow: https://stackoverflow.com/questions/517923/what-is-the-best-way-to-remove-accents-in-a-python-unicode-string # noqa: E501
+    """
+    return "".join(
+        l
+        for l in unicodedata.normalize("NFD", col_name)
+        if not unicodedata.combining(l)
+    )
+
+
 @pf.register_dataframe_method
 def remove_empty(df: pd.DataFrame) -> pd.DataFrame:
     """

tests/functions/test_clean_names.py

@@ -104,6 +104,14 @@ def test_clean_names_strip_underscores(
     assert set(df.columns) == set(expected_columns)
 
 
+@pytest.mark.functions
+def test_clean_names_strip_accents():
+    df = pd.DataFrame({"João": [1, 2], "Лука́ся": [1, 2], "Käfer": [1, 2]})
+    df = df.clean_names(strip_accents=True)
+    expected_columns = ["joao", "лукася", "kafer"]
+    assert set(df.columns) == set(expected_columns)
+
+
 @pytest.mark.functions
 def test_incorrect_strip_underscores(multiindex_dataframe):
     with pytest.raises(JanitorError):