[BUG] Avoid change_type mutating original DataFrame (#1162)

* lint via isort

* move function to outer

* avoid mutating original data

* add blank lines

* Add annotations for `_convert`

* Test it

* Update CHANGELOG.md

* convert each element to dtype

x isn't Series. it's the element of Series

* Docs update

* fix for health check

* Update test_case_when.py

Co-authored-by: Samuel Oranyeli <[email protected]>
Co-authored-by: Eric Ma <[email protected]>

Author: 3 people

CHANGELOG.md

@@ -21,6 +21,7 @@
 -   [ENH] Enable `encode_categorical` handle 2 (or more ) dimensions array. PR #1153 @Zeroto521
 -   [INF] Cancel old workflow runs via Github Action `concurrency`. PR #1161 @Zeroto521
 -   [ENH] Faster computation for non-equi join, with a numba engine. Issue #1102 @samukweku
+-   [BUG] Avoid `change_type` mutating original `DataFrame`. PR #1162 @Zeroto521
 
 ## [v0.23.1] - 2022-05-03
 

janitor/functions/change_type.py

@@ -1,4 +1,5 @@
 from typing import Hashable
+
 import pandas as pd
 import pandas_flavor as pf
 
@@ -15,7 +16,7 @@ def change_type(
 ) -> pd.DataFrame:
     """Change the type of a column.
 
-    This method mutates the original DataFrame.
+    This method does not mutate the original DataFrame.
 
     Exceptions that are raised can be ignored. For example, if one has a mixed
     dtype column that has non-integer strings and integers, and you want to
@@ -57,20 +58,24 @@ def change_type(
     :raises ValueError: If unknown option provided for
         `ignore_exception`.
     """
+
+    df = df.copy()  # avoid mutating the original DataFrame
     if not ignore_exception:
         df[column_name] = df[column_name].astype(dtype)
     elif ignore_exception == "keep_values":
         df[column_name] = df[column_name].astype(dtype, errors="ignore")
     elif ignore_exception == "fillna":
-
-        def convert(x, dtype):
-            """Casts item `x` to `dtype` or None if not possible."""
-            try:
-                return dtype(x)
-            except ValueError:
-                return None
-
-        df[column_name] = df[column_name].apply(lambda x: convert(x, dtype))
+        df[column_name] = df[column_name].apply(lambda x: _convert(x, dtype))
     else:
         raise ValueError("Unknown option for ignore_exception")
+
     return df
+
+
+def _convert(x, dtype: type):
+    """Casts item `x` to `dtype` or None if not possible."""
+
+    try:
+        return dtype(x)
+    except ValueError:
+        return None

tests/functions/test_case_when.py

@@ -76,8 +76,8 @@ def test_default_ndim(df):
 
 
 @pytest.mark.turtle
-@settings(deadline=None)
 @given(df=df_strategy())
+@settings(deadline=None)
 def test_default_length(df):
     """Raise ValueError if `default` length != len(df)."""
     assume(len(df) > 10)

tests/functions/test_change_type.py

@@ -1,6 +1,7 @@
 import numpy as np
 import pandas as pd
 import pytest
+from pandas.testing import assert_frame_equal
 
 
 @pytest.mark.functions
@@ -43,3 +44,14 @@ def test_change_type_raise_exception():
         df = df.change_type(
             column_name="col1", dtype=float, ignore_exception=False
         )
+
+
+@pytest.mark.functions
+def test_original_data_type(dataframe):
+    df = pd.DataFrame(range(3), columns=["col1"])
+    df_original = df.copy()
+
+    df.change_type("col1", dtype=str)
+
+    # 'cols' is still int type not str type
+    assert_frame_equal(df, df_original)