Lint and parameter description

Author: Jaspvr

pandas/core/generic.py

@@ -3859,6 +3859,11 @@ def to_csv(
 
         {storage_options}
 
+        preserve_complex : bool, default False
+            If True, arrays (e.g. NumPy arrays) or complex data are serialized and
+            reconstructed in a custom manner. If False (default), standard CSV
+            behavior is used.
+
         Returns
         -------
         None or str

pandas/io/parsers/readers.py

@@ -455,6 +455,11 @@ class _read_shared(TypedDict, Generic[HashableT], total=False):
 
 {storage_options}
 
+preserve_complex : bool, default False
+    If True, arrays (e.g. NumPy arrays) or complex data are serialized and
+    reconstructed in a custom manner. If False (default), standard CSV
+    behavior is used.
+
 dtype_backend : {{'numpy_nullable', 'pyarrow'}}
     Back-end data type applied to the resultant :class:`DataFrame`
     (still experimental). If not specified, the default behavior

pandas/tests/io/formats/test_to_csv.py

@@ -740,4 +740,4 @@ def test_to_csv_iterative_compression_buffer(compression):
         tm.assert_frame_equal(
             pd.read_csv(buffer, compression=compression, index_col=0), df
         )
-        assert not buffer.closed
+        assert not buffer.closed

scripts/tests/test_csv.py

@@ -1,7 +1,7 @@
-import os
 import tempfile
 
 import numpy as np
+
 import pandas as pd
 
 
@@ -11,42 +11,39 @@ def test_preserve_numpy_arrays_in_csv():
         "id": [1, 2],
         "embedding": [
             np.array([0.1, 0.2, 0.3]),
-            np.array([0.4, 0.5, 0.6])
+            np.array([0.4, 0.5, 0.6]),
         ],
     })
 
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as tmp:
+    with tempfile.NamedTemporaryFile(suffix=".csv") as tmp:
         path = tmp.name
-
-    try:
         df.to_csv(path, index=False, preserve_complex=True)
         df_loaded = pd.read_csv(path, preserve_complex=True)
-        assert isinstance(df_loaded["embedding"][0], np.ndarray), (
-            "Test Failed: The CSV did not preserve embeddings as NumPy arrays!"
-        )
-        print("PASS: test_preserve_numpy_arrays_in_csv")
-    finally:
-        os.remove(path)
+
+        assert isinstance(
+            df_loaded["embedding"][0], np.ndarray
+        ), "Test Failed: The CSV did not preserve embeddings as NumPy arrays!"
+
+    print("PASS: test_preserve_numpy_arrays_in_csv")
 
 
 def test_preserve_numpy_arrays_in_csv_empty_dataframe():
     print("\nRunning: test_preserve_numpy_arrays_in_csv_empty_dataframe")
     df = pd.DataFrame({"embedding": []})
     expected = "embedding\n"
 
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as tmp:
+    with tempfile.NamedTemporaryFile(suffix=".csv") as tmp:
         path = tmp.name
-
-    try:
         df.to_csv(path, index=False, preserve_complex=True)
         with open(path, encoding="utf-8") as f:
             result = f.read()
-        assert result == expected, (
-            f"CSV output mismatch for empty DataFrame.\nGot:\n{result}\nExpected:\n{expected}"
-        )
-        print("PASS: test_preserve_numpy_arrays_in_csv_empty_dataframe")
-    finally:
-        os.remove(path)
+
+    msg = (
+        f"CSV output mismatch for empty DataFrame.\n"
+        f"Got:\n{result}\nExpected:\n{expected}"
+    )
+    assert result == expected, msg
+    print("PASS: test_preserve_numpy_arrays_in_csv_empty_dataframe")
 
 
 def test_preserve_numpy_arrays_in_csv_mixed_dtypes():
@@ -56,30 +53,33 @@ def test_preserve_numpy_arrays_in_csv_mixed_dtypes():
         "name": ["alice", "bob"],
         "scores": [
             np.array([95.5, 88.0]),
-            np.array([76.0, 90.5])
+            np.array([76.0, 90.5]),
         ],
         "age": [25, 30],
     })
 
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".csv") as tmp:
+    with tempfile.NamedTemporaryFile(suffix=".csv") as tmp:
         path = tmp.name
-
-    try:
         df.to_csv(path, index=False, preserve_complex=True)
         df_loaded = pd.read_csv(path, preserve_complex=True)
-        assert isinstance(df_loaded["scores"][0], np.ndarray), (
-            "Failed: 'scores' column not deserialized as np.ndarray."
+
+        err_scores = "Failed: 'scores' column not deserialized as np.ndarray."
+        assert isinstance(df_loaded["scores"][0], np.ndarray), err_scores
+        assert df_loaded["id"].dtype == np.int64, (
+            "Failed: 'id' should still be int."
+        )
+        assert df_loaded["name"].dtype == object, (
+            "Failed: 'name' should still be object/string."
+        )
+        assert df_loaded["age"].dtype == np.int64, (
+            "Failed: 'age' should still be int."
         )
-        assert df_loaded["id"].dtype == np.int64, "Failed: 'id' should still be int."
-        assert df_loaded["name"].dtype == object, "Failed: 'name' should still be object/string."
-        assert df_loaded["age"].dtype == np.int64, "Failed: 'age' should still be int."
 
-        print("PASS: test_preserve_numpy_arrays_in_csv_mixed_dtypes")
-    finally:
-        os.remove(path)
+    print("PASS: test_preserve_numpy_arrays_in_csv_mixed_dtypes")
 
 
 if __name__ == "__main__":
     test_preserve_numpy_arrays_in_csv()
     test_preserve_numpy_arrays_in_csv_empty_dataframe()
     test_preserve_numpy_arrays_in_csv_mixed_dtypes()
+    print("\nDone.")