Backport PR #32577: REG: Restore read_csv function for some file-likes (#32609)

Co-authored-by: gfyoung <[email protected]>

Author: meeseeksmachine

doc/source/whatsnew/v1.0.2.rst

@@ -25,6 +25,7 @@ Fixed regressions
 - Fixed regression in :meth:`pandas.core.groupby.GroupBy.agg` calling a user-provided function an extra time on an empty input (:issue:`31760`)
 - Joining on :class:`DatetimeIndex` or :class:`TimedeltaIndex` will preserve ``freq`` in simple cases (:issue:`32166`)
 - Fixed bug in the repr of an object-dtype :class:`Index` with bools and missing values (:issue:`32146`)
+- Fixed regression in :meth:`read_csv` in which the ``encoding`` option was not recognized with certain file-like objects (:issue:`31819`)
 -
 
 .. ---------------------------------------------------------------------------

pandas/_libs/parsers.pyx

@@ -638,7 +638,8 @@ cdef class TextReader:
                 raise ValueError(f'Unrecognized compression type: '
                                  f'{self.compression}')
 
-            if self.encoding and isinstance(source, (io.BufferedIOBase, io.RawIOBase)):
+            if (self.encoding and hasattr(source, "read") and
+                    not hasattr(source, "encoding")):
                 source = io.TextIOWrapper(
                     source, self.encoding.decode('utf-8'), newline='')
 

pandas/io/parsers.py

@@ -5,7 +5,7 @@
 from collections import abc, defaultdict
 import csv
 import datetime
-from io import BufferedIOBase, RawIOBase, StringIO, TextIOWrapper
+from io import StringIO, TextIOWrapper
 import re
 import sys
 from textwrap import fill
@@ -1876,7 +1876,7 @@ def __init__(self, src, **kwds):
 
             # Handle the file object with universal line mode enabled.
             # We will handle the newline character ourselves later on.
-            if isinstance(src, (BufferedIOBase, RawIOBase)):
+            if hasattr(src, "read") and not hasattr(src, "encoding"):
                 src = TextIOWrapper(src, encoding=encoding, newline="")
 
             kwds["encoding"] = "utf-8"

pandas/tests/io/parser/test_encoding.py

@@ -175,3 +175,25 @@ def test_encoding_temp_file(all_parsers, utf_value, encoding_fmt, pass_encoding)
 
         result = parser.read_csv(f, encoding=encoding if pass_encoding else None)
         tm.assert_frame_equal(result, expected)
+
+
+def test_encoding_named_temp_file(all_parsers):
+    # see gh-31819
+    parser = all_parsers
+    encoding = "shift-jis"
+
+    if parser.engine == "python":
+        pytest.skip("NamedTemporaryFile does not work with Python engine")
+
+    title = "てすと"
+    data = "こむ"
+
+    expected = DataFrame({title: [data]})
+
+    with tempfile.NamedTemporaryFile() as f:
+        f.write(f"{title}\n{data}".encode(encoding))
+
+        f.seek(0)
+
+        result = parser.read_csv(f, encoding=encoding)
+        tm.assert_frame_equal(result, expected)