fix: allow RawIOBase streams in identify_stream method

python/src/magika/magika.py

@@ -192,7 +192,7 @@ def identify_stream(self, stream: BinaryIO) -> MagikaResult:
                 "Input stream must be opened in bytes mode, not in text mode."
             )
 
-        if not isinstance(stream, io.BufferedIOBase):
+        if not isinstance(stream, (io.BufferedIOBase, io.RawIOBase)):
             raise TypeError("Input stream must be a readable BinaryIO object.")
 
         if (

python/tests/test_magika_python_module.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 
 import io
+import os
 import signal
 import tempfile
 from pathlib import Path
@@ -242,6 +243,29 @@ def test_magika_module_identify_stream_does_not_alter_position() -> None:
         assert stream.tell() == pos
 
 
+def test_magika_module_identify_stream_accepts_raw_fileio() -> None:
+    m = Magika()
+
+    fd, tmp_path = tempfile.mkstemp()
+    os.close(fd)
+    try:
+        with open(tmp_path, "wb") as f:
+            f.write(b"print('hello')\n")
+
+        stream = io.FileIO(tmp_path, "rb")
+        try:
+            # Verify raw binary streams are accepted and position is restored.
+            expected_pos = 2
+            stream.seek(expected_pos)
+            res = m.identify_stream(stream)
+            assert res.ok
+            assert stream.tell() == expected_pos
+        finally:
+            stream.close()
+    finally:
+        os.unlink(tmp_path)
+
+
 def test_magika_module_with_whitespaces() -> None:
     m = Magika()