GH-39440: [Python] Calling pyarrow.dataset.ParquetFileFormat.make_write_options as a class method results in a segfault (#40976)

AlenkaF · pitrou · web-flow · commit 074d45f1c133 · 2024-04-05T10:54:14.000+02:00
### Rationale for this change Calling `make_write_options()` method as class instead of instance method results in segfault. ### What changes are included in this PR? Adds a type check on `self` and raises an error if not `ParquetFileFormat`. ### Are these changes tested? Yes. ### Are there any user-facing changes? No. * GitHub Issue: #39440 Lead-authored-by: AlenkaF <frim.alenka@gmail.com> Co-authored-by: Alenka Frim <AlenkaF@users.noreply.github.com> Co-authored-by: Antoine Pitrou <pitrou@free.fr> Signed-off-by: AlenkaF <frim.alenka@gmail.com>
diff --git a/python/pyarrow/_dataset_parquet.pyx b/python/pyarrow/_dataset_parquet.pyx
@@ -198,6 +198,10 @@ cdef class ParquetFileFormat(FileFormat):
         -------
         pyarrow.dataset.FileWriteOptions
         """
+        # Safeguard from calling make_write_options as a static class method
+        if not isinstance(self, ParquetFileFormat):
+            raise TypeError("make_write_options() should be called on "
+                            "an instance of ParquetFileFormat")
         opts = FileFormat.make_write_options(self)
         (<ParquetFileWriteOptions> opts).update(**kwargs)
         return opts
diff --git a/python/pyarrow/tests/test_dataset.py b/python/pyarrow/tests/test_dataset.py
@@ -5630,3 +5630,16 @@ def test_checksum_write_dataset_read_dataset_to_table(tempdir):
             corrupted_dir_path,
             format=pq_read_format_crc
         ).to_table()
+
+
+def test_make_write_options_error():
+    # GH-39440
+    msg = ("make_write_options\\(\\) should be called on an "
+           "instance of ParquetFileFormat")
+    with pytest.raises(TypeError, match=msg):
+        pa.dataset.ParquetFileFormat.make_write_options(43)
+
+    pformat = pa.dataset.ParquetFileFormat()
+    msg = "make_write_options\\(\\) takes exactly 0 positional arguments"
+    with pytest.raises(TypeError, match=msg):
+        pformat.make_write_options(43)
