Random qualifiers being added to dataframes since 21.0.0

[andygrove]

Describe the bug

I am trying to upgrade the DataFusion Python bindings and running into the following error. The schema name changes on each run, so it seems to be auto-generated somehow.

E       assert "Schema error: No field named 'c'" in 'Schema error: No field named "c". Valid fields are "c056f9f370d364d4ea08482ce43090931"."a", "c056f9f370d364d4ea08482ce43090931"."b".'

Here is the code that is used to create the dataframe:

    batch = pa.RecordBatch.from_arrays(
        [pa.array([1, 2, 3]), pa.array([4, 4, 6])],
        names=["a", "b"],
    )
    return ctx.create_dataframe([[batch]])

The column names are not qualified, so I am not sure why these random qualifiers are being added.

Perhaps I need to make changes in the Python context?

To Reproduce

See apache/datafusion-python#301

Expected behavior

No response

Additional context

No response

[andygrove]

Never mind, the uuids are coming from the Python context.