>>> import numpy as np
>>> dt = np.dtype([('x', np.int8), ('y', np.float32)])
>>> arr = np.arange(5*10, dtype=np.int8).view(dt)
>>> pa.deserialize(pa.serialize(arr).to_buffer())
Traceback (most recent call last):
File "<ipython-input-13-6810c780d033>", line 1, in <module>
pa.deserialize(pa.serialize(arr).to_buffer())
File "serialization.pxi", line 441, in pyarrow.lib.deserialize
File "serialization.pxi", line 404, in pyarrow.lib.deserialize_from
File "serialization.pxi", line 257, in pyarrow.lib.SerializedPyObject.deserialize
File "serialization.pxi", line 174, in pyarrow.lib.SerializationContext._deserialize_callback
File "/home/antoine/arrow/python/pyarrow/serialization.py", line 44, in _deserialize_numpy_array_list
return np.array(data[0], dtype=np.dtype(data[1]))
TypeError: a bytes-like object is required, not 'int'
Bug
Major
