[BUG] Unable to create an IntervalIndex from a list of Interval scalars

[galipremsagar]

Steps/Code to reproduce bug

In [2]: import numpy as np

In [3]: from pandas._libs.interval import Interval

In [4]: x = [np.nan, Interval(2.0, 3.0, closed='right'), Interval(3.0, 4.0, closed='right')]

In [5]: import pandas as pd

In [6]: pd.Index(x)
Out[6]: IntervalIndex([nan, (2.0, 3.0], (3.0, 4.0]], dtype='interval[float64, right]')

In [7]: import cudf
cudf.
In [8]: cudf.Index(x)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/column/column.py:2268, in as_column(arbitrary, nan_as_null, dtype, length)
   2266 try:
   2267     data = as_column(
-> 2268         memoryview(arbitrary), dtype=dtype, nan_as_null=nan_as_null
   2269     )
   2270 except TypeError:

TypeError: memoryview: a bytes-like object is required, not 'list'

During handling of the above exception, another exception occurred:

ArrowInvalid                              Traceback (most recent call last)
File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/column/column.py:2396, in as_column(arbitrary, nan_as_null, dtype, length)
   2392         pa_type = np_to_pa_dtype(
   2393             _maybe_convert_to_default_type("float")
   2394         )
-> 2396 pyarrow_array = pa.array(
   2397     arbitrary,
   2398     type=pa_type,
   2399     from_pandas=True if nan_as_null is None else nan_as_null,
   2400 )
   2402 if (
   2403     isinstance(arbitrary, pd.Index)
   2404     and arbitrary.dtype == cudf.dtype("object")
   (...)
   2407     )
   2408 ):

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/array.pxi:327, in pyarrow.lib.array()

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/array.pxi:39, in pyarrow.lib._sequence_to_array()

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/error.pxi:144, in pyarrow.lib.pyarrow_internal_check_status()

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/error.pxi:100, in pyarrow.lib.check_status()

ArrowInvalid: Could not convert Interval(2.0, 3.0, closed='right') with type pandas._libs.interval.Interval: did not recognize Python value type when inferring an Arrow data type

During handling of the above exception, another exception occurred:

ArrowInvalid                              Traceback (most recent call last)
Cell In[8], line 1
----> 1 cudf.Index(x)

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/nvtx/nvtx.py:101, in annotate.__call__.<locals>.inner(*args, **kwargs)
     98 @wraps(func)
     99 def inner(*args, **kwargs):
    100     libnvtx_push_range(self.attributes, self.domain.handle)
--> 101     result = func(*args, **kwargs)
    102     libnvtx_pop_range(self.domain.handle)
    103     return result

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/index.py:3441, in Index.__new__(cls, data, dtype, copy, name, tupleize_cols, nan_as_null, **kwargs)
   3436 if tupleize_cols is not True:
   3437     raise NotImplementedError(
   3438         "tupleize_cols != True is not yet supported"
   3439     )
-> 3441 return as_index(
   3442     data,
   3443     copy=copy,
   3444     dtype=dtype,
   3445     name=name,
   3446     nan_as_null=nan_as_null,
   3447     **kwargs,
   3448 )

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/nvtx/nvtx.py:101, in annotate.__call__.<locals>.inner(*args, **kwargs)
     98 @wraps(func)
     99 def inner(*args, **kwargs):
    100     libnvtx_push_range(self.attributes, self.domain.handle)
--> 101     result = func(*args, **kwargs)
    102     libnvtx_pop_range(self.domain.handle)
    103     return result

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/index.py:3340, in as_index(arbitrary, nan_as_null, **kwargs)
   3337 elif isinstance(arbitrary, cudf.DataFrame):
   3338     return cudf.MultiIndex.from_frame(arbitrary)
   3339 return as_index(
-> 3340     column.as_column(
   3341         arbitrary, dtype=kwargs.get("dtype", None), nan_as_null=nan_as_null
   3342     ),
   3343     **kwargs,
   3344 )

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/column/column.py:2443, in as_column(arbitrary, nan_as_null, dtype, length)
   2437     return cudf.core.column.ListColumn.from_sequences(
   2438         arbitrary
   2439     )
   2440 elif isinstance(arbitrary, abc.Iterable) or isinstance(
   2441     arbitrary, abc.Sequence
   2442 ):
-> 2443     data = as_column(
   2444         _construct_array(arbitrary, dtype),
   2445         dtype=dtype,
   2446         nan_as_null=nan_as_null,
   2447     )
   2448 else:
   2449     raise e

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/cudf/core/column/column.py:2171, in as_column(arbitrary, nan_as_null, dtype, length)
   2169         data = data.astype(dtype)
   2170 elif arb_dtype.kind in ("O", "U"):
-> 2171     data = as_column(pa.array(arbitrary), dtype=arbitrary.dtype)
   2172     # There is no cast operation available for pa.Array from int to
   2173     # str, Hence instead of handling in pa.Array block, we
   2174     # will have to type-cast here.
   2175     if dtype is not None:

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/array.pxi:323, in pyarrow.lib.array()

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/array.pxi:83, in pyarrow.lib._ndarray_to_array()

File /nvme/0/pgali/envs/cudfdev/lib/python3.10/site-packages/pyarrow/error.pxi:100, in pyarrow.lib.check_status()

ArrowInvalid: Could not convert Interval(2.0, 3.0, closed='right') with type pandas._libs.interval.Interval: tried to convert to double

Expected behavior

In [6]: cudf.Index(x)
Out[6]: IntervalIndex([nan, (2.0, 3.0], (3.0, 4.0]], dtype='interval[float64, right]')

Environment overview (please complete the following information)

• Environment location: [Bare-metal]
• Method of cuDF install: [from source]