使用set_index使用dask系列在dask dataframe上使用set_index时长度不匹配错误

Question

import pandas as pd
import numpy as np
import dask.array as dka
import dask.dataframe as dkf

vals = np.arange(200).reshape(20,10)
vals = dka.from_array(vals)
columns = [f"col_{i}" for i in range(10)]

ser = pd.Series(range(20), index=np.arange(20)+100, name='idx')

ddf = dkf.from_dask_array(vals, columns=columns)
dks = dkf.from_pandas(ser, npartitions=ddf.npartitions)

ddf.set_index(dks, sorted=True).compute()
-----------------
ValueError: Length mismatch: Expected 19 rows, received array of length 0

我对这个错误感到困惑，dks有20个元素，为什么dask抱怨它的长度为0？当ddf有20行时，为什么要期望19行？