System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Mac OS Sierra 10.12.6
• Ray installed from (source or binary): Installed with pip pip install ray
• Ray version: 0.4.0
• Python version: Python 3.6.4 |Anaconda, Inc.|

Describe the problem

I am unable to get the ray.dataframe.read_csv to properly read a pipe delimited file into a dataframe. Reading a comma-separated file works fine.

Here are two dummy text files I created:

pipe_delimited.txt

a|b|c
1|2|3
4|5|6
7|8|9
10|11|12

comma_delimited.txt

a,b,c
1,2,3
4,5,6
7,8,9
10,11,12

>>> import ray.dataframe as pd

>>> df = pd.read_csv('pipe_delimited.txt', delimiter='|')
>>> print (df.shape)
>>> print (df.columns)
(4, 1)
Index(['a|b|c'], dtype='object')

>>> df = pd.read_csv('comma_delimited.txt', delimiter=',')
>>> print (df.shape)
>>> print (df.columns)
(4, 3)
Index(['a', 'b', 'c'], dtype='object')