TypeError: a bytes-like object is required, not 'str' when loading pickle in text mode

[L3ster1337]

I am encountering a TypeError when attempting to load a pickle file using lsf_runner.py. The code in lsf_runner.py opens the job-instance.pickle file in text mode ("r") and then calls pickle.load(). In Python 3, pickle.load() expects a binary stream, and opening the file in text mode results in a TypeError: a bytes-like object is required, not 'str'.

Steps to Reproduce:

Run lsf_runner.py with a job-instance.pickle file that has been generated in any manner (binary or ASCII protocol).
Observe that lsf_runner.py opens job-instance.pickle using open("job-instance.pickle", "r").
pickle.load() (or dill.load()) fails with TypeError: a bytes-like object is required, not 'str'.

Expected Behavior:
pickle.load() should successfully unpickle the object.

Actual Behavior:
The unpickling fails with a TypeError due to the file being opened in text mode.

Environment:

Python 3.x
dill / pickle module
lsf_runner.py script as provided by the repository.

Possible Solution:
Change the file opening mode in lsf_runner.py from "r" to "rb". For example:

with open("job-instance.pickle", "rb") as pickle_file_handle:
    job = pickle.load(pickle_file_handle)