Python 3.7 considers r'\s*' an empty pattern match, instead of the previous behavior of matching whitespace, consequently, the pattern r'\s+' should be used in equuleus.
The had been addressed in vyos-1x, commit 69a14678de1c4f9b35414cbb00f33330478e4c58, but is documented here for reference.
Example:
Python 3.6 and below:
>>> import re >>> path = 'some path' >>> path_list = re.split(r'\s*', path) >>> path_list ['some', 'path']
Python 3.7 (equuleus):
>>> import re >>> path = 'some path' >>> path_list = re.split(r'\s*', path) >>> path_list ['', 's', 'o', 'm', 'e', '', 'p', 'a', 't', 'h', '']