Inconsistency in strftime between python 3.6 and 3.7 when parsing first-century dates.

Python 3.6
>>> datetime.datetime.strptime(d, "%Y-%m-%d").strftime("%Y-%m-%d")
'0020-10-05'

Python 3.7
>>> datetime.datetime.strptime("0020-10-05", "%Y-%m-%d").strftime("%Y-%m-%d")
'20-10-05'

This means that the following would work for 3.6 but raise a ValueError for 3.7:

>>> d = "0020-10-05"
>>> d = datetime.datetime.strptime(d, "%Y-%m-%d").strftime("%Y-%m-%d")
>>> d = datetime.datetime.strptime(d, "%Y-%m-%d").strftime("%Y-%m-%d")

3.8.2 (on Alpine Linux under WSL) produces '0020-10-05', just like your 3.6 example. Not seeing anything obvious in commit history that would break it for 3.7. That said, 3.7 is in security fix only mode at this point (see https://devguide.python.org/#status-of-python-branches ); as this works on the latest release, I'm thinking this won't be fixed for 3.7.

The behavior you see with a Python 3.7 is not universal. For example, on macOS:

Python 3.7.9 (v3.7.9:13c94747c7, Aug 15 2020, 01:31:08)
[Clang 6.0 (clang-600.0.57)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import datetime
>>> datetime.datetime.strptime("0020-10-05", "%Y-%m-%d").strftime("%Y-%m-%d")
'0020-10-05'

So there is apparently something different in the environments between the Python 3.6 and 3.7 you are using, rather than an issue in Python itself. Perhaps a comparison of the outputs between:

python3.6 -m test.pythoninfo
python3.7 -m test.pythoninfo

will suggest something. In any case, as Josh notes, Python 3.7 is in the security-fix-only phase of its life cycle so even if there were an issue in Python itself it would likely not meet the criteria to be fixed in 3.7.