Any ideas on why I get this error?
My project was working fine. I copied it to an external drive and onto my laptop to work on the road; it worked fine. I copied it back to my desktop and had a load of issues with invalid interpreters etc, so I made a new project and copied just the scripts in, made a new requirements.txt and installed all the packages, but when I run it, I get this error:
Traceback (most recent call last):
File "E:\Dev\spot_new\flask_blog\run.py", line 1, in <module>
from flaskblog import app
File "E:\Dev\spot_new\flask_blog\flaskblog\__init__.py", line 3, in <module>
from flask_bcrypt import Bcrypt
File "E:\Dev\spot_new\venv\lib\site-packages\flask_bcrypt.py", line 21, in <module>
from werkzeug.security import safe_str_cmp
ImportError: cannot import name 'safe_str_cmp' from 'werkzeug.security' (E:\Dev\spot_new\venv\lib\site-packages\werkzeug\security.py)
I've tried uninstalling Python, Anaconda, PyCharm, deleting every reg key and environment variable I can find that looks pythonic, reinstalling all from scratch but still no dice.
Werkzeug released v2.1.0 today, removing werkzeug.security.safe_str_cmp.
You can probably resolve this issue by pinning Werkzeug~=2.0.0 in your requirements.txt file (or similar).
pip install Werkzeug==2.0.0
After that it is likely that you will also have an AttributeError related to the jinja package, so if you have it, run:
pip install jinja2==3.0.3
Werkzeug 2.1.0 release notes recommend using the hmap equivalent. For reference, here is the implementation of safe_str_cmp from wekzeug 2.0.x, and here is a stripped-down version:
import hmac
def safe_str_cmp(a: str, b: str) -> bool:
"""This function compares strings in somewhat constant time. This
requires that the length of at least one string is known in advance.
Returns `True` if the two strings are equal, or `False` if they are not.
"""
if isinstance(a, str):
a = a.encode("utf-8") # type: ignore
if isinstance(b, str):
b = b.encode("utf-8") # type: ignore
return hmac.compare_digest(a, b)
or even more stripped-down one:
import hmac
str_to_bytes = lambda s: s.encode("utf-8") if isinstance(s, str) else s
safe_str_cmp = lambda a, b: hmac.compare_digest(str_to_bytes(a), str_to_bytes(b))