Often when defining custom objects we end up testing that the __eq__ raises a particular error for unsupported comparisons. In practice, this means we have an unused comparison that triggers the B015 rule—see the MWE below. I wonder if there's a recommended way to get around this (without suppressing B015 altogether), or whether it could say be disabled within a with pytest.raises context.
Minimal working example:
# B015.py
import pytest
class MyClass:
def __init__(self, x):
self.x = x
def __eq__(self, other):
if not isinstance(other, MyClass):
raise TypeError(f"Cannot compare with {type(other)}")
return self.x == other.x
a = MyClass(5)
b = MyClass(5)
c = MyClass(4)
assert a == b
assert a != c
with pytest.raises(TypeError):
a == a.x
$ flake8 B015.py
B015.py:21:5: B015 Result of comparison is not used. This line doesn't do anything. Did you intend to prepend it with assert?
Often when defining custom objects we end up testing that the
__eq__
raises a particular error for unsupported comparisons. In practice, this means we have an unused comparison that triggers theB015
rule—see the MWE below. I wonder if there's a recommended way to get around this (without suppressingB015
altogether), or whether it could say be disabled within awith pytest.raises
context.Minimal working example: