3 reasons why I think this approach should not be encouraged and should have 0 on Best Practices:

1. Security risk: Using eval() can be dagerous, especially if the input comes from an untrusted source. It can execute arbitrary code, leading to potential security vulnerabilities.
2. Hard to Debug: Errors in dynamically evaluated expressions can be harder to trace compared to explicit conditional branches.
3. Performance: eval() can be slower than direct operations due to the overhead of parsing and interpreting the string.

f meams that it is f-string and all variabels in it convert to string

i would suggest you take a deep look for logical errors... it happened to me also ... errors are most likely to be found in expressions where your variable names are different than the code youre comparing
... otherwise ...
use chat gpt ... it may suggest some minor changes that you may have missed

I've found something similar too!