cyschneck
commented
7 months ago Decorators are a common method to validate input
def validate_user_input(func):
def validate_wrapper(x, y):
if x < 0:
raise ValueError(f"x should be postive {func}")
return func(x, y)
return validate_wrapper
@validate_user_input
def average(x, y):
return (x+y)/2
average(-1, 0)Validation output:
File "/home/user/Desktop/testing.py", line 12, in <module>
average(-1, 0)
File "/home/user/Desktop/testing.py", line 4, in validate_wrapper
raise ValueError(f"x should be postive {func}")
ValueError: x should be postive <function average at 0x7f485d119f80>Can be expanded for **kwargs and class inheritance
def validate_user_input(*types):
def validate_acceptable_inputs(func):
def validate_wrapper(*args, **kwargs):
for (a, t) in zip(args, types):
assert isinstance(a, t), f"arg {a} does not match expected type {t}"
return func(*args, **kwargs)
return validate_wrapper
return validate_acceptable_inputs
@validate_user_input((int, float), (int, float))
def testing(x, y):
return (x * y)
print(testing("3", 2))Without validation, output=33, with validation that only accepts int and float types
File "/home/user/Desktop/testing.py", line 14, in <module>
print(testing("3", 2))
^^^^^^^^^^^^^^^
File "/home/user/Desktop/testing.py", line 5, in validate_wrapper
assert isinstance(a, t), f"arg {a} does not match expected type {t}"
AssertionError: arg 3 does not match expected type (<class 'int'>, <class 'float'>)
Describe the feature you'd like added to this project Expand input validation for existing geoCAT-comp functions with new associated testing
Describe the solution you'd like Expand existing input validation (i.e., float/int vs xarray) across geoCAT-comp functionality with some reasonable custom messages for common errors. New validation should be associated with new testing
Additional context Opened in favor of #254