Decorators with Arguments¶
If decorators are to be configurable via additional decorator parameters, the static decoration scheme according to Pyc. 2 suggests the extended form,
@Decorator(*deco_args, **deco_kwargs)
def function():
pass # Some code ...
which is understood by the Python compiler in its dynamic form:
function = Decorator(*deco_args, **deco_kwargs)(function)
Pyc. 5 is a special case of the most general dynamic decoration scheme.
modified = Decorator(*deco_args, **deco_kwargs)(original, *args, **kwargs)
Therein original and modified stand respectively for a function or
class. Both extra parameter arguments *args and **kwargs refer to
the original, i.e. they define parameter values for an original function
or for the initializer of the original class, respectively.
Correspondingly, any direct class decoration like
@Decorator(*deco_args, **deco_kwargs)
class Classname:
def __init__(self, *args, **kwargs)
pass # Some code ...
should be understood as decoration of the class initializer.
class Classname:
@Decorator(*deco_args, **deco_kwargs)
def __init__(self, *args, **kwargs)
pass # Some code ...
On the other hand, the decorator parameter arguments *deco_args and
**deco_kwargs are used to pass information to the decorator itself, e.g.
to parameterize or configure the decorator.
Since the decoration of a class also refers to the class initializer, and
thus to a function, a symbolic function is used below as an example
representative of an arbitrary decoratored object.
© Copyright 2020-, Martin Abel, eVation. All Rights Reserved. |
|||
