[Python-ideas] Re: Add decorator_with_params function to functools module
Sorry for the long response. After investigation, I have found that there is a third-party library called `decorator` which do smth similar to what I propose (I didn't know about that library when I created this thread). As I said, the idea of `decorator_factory` is to create a unified way of how to deal with simple decorator with parameters. For me, this is smth similar to `functools.lru_cache`, before it was introduced I written such functionality by myself many times. Thank you for your opinion, and sorry about the long response. ___ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-le...@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/GXSJ2XU4SPQ66DRW6AAAMAF2AG3SSYPD/ Code of Conduct: http://python.org/psf/codeofconduct/
[Python-ideas] Re: Add decorator_with_params function to functools module
Is it really worth it? Fact is, while it can shave off some lines of code, I think it is interesting to know _which_ lines of code - Usually when one writes a decorator, it is expected that they will know what they are writing, and will want to be in control of their code. Delegating this to a decorator-decorator that is to be copied and pasted, and will definitely change call-order, and when your decorator is called, is something that I, at least, would be wary to use. Meanwhile, when I want this pattern, it really takes me 2 LoC inside the decorator to have the same functionality, and still be 100% in control of when my function is called: ``` from functools import partial def mydecorator(func=None, /, *, param1=None, **kwargs): if func is None: return partial(mydcorator, param1=None, **kwargs) # decorator code goes here ... ``` So, yes, your proposal has some utility - but I consider it to be marginal - it is the kind of stuff that I'd rather see on a 3rdy party package with extra-stuff to help building decorators than on stdlib. On Mon, 30 Nov 2020 at 17:04, Yurii Karabas <1998uri...@gmail.com> wrote: > The idea of `decorator_factory` is to eliminate boilerplate code that used > to create a decorator with parameters. > > A perfect example is how `dataclass` decorator can be simplified from this: > ``` > def dataclass(cls=None, /, *, init=True, repr=True, eq=True, order=False, > unsafe_hash=False, frozen=False): > def wrap(cls): > return _process_class(cls, init, repr, eq, order, unsafe_hash, > frozen) > > # See if we're being called as @dataclass or @dataclass(). > if cls is None: > # We're called with parens. > return wrap > > # We're called as @dataclass without parens. > return wrap(cls) > ``` > To this: > @functools.decorator_factory > def dataclass(cls, /, *, init=True, repr=True, eq=True, order=False, > unsafe_hash=False, frozen=False): > return _process_class(cls, init, repr, eq, order, unsafe_hash, frozen) > ``` > ___ > Python-ideas mailing list -- python-ideas@python.org > To unsubscribe send an email to python-ideas-le...@python.org > https://mail.python.org/mailman3/lists/python-ideas.python.org/ > Message archived at > https://mail.python.org/archives/list/python-ideas@python.org/message/NMCDVBOYFXEKZ4L3ASLNYL2NBATJ3VCX/ > Code of Conduct: http://python.org/psf/codeofconduct/ > ___ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-le...@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/OU4VVMBE74KXK6HF6G7LEFP735AJM4G2/ Code of Conduct: http://python.org/psf/codeofconduct/
[Python-ideas] Re: Add decorator_with_params function to functools module
The idea of `decorator_factory` is to eliminate boilerplate code that used to create a decorator with parameters. A perfect example is how `dataclass` decorator can be simplified from this: ``` def dataclass(cls=None, /, *, init=True, repr=True, eq=True, order=False, unsafe_hash=False, frozen=False): def wrap(cls): return _process_class(cls, init, repr, eq, order, unsafe_hash, frozen) # See if we're being called as @dataclass or @dataclass(). if cls is None: # We're called with parens. return wrap # We're called as @dataclass without parens. return wrap(cls) ``` To this: @functools.decorator_factory def dataclass(cls, /, *, init=True, repr=True, eq=True, order=False, unsafe_hash=False, frozen=False): return _process_class(cls, init, repr, eq, order, unsafe_hash, frozen) ``` ___ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-le...@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/NMCDVBOYFXEKZ4L3ASLNYL2NBATJ3VCX/ Code of Conduct: http://python.org/psf/codeofconduct/
[Python-ideas] Re: Add decorator_with_params function to functools module
After discussion at the python issue tracker, the better name
`decorator_factory` was proposed.
I have just gone through the standard library code to find places where
`decorator_factory` can be used.
1. `dataclasses.dataclass`
```
def dataclass(cls=None, /, *, init=True, repr=True, eq=True, order=False,
unsafe_hash=False, frozen=False):
"""Returns the same class as was passed in, with dunder methods
added based on the fields defined in the class.
Examines PEP 526 __annotations__ to determine fields.
If init is true, an __init__() method is added to the class. If
repr is true, a __repr__() method is added. If order is true, rich
comparison dunder methods are added. If unsafe_hash is true, a
__hash__() method function is added. If frozen is true, fields may
not be assigned to after instance creation.
"""
def wrap(cls):
return _process_class(cls, init, repr, eq, order, unsafe_hash, frozen)
# See if we're being called as @dataclass or @dataclass().
if cls is None:
# We're called with parens.
return wrap
# We're called as @dataclass without parens.
return wrap(cls)
```
```
@functools.decorator_factory
def dataclass(cls, /, *, init=True, repr=True, eq=True, order=False,
unsafe_hash=False, frozen=False):
"""Returns the same class as was passed in, with dunder methods
added based on the fields defined in the class.
Examines PEP 526 __annotations__ to determine fields.
If init is true, an __init__() method is added to the class. If
repr is true, a __repr__() method is added. If order is true, rich
comparison dunder methods are added. If unsafe_hash is true, a
__hash__() method function is added. If frozen is true, fields may
not be assigned to after instance creation.
"""
return _process_class(cls, init, repr, eq, order, unsafe_hash, frozen)
```
2. `functools.lru_cache`
```
def lru_cache(maxsize=128, typed=False):
"""Least-recently-used cache decorator.
If *maxsize* is set to None, the LRU features are disabled and the cache
can grow without bound.
If *typed* is True, arguments of different types will be cached separately.
For example, f(3.0) and f(3) will be treated as distinct calls with
distinct results.
Arguments to the cached function must be hashable.
View the cache statistics named tuple (hits, misses, maxsize, currsize)
with f.cache_info(). Clear the cache and statistics with f.cache_clear().
Access the underlying function with f.__wrapped__.
See:
http://en.wikipedia.org/wiki/Cache_replacement_policies#Least_recently_used_(LRU)
"""
# Users should only access the lru_cache through its public API:
# cache_info, cache_clear, and f.__wrapped__
# The internals of the lru_cache are encapsulated for thread safety and
# to allow the implementation to change (including a possible C version).
if isinstance(maxsize, int):
# Negative maxsize is treated as 0
if maxsize < 0:
maxsize = 0
elif callable(maxsize) and isinstance(typed, bool):
# The user_function was passed in directly via the maxsize argument
user_function, maxsize = maxsize, 128
wrapper = _lru_cache_wrapper(user_function, maxsize, typed, _CacheInfo)
wrapper.cache_parameters = lambda : {'maxsize': maxsize, 'typed': typed}
return update_wrapper(wrapper, user_function)
elif maxsize is not None:
raise TypeError(
'Expected first argument to be an integer, a callable, or None')
def decorating_function(user_function):
wrapper = _lru_cache_wrapper(user_function, maxsize, typed, _CacheInfo)
wrapper.cache_parameters = lambda : {'maxsize': maxsize, 'typed': typed}
return update_wrapper(wrapper, user_function)
return decorating_function
```
```
@decorator_factory
def lru_cache(user_function, /, maxsize=128, typed=False):
"""Least-recently-used cache decorator.
If *maxsize* is set to None, the LRU features are disabled and the cache
can grow without bound.
If *typed* is True, arguments of different types will be cached separately.
For example, f(3.0) and f(3) will be treated as distinct calls with
distinct results.
Arguments to the cached function must be hashable.
View the cache statistics named tuple (hits, misses, maxsize, currsize)
with f.cache_info(). Clear the cache and statistics with f.cache_clear().
Access the underlying function with f.__wrapped__.
See:
http://en.wikipedia.org/wiki/Cache_replacement_policies#Least_recently_used_(LRU)
"""
# Users should only access the lru_cache through its public API:
# cache_info, cache_clear, and f.__wrapped__
# The internals of the lru_cache are encapsulated for thread safety and
# to allow the implementation to change (including a possible C version).
if
[Python-ideas] Re: Add decorator_with_params function to functools module
You can use classes as decorators, it's quite more simple: https://towardsdatascience.com/using-class-decorators-in-python-2807ef52d273?gi=ea5091974462 ___ Python-ideas mailing list -- python-ideas@python.org To unsubscribe send an email to python-ideas-le...@python.org https://mail.python.org/mailman3/lists/python-ideas.python.org/ Message archived at https://mail.python.org/archives/list/python-ideas@python.org/message/I6GUPJDYILR7SBSWATAULHYL2X4GGBRL/ Code of Conduct: http://python.org/psf/codeofconduct/
