> On 4 Dec 2021, at 21:21, Chris Angelico <ros...@gmail.com> wrote: > > On Sun, Dec 5, 2021 at 5:29 AM Barry Scott <ba...@barrys-emacs.org > <mailto:ba...@barrys-emacs.org>> wrote: >> >> >> >>> On 1 Dec 2021, at 06:16, Chris Angelico <ros...@gmail.com> wrote: >>> >>> I've just updated PEP 671 https://www.python.org/dev/peps/pep-0671/ >>> with some additional information about the reference implementation, >>> and some clarifications elsewhere. >> >> (I suspect that there was a reply that I should be replying to but, cannot >> find one appropriate) >> >> I have a lot of code that exploits the fact that passing an explicit None >> will cause the early bound default idiom to set the default for me. >> >> def inner(timestamp=None): >> if timestamp is None: >> timestamp = time.time() >> do_stuff... >> >> def outer(timestamp=None): >> inner(timestamp=timestamp) >> >> outer can in an idiomatic way have inner default timestamp and not have to >> know what that means. > > If you need outer() to be able to have a value that means "use the > default", then there are three options: > > 1) Don't pass timestamp at all. In simple cases where it will only and > always specify the default, this is fine. > 2) Define a sentinel that is indeed part of your API. > 3) Use *args or **kwargs to choose whether to pass it or not (best if > there are multiple of them). > > You can continue to use the existing system of "if none, do this", or > you can flip it around and have the sentinel as a special token within > your code: > > def inner(timestamp=>time.time()): > if timestamp is None: timestamp = time.time()
And, obviously, if you end up needing the write the explicit check for None there is no advantage to using late bound default. > > Depends on how important this feature is outside of your own helper > functions. (I would probably not do this for None specifically - if > it's purely internal, I'm more likely to use a dedicated local > sentinel object.) > > But as soon as there are two or three arguments that "might have to be > passed, might not", it's far more readable to use kwargs to pass just > the ones you want. > > def outer(**kwargs): > inner(**kwargs) > > That way, if something changes in inner(), you don't have to worry > about breaking your caller's API. Yes that's a good point. Use the *kwargs style to pass down stuff. > >> With late bound I cannot do this without more complex pattern of building an >> arg list. >> >> What if passing None still worked? I know the argument that there are more >> sentinels then None. >> >> def inner(timestamp=>time.time()) >> do_stuff... >> >> def outer(timestamp=None): >> inner(timestamp=timestamp) >> >> The code in inner that decides to when to allow the default could check for >> timestamp being >> missing or arg present and None. >> >> Would the lack of support for other sentinels out weight the simple way to >> get the default applied? >> > > None is most assuredly not going to trigger a late-bound default. Are you state that this is because in most of the cases where I might think that I need this behaviour there are better patterns to use like *kwargs? Is that worth stating in the PEP in the rejected ideas? > Python is not JavaScript :) Thank your choice of deity for that! Barry > > ChrisA > _______________________________________________ > Python-ideas mailing list -- python-ideas@python.org > <mailto:python-ideas@python.org> > To unsubscribe send an email to python-ideas-le...@python.org > <mailto:python-ideas-le...@python.org> > https://mail.python.org/mailman3/lists/python-ideas.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/QWYXRITR56CKURYKE7CKQ7A4WVNTUVJL/ > > <https://mail.python.org/archives/list/python-ideas@python.org/message/QWYXRITR56CKURYKE7CKQ7A4WVNTUVJL/> > Code of Conduct: http://python.org/psf/codeofconduct/ > <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/F2ILJ42IWMDH6V3T2AJDMLWA2AIV43MY/ Code of Conduct: http://python.org/psf/codeofconduct/
