On 3 November 2016 at 23:06, Steven D'Aprano <st...@pearwood.info> wrote:
> On Thu, Nov 03, 2016 at 12:35:07PM -0700, Chris Barker wrote: > > On Thu, Nov 3, 2016 at 12:00 PM, MRAB <pyt...@mrabarnett.plus.com> > wrote: > > > > > self.an_arg = the_default if an_arg is None else an_arg > > > > > No, ?? is a bit like 'or', except that only None is falsey, so it > would be: > > > > > > self.an_arg = an_arg ?? the_default > > > > > > thanks! and actually, that reads much better to me. > > That suggests a possible different colour for this operator: `?or`. > > (Apologies if that's already been suggested and rejected earlier.) > > The None-aware "safe navigation" operators ?. and ?[] will be used where > Python already uses punctuation: > > spam.eggs # ordinary attribute lookup > spam?.eggs # None-aware attribute lookup > > spam[eggs] # ordinary item lookup > spam?[eggs] # None-aware item lookup > > > which is simple enough to remember: just prefix your usual operator with > a question mark to make it None-aware. But one of the disadvantages of > ?? as an operator is that it replaces a keyword with completely > unrelated punctuation: > > spam or default # default only if spam is any Falsey value > spam ?? default # default only if spam is None > > Compared to: > > spam ?or default > > gives us the same rule: prefix the `or` operator with ? to make it None- > aware. > Agree 100%. Prefixing an operator with ? meaning that it becomes None-aware is easy to remember. Just think of `?or` the same as `or`, but while `or` is truth-aware, `?or` would be None-aware. Similarly to other operators like . and []. How many times have I wanted to do something like `foo or bar`, but then again this is not always correct in case foo is an object that can be true or false, in which case it has to be expanded to `foo if foo is not None else bar`. With the proposal, we could write `foo ?or bar`, which is intuitively similar to the well known `foo or bar` expression and therefore easy to pick up. > > > -- > Steve > _______________________________________________ > Python-ideas mailing list > Python-ideas@python.org > https://mail.python.org/mailman/listinfo/python-ideas > Code of Conduct: http://python.org/psf/codeofconduct/ > -- Gustavo J. A. M. Carneiro Gambit Research "The universe is always one step beyond logic." -- Frank Herbert
_______________________________________________ Python-ideas mailing list Python-ideas@python.org https://mail.python.org/mailman/listinfo/python-ideas Code of Conduct: http://python.org/psf/codeofconduct/
