Re: [RFC] Skipping parameters take 2

Hey Marcio, > On 18 Jan 2015, at 02:22, Marcio Almada <marcio.web2@gmail.com> wrote: > > Andrea, > >> For consistency with list(), we could also just put nothing: >> >> foo($bar, , $baz); >> >> Which is like: >> >> list($bar, , $baz) = $arr; >> >> Thoughts? > > Not sure. Do we consider both contexts (list assignment skipping and > parameter skipping) as assignments? If so, then the consistency > arguments seems very strong to me. Not really. But, list() is very function-like in syntax, so it’d make sense to follow its lead for actual functions. > Only point against the consistency path is that in the context of > list() the blank identifier is used to discard a value, while on > parameter skipping the blank identifier inherits a previously declared > value. Yes, I suppose that’s a fair point. Though if you were to implement a list() clone as a function (purely for the sake of example, you wouldn’t *actually* want to do this), you might actually do something similar: function listLike(&$a = 0, &$b = 0, &$c = 0, array $arr) { if ($a !== 0) { $a = $arr[0]; } if ($b !== 0) { $b = $arr[1]; } if ($c !== 0) { $c = $arr[2]; } } listLike($bar, 0, $baz, $arr); list()’s value skipping is a little bit like function argument skipping, in that way, right? With this list()-like default value syntax, it’d even work the same: listLike($bar, , $baz, $arr); …that probably made no sense. But I think there’s a case to be made that since list() follows this syntax, we should for function calls to. By the way, while I was opposed to this RFC before, I am seeing its value somewhat when using badly-designed APIs like json_decode. That said, I’m still not sure I like the RFC concept… better to make new, improved APIs than add kludges to make bad APIs more usable. -- Andrea Faulds http://ajf.me/