Re: [VOTE] Scalar Type Hints

Hi Ferenc, > On 8 Feb 2015, at 22:18, Ferenc Kovacs <tyra3l@gmail.com> wrote: > > I've voted no for two reasons: > 1, I don't really like the the proposed declare solution, I know that you put much thought > and effort with finding a way to make the strict hints opt-in for the caller, but I think it is > still not a good compromise: people in favor of strict hints will hate it because it is opt-in from > the caller, so they can't use scalar hints to save the boilerplate for argument validation Hmm. Personally, I don’t think types are really a substitute for argument validation. This doesn’t provide a means to force strict types upon people who don’t want them, sure… but the proposal is designed precisely to prevent that, since I don’t see someone like Zeev (say) wanting to deal with strict types any time soon. > , and other people like me will unlike it because they think that the blockless declare will > cause many errors and debugging headache when somebody moves some code between files (or > concatenating files as part of their deployment process). It’s worth noting that I intend to change this to have the “<?php strict” syntax rather than declare(). That’ll have to be a subsequent RFC, as I don’t want to cancel the vote. But it looks like it will pass since all the feedback I’ve got about that so far has been positive. > 2, As this is a really polarizing topic I would like if the eventual solution would have wider > support from the current php-src regular contributors, and not just pushed through by > "force". (And I know that status quo sucks when somebody trying to do innovative stuff, > but it is a safe thing to default to). I’m not sure. There are some people completely opposed to compromise, that can’t be helped. For everyone else, I’m not sure how I feel about contentious votes. If it gets through by a 2/3 majority then presumably it does have significant support, though I worry about lasting impact. Votes do, at least, provide a way to resolve contentious issues, though. > Thanks for your efforts on such a hard topic and I hope you can eventually succeed where > everyone else failed thus far! Thanks. -- Andrea Faulds http://ajf.me/