Oh and for the muscle memory argument: Switching from NSString * var to let var:String
Was a bigger effort :D Joao Sent from my iPhone > On 15 Dec 2015, at 23:28, Chris Lattner <[email protected]> wrote: > > >> On Dec 15, 2015, at 12:11 AM, João Nunes <[email protected]> wrote: >> >> In that case why did you accept to remove for loop c style and ++/-- >> operators? It is the same reasoning. > > The rationale for removing them is outlined here: > https://github.com/apple/swift-evolution/blob/master/proposals/0004-remove-pre-post-inc-decrement.md > > AFAICT, none of the disadvantages there apply to semicolons, except perhaps > the first one (which I’d consider a stretch). > > -Chris > >> >> Joao >> >> Sent from my iPhone >> >>> On 15 Dec 2015, at 06:56, Chris Lattner <[email protected]> wrote: >>> >>> >>>>> On Dec 14, 2015, at 3:44 PM, Douglas Gregor <[email protected]> wrote: >>>>> >>>>> >>>>> On Dec 14, 2015, at 11:01 AM, Chris Lattner via swift-evolution >>>>> <[email protected]> wrote: >>>>> >>>>> There are two different topics here, and I’d suggest exploring them >>>>> separately. >>>>> >>>>> - The semicolon within a line is a expressivity feature. >>>>> - The semicolon at the end of the line is accepted, but generally ignored. >>>>> >>>>> Most people seem to be focusing on the second one. IMO, I think that it >>>>> adds value to the language for people coming from semi-colon oriented >>>>> languages or bouncing between multiple languages (that muscle memory >>>>> takes awhile to break). OTOH, it is just syntactic noise (along with >>>>> redundant parens in conditions and many other things), and so having a >>>>> warning (probably opt-in) for it would make sense to me. >>>> >>>> This seems firmly in linter/style warning territory. Unlike with other >>>> features that are being removed in Swift 3.0, optional semicolons at the >>>> end of are utterly harmless: nobody is going to have to reason about them >>>> beyond “oh, I can ignore that. >>> >>> I agree. I am missing the motivation for what harm they are causing. >>> >>> -Chris >
_______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
