> On Oct 14, 2016, at 3:42 PM, Alex Martini via swift-evolution > <[email protected]> wrote: > >> On Oct 14, 2016, at 1:53 PM, Hooman Mehr via swift-evolution >> <[email protected] <mailto:[email protected]>> wrote: >> >>> On Oct 14, 2016, at 1:49 PM, Daniel Duan via swift-evolution >>> <[email protected] <mailto:[email protected]>> wrote: >>> >>> Agree with Robert here. I'd rather be able to use it as part of operators. >>> Currently the character set for operators and identifier head are mutually >>> exclusive. So this proposal will remove that possibility. This deserves >>> some discussion. >>> >>> Daniel Duan >>> Sent from my iPhone >> > >> I don’t think $ will be become available to be used as an operator if we >> remove its identifier use. > > > If $ is an operator, then "$2" looks like the custom unary prefix operator > "$" applied to "2". That's a problem; it needs to be parsed as the > identifier "$2" instead. > > We could prevent "$" from being allowed a prefix operator. There is > precedent for the language reserving operators that would otherwise be > allowed, such as postfix "?" and postfix "!". >
Yes, we might have to special case the parser for this. I imagine some flag such as “isParsingClosureWithoutExplicitParameter”. At least it’s not impossible 😉 > — Alex > _______________________________________________ > swift-evolution mailing list > [email protected] > https://lists.swift.org/mailman/listinfo/swift-evolution
_______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
