On 9/16/20 1:59 AM, Corentin Jabot wrote:
>
>
> On Tue, Sep 15, 2020, 23:53 Tom Honermann <tom_at_[hidden]
> <mailto:tom_at_[hidden]>> wrote:
>
> On 9/15/20 12:09 PM, Corentin Jabot via SG16 wrote:
>>
>>
>> On Tue, Sep 15, 2020, 18:03 Steve Downey via SG16
>> <sg16_at_[hidden] <mailto:sg16_at_[hidden]>> wrote:
>>
>> Thanks Corentin!
>> I'll have these to chase down now. The problem I keep running
>> into is
>> making sure that the various proposals are adopted, and which
>> ones
>> have so many caveats that they aren't good examples. Python,
>> Rust, and
>> JS look solid.
>>
>>
>> My understanding is that rust hasn't adopted it yet.
>>
>> Swift is a really good example of what NOT to do: some emojis are
>> considered symbols ( swift has arbitrary symbols) while some are
>> considered variable name.
>
> Swift does exactly what C++ currently does for identifiers. Compare:
>
> * https://docs.swift.org/swift-book/ReferenceManual/LexicalStructure.html#ID412
>
> At the very bottom of the same page, swift describes operators.
> Which is not an issue C++ has
> So in Swift, ☹️ is an operator, 🙂 is an identifier.
> More description of the issue here
> https://lists.swift.org/pipermail/swift-evolution/Week-of-Mon-20160912/027112.html
>
> The takeaway is that a careless emoji support in identifiers impacts
> C++ future ability to give special meaning to codepoints recognized as
> symbols, if we ever wanted to do that.

Thanks, Corentin. That is a good point and there is some great analysis
in the link you provided.

Steve, I think we could benefit from blatantly stealing (and crediting
of course!) some motivation from
https://gist.github.com/jtbandes/c0b0c072181dcd22c3147802025d0b59.
Specifically, noting that the current identifier allowances permit
identifiers that look like operators and, as Corentin indicates, we may
want to reserve those for use as future operators.

Also, Steve, I recall you stating that you planned to put together some
slides for the next EWG presentation. I would be happy to review and
provide feedback/updates on those if you like.

Tom.

>
>
> * http://eel.is/c++draft/lex.name#1
>
> Tom.
>
>>
>> On Tue, Sep 15, 2020 at 11:44 AM Corentin Jabot via SG16
>> <sg16_at_[hidden] <mailto:sg16_at_[hidden]>> wrote:
>> >
>> > C#
>> https://docs.microsoft.com/en-us/dotnet/csharp/language-reference/language-specification/lexical-structure#identifiers
>> (not exactly uax)
>> >
>> > Rust
>> https://rust-lang.github.io/rfcs/2457-non-ascii-idents.html
>> >
>> > Swift
>> https://forums.swift.org/t/a-path-forward-on-rationalizing-unicode-identifiers-and-operators/6735
>> (not exactly uax)
>> >
>> > Erlang
>> https://www.erlang.org/erlang-enhancement-proposals/eep-0040.html
>> >
>> > Julia https://docs.julialang.org/en/v1/manual/variables/
>> (not exactly uax)
>> >
>> > Python 3 https://www.python.org/dev/peps/pep-3131/
>> >
>> > JS https://tc39.es/ecma262/
>> >
>> >
>> > On Tue, Sep 15, 2020, 17:23 Steve Downey via SG16
>> <sg16_at_[hidden] <mailto:sg16_at_[hidden]>> wrote:
>> >>
>> >> I know we've discussed them before, but when trying to
>> track down
>> >> actual references for purposes of including in a paper, I
>> keep failing
>> >> at Google.
>> >>
>> >> Asking for help so that R7 will be better.
>> >>
>> >> PR's also welcome on
>> https://github.com/steve-downey/papers/blob/master/d1949.md
>> >> --
>> >> SG16 mailing list
>> >> SG16_at_[hidden] <mailto:SG16_at_[hidden]>
>> >> https://lists.isocpp.org/mailman/listinfo.cgi/sg16
>> >
>> > --
>> > SG16 mailing list
>> > SG16_at_[hidden] <mailto:SG16_at_[hidden]>
>> > https://lists.isocpp.org/mailman/listinfo.cgi/sg16
>> --
>> SG16 mailing list
>> SG16_at_[hidden] <mailto:SG16_at_[hidden]>
>> https://lists.isocpp.org/mailman/listinfo.cgi/sg16
>>
>>
>