Fair enough. Trying to keep in line with “UTF-8 all the things” so that we can
1) not restrict languages (for instance let’s say the German company tüv produced a toolchain and required the umlaut’d U).
2) Allowing these target tuplets to exist in JSON files safely for other proposals I’m writing like linker scripts and response/option files

On Mon, Sep 9, 2019 at 08:08, Tom Honermann <tom@honermann.net> wrote:
I think this is a useful endeavor and I agree with the direction.

My only question is why there is a specification for the encoding of the target tuplet at all.  I would expect that we instead specify the allowable characters, perhaps deferring to Unicode UAX#31 (https://www.unicode.org/reports/tr31/tr31-31.html).  Assuming that the basic source character set doesn't suffice for this.  The target tuplet appears on command lines and we don't control that environment.

Tom.

On 9/9/19 8:23 AM, Isabella Muerte via SG15 wrote:
Hello all,

As today is the cut off for papers for CppCon, I've written one small brief one I'd like to have discussed regarding "target tuplets". I know we've been on a modules kick recently, but this is tooling related, and I would argue deserves some time. I've attached a copy and (following discussion) a P version will be mailed for Belfast.



Izzy

_______________________________________________
SG15 mailing list
SG15@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/sg15