Feedback Thread: DIP 1036--String Interpolation Tuple Literals--Community Review Round 2

[Steven Schveighoffer]

On 1/27/21 7:09 AM, Atila Neves wrote:
> On Wednesday, 27 January 2021 at 10:33:53 UTC, Mike Parker wrote:
>> This is the feedback thread for the second round of Community Review 
>> of DIP 1036, "String Interpolation Tuple Literals".
>>
>> [...]
> 
> I thought "the library" was confusing. I guess Phobos is what's meant? 
> How would that work without an import? Shouldn't it be "the runtime" 
> instead?

Yes, it's meant to be the piece of the language that lives in the 
library. I meant the runtime. Will change.

> The DIP says "// auto variables cannot be assigned to a interpolation 
> sequence" but earlier on there's this:
> 
> auto msg1 = i"Hello, ${name}, this is your ${visits}${post(visits)} time 
> visiting";
> auto msg2 = ir"Hello, ${name}, this is your ${visits}${post(visits)} 
> time visiting";
> auto msg3 = i`Hello, ${name}, this is your ${visits}${post(visits)} time 
> visiting`;
> auto msg3 = q{Hello, ${name}, this is your ${visits}${post(visits)} time 
> visiting};

I can see how the statement is awkward. I'm trying to convey that the 
*direct translation* of the interpolation tuple literal to interleaving 
InterpolationLiterals and expressions cannot be assigned to an auto 
variable, therefore the rewrite to idup is used.

> 
> I'm guessing msg3 was supposed to be `iq{...}`?

Yes, it should be that. Will correct.

-Steve