On Wednesday, 5 August 2020 at 16:35:12 UTC, Paul Backus wrote:
> On Wednesday, 5 August 2020 at 16:22:48 UTC, Steven Schveighoffer wrote:
>> On 8/5/20 12:10 PM, Paul Backus wrote:
>>> 
>>> Not a fan. There are cases where I've done this intentionally--not with eponymous template members, but with private helper templates:
>>> 
>>> template foo(T) {
>>>      enum helper(T) = ...; // error?
>>>      void foo(T i) {
>>>          static if (helper!T) ...;
>>>          else ...;
>>>      }
>>> }

Can you explain the "intentionally"? What purpose did it serve to name the template parameters the same? (in the code given, if you only ever used helper!T, you could have made "helper" a non-templated enum)

template foo(T) {
     enum helper = ...uses T...;
     void foo(T i) {
         static if (helper) ...;
         else ...;
     }
}

>>
>> I've intentionally done that, and even though it worked, rewrote the template parameter names because it's confusing to read.
>>
>> It would have to be a long deprecation cycle.
>>
>> I would be in favor of the change.
>>
>> -Steve
>
> "Confusing to read" does not justify a compiler warning.

Actually, I think it does. There are not many other reasons I can think of for why we have the existing "warnings".

The bigger problem is that warnings are all or nothing, instead of being able to disable/enable warnings to what the user likes [*]. So there is no way to gently add warnings and get user experience and feedback.

-Johan

[*] https://github.com/dlang/dmd/pull/5592