```
void main()
{
  import std.stdio;
  auto f = (){
    string[] t;
    { // inner scope
      t ~= "hello";
      scope( exit ) t ~= "world";
    } // inner scope exit
    return t;
  };

  f().writeln; // ["hello", "world"]
}
```
removing the inner scope in f() gives ["hello"]

So when no inner scope is present, the scope exit 'runs' after the return? Is that indeed expected behavior according to the specification?

On Monday, 29 June 2020 at 22:31:12 UTC, Arjan wrote:

> So when no inner scope is present, the scope exit 'runs' after the return? Is that indeed expected behavior according to the specification?

Yes. A scope ends at the '}'. Destructors and scope guards execute then, after the return.

On 6/29/20 6:31 PM, Arjan wrote:
> ```
> void main()
> {
>    import std.stdio;
>    auto f = (){
>      string[] t;
>      { // inner scope
>        t ~= "hello";
>        scope( exit ) t ~= "world";
>      } // inner scope exit
>      return t;
>    };
> 
>    f().writeln; // ["hello", "world"]
> }
> ```
> removing the inner scope in f() gives ["hello"]
> 
> So when no inner scope is present, the scope exit 'runs' after the return? Is that indeed expected behavior according to the specification?

Yes. The return statement is inside the scope of the function, so it runs before the scope is exited. Are you saying the spec doesn't say that?

-Steve

On Monday, 29 June 2020 at 22:47:16 UTC, Steven Schveighoffer wrote:
> Yes. The return statement is inside the scope of the function, so it runs before the scope is exited. Are you saying the spec doesn't say that?

Thanks for the assurance. The spec does state it like this:
```
The ScopeGuardStatement executes NonEmptyOrScopeBlockStatement at the close of the current scope, rather than at the point where the ScopeGuardStatement appears.
```
Which is correct, but there is no single example with a return where the ScopeBlockStatement interferes with the return.

I started wondering about this since I hit a bug in a piece of code.

On 6/30/20 2:56 AM, Arjan wrote:
> On Monday, 29 June 2020 at 22:47:16 UTC, Steven Schveighoffer wrote:
>> Yes. The return statement is inside the scope of the function, so it runs before the scope is exited. Are you saying the spec doesn't say that?
> 
> Thanks for the assurance. The spec does state it like this:
> ```
> The ScopeGuardStatement executes NonEmptyOrScopeBlockStatement at the close of the current scope, rather than at the point where the ScopeGuardStatement appears.
> ```
> Which is correct, but there is no single example with a return where the ScopeBlockStatement interferes with the return.
> 
> I started wondering about this since I hit a bug in a piece of code.

I can see where it would be confusing, and it could probably contain an example and clarification.

-steve

On Tuesday, 30 June 2020 at 12:18:14 UTC, Steven Schveighoffer wrote:
> On 6/30/20 2:56 AM, Arjan wrote:
>> On Monday, 29 June 2020 at 22:47:16 UTC, Steven Schveighoffer wrote:
>>> [...]
>> 
>> Thanks for the assurance. The spec does state it like this:
>> ```
>> The ScopeGuardStatement executes NonEmptyOrScopeBlockStatement at the close of the current scope, rather than at the point where the ScopeGuardStatement appears.
>> ```
>> Which is correct, but there is no single example with a return where the ScopeBlockStatement interferes with the return.
>> 
>> I started wondering about this since I hit a bug in a piece of code.
>
> I can see where it would be confusing, and it could probably contain an example and clarification.
>
> -steve

That would certainly be helpfull.

On Tuesday, 30 June 2020 at 12:18:14 UTC, Steven Schveighoffer wrote:
> I can see where it would be confusing, and it could probably contain an example and clarification.

https://issues.dlang.org/show_bug.cgi?id=20997