www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - Goto skipping declarations

reply Ben Jones <fake fake.fake> writes:
In general, you can't skip a declaration with goto, but it seems 
to be allowed if the declaration you're skipping is labelled... 
Is that expected or an accepts invalid bug?

https://godbolt.org/z/4qx8Pf6G7

```d
void f1(){ //fails with error about skipping a declaration
     int x;
     goto Label;
     int y;
     Label:
     int z;
}

void f2(){ //compiles fine
     int x;
     goto Label;
     Dummy:
     int y;
     Label:
     int z;
}
```
May 03
next sibling parent reply Jonathan M Davis <newsgroup.d jmdavisprog.com> writes:
On Friday, May 3, 2024 1:15:16 PM MDT Ben Jones via Digitalmars-d-learn wrote:

 In general, you can't skip a declaration with goto, but it seems
 to be allowed if the declaration you're skipping is labelled...
 Is that expected or an accepts invalid bug?

 https://godbolt.org/z/4qx8Pf6G7

 ```d
 void f1(){ //fails with error about skipping a declaration
      int x;
      goto Label;
      int y;
      Label:
      int z;
 }

 void f2(){ //compiles fine
      int x;
      goto Label;
      Dummy:
      int y;
      Label:
      int z;
 }
 ```


It has to be a bug, and taking it a step further shows that. If you print
out y, you'll get a seemingly random number. E.G. On the first run, I got

554440803

and on the second I got

549310547

Presumably, it's a garbage value from whatever happened to be on the stack.

I'm quite sure that the spec doesn't have anything about being allowed to
skip a declaration just because it has a label on it (honestly, if we _did_
want that to be the case, the spec would probably be missing it, since it
tends to fall on the side of having too few details rather than too many),
but even if it did, the code is clearly doing something that should not be
happening with initialization without explicitly using = void. So,
_something_ here would nee to be fixed.

In any case, I expect that the compiler is just going dumb here because of
the label for some reason, and one or more of the checks that it's supposed
to be doing is being missed.

- Jonathan M Davis
May 03
next sibling parent Ben Jones <fake fake.fake> writes:
On Friday, 3 May 2024 at 20:38:31 UTC, Jonathan M Davis wrote:


https://issues.dlang.org/show_bug.cgi?id=24535



 May 03



prev sibling  parent  "Richard (Rikki) Andrew Cattermole" <richard cattermole.co.nz>  writes:


On 04/05/2024 8:38 AM, Jonathan M Davis wrote:

 In any case, I expect that the compiler is just going dumb here because of
 the label for some reason, and one or more of the checks that it's supposed
 to be doing is being missed.


It is very simple code.

A reverse search over the double linked list for the goto from the label.

I looked into it fairly recently as an example of type state analysis D 
is already designed against.



 May 03



prev sibling  parent  Jonathan M Davis <newsgroup.d jmdavisprog.com>  writes:


On Friday, May 3, 2024 2:38:31 PM MDT Jonathan M Davis via Digitalmars-d-learn 
wrote:

 On Friday, May 3, 2024 1:15:16 PM MDT Ben Jones via Digitalmars-d-learn 


wrote:

 In general, you can't skip a declaration with goto, but it seems
 to be allowed if the declaration you're skipping is labelled...
 Is that expected or an accepts invalid bug?

 https://godbolt.org/z/4qx8Pf6G7

 ```d
 void f1(){ //fails with error about skipping a declaration

      int x;
      goto Label;
      int y;
      Label:
      int z;

 }

 void f2(){ //compiles fine

      int x;
      goto Label;
      Dummy:
      int y;
      Label:
      int z;

 }
 ```


 It has to be a bug, and taking it a step further shows that. If you print
 out y, you'll get a seemingly random number. E.G. On the first run, I got

 554440803

 and on the second I got

 549310547

 Presumably, it's a garbage value from whatever happened to be on the stack.

 I'm quite sure that the spec doesn't have anything about being allowed to
 skip a declaration just because it has a label on it (honestly, if we _did_
 want that to be the case, the spec would probably be missing it, since it
 tends to fall on the side of having too few details rather than too many),
 but even if it did, the code is clearly doing something that should not be
 happening with initialization without explicitly using = void. So,
 _something_ here would nee to be fixed.

 In any case, I expect that the compiler is just going dumb here because of
 the label for some reason, and one or more of the checks that it's supposed
 to be doing is being missed.


Here. I reported it:

https://issues.dlang.org/show_bug.cgi?id=24534

- Jonathan M Davis



 May 03