http://ftp.digitalmars.com/dmd1beta.zip http://ftp.digitalmars.com/dmd2beta.zip

I noticed that the two RDMD pull requests were merged in (thanks!), but the new RDMD binaries in beta3 seem to have been built from the old rdmd.d source. And the Windows RDMD wan't rebuilt at all.

On 10/24/2011 8:53 PM, Nick Sabalausky wrote:
> I noticed that the two RDMD pull requests were merged in (thanks!), but the new RDMD binaries in beta3 seem to have been built from the old rdmd.d source. And the Windows RDMD wan't rebuilt at all.
>

Darn it. I will try and figure out what happened to my build/install script.

Ok, put up a new one. Once I get the build/install script right, this should go smoother :-)

On 10/24/2011 8:53 PM, Nick Sabalausky wrote:
> I noticed that the two RDMD pull requests were merged in (thanks!), but the new RDMD binaries in beta3 seem to have been built from the old rdmd.d source. And the Windows RDMD wan't rebuilt at all.
>

>From a thread "Another Shared Bug?" by Andrew Wiley, I found two bugs
around typeof(super).
Issue 6847 - typeof(super) doesn't work outside member function
Issue 6848 - typeof(super) does not take into account const/immutable
attributes inside member functions

These are similar to special cases of typeof(this), that have been
already fixed in 2.056.

If there is a chance to fix it before releasing betas, please merge dmd/pull/475.

Kenji Hara

2011/10/25 Walter Bright <walter at digitalmars.com>:
> http://ftp.digitalmars.com/dmd1beta.zip
> http://ftp.digitalmars.com/dmd2beta.zip
> _______________________________________________
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>

Verified. Everything's working fine for me now.

From: "Walter Bright"
> Ok, put up a new one. Once I get the build/install script right, this should go smoother :-)
>
> On 10/24/2011 8:53 PM, Nick Sabalausky wrote:
>> I noticed that the two RDMD pull requests were merged in (thanks!), but the new RDMD binaries in beta3 seem to have been built from the old rdmd.d source. And the Windows RDMD wan't rebuilt at all.
>>

On 25 October 2011 02:56, Walter Bright <walter at digitalmars.com> wrote:
> http://ftp.digitalmars.com/dmd1beta.zip http://ftp.digitalmars.com/dmd2beta.zip

Two D1 issues, both very quick to fix:
(1)  From github, d1 doesn't compile. expression.h hasn't been updated
(the signature for RemoveExp::interpret is missing).
(2) The regression 6235 still fails on D1, it hasn't been merged
across from D2. It's just a couple of lines to change in dsymbol.c
around line 1150.

FWIW also missing from the Phobos changelog:

6819 BigInt ^^ fails for some big numbers (powers)

On 10/25/2011 2:02 AM, Don Clugston wrote:
> On 25 October 2011 02:56, Walter Bright<walter at digitalmars.com>  wrote:
>> http://ftp.digitalmars.com/dmd1beta.zip http://ftp.digitalmars.com/dmd2beta.zip
> Two D1 issues, both very quick to fix:
> (1)  From github, d1 doesn't compile. expression.h hasn't been updated
> (the signature for RemoveExp::interpret is missing).

Done.

> (2) The regression 6235 still fails on D1, it hasn't been merged across from D2. It's just a couple of lines to change in dsymbol.c around line 1150.

Can you be more specific?

> FWIW also missing from the Phobos changelog:
>
> 6819 BigInt ^^ fails for some big numbers (powers)

Done.

On 25 October 2011 20:47, Walter Bright <walter at digitalmars.com> wrote:
>
>
> On 10/25/2011 2:02 AM, Don Clugston wrote:
>>
>> On 25 October 2011 02:56, Walter Bright<walter at digitalmars.com> ?wrote:
>>>
>>> http://ftp.digitalmars.com/dmd1beta.zip http://ftp.digitalmars.com/dmd2beta.zip
>>
>> Two D1 issues, both very quick to fix:
>> (1) ?From github, d1 doesn't compile. expression.h hasn't been updated
>> (the signature for RemoveExp::interpret is missing).
>
> Done.
>
>> (2) The regression 6235 still fails on D1, it hasn't been merged across from D2. It's just a couple of lines to change in dsymbol.c around line 1150.
>
> Can you be more specific?

The last pull request (#474) didn't get merged into D1. It was only 3 lines:

            else
            {   /* For arrays, $ will either be a compile-time constant
                 * (in which case its value in set during constant-folding),
                 * or a variable (in which case an expression is created in
                 * toir.c).
                 */
-                v->init = new VoidInitializer(0);
+                VoidInitializer *e = new VoidInitializer(0);
+                e->type = Type::tsize_t;
 +               v->init = e;
            }
            *pvar = v;

But it's hardly worth delaying the release just for this.

On 10/25/2011 11:45 PM, Don Clugston wrote:
>
> The last pull request (#474) didn't get merged into D1. It was only 3 lines:
>
>              else
>              {   /* For arrays, $ will either be a compile-time constant
>                   * (in which case its value in set during constant-folding),
>                   * or a variable (in which case an expression is created in
>                   * toir.c).
>                   */
> -                v->init = new VoidInitializer(0);
> +                VoidInitializer *e = new VoidInitializer(0);
> +                e->type = Type::tsize_t;
>   +               v->init = e;
>              }
>              *pvar = v;
>
> But it's hardly worth delaying the release just for this. _______________________________________________
>

I see the problem, it was pulled in. I just hadn't pushed it.