On 1/15/2011 12:47 PM, Walter Bright wrote:
> 
> 
> Brad Roberts wrote:
>>
>> I haven't tested with d1, but I don't recall seeing any changes to the alignment
>> code to correct the issues that test23.d's test22 exposed.  64bit c sizes Bar22
>> at 8 bytes, d2 at 16.
>> 
> 
> I'll check it in. I would have earlier, except that D2 fails its test suite now, even for 32 bits.
> 
> Aside: why aren't the windows dll tests in the checked in test suite? Naturally, they all fail now.

There's a small subset of the old/private test suite that isn't part of the public one.  The nature of them:

1) elaborate not-obviously-non-copyrighted
2) elaborate build steps
3) platform specific

The latter two sets can be addressed, just haven't.  I got the first 95+% done to make it easy for others to test.  The other 5% (both numbers are pure guestimates) still need to be handled.  Frankly, part of why I haven't is that it feels like I'm the only one that uses the public tests.  Neither your nor Don do last I knew, so my motivation hasn't been particularly high to invest the extra effort to finish off those last sets of tests.

On 15 January 2011 22:09, Brad Roberts <braddr at puremagic.com> wrote:
> On 1/15/2011 12:47 PM, Walter Bright wrote:
>>
>>
>> Brad Roberts wrote:
>>>
>>> I haven't tested with d1, but I don't recall seeing any changes to the alignment code to correct the issues that test23.d's test22 exposed. ?64bit c sizes Bar22 at 8 bytes, d2 at 16.
>>>
>>
>> I'll check it in. I would have earlier, except that D2 fails its test suite now, even for 32 bits.
>>
>> Aside: why aren't the windows dll tests in the checked in test suite? Naturally, they all fail now.
>
> There's a small subset of the old/private test suite that isn't part of the public one. ?The nature of them:
>
> 1) elaborate not-obviously-non-copyrighted
> 2) elaborate build steps
> 3) platform specific
>
> The latter two sets can be addressed, just haven't. ?I got the first 95+% done to make it easy for others to test. ?The other 5% (both numbers are pure guestimates) still need to be handled. ?Frankly, part of why I haven't is that it feels like I'm the only one that uses the public tests. ?Neither your nor Don do last I knew, so my motivation hasn't been particularly high to invest the extra effort to finish off those last sets of tests.

Yes, I still can't compile the D autotester on Windows.
I spent about an hour trying to get it to work, then gave up.

On 1/15/2011 1:47 PM, Don Clugston wrote:
> 
> Yes, I still can't compile the D autotester on Windows.
> I spent about an hour trying to get it to work, then gave up.

What sort of errors are you seeing?  You can look at the auto testers code to see what it's doing.  Obviously it does work on windows.

    http://github.com/braddr/d-tester

On 15 January 2011 23:03, Brad Roberts <braddr at puremagic.com> wrote:
> On 1/15/2011 1:47 PM, Don Clugston wrote:
>>
>> Yes, I still can't compile the D autotester on Windows.
>> I spent about an hour trying to get it to work, then gave up.
>
> What sort of errors are you seeing? ?You can look at the auto testers code to see what it's doing. ?Obviously it does work on windows.

Path errors. And I'm not sure that the environment variables worked correctly.

>
> ? ?http://github.com/braddr/d-tester
>
> _______________________________________________
> dmd-beta mailing list
> dmd-beta at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>

On 1/15/2011 2:22 PM, Don Clugston wrote:
> On 15 January 2011 23:03, Brad Roberts <braddr at puremagic.com> wrote:
>> On 1/15/2011 1:47 PM, Don Clugston wrote:
>>>
>>> Yes, I still can't compile the D autotester on Windows.
>>> I spent about an hour trying to get it to work, then gave up.
>>
>> What sort of errors are you seeing?  You can look at the auto testers code to see what it's doing.  Obviously it does work on windows.
> 
> Path errors. And I'm not sure that the environment variables worked correctly.
> 
>>
>>    http://github.com/braddr/d-tester

I'd like to help, but you're not giving me a lot of details to work with..