Please take some time to reduce it to a standalone reproducible test case and file a bugzilla entry for it.

On 9/17/2010 6:23 PM, Shin Fujishiro wrote:
> It depends on the order of compilation.  The following succeeds:
> 
> % dmd -c std/concurrency.d std/string.d
> % _
> 
> While the following fails (note the order of arguments):
> 
> % dmd -c std/string.d std/concurrency.d
> std/typecons.d(336): Error: cannot evaluate empty("") at compile time
> std/typecons.d(371): Error: cannot evaluate injectNamedFields() at compile time
> std/typecons.d(371): Error: argument to mixin must be a string, not (injectNamedFields())
> 
> Anyways, I'll check in a workaround.
> 
> 
> Shin

I just looked at the tester and see that the build failure has been resolved, but the unittest step is failing with an Access Violation.

http://d.puremagic.com/test-results/test_data.ghtml?dataid=3420

On 9/17/2010 8:15 PM, Brad Roberts wrote:
> Please take some time to reduce it to a standalone reproducible test case and file a bugzilla entry for it.
> 
> On 9/17/2010 6:23 PM, Shin Fujishiro wrote:
>> It depends on the order of compilation.  The following succeeds:
>>
>> % dmd -c std/concurrency.d std/string.d
>> % _
>>
>> While the following fails (note the order of arguments):
>>
>> % dmd -c std/string.d std/concurrency.d
>> std/typecons.d(336): Error: cannot evaluate empty("") at compile time
>> std/typecons.d(371): Error: cannot evaluate injectNamedFields() at compile time
>> std/typecons.d(371): Error: argument to mixin must be a string, not (injectNamedFields())
>>
>> Anyways, I'll check in a workaround.
>>
>>
>> Shin
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos