http://d.puremagic.com/issues/show_bug.cgi?id=5718

std.traits.mangledName is broken. Being a regression, this should get fixed.

Who wants to sign up for it?

You may want to make a card for this in Trello so someone can grab it.

Thanks,

Andrei

On 1/31/12 1:58 AM, Walter Bright wrote:
> http://d.puremagic.com/issues/show_bug.cgi?id=5718
>
> std.traits.mangledName is broken. Being a regression, this should get fixed.
>
> Who wants to sign up for it?
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Why was the prefix for that symbol changed from "_D21demangle?" to "nt11__unittest1FZv59__T5DummyS46_D21demangle?"?  What does the "nt" signify?  The rest of the new symbol name looks like it will demangle correctly, but core.demangle does expect the symbol to begin with "_D".

On Jan 31, 2012, at 1:58 AM, Walter Bright wrote:

> http://d.puremagic.com/issues/show_bug.cgi?id=5718
> 
> std.traits.mangledName is broken. Being a regression, this should get fixed.
> 
> Who wants to sign up for it?
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

I don't know. The Dummy part is coming from the unmangle template, not the compiler. I don't know where the nt is coming from.


On 1/31/2012 2:27 PM, Sean Kelly wrote:
> Why was the prefix for that symbol changed from "_D21demangle?" to "nt11__unittest1FZv59__T5DummyS46_D21demangle?"?  What does the "nt" signify?  The rest of the new symbol name looks like it will demangle correctly, but core.demangle does expect the symbol to begin with "_D".
>
> On Jan 31, 2012, at 1:58 AM, Walter Bright wrote:
>
>> http://d.puremagic.com/issues/show_bug.cgi?id=5718
>>
>> std.traits.mangledName is broken. Being a regression, this should get fixed.
>>
>> Who wants to sign up for it?

Huh. Then I suggest fixing the mangling rather than the demangling.

On Jan 31, 2012, at 4:51 PM, Walter Bright <walter at digitalmars.com> wrote:

> I don't know. The Dummy part is coming from the unmangle template, not the compiler. I don't know where the nt is coming from.
> 
> 
> On 1/31/2012 2:27 PM, Sean Kelly wrote:
>> Why was the prefix for that symbol changed from "_D21demangle?" to "nt11__unittest1FZv59__T5DummyS46_D21demangle?"?  What does the "nt" signify?  The rest of the new symbol name looks like it will demangle correctly, but core.demangle does expect the symbol to begin with "_D".
>> 
>> On Jan 31, 2012, at 1:58 AM, Walter Bright wrote:
>> 
>>> http://d.puremagic.com/issues/show_bug.cgi?id=5718
>>> 
>>> std.traits.mangledName is broken. Being a regression, this should get fixed.
>>> 
>>> Who wants to sign up for it?
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

It is a limitation of current mangledName implementation.
But we cannot completely replace it to built-in mangleof property, see:
http://d.puremagic.com/issues/show_bug.cgi?id=5718#c5

And also, I've posted a pull to fix the original issue. https://github.com/D-Programming-Language/phobos/pull/414

Thanks.

Kenji Hara

2012/2/1 Walter Bright <walter at digitalmars.com>:
> I don't know. The Dummy part is coming from the unmangle template, not the compiler. I don't know where the nt is coming from.
>
>
>
> On 1/31/2012 2:27 PM, Sean Kelly wrote:
>>
>> Why was the prefix for that symbol changed from "_D21demangle?" to "nt11__unittest1FZv59__T5DummyS46_D21demangle?"? ?What does the "nt" signify? ?The rest of the new symbol name looks like it will demangle correctly, but core.demangle does expect the symbol to begin with "_D".
>>
>> On Jan 31, 2012, at 1:58 AM, Walter Bright wrote:
>>
>>> http://d.puremagic.com/issues/show_bug.cgi?id=5718
>>>
>>> std.traits.mangledName is broken. Being a regression, this should get fixed.
>>>
>>> Who wants to sign up for it?
>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos