Comment #2 by peter.alexander.au — 2013-09-14T13:45:09Z
(In reply to comment #0)
> As the .stringof representation often changes between compiler versions, it is
> not suitable for code generation because it is unreliable.
Wouldn't a better solution be to define exactly what .stringof should return? Implementation-defined semantics should be avoided where possible since it breaks portability.
Comment #3 by andrej.mitrovich — 2013-09-14T13:51:20Z
(In reply to comment #2)
> (In reply to comment #0)
> > As the .stringof representation often changes between compiler versions, it is
> > not suitable for code generation because it is unreliable.
>
> Wouldn't a better solution be to define exactly what .stringof should return?
> Implementation-defined semantics should be avoided where possible since it
> breaks portability.
See Kenji's comment here:
http://d.puremagic.com/issues/show_bug.cgi?id=10722#c6
Comment #4 by peter.alexander.au — 2013-09-14T14:43:06Z
(In reply to comment #3)
> (In reply to comment #2)
> > (In reply to comment #0)
> > > As the .stringof representation often changes between compiler versions, it is
> > > not suitable for code generation because it is unreliable.
> >
> > Wouldn't a better solution be to define exactly what .stringof should return?
> > Implementation-defined semantics should be avoided where possible since it
> > breaks portability.
>
> See Kenji's comment here:
> http://d.puremagic.com/issues/show_bug.cgi?id=10722#c6
I can't see anything in Kenji's comment that suggests it couldn't be properly defined?
Comment #5 by andrej.mitrovich — 2013-09-16T08:25:30Z