[Spirit-general] Re: Phoenix: VC 7.1 workaround needed

[Stefan S. <ste...@sl...>]

Joao Abecasis wrote:

> 
> g++ 3.4.2 also complains about an invalid conversion from const
> char * to int. I guess this has something to do with the type
> deduction mechanism.
> 

that seems to be the same as the debugger told me the Intel compiler 
executable: the std::string is constructed by two int 'iterators' that 
are the result of the "char const* + int" operations.

I wanted to explorer that in a small test cast but I wasn't successful 
due to lack of time; higher_rank<const char*, int>::type is const char* 
as it's supposed to be.

> Have you given phoenix-2 a try with this?
> 

no, maybe I can make it tomorrow; anyway, I'll have to get this running 
with phoenix-1.

> I'm attaching a phoenix-2-ified version of your test. With the
> same compiler I now get a warning about returning a reference to
> a temporary (Joel, is this expected?).
> 
> regards,
> 
> 
> João