Re: [GD-Windows] VC 2005 Warning C4244

[Jon W. <hp...@mi...>]

In the past, I've silenced that kind of compiler warning by explicitly 
masking with the range. I e:

u_short n = htons(x) & 0xffff;

The compiler, at the time, was smart enough to realize that the range of 
the masked value could never go outside the range of the stored variable.

I honestly forget which compiler it was, though, as it was a while ago, 
but I suggest you try it! If it doesn't work for MSVC, then you'll have 
to live with disabling 4244.

Cheers,

          / h+


Dan Glastonbury wrote:
> G'Day,
>   I'm trying to compile at warning level 4 and I'm getting a lot of
> spurious C4244.  This warning is about an integer narrow storing
> possibly loosing information.
>
> The problem is with this kind of code:
>
> u_short n = htons(x);
>
> htons returns a u_short but, as the standard says, the result of
> integer expressions is promoted to int (or unsigned) and is causing
> the warning.  Is there anyway I can code it to shut up VC 2005?
>
> u_short n = (u_short)htons(x) doesn't work, because the result is
> still promoted in the context of an expression!
>
> cheers
> DanG
>
>   

-- 
-- Go is to Western chess what philosophy is to double entry accounting.