Re: strtod bugs

[Jim Meyering]

Eric Blake <address@hidden> wrote:

> According to Jim Meyering on 3/31/2008 11:24 AM:
> |
> | Actually, strtod is doing the right thing: it produces -0,
> | with both the libc version and the gnulib-supplied function.
> |
> | The trouble is that signbit(result) returns INT_MIN,
> | while signbit(-0.0) returns 1.  Both seem to be allowed, so
> | how about this change?  It solves the problem for me.
>
> Good catch.  Yes, signbit is allowed to return any non-zero value when the
> argument is negative, so your patch looks correct.  However,
>
> | +    ASSERT (!!signbit (result) == !!signbit (-(double)0.0)); /* IRIX 6.5 */
>
> Any reason you added the cast to double here?

Oh!  I knew there was something else to remove.
I added that when debugging.  Will remove.  Thanks!