On Fri, Apr 18, 2014 at 2:06 PM, Martin Storsjö <[email protected]> wrote: > On Fri, 18 Apr 2014, Janne Grunau wrote: > >> On 2014-04-17 20:17:42 +0200, Anton Khirnov wrote: >>> >>> >>> On Thu, 17 Apr 2014 12:57:35 +0200, Janne Grunau <[email protected]> >>> wrote: >>>> >>>> On 2014-04-17 12:46:37 +0200, Anton Khirnov wrote: >>>>> >>>>> --- >>>>> tests/tiny_psnr.c | 5 +++++ >>>>> 1 file changed, 5 insertions(+) >>>>> >>>>> diff --git a/tests/tiny_psnr.c b/tests/tiny_psnr.c >>>>> index 66eaf82..5f819bb 100644 >>>>> --- a/tests/tiny_psnr.c >>>>> +++ b/tests/tiny_psnr.c >>>>> @@ -223,6 +223,11 @@ int main(int argc, char *argv[]) >>>>> a = get_f32l(buf[0] + j); >>>>> b = get_f32l(buf[1] + j); >>>>> } >>>>> + if (isnan(a) || isnan(b)) { >>>>> + fprintf(stderr, "Invalid values in the input.\n"); >>>>> + return 1; >>>>> + } >>>> >>>> >>>> (-)inf is handled as expected? just asking since you asked about >>>> isfinite >>>> first >>>> >>> >>> Actually no, seems I tested it wrong. >>> >>> When it's converted to fixed point, the inf becomes zero. >>> >>> Shall I use isfinite() or a separate isinf() call? >> >> >> isfinite is nicer we could try our luck if everything supports that c99 >> feature. if you want to be careful use !isnan/!isinf. Testing the result >> fabs(b - a) is preferable in that case. > > > Unsurprisingly, MSVC has not got any of them (isinf, isnan, isfinite). >
There is _isnan and _finite, not C99 names, but compatible functions. _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
