Avoid cast of double nan to int.

It may cause exceptions on some platform.

Originally committed as revision 25311 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavutil/rational.c

@@ -96,10 +96,12 @@ AVRational av_sub_q(AVRational b, AVRational c){
 AVRational av_d2q(double d, int max){
     AVRational a;
 #define LOG2  0.69314718055994530941723212145817656807550013436025
-    int exponent= FFMAX( (int)(log(fabs(d) + 1e-20)/LOG2), 0);
-    int64_t den= 1LL << (61 - exponent);
+    int exponent;
+    int64_t den;
     if (isnan(d))
         return (AVRational){0,0};
+    exponent = FFMAX( (int)(log(fabs(d) + 1e-20)/LOG2), 0);
+    den = 1LL << (61 - exponent);
     av_reduce(&a.num, &a.den, (int64_t)(d * den + 0.5), den, max);
 
     return a;