bessel function: bug or not?



> (%i1) bessel_j(0,1);
> (%o1) bessel_j(0,1);
>
> other case
> (%i2) bessel_j(0,1.0);
> (%o2) 0.76519768655797

It's not a bug - it's the correct Maxima's behavior - an approximate  
floating point value is computed if the function is given approximate  
floating point arguments - the same as with

sin(1);

vs

sin(1.0);

Maxima doesn't give approximate numerical results unless it's told to  
do so (either with floating point arguments or with float() function).

Regards,
Ziga