Exercise 5.3 / 5.4

[pbrtfan]

In these exercises, should we assume that the cos(theta) and h^2 terms
are constant over the unit disk / quadrilateral? If so the integral
obviously becomes very simple as all the constant terms can be moved
outside it.

However if they are varying over the surface I'm not sure the best way
to perform the integration without resorting to numerical methods.
Could someone point me in the right direction? I feel like I'm
missing something very obvious..

Thanks

[pbrtfan]

For the record I found a nice derivation on Iniqigo Quilez's site:

http://iquilezles.org/www/articles/sphereao/sphereao.htm

If anyone can recommend some other resources I'd appreciate it.