On Sat, 07 Mar 2009 20:56:57 -0600, Barton Willis wrote:
Unfortunately, it doesn't work. I simplified to a toy example:
(%i1) exp: u(e)^(a-1)*v(e)^(1-a)+u(e)^a*v(e)^a;
a a a - 1 1 - a
(%o1) u (e) v (e) + u(e) v(e)
(%i2) facsum(diff(exp,e),diff(v(e),e), diff(u(e),e));
a + 1 - a - 1 2 a d
(%o2) (u(e) v(e) (a u(e) v(e) - a v(e) + v(e)) (-- (v(e)))
de
a 2 a d
u (e) (a u(e) v(e) + a v(e) - v(e)) (-- (u(e)))
de 2
+ --------------------------------------------------)/
u (e)
a
v (e)
What I would like to get as the result is the coefficients for v_e and
u_e, ie
[(1-a)*u(e)^(a-1)*v(e)^(-a)+a*u(e)^a*v(e)^(a-1),
(a-1)*u(e)^(a-2)*v(e)^(1-a)+a*u(e)^(a-1)*v(e)^a];
I am sure Maxima can do this easily, I just don't know how to achieve
it:-)
Tamas