Calculating the optimal continued fraction of a quadratic irrational
This program finds the first n terms of the optimal continued fraction expansion
x=a0+
ε1/a1+
ε2/a2+ ···
of a quadratic irrational x = (u+t√d)/v, where d,t,u,v are integers, d >1 and nonsquare, t and v nonzero.
We first convert x to (P0+√d)/Q0 where Q0 divides d - P02.
Our account is based on the paper Optimal continued fractions by Wieb Bosma, Indag. Math. 49 (1987) 353-379. (See pseudo-code.)
We output the partial numerators and denominators εk and ak, the complete quotients ξk=(Pk+√d)/Qk and the convergents Ak/Bk for k = 0,,,n.
e=1 prints the bk and uk/vk.
Last modified 2nd April 2008
Return to main page