#474 rootscontract expands unnecessarily

[Stavros Macrakis]

rootscontract((4*t+4)^(1/2)/(32*t+32)^(1/5)),
rootsconmode:all
=> (t^3+3*t^2+3*t+1)^(1/10)
instead of
(t+1)^(3/10)

There is no good reason for it to be expanding the wole
thing when there is a non-trivial gcd. At worst, it could
give the result in the form:

(4*t+4)^(3/10)/8^(1/5)

Then again, there doesn't seem to be any equivalent of
rootscontract for *integral* powers. How do you simplify

(4*t+4)^20/(t+1)^20

to 2^40

without using factor (which in general of course is much
more expensive than GCD)?

[Nobody/Anonymous]

Logged In: NO

ratsubst helps:

ex1:(4*t+4)^20/(t+1)^20
ratsubst(z, t+1, %);

[Robert Dodier]

Logged In: YES
user_id=501686

Observed in 5.9.3cvs.