On Tue, Jan 14, 2025 at 02:19:46PM +0100, 'Ralf Hemmecke' via FriCAS - computer
algebra system wrote:
> Hi Waldek,
>
> I know that rsimp isn't perfect and that you will improve it in the future.
> But somehow I find this result a bit weird given that 31^2=961.
>
> Ralf
>
> %%% (240) -> xx := sqrt(-155*sqrt(5)+543)
>
> +----------------+
> | +-+
> (240) \|- 155 \|5 + 543
> %%% (241) -> (rsimp(xx)$RootSimplification)
>
> +---+
> +-+ |961
> (31 \|5 - 25) |---
> \| 10
> (241) --------------------
> 31
The attached patch makes this simpler: when numerator or denominator
is an exact power it pulls corresponding factor before root sign.
--
Waldek Hebisch
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to fricas-devel+unsubscr...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/fricas-devel/Z4bcfUCckf032oxp%40fricas.org.
diff --git a/src/algebra/rsimp.spad b/src/algebra/rsimp.spad
index 8f2e867e..9ce27902 100644
--- a/src/algebra/rsimp.spad
+++ b/src/algebra/rsimp.spad
@@ -967,7 +967,13 @@ RootSimplification() : Exports == Implementation where
"failed"
my_root(s1 : eI, k : Integer, opr : BasicOperator) : eI ==
- numer(s1) = 1 => 1/kernel(opr, [1/s1, k::eI])
+ n1 := numer(s1)::eI
+ n1 = 1 => 1/kernel(opr, [1/s1, k::eI])
+ d1 := denom(s1)::eI
+ (nu1 := rsimp1(n1, k)) case eI =>
+ nu1::eI/kernel(opr, [d1, k::eI])
+ (du1 := rsimp1(d1, k)) case eI =>
+ kernel(opr, [n1, k::eI])/du1::eI
kernel(opr, [s1, k::eI])
eval_rl(rl : List(eI), kf : eI, k : Integer) : eI ==
