Just found another bug in simplifyExp. In
p1 := numer p
there is risk of losing information because 'smpexp' can return a value
with 'denom' not equal to 1.
- Qian
diff --git a/src/algebra/manip.spad b/src/algebra/manip.spad
index 873fd5bf..3e6c163e 100644
--- a/src/algebra/manip.spad
+++ b/src/algebra/manip.spad
@@ -831,30 +831,30 @@ TranscendentalManipulations(R, F) : Exports ==
Implementation where
smp2htrigs p == map((k1 : K) : F +-> htrigs(k1::F), (r1 : R) : F
+-> r1::F, p)
simplifyExp x ==
- p := smpexp numer x
- q := smpexp denom x
- not (monomial?(p1 := numer p) and monomial?(q1 := numer q)) => p/q
+ f := smpexp numer x / smpexp denom x
+ p := numer f
+ q := denom f
+ not (monomial? p and monomial? q) => f
-- currently only fraction of monomials are simplified, e.g:
-- (exp(a)*b^c*d^e*...)/(exp(f)*b^g*h^i*...)
- lk1 := variables p1
- lk2 := variables q1
- term : F := 1
+ lk1 := variables p
+ lk2 := variables q
+ term : P := 1
for k2 in lk2 repeat
- -- assert!(degree(q1, k) = 1)
op := operator k2
args := argument k2
h := height k2
if op = opexp then
for k1 in lk1 repeat
if is?(k1, opexp) then
- term := term * kernel(op, [-(args.1)], h)::F
+ term := term * kernel(op, [-(args.1)], h)::P
break
if is?(op, POWER) then
for k1 in lk1 repeat
if is?(k1, POWER) and args.1 = first argument k1 then
- term := term * kernel(op, [args.1, -(args.2)],
h)::F
+ term := term * kernel(op, [args.1, -(args.2)],
h)::P
break
- termexp(numer(p*term))/termexp(numer(q*term))
+ termexp(p*term)/termexp(q*term)
htrigs f ==
(m := mainKernel f) case "failed" => f
diff --git a/src/input/bugs2023.input b/src/input/bugs2023.input
index 07008c6a..acb205e4 100644
--- a/src/input/bugs2023.input
+++ b/src/input/bugs2023.input
@@ -118,5 +118,6 @@ testcase "simplifyExp"
testEquals("height simplifyExp sin(exp(a)*exp(-a))", "1")
testEquals("simplifyExp((exp(a)*exp(b))^c)", "(exp(a+b))^c")
+testEquals("simplifyExp(exp(a-1)*exp(-a))", "exp(-1)")
statistics()
--
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 [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/fricas-devel/bca8e3aa-1024-4ad6-94e0-e959f604f59d%40gmail.com.
