#2341 Invalid simplification with radcan

[Michael Orlitzky]

Is this a valid way to simplify logarithms of potentially negative values? If so, it looks like radcan() is assuming t is positive? It does the same thing even if I add assume(t<0).

(%i1) display2d: false;

(%o1) false
(%i2) domain: complex;

(%o2) complex
(%i3) declare(t, real);

(%o3) done
(%i4) f(t) := log(2*t) + log(1/t);

(%o4) f(t):=log(1/t)+log(2*t)
(%i5) radcan(f(t));

(%o5) log(2)
(%i6) radcan(f(-1));

(%o6) log(-1)+log(-2)
(%i7) bug_report();

...

Maxima version: 5.26.0
Maxima build date: 18:36 1/18/2012
Host type: x86_64-pc-linux-gnu
Lisp implementation type: ECL
Lisp implementation version: 11.1.1