Fix for bug [ 1370433 ] trigsimp(sqrt(%i2)) !=sqrt(trigsimp(%i2))?

Richard Fateman wrote:
> I think the issue is more basic.
>
> the bug is in the simplifier, which transforms sqrt(x^2) to abs(x).
>
>   
Isn't that because radexpand defaults to true?  So perhaps radcan isn't 
really the problem but that radexpand is true?  sqrt is documented to 
return abs(x) if radexpand is true.
>
> There are two square roots. Which do you mean?  { there is a typo-- it is
> either 2*abs(cos(x))  or -2*abs(cos(x)).}
>
>   
Oops.  You're right.
> Since you have not specified which one, how is Maxima to know?
>   
Doesn't sqrt(x) mean the principal square root?

Ray