Bill Page wrote:
>
> The following sequence of commands:
>
> --
> integrate(sin(sqrt(x*x)),x)
> )trace INTEF )math
> integrate(sin(sqrt(x*x)),x)
> )trace INTEF )off
> integrate(sin(sqrt(x*x)),x)
> D(%,x)
> %-sin(sqrt(x*x))
> normalize %
> --
>
> produces unexpected results. For example:
>
> http://axiom-wiki.newsynthesis.org/SandBoxTraceAffectsResults
>
> I get slightly different but still wrong results in different versions
> of FriCAS.
I get error already when computing last integral. This probably
is the same problem that I described with subject 'Troubles with
kernels'. I consider this a bug...
More precisely, ')trace' seem to cause no troubles. However
')untrace' causes re-evaluation of 'Expression' and 'Kernel'.
>From one point of view re-evaluation is reasonable. But in this
case apparently we get a kernel entered in old cache and assigned
a position. Then we create a new cache and for new kernels
use positions in new chache. This is incorrect, becuse comparisons
of kernels compare positions. Of course postions in old cache
and new one are independent and we get all kind of weird things
like x which is not equal to x.
ATM doing ')clear completely' after ')untrace' seems to
get rid of the problem.
--
Waldek Hebisch
[email protected]
--
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 post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
