On Dec 11, 2007 8:39 AM, Joel B. Mohler <[EMAIL PROTECTED]> wrote:
>
> Hi,
>
> I've noticed a very recent regression -- it worked 2 months ago.
>
> sage: t=var('t')
> sage: f=t*cos(0)
> sage: float(f(1))
> 1.0
> sage: f=t*sin(0)
> sage: float(f(1))
> Traceback...
> <type 'exceptions.TypeError'>: float() argument must be a string or a number
>
> --
It is actually hard to decide how to fix this. This is a result of
several significant fixes
and optimizations recently. What is happening is that for t*sin(0)
the simplified
form is 0, so (t*sin(0)).variables() is [].
sage: t=var('t')
sage: f = t*cos(0)
sage: f.variables()
(t,)
sage: g = t*sin(0)
sage: g.variables()
()
sage: float(f(1))
1.0
sage: float(g(t=1))
0.0
Both f(1) and g(1) are formal products. However:
sage: g(1)._operands
[t, 0]
sage: f(1)._operands
[1, 1]
Notice the [t, 0].
One possible solution would be to call simplify before
doing float(...) -- but that could greatly slow symbolic calculus
down in some cases. Another possibility would be to change
the definition of variables() to return all variables, even the ones
that are simplified away:
sage: (x - x).variables() # fake
(x,)
That would be very confusing.
A third possibility would be to make implicit calling use variables
in the unsimplified expression if the simplified expression has
no variables. This would cleanly deal with your case above.
Thoughts?
-- William
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
