#6870: [with third patch, needs review] Bug in binomial
------------------------------+---------------------------------------------
Reporter: hgranath | Owner: somebody
Type: defect | Status: new
Priority: minor | Milestone: sage-4.1.2
Component: basic arithmetic | Keywords: binomial
Reviewer: | Author:
Merged: |
------------------------------+---------------------------------------------
Comment(by kcrisman):
Replying to [comment:8 hgranath]:
> I really have no idea what to do. So assume we have something like
>
> {{{
> if isinstance(x,Expression):
> }}}
>
> then what to do? We have no idea at this point if x happens to be
something
> like SR(10**7) or n+1, and we will not know until we try x=ZZ(x). If
> that try should fail, it is already too late to avoid the time penalty
> that is the point of this discussion.
Understood; I think I didn't catch what the salient point was in your
earlier comment, but now I do, and I agree that there is no easy way
around it. My apologies.
Here is my last idea. I think that the .pyobject() method of Expression
can catch this - because it returns an error for anything which is not a
bare coefficient:
{{{
sage: def h():
try:
a.pyobject()
return ZZ(a)
except:
return a
sage: a = n+1
sage: h()
n + 1
sage: timeit('h()')
625 loops, best of 3: 5.98 µs per loop
sage: a = SR(10^7)
sage: timeit('h()')
625 loops, best of 3: 1.89 µs per loop
}}}
The docstring confirms this performs as advertised. If that doesn't speed
things up, then I guess it's not possible :(
So if that doesn't work, your original solution stands as a definite
improvement, with the changes you noted - put the rational check inside
the already existing not isinstance(m, int etc.) check, before trying m =
ZZ(m), do the floating point fix where you have it. Oh yeah, be sure to
add doctests for the SR(10**7) etc.
Because in the long run, we should have more correct cases, I think. If
someone notices a really bad slowdown, we will have to write a very fast
symbolic binomial or something. But all of these are an improvement on
before. Thanks for all of it!
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/6870#comment:9>
Sage <http://sagemath.org/>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en
-~----------~----~----~----~------~----~------~--~---
