bad binomial sum

[Harald Schilly]

Hi, I got this from the report a problem link:

Typing (in the inotebook)

var('t,k,i')
sum(binomial(i+t,t),i,0,k)

results in

binomial(k + t + 1, t + 1) - 1

which is false, the well-known answer is binomial(k + t + 1, t + 1)

[andrejv]

There is a fix for this bug in maxima cvs. If you don't want to wait
for the next release I can provide a patch.

Andrej

[Burcin Erocal]

Hi Andrej,

Do you know when the next release will be?

If it will be more than 2 - 3 weeks, then we should patch our maxima
package in the next Sage release (4.3.2).

I opened a ticket for this:

http://trac.sagemath.org/sage_trac/ticket/7952

Thank you.

Burcin

[andrejv]

> andrejv <andrej.vodopi...@gmail.com> wrote:
> > On Jan 10, 6:15 pm, Harald Schilly <harald.schi...@gmail.com> wrote:
> > > Hi, I got this from the report a problem link:
>
> > > Typing (in the inotebook)
>
> > > var('t,k,i')
> > > sum(binomial(i+t,t),i,0,k)
>
> > > results in
>
> > > binomial(k + t + 1, t + 1) - 1
>
> > > which is false, the well-known answer is binomial(k + t + 1, t + 1)
>
> > There is a fix for this bug in maxima cvs. If you don't want to wait
> > for the next release I can provide a patch.
>
> Do you know when the next release will be?
>
> If it will be more than 2 - 3 weeks, then we should patch our maxima
> package in the next Sage release (4.3.2).
>
> I opened a ticket for this:
>
> http://trac.sagemath.org/sage_trac/ticket/7952

The next Maxima release will be sometime in April I think. The patch
is very simple:

http://maxima.cvs.sourceforge.net/viewvc/maxima/maxima/src/combin.lisp?r1=1.39&r2=1.40&view=patch

Andrej