Name: V. Shtabovenko Date: 08/21/19-11:12:33 PM Z

Hi,

thanks for the bug report. In fact, I fixed it in the dev version almost
two and a half years ago, but somehow forgot to propagate it back into
the stable branch:

https://github.com/FeynCalc/feyncalc/commit/4abb8e06c1976055875397911c0396bdf1146114

It’s about the Dimension option of SetMandelstam, which should be set
to to {4,D} instead of {4,D,___} since the latter sets scalar products for
all possible dimensions.

I have just added that old commit to the stable branch, so the bug
should be now fixed upon reinstalling FeynCalc.

The D-4 scalar products (and 4-vectors) are not set to zero by default,
since they can be useful for cross checks. One can always do something like

Momentum[p1,D-4]=0;

though.

Cheers,
Vladyslav

Am 21.08.19 um 12:04 schrieb Xiangpeng Wang:
> Hi,
>
> I think there might be a bug about SetMandelstam in D Dimension.
>
>
> In[5]:= $BreitMaison = True;
>
> In[13]:= SetMandelstam[s, t, u, p1, p2, -k1, -k2, 0, 0, m, m];
>
> In[17]:= Pair[Momentum[k1, -4 + D], Momentum[k1, -4 + D]]
>
> Out[17]= m^2
>
> Which should be 0.
>
> In some gamma5 involved D-dimensional traces, there would lead wrong
> results:
>
>
> In[16]:= test1 = DiracTrace[(m + DiracGamma[Momentum[k2, D],
> D]).DiracGamma[ LorentzIndex[mu3, D], D].(-m +
> DiracGamma[Momentum[k1, D], D]).DiracGamma[ 5].(m +
> DiracGamma[Momentum[k1, D], D]).DiracGamma[ LorentzIndex[mu1, D],
> D].(m + DiracGamma[Momentum[-k1 - k2, D], D]).DiracGamma[
> LorentzIndex[mu3, D], D].(-m + DiracGamma[Momentum[k2, D],
> D]).DiracGamma[5]] /. DiracTrace -> Tr /. D -> 4
>
> Out[16]= 0
>
> In[15]:= test2 = ChangeDimension[ DiracTrace[(m +
> DiracGamma[Momentum[k2, D], D]).DiracGamma[ LorentzIndex[mu3, D],
> D].(-m + DiracGamma[Momentum[k1, D], D]).DiracGamma[ 5].(m +
> DiracGamma[Momentum[k1, D], D]).DiracGamma[ LorentzIndex[mu1, D],
> D].(m + DiracGamma[Momentum[-k1 - k2, D], D]).DiracGamma[
> LorentzIndex[mu3, D], D].(-m + DiracGamma[Momentum[k2, D],
> D]).DiracGamma[5]], 4] /. DiracTrace -> Tr
>
> Out[15]= -32 m^2 (m^2-s) Overscript[k1, _]^mu1
>