Hi,
it is a bug, but I'm afraid that it will take me some time to fix it:
The source code of TARCER is very poorly documented and I have
my PhD defense next week.
This particular case uses special notation
TFI[d, pp, dp, {a,b}, {{n1,m1},{n2,m2},{n3,m3},{n4,m4},{n5,m5}}]
for including numerators with OPEDelta's, so it's more Rolf's expertise.
Anyhow, I'll look at it end of next week.
Cheers,
Vladyslav
Am 26.05.2017 um 16:52 schrieb zhang:
> Hi, all,
>
> When I run the examples notebook for Tarcer in
> FeynCalc 9.1.0 and FeynCalc 9.2.0.
>
> In[3]:= eg2[m_] := TFI[d, M^2, \[CapitalDelta]p, {m, 0},
> {{2, M}, 1, 1, {1, M},
> {1, M}}]
>
> In[4]:= TarcerRecurse[eg2[1]]
>
> Out[4]= -(((d-2)^2 (27 d^4-404 d^3+2175 d^2-4902 d+3776)
> Tarcer`Private`dp (Subsuperscript[A, {1,M}, (d)])^2)/
> (256 (d-5)^2 (d-4)^2 (d-3) M^8))-((d-2) (3 d-10) (3 d-8)
> Tarcer`Private`dp Subsuperscript[J, {1,M}{1,0}{1,0}, (d)])
> /(32 (d-5) (d-4) (2 d-7) M^6)+(d (3 d-10) (3 d-8)
> Tarcer`Private`dp Subsuperscript[J, {1,M}{1,M}{1,M}, (d)])
> /(128 (d-4)^2 M^6)
>
> An internal parameter Tarcer`Private`dp shows up in the final expression rather than the expected \[CapitalDelta]p.
> Is this a bug?
>
>
> Thanks!
>
> Zhang
>
This archive was generated by hypermail 2b29 : 09/04/20-12:55:05 AM Z CEST