Anti5 does not work in some cases.

[ygtw1]

In[30]:= GAD[5].(a GAD[u]+b GAD[u].GSD[p]+c GSD[p].GAD[u].GSD[q])//Anti5[#,Infinity]&

During evaluation of In[30]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of DiracGamma[6].

During evaluation of In[30]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of LC[FeynCalcFeynCalcInternalPrivatea___][FeynCalcFeynCalcInternalPrivateb___].

During evaluation of In[30]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of DiracGamma[6].

During evaluation of In[30]:= General::stop: Further output of $RecursionLimit::reclim2 will be suppressed during this calculation.

Out[30]= DiracGamma[
5].(a DiracGamma[LorentzIndex[u, D], D] +
b DiracGamma[LorentzIndex[u, D], D].DiracGamma[Momentum[p, D], D] +
c DiracGamma[Momentum[p, D], D].DiracGamma[LorentzIndex[u, D],
D].DiracGamma[Momentum[q, D], D])

[vsht]

Should be now fixed, thanks.

[ygtw1]

In some cases, Anti5 still does not work properly unless DotExpand is applied beforehand.

In[1]:= << FeynCalc`

In[2]:= GAD[5].(a GSD[p] + b GSD[q]).GAD[u] // Anti5[#, Infinity] &

During evaluation of In[2]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of {FeynCalcInternal->True}.

During evaluation of In[2]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of {FeynCalcInternal->True}.

During evaluation of In[2]:= $RecursionLimit::reclim2: Recursion depth of 1024 exceeded during evaluation of DiracGamma[6].

During evaluation of In[2]:= General::stop: Further output of $RecursionLimit::reclim2 will be suppressed during this calculation.

Out[2]= DiracGamma[
5].(a DiracGamma[Momentum[p, D], D] +
b DiracGamma[Momentum[q, D], D]).DiracGamma[LorentzIndex[u, D], D]

In[3]:= GAD[5].(a GSD[p] + b GSD[q]).GAD[u] //
Anti5[# // DotExpand, Infinity] &

Out[3]= a DiracGamma[Momentum[p, D], D].DiracGamma[LorentzIndex[u, D],
D].DiracGamma[5] +
b DiracGamma[Momentum[q, D], D].DiracGamma[LorentzIndex[u, D],
D].DiracGamma[5]

[vsht]

Should be now also fixed. TBH, I never really used Anti5 for anything because DiracSimplify was more useful for most situations.

[ygtw1]

There are still some minor issues with Anti5.

In[64]:= GAD[5].(a + GSD[p]). (b + GSD[p]);
% // Anti5[#, Infinity] & // FCE
%% // Anti5[#, Infinity] & // Anti5[#, Infinity] & // FCE

Out[65]= -a GSD[p].GA[5] - b GSD[p].GA[5] + GA[5].GSD[p].GSD[p] +
a b GA[5]

Out[66]= (-a - b) GSD[p].GA[5] + GSD[p].GSD[p].GA[5] + a b GA[5]

For certain complicated DiracGamma chains, Anti5 can be very fast, while DiracSimplify may take a long time.

[vsht]

I added a Collect2 to the result, so that the output should be now more consistent.

[ygtw1]

In[64]:= GAD[5].(a + GSD[p]). (b + GSD[p]);

In the example above, Anti5[#, Infinity] & must be applied twice before γ5 is fully moved to the rightmost position.

[vsht]

Thanks, should be now fixed.

[ygtw1]

OK, thanks.