What is the purpouse of DiracChainEvalSign and why does FCFAconvert fail if it is present?

[SamuelJosephs]

I have run into this issue and I am not sure how to proceed, I cannot find where DiracChainEvalSign enters into diracObjectsEval which triggers the error in the source code. Any pointers would be much appreciated!

For context this is my code with the .gen and .mod files attached:

tops = CreateTopologies[0, 1 -> 3]
procs = InsertFields[tops, S[1] -> {F[2, {1}], -F[2, {1}], V[1]}, 
  Model -> "SMEFT_d6_UFO_FULL", GenericModel -> "SMEFT_d6_UFO_FULL"]
Paint[procs]

amp[0] = 
 FCFAConvert[CreateFeynAmp[procs], IncomingMomenta -> {pH}, 
  OutgoingMomenta -> {p1, p2, k}, List -> False, ChangeDimension -> 4,
   TransversePolarizationVectors -> {k}, DropSumOver -> True, 
  SMP -> True, Contract -> True, UndoChiralSplittings -> True]

SMEFT_d6_UFO_FULL.gen.txt
SMEFT_d6_UFO_FULL.mod.txt

And this is the error:
FCFADiracChainJoin has encountered a fatal problem and must abort the
computation. The problem reads: Evaluation of isolated objects failed.

Any Help is much appreciated!

[vsht]

Your model contains undefined symbols carrying Dirac indices and depending on a Dirac matrix:

$LoadAddOns = {"FeynArts"};
FAPatch[FAModelsDirectory -> "SMEFT", PatchModelsOnly -> True];
<< FeynCalc`

$FAVerbose = False;
tops = CreateTopologies[0, 1 -> 3];
procs = InsertFields[tops, S[1] -> {F[2, {1}], -F[2, {1}], V[1]}, 
   Model -> 
    FileNameJoin[{NotebookDirectory[], "SMEFT", "SMEFT_d6_UFO_FULL"}],
    GenericModel -> 
    FileNameJoin[{NotebookDirectory[], "SMEFT", "SMEFT_d6_UFO_FULL"}]];
Paint[procs];

amp[0] = 
  FCFAConvert[CreateFeynAmp[DiagramExtract[procs, {6}]], 
   IncomingMomenta -> {pH}, OutgoingMomenta -> {p1, p2, k}, 
   List -> False, ChangeDimension -> 4, 
   TransversePolarizationVectors -> {k}, DropSumOver -> True, 
   SMP -> True, Contract -> True, UndoChiralSplittings -> True, 
   FCFADiracChainJoin -> False];

SelectNotFree[amp[0][[1]], G]
% // InputForm

G[-1][0][-F[2, {1}], F[Index[Generic, 5]], V[1]][DiracObject[DiracGamma[LorentzIndex[KI1[3]]], DiracGamma[6]][KI1[1], KI1[2]]]*
 (DiracChain[DiracGamma[6], DiracIndex[Dir2], DiracIndex[Dir3]]*G[-1][0][F[2, {1}], -F[Index[Generic, 5]], S[1]][DiracObject[DiracGamma[6]][KI1[1], KI1[2]]] + 
  DiracChain[DiracGamma[6], DiracIndex[Dir3], DiracIndex[Dir2]]*G[-1][0][F[2, {1}], -F[Index[Generic, 5]], S[1]][DiracObject[DiracGamma[6]][KI1[2], KI1[1]]] + 
  DiracChain[DiracGamma[7], DiracIndex[Dir2], DiracIndex[Dir3]]*G[-1][0][F[2, {1}], -F[Index[Generic, 5]], S[1]][DiracObject[DiracGamma[7]][KI1[1], KI1[2]]] + 
  DiracChain[DiracGamma[7], DiracIndex[Dir3], DiracIndex[Dir2]]*G[-1][0][F[2, {1}], -F[Index[Generic, 5]], S[1]][DiracObject[DiracGamma[7]][KI1[2], KI1[1]]])

Unless those G's are spelled out explicitly in terms of Dirac matrices, FeynCalc is unable to recover a closed Dirac chain from this kind of input.

[SamuelJosephs]

No I don't think FAPatch is introducing them, but I'm still really not sure how they are ending up in the un diracChainJoined output.

[SamuelJosephs]

As they just aren't in the input .mod/.gen files in a way that isn't in the feynArts specification as detailed here on page 56

[SamuelJosephs]

Ah, it seems that CreateFeynAmp[] is inserting them, curious.

[SamuelJosephs]

it was Insertion level, I needed to mandate that InsertionLevel->{classes} then it works. Thanks for your help!