LToolsEvaluate does not work with DB0 or DB1

[kMoehling]

Using FeynCalc 10.1.0 and FeynHelpers 2.0.0 I am currently trying to numerically evaluate expressions containing DB0, DB1 (and ideally also DB00 which exists in LoopTools but not FeynCalc).

However, applying LToolsEvaluate to e.g. DB1 gives the error:

"List of unique scalar integrals contains integrals that are not written as PaVe functions"

As far as I can see, the issue is that DB0 and DB1 are not converted to PaVe[] functions by ToPaVe2 (as intended),
but the algorithm then only replaces PaVe[] functions by their LoopTools equivalent.

By the way, the option PaVeIntegralHeads of FCLoopSplit and FCLoopExtract (used by LToolsEvaluate to extract the loop integrals present in the expression) by default contains DB1, but not DB0, such that DB1 is recognized by LToolsEvaluate as
as loop function, but DB0 is not.

[vsht]

Thanks for the comment. I added the support for evaluating DB0 and DB1 in the master branch, so this should work now

LToolsEvaluate[DB0[1, 2, 3]]
LToolsEvaluate[DB1[1, 2, 3, BReduce -> False]]

DB00 indeed doesn't exist in FeynCalc, just as DB11, DB001 and DB111. TBH I never used DB0 and DB1 in FeynCalc.

If I need to differentiate a loop integral, I'd rather do it with FCLoopGLIDifferentiate and then run IBPs to simplify it. I guess that people use DB0 and DB1 to calculate something in the on-shell scheme, but otherwise it's hard for me to think of relevant use cases.

[kMoehling]

Thanks for fixing this! I am indeed doing some calculations in the on-shell scheme and DB0, DB1 (and also B00 / DB00) are quite useful to write the field renormalization compactly.

[vsht]

At the moment I don't really have time to add more of these explicit PaVe functions, but if you want to make a pull request I could add it to code.