A bug when calculating FCLoopFindIntegralMappings[{GLI[...,{0...}],...}, topos]

[ygtw1]

topos = {
FCTopology[topo1, {SFAD[{p1, m^2}], SFAD[{p2, m^2}]}, {p1, p2}, {}, {}, {}],
FCTopology[topo2, {SFAD[{p3, m^2}], SFAD[{p4, m^2}]}, {p3, p4}, {}, {}, {}]
};
glis = {GLI[topo1, {0, 0}], GLI[topo1, {0, -1}], GLI[topo1, {1, 1}], GLI[topo1, {1, 2}], GLI[topo1, {2, 1}], GLI[topo2, {1, 1}],
GLI[topo2, {2, 2}]};

FCLoopFindIntegralMappings[glis, topos]

FCFeynmanPrepare::failmsg: Error! FCFeynmanPrepare has encountered a fatal problem and must abort the computation. The problem reads: The input expression is not a proper integral or list of propagators

$Aborted

[ygtw1]

It might be possible for FCLoopFindIntegralMappings to output the following result：
{{GLI[topo1, {0, 0}]->0,GLI[topo1, {0, -1}]->0,
GLI[topo2, {1, 1}] -> GLI[topo1, {1, 1}],
GLI[topo1, {2, 1}] -> GLI[topo1, {1, 2}]}, {
GLI[topo1, {1, 1}], GLI[topo1, {1, 2}], GLI[topo2, {2, 2}]}} for Options DropScaleless -> True,
or
{{GLI[topo2, {1, 1}] -> GLI[topo1, {1, 1}],
GLI[topo1, {2, 1}] -> GLI[topo1, {1, 2}]}, {GLI[topo1, {0, 0}],GLI[topo1, {0, -1}],
GLI[topo1, {1, 1}], GLI[topo1, {1, 2}], GLI[topo2, {2, 2}]}} for Options DropScaleless -> False.

[vsht]

As a quick fix: It's due to GLI[topo1, {0, 0}]. If you run FCLoopFindIntegralMappings after a reduction something like that will never appear, since FIRE/KIRA will set it to zero before head.

[vsht]

It should be fixed now, thanks.

FCLoopFindIntegralMappings will not explicitly map vanishing integrals to 0 as it is not its primary job.
However, it will also not get confused any more if there are scaleless integrals in the list.

[ygtw1]

OK, thanks!