I’d like to raise your attention to this issue that appeared in CMSSW:
Questions for the ROOT team are:
Was this behaviour recently changed? I could browse the history myself, but maybe somebody remembers.
Shouldn’t TMath::AreEqualAbs use a less-than-or-equal comparison, so that
epsilon=0can still succeed?
Any suggestions on how to check the compatibility of histograms for add/merge? Catching the exception is sort-of ok, but discouraged; the helpers like
CheckConsistencyare not public.
ROOT Version: 6.12/07 via CMSSW