Closed odow closed 1 year ago
thanks a lot, yes indeed I did not implement the behavior according to the MOI standard, not a bug per se but this is a very peculiar expectation and I was avoiding to move to far away from the solver behaviour
but more coverage and consistency is a positive thing
x-ref: https://github.com/jump-dev/MathOptInterface.jl/pull/2085
There are actually a bunch of excluded tests in SCIP. That's usually suggestive of bugs in the implementation.