I'm looking to apply Newton's method with a positivity constraint, using what I suppose will be an interior point (log-barrier) method, I have both the gradient and Hessian in analytic form, has this been implemented in any of the Julia optimization packages? I looked at the documentation for Optim.jl, but its Fminbox() function doesn't appear to use Newton's method for its optimization procedure.