I am lost in upgrading from v0.1.x to v0.2.x

[liuyxpp]

How can I upgrade following code to be v0.2.x compatible?

        F(x, p) = collect(binodal_condition(p.Aα, x..., p.model))
	J(x, p) = ForwardDiff.jacobian(z->F(z,p), x)
	par = (Aα=ϕAα, model=model)
	x0 = zeros(3)
	prob = BifurcationProblem(F, x0, par, (@lens _.Aα))  # for v0.2.x
	
	opts = BK.ContinuationPar(dsmax=0.02, dsmin=1e-5, ds=0.001, maxSteps=2000,
							pMin=ϕAα_min,
							pMax=ϕAα_max,
							saveSolEveryStep=1,
							maxBisectionSteps=50,
							detectBifurcation=3)
	opts = BK.ContinuationPar(opts; kwargs...)

	# br, = BK.continuation(F, J, x0, par, (@lens _.Aα), opts,
	# 		recordFromSolution = (x,p) -> (ϕBα=x[1], ϕAβ=x[2], ϕBβ=x[3], ϕAα=p),
	# 		bothside=bothside, verbosity=3)  # BifurcationKit version 0.1.X, what is the algo used here?
	br, = BK.continuation(prob, Natural(), opts,
			recordFromSolution = (x,p) -> (ϕBα=x[1], ϕAβ=x[2], ϕBβ=x[3], ϕAα=p),
			bothside=bothside, verbosity=3)  # Try v0.2.x but errored

The error information is

MethodError: no method matching iterate(::BifurcationKit.ContResult{BifurcationKit.EquilibriumCont, NamedTuple{(:x, :param, :itnewton, :itlinear, :ds, :θ, :n_unstable, :n_imag, :stable, :step), Tuple{Float64, Float64, Int64, Int64, Float64, Float64, Int64, Int64, Bool, Int64}}, Vector{ComplexF64}, Matrix{ComplexF64}, BifurcationKit.SpecialPoint{Float64, NamedTuple{(:x,), Tuple{Float64}}, Vector{Float64}, Vector{Float64}}, Vector{NamedTuple{(:x, :p, :step), Tuple{Vector{Float64}, Float64, Int64}}}, BifurcationKit.ContinuationPar{Float64, BifurcationKit.DefaultLS, BifurcationKit.DefaultEig{typeof(real)}}, BifurcationKit.BifurcationProblem{BifurcationKit.BifFunction{Main.var"workspace#21".var"#F#4", BifurcationKit.var"#6#21"{Main.var"workspace#21".var"#F#4"}, Nothing, BifurcationKit.var"#4#19"{Main.var"workspace#21".var"#F#4"}, Nothing, BifurcationKit.var"#9#25"{BifurcationKit.var"#d1Fad#23"{Main.var"workspace#21".var"#F#4"}}, BifurcationKit.var"#11#27", BifurcationKit.var"#13#29", BifurcationKit.var"#15#31", Bool, Float64}, Vector{Float64}, NamedTuple{(:Aα, :model), Tuple{Float64, NamedTuple{(:αA, :αB, :αS, :χABN, :χASN, :χBSN), NTuple{6, Float64}}}}, Setfield.PropertyLens{:Aα}, typeof(BifurcationKit.plotDefault), typeof(BifurcationKit.recordSolDefault)}, BifurcationKit.Natural})

Closest candidates are:

iterate(!Matched::Union{LinRange, StepRangeLen}) at range.jl:872

iterate(!Matched::Union{LinRange, StepRangeLen}, !Matched::Integer) at range.jl:872

iterate(!Matched::T) where T<:Union{Base.KeySet{<:Any, <:Dict}, Base.ValueIterator{<:Dict}} at dict.jl:712

...

indexed_iterate(::BifurcationKit.ContResult{BifurcationKit.EquilibriumCont,

...

Sorry for the bother. But I can not find any tips for migration in the doc.

[rveltz]

Yes the error is not very helpful. It should not be br, = continuation... but br = continuation...

prob = BifurcationProblem(F, x0, par, (@lens _.Aα) ; recordFromSolution = (x,p) -> (ϕBα=x[1], ϕAβ=x[2], ϕBβ=x[3], ϕAα=p),)
br = BK.continuation(prob, Natural(), opts,
			bothside=bothside, verbosity=3) 

[liuyxpp]

Ah, it is my bad, thanks! However, I do think it is a good idea to have a migration guide in the doc somewhere to ease transition.