get_fsalfirstlast method error

[ParamThakkar123]

solving the ODEProblem in SciMLBenchmarks for Henon_Heiles_energy_conservation_benchmarks using TaylorMethod() function gives the following method call error which didn't occur in the previous versions

MethodError: no method matching get_fsalfirstlast(::TaylorIntegrationDiffEqExt.TaylorMethodCache{Vector{Float64}, Vector{Float64}, Taylor1{Float64}, Vector{Taylor1{Float64}}}, ::Vector{Float64})



Closest candidates are:

  get_fsalfirstlast(!Matched::OrdinaryDiffEqQPRK.QPRK98Cache, ::Any)

   @ OrdinaryDiffEqQPRK C:\Users\Hp\.julia\packages\OrdinaryDiffEqQPRK\7AUMg\src\qprk_caches.jl:33

  get_fsalfirstlast(!Matched::OrdinaryDiffEqLowOrderRK.MSRK6Cache, ::Any)

   @ OrdinaryDiffEqLowOrderRK C:\Users\Hp\.julia\packages\OrdinaryDiffEqLowOrderRK\CPmMP\src\low_order_rk_perform_step.jl:1455

  get_fsalfirstlast(!Matched::OrdinaryDiffEqLowOrderRK.SIR54Cache, ::Any)

   @ OrdinaryDiffEqLowOrderRK C:\Users\Hp\.julia\packages\OrdinaryDiffEqLowOrderRK\CPmMP\src\low_order_rk_perform_step.jl:1696

  ...





Stacktrace:
...
   @ c:\Users\Hp\Desktop\SciMLBenchmarks.jl\benchmarks\DynamicalODE\jl_notebook_cell_df34fa98e69747e1a8f8a730347b8e2f_W3sZmlsZQ==.jl:4 [inlined]

 [6] top-level scope

   @ c:\Users\Hp\Desktop\SciMLBenchmarks.jl\benchmarks\DynamicalODE\jl_notebook_cell_df34fa98e69747e1a8f8a730347b8e2f_W5sZmlsZQ==.jl:1

Code to reproduce the issue :

https://github.com/SciML/SciMLBenchmarks.jl/blob/master/benchmarks/DynamicalODE/Henon-Heiles_energy_conservation_benchmark.jmd

on Julia version 1.10.4 and TaylorIntegration version 0.16.1, OrdinaryDiffEq version 6

[ChrisRackauckas]

TaylorIntegration.jl needs a dispatch for the method.

[ParamThakkar123]

Should I raise an issue on the TaylorIntegration.jl repository for this ??

[ChrisRackauckas]

yes or PR

[ParamThakkar123]

raised an issue in TaylorIntegration.jl PerezHz/TaylorIntegration.jl#200

will start with a PR soon.

[ParamThakkar123]

Should the dispatch method be implemented in OrdinaryDiffEq.jl or should it be in TaylorIntegration.jl. I implemented the dispatch methods in TaylorIntegration.jl code but they still error here :

 Expression: solve(prob, TaylorMethod(20), abstol = 1.0e-20)
  MethodError: no method matching get_fsalfirstlast(::TaylorIntegrationDiffEqExt.TaylorMethodCache{Vector{Float64}, Vector{Float64}, Taylor1{Float64}, Vector{Taylor1{Float64}}}, ::Vector{Float64})
  The function `get_fsalfirstlast` exists, but no method is defined for this combination of argument types.

  Closest candidates are:
    get_fsalfirstlast(::OrdinaryDiffEqIMEXMultistep.IMEXMutableCache, ::Any)
     @ OrdinaryDiffEqIMEXMultistep C:\Users\Hp\.julia\packages\OrdinaryDiffEqIMEXMultistep\2dPDm\src\imex_multistep_caches.jl:3
    get_fsalfirstlast(::OrdinaryDiffEqLowOrderRK.MidpointCache, ::Any)
     @ OrdinaryDiffEqLowOrderRK C:\Users\Hp\.julia\packages\OrdinaryDiffEqLowOrderRK\CPmMP\src\fixed_timestep_perform_step.jl:192
    get_fsalfirstlast(::OrdinaryDiffEqFunctionMap.FunctionMapCache, ::Any)
     @ OrdinaryDiffEqFunctionMap C:\Users\Hp\.julia\packages\OrdinaryDiffEqFunctionMap\15mZr\src\functionmap_caches.jl:6
    ...

  Stacktrace:
    [1] __init(prob::ODEProblem{Vector{Float64}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, SciMLBase.AutoSpecialize, var"#f!#382", LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}, @Kwargs{}, SciMLBase.StandardODEProblem}, alg::TaylorIntegrationDiffEqExt.TaylorMethodParams, timeseries_init::Tuple{}, ts_init::Tuple{}, ks_init::Tuple{}, recompile::Type{Val{true}}; saveat::Tuple{}, tstops::Tuple{}, d_discontinuities::Tuple{}, save_idxs::Nothing, save_everystep::Bool, save_on::Bool, save_start::Bool, save_end::Nothing, callback::Nothing, dense::Bool, calck::Bool, dt::Float64, dtmin::Float64, dtmax::Float64, force_dtmin::Bool, adaptive::Bool, gamma::Rational{Int64}, abstol::Float64, reltol::Nothing, qmin::Rational{Int64}, qmax::Int64, qsteady_min::Int64, qsteady_max::Int64, beta1::Nothing, beta2::Nothing, qoldinit::Rational{Int64}, controller::Nothing, fullnormalize::Bool, failfactor::Int64, maxiters::Int64, internalnorm::typeof(DiffEqBase.ODE_DEFAULT_NORM), internalopnorm::typeof(LinearAlgebra.opnorm), isoutofdomain::typeof(DiffEqBase.ODE_DEFAULT_ISOUTOFDOMAIN), unstable_check::typeof(DiffEqBase.ODE_DEFAULT_UNSTABLE_CHECK), verbose::Bool, timeseries_errors::Bool, dense_errors::Bool, advance_to_tstop::Bool, stop_at_next_tstop::Bool, initialize_save::Bool, progress::Bool, progress_steps::Int64, progress_name::String, progress_message::typeof(DiffEqBase.ODE_DEFAULT_PROG_MESSAGE), progress_id::Symbol, userdata::Nothing, allow_extrapolation::Bool, initialize_integrator::Bool, alias_u0::Bool, alias_du0::Bool, initializealg::OrdinaryDiffEqCore.DefaultInit, kwargs::@Kwargs{})
      @ OrdinaryDiffEqCore C:\Users\Hp\.julia\packages\OrdinaryDiffEqCore\NnA60\src\solve.jl:475
    [2] solve(::ODEProblem{Vector{Float64}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, SciMLBase.AutoSpecialize, var"#f!#382", LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}, @Kwargs{}, SciMLBase.StandardODEProblem}, ::TaylorIntegrationDiffEqExt.TaylorMethodParams; verbose::Bool, kwargs::@Kwargs{abstol::Float64})
      @ TaylorIntegrationDiffEqExt C:\Users\Hp\.julia\packages\TaylorIntegration\CrO67\ext\TaylorIntegrationDiffEqExt.jl:244
    [3] (::var"#344#390"{ODEProblem{Vector{Float64}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, SciMLBase.AutoSpecialize, var"#f!#382", LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}, @Kwargs{}, SciMLBase.StandardODEProblem}})()
      @ Main .\none:0
    [4] with_logstate(f::var"#344#390"{ODEProblem{Vector{Float64}, Tuple{Float64, Float64}, true, SciMLBase.NullParameters, ODEFunction{true, SciMLBase.AutoSpecialize, var"#f!#382", LinearAlgebra.UniformScaling{Bool}, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, Nothing, typeof(SciMLBase.DEFAULT_OBSERVED), Nothing, Nothing, Nothing, Nothing, Nothing, Nothing}, @Kwargs{}, SciMLBase.StandardODEProblem}}, logstate::Base.CoreLogging.LogState)
      @ Base.CoreLogging .\logging\logging.jl:522
    [5] with_logger(f::Function, logger::TestLogger)
      @ Base.CoreLogging .\logging\logging.jl:632
    [6] #collect_test_logs#75
      @ C:\Users\Hp\.julia\juliaup\julia-1.11.1+0.x64.w64.mingw32\share\julia\stdlib\v1.11\Test\src\logging.jl:126 [inlined]
    [7] collect_test_logs
      @ C:\Users\Hp\.julia\juliaup\julia-1.11.1+0.x64.w64.mingw32\share\julia\stdlib\v1.11\Test\src\logging.jl:124 [inlined]
    [8] #match_logs#76
      @ C:\Users\Hp\.julia\juliaup\julia-1.11.1+0.x64.w64.mingw32\share\julia\stdlib\v1.11\Test\src\logging.jl:274 [inlined]
    [9] macro expansion
      @ C:\Users\Hp\Desktop\TaylorIntegration.jl\test\common.jl:75 [inlined]
   [10] macro expansion
      @ C:\Users\Hp\.julia\juliaup\julia-1.11.1+0.x64.w64.mingw32\share\julia\stdlib\v1.11\Test\src\Test.jl:1700 [inlined]
   [11] macro expansion
      @ C:\Users\Hp\Desktop\TaylorIntegration.jl\test\common.jl:73 [inlined]
   [12] macro expansion
      @ C:\Users\Hp\.julia\juliaup\julia-1.11.1+0.x64.w64.mingw32\share\julia\stdlib\v1.11\Test\src\Test.jl:1700 [inlined]
   [13] top-level scope
      @ C:\Users\Hp\Desktop\TaylorIntegration.jl\test\common.jl:10

[ParamThakkar123]

Dispatch method I implemented :

function get_fsalfirstlast(cache::TaylorIntegrationDiffEqExt.TaylorMethodCache{Matrix{Float64}, Matrix{Float64}, Taylor1{Float64}, Matrix{Taylor1{Float64}}}, 
    u::Matrix{Float64})
    # Extract the FSAL components from the cache
    if isdefined(cache, :fsalfirst) && isdefined(cache, :k)
        return cache.fsalfirst, cache.k
    else
        # Fallback for cases where FSAL components might not be defined
        error("FSAL components not defined in the cache")
    end
end

function get_fsalfirstlast(cache::TaylorMethodCache{Vector{Float64}, Vector{Float64}, Taylor1{Float64}, Vector{Taylor1{Float64}}}, u::Vector{Float64})
    # Extract the FSAL components from the cache
    if isdefined(cache, :fsalfirst) && isdefined(cache, :k)
        return cache.fsalfirst, cache.k
    else
        # Fallback for cases where FSAL components might not be defined
        error("FSAL components not defined in the cache")
    end
end

[ChrisRackauckas]

That's shadowing not extending. It should be OrdinaryDiffEqCore. get_fsalfirstlast and it has to be implemented in TaylorIntegrationDiffEqExt

[ParamThakkar123]

Yeah 👍 . Got it. Thank you !