isfinite error with lyapunov but not lyapunovspectrum

[ianwineman]

Describe the bug
Using the lyapunov function returns a method error for isfinite where lyapunovspectrum works as expected.

Example error output

julia> lyapunov(sm, 10_000)
ERROR: MethodError: no method matching isfinite(::SVector{2, Float64})
The function `isfinite` exists, but no method is defined for this combination of argument types.

Closest candidates are:
  isfinite(::Missing)
   @ Base missing.jl:101
  isfinite(::BigFloat)
   @ Base mpfr.jl:1048
  isfinite(::Measurements.Measurement)
   @ Measurements ~/.julia/packages/Measurements/pivk0/src/comparisons-tests.jl:52
  ...

Stacktrace:
 [1] (::DynamicalSystemsBase.var"#1#2")(x::SVector{2, Float64})
   @ DynamicalSystemsBase ~/.julia/packages/DynamicalSystemsBase/kc7rt/src/core/dynamicalsystem_interface.jl:345
 [2] _all
   @ ./reduce.jl:1310 [inlined]
 [3] all
   @ ./reducedim.jl:995 [inlined]
 [4] successful_step
   @ ~/.julia/packages/DynamicalSystemsBase/kc7rt/src/core/dynamicalsystem_interface.jl:345 [inlined]
 [5] successful_step
   @ ~/.julia/packages/DynamicalSystemsBase/kc7rt/src/derived_systems/parallel_systems.jl:136 [inlined]
 [6] lyapunov(pds::DynamicalSystemsBase.ParallelDynamicalSystemAnalytic{…}, T::Int64; Ttr::Int64, Δt::Int64, d0::Float64, d0_upper::Float64, d0_lower::Float64, show_progress::Bool)
   @ ChaosTools ~/.julia/packages/ChaosTools/PqsgZ/src/chaosdetection/lyapunovs/lyapunov.jl:113
 [7] lyapunov(ds::DeterministicIteratedMap{…}, T::Int64; u0::SVector{…}, d0::Float64, inittest::ChaosTools.var"#64#65"{…}, kwargs::@Kwargs{})
   @ ChaosTools ~/.julia/packages/ChaosTools/PqsgZ/src/chaosdetection/lyapunovs/lyapunov.jl:64
 [8] lyapunov(ds::DeterministicIteratedMap{…}, T::Int64)
   @ ChaosTools ~/.julia/packages/ChaosTools/PqsgZ/src/chaosdetection/lyapunovs/lyapunov.jl:55
 [9] top-level scope
   @ REPL[10]:1
Some type information was truncated. Use `show(err)` to see complete types.

Works:

using DynamicalSystems, PredefinedDynamicalSystems
sm = PredefinedDynamicalSystems.standardmap([0.1, 0.11]; k=0.6)
lyapunovspectrum(sm, 10_000)

Does not work:

using DynamicalSystems, PredefinedDynamicalSystems
sm = PredefinedDynamicalSystems.standardmap([0.1, 0.11]; k=0.6)
lyapunov(sm, 10_000)

Package versions

[f3fd9213] Attractors v1.29.1
[608a59af] ChaosTools v3.5.0
[ab4b797d] ComplexityMeasures v3.8.5
[5732040d] DelayEmbeddings v2.9.0
[61744808] DynamicalSystems v3.6.3
[6e36e845] DynamicalSystemsBase v3.15.0
[4665ce21] FractalDimensions v1.9.5
[31e2f376] PredefinedDynamicalSystems v1.5.0
[639c3291] RecurrenceAnalysis v2.1.2
[40b095a5] StateSpaceSets v2.5.2
[c804724b] TimeseriesSurrogates v2.7.3

[Datseris]

eh...? I've never heard this before. What's your Julia version? Can you also do st -m StaticArrays?

[ianwineman]

Julia version: 1.11.7 (2025-09-08) & [90137ffa] StaticArrays v1.9.15

[Datseris]

Okay, I don't have the time to fix this now but I believe the error occurs when checking if a successful step has been taken for the discrete time system, as this is the only reason to check whether the state is finite. Should be easy to resolve if you'd like to make a PR.