sol.retcode == :Success is not recognized correctly

[Denis-Titov]

I run the following code from the readme:

f(u,p,t) = 1.01*u
u0=1/2
tspan = (0.0,1.0)
prob = ODEProblem(f,u0,tspan)
sol = solve(prob,Tsit5(),reltol=1e-8,abstol=1e-8)
sol.retcode == :Success #false
Symbol(sol.retcode) == :Success #true

sol.retcode == :Success is false even though Symbol(sol.retcode) == :Success is true

This used to work correctly but stopped working now.

I used Julia 1.9.0-rc1 and DifferentialEquations v7.7.0

[oscardssmith]

It's been deprecated for a while and should now be sol.retcode == ReturnCode.Success.

[ChrisRackauckas]

That's not in the README? I don't see any sol.retcode in the README? Let me know if there's a spot that's still using the old check but I can't find it.

[Denis-Titov]

I saw sol.retcode != :Success in documentation here: https://docs.sciml.ai/DiffEqDocs/stable/basics/faq/#My-ODE-is-solving-really-slow

And that's why I started using sol.retcode == :Success, which stopped working.
Sorry sol.retcode is not in README. I meant the ODE example is from README not sol.retcode part

[ChrisRackauckas]

Cool fixed SciML/DiffEqDocs.jl@3d5ee28

[hhuber2512]

Hello,

There's another usage of

sol.retcode != :Success

in the SciML documentation, here: https://docs.sciml.ai/DiffEqParamEstim/stable/methods/optimization_based_methods/

It also comes up in the Turing.jl documentation, when they're discussing Bayesian inference with Differential Equations: https://turinglang.org/v0.25/tutorials/10-bayesian-differential-equations/

[ChrisRackauckas]

in the SciML documentation, here: https://docs.sciml.ai/DiffEqParamEstim/stable/methods/optimization_based_methods/

Thanks, just fixed that one.

[ChrisRackauckas]

It's been deprecated for a while and should now be sol.retcode == ReturnCode.Success.

SciMLBase.successful_retcode(sol). See https://docs.sciml.ai/SciMLBase/stable/interfaces/Solutions/#Return-Code-Traits