Oscar Smith

fix Rosenbrock methods when `eltype(u)!=eltype(tspan)`

6

MWE ``` using OrdinaryDiffEq ode_f(du, u, p, t) = du[1] = -u[1] prob = ODEProblem(ode_f, [1f0], (0.0,1.0)) solve(prob, Rodas5P()) ```

``` julia> using OrdinaryDiffEq julia> using SparseArrays julia> f(out, du, u, p, t) = out .= du-u f (generic function with 1 method) julia> func = DAEFunction(f; jac_prototype=sparse(ones(3,3))) (::DAEFunction{true, SciMLBase.FullSpecialize,...

`dualT4` should not be promoted with `T` since `T` might already be a dual

4

this is the second part of the fix for https://github.com/SciML/DifferentialEquations.jl/issues/994

`gamma(0)` returns the wrong value.

1

This issue is the opposite issue of https://github.com/JuliaMath/SpecialFunctions.jl/issues/163. Currently `gamma(0.0)` returns `Inf` as does `gamma(0.0+0.0im)`, and `gamma(0.0-0.0im)` This isn't the correct limit because ``` gamma(0. + 1e-16im) 0.6123233995736787 - 1.0000000000000034e16im...

Fearless taylor version

2

If we're willing to use the information that derivatives are linear operators, we can remove a lot of complexity from forward mode differentiation (and improve type stability in the process)....

In Julia 1.10, tab completion has been massively powered up to allow things like tab complete through array getindex, but in the debugger's evaluation mode, tab completion seems a lot...

For code that is transofrmed by a marco, there sometimes isn't line information which makes it impossible to add a breakpoint. The ability to add a breakpoint to the current...

Stop using `Core.Compiler.return_type`

3

`Base.promte_op` is a strictly less sketchy version.

`ldiv` of `LUFactorization` can throw `SingluarException`

1

``` using LinearAlgebra, SparseArrays F = lu(spzeros(2,2), check=false) F \ rand(2) ERROR: SingularException(0) Stacktrace: [1] umferror(status::Int32) @ SparseArrays.UMFPACK ~/.julia/juliaup/julia-1.10.1+0.x64.linux.gnu/share/julia/stdlib/v1.10/SparseArrays/src/solvers/umfpack.jl:105 [2] macro expansion @ ~/.julia/juliaup/julia-1.10.1+0.x64.linux.gnu/share/julia/stdlib/v1.10/SparseArrays/src/solvers/umfpack.jl:665 [inlined] [3] macro expansion @ ./lock.jl:267...

https://github.com/JuliaMath/DoubleFloats.jl/pull/136 gives `DoubleFloats` an `exp` (and `exp2`, `exp10`) implementation that is around 4x faster than what `MultiFloats` is using. If you wanted this version to work for all levels of...