JuliaMath / Interpolations.jl

Fast, continuous interpolation of discrete datasets in Julia
http://juliamath.github.io/Interpolations.jl/
Other
518 stars 110 forks source link

Fix `checkbounds` for trailing/missing dimensions. #566

Closed N5N3 closed 9 months ago

N5N3 commented 9 months ago

close #562.

mkitti commented 9 months ago

Why is CI stalled?

codecov[bot] commented 9 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (4b617f6) 87.85% compared to head (4113d66) 87.87%.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #566 +/- ## ========================================== + Coverage 87.85% 87.87% +0.01% ========================================== Files 28 28 Lines 1878 1880 +2 ========================================== + Hits 1650 1652 +2 Misses 228 228 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

N5N3 commented 9 months ago

Looks like JLArray tests failed on master due to https://github.com/JuliaLang/julia/pull/51764. We can do nothing here. JLArrays.jl needs a patch like

Base.cconvert(::Type{Ptr{T}}, x::JLArray{T}) where {T} = Base.cconvert(Ptr{T}, x.data)

CC @@maleadt

maleadt commented 9 months ago

https://github.com/JuliaGPU/GPUArrays.jl/pull/500, https://github.com/JuliaRegistries/General/pull/94501

N5N3 commented 9 months ago

bump

mkitti commented 9 months ago

Are we dropping Julia pre-1.6 compatability?

ChainRulesCore: Error During Test at D:\a\Interpolations.jl\Interpolations.jl\test\chainrules.jl:12
  Test threw exception
  Expression: Zygote.gradient(itp, 1) == Tuple(Interpolations.gradient(itp, 1))
  MethodError: no method matching _methods_by_ftype(::Type{Tuple{typeof(rrule),Zygote.ZygoteRuleConfig{Zygote.Context},Interpolations.BSplineInterpolation{Float64,1,Array{Float64,1},BSpline{Linear{Throw{OnGrid}}},Tuple{Base.OneTo{Int64}}},Int64}}, ::Int64, ::UInt64, ::Bool, ::Base.RefValue{UInt64}, ::Base.RefValue{UInt64}, ::Ptr{Int32})
  Closest candidates are:
    _methods_by_ftype(::Any, ::Int64, ::UInt64) at reflection.jl:835
    _methods_by_ftype(::Any, ::Int64, ::UInt64, !Matched::Array{UInt64,1}, !Matched::Array{UInt64,1}) at reflection.jl:838
mkitti commented 9 months ago

If we are, please adjust the CI here to drop 1.3.

N5N3 commented 9 months ago

No idea here. I'm not familiar with CI setting. And the error has been there on master branch for quite a while.

mkitti commented 9 months ago

Remove this line: https://github.com/JuliaMath/Interpolations.jl/blob/4b617f6f85e5ee62265f4d2ac9a559f092161601/.github/workflows/CI.yml#L16

mkitti commented 9 months ago

Change this line to Julia 1.6 https://github.com/JuliaMath/Interpolations.jl/blob/4b617f6f85e5ee62265f4d2ac9a559f092161601/Project.toml#L35