Problem with Julia packages after updating to 1.9

General Usage
, ,
1

I update my Julia version to 1.9 (previously 1.8.3) and now when I import the following packages

using CSV, DataFrames, Statistics, LinearAlgebra, Plots, SharedArrays, LaTeXStrings, Measures, Interpolations, LsqFit

I get the following error message

ERROR: LoadError: 
ArgumentError: Package AdaptStaticArraysCoreExt does not have Adapt in its dependencies:
- You may have a partially installed environment. Try `Pkg.instantiate()`
  to ensure all packages in the environment are installed.
- Or, if you have AdaptStaticArraysCoreExt checked out for development and have
  added Adapt as a dependency but haven't updated your primary
  environment's manifest file, try `Pkg.resolve()`.
- Otherwise you may need to report an issue with AdaptStaticArraysCoreExt

If I attempt to follow the prompt recommendation, it does not work, and I keep getting the same mistake. Although if I run the using [insert Packages] twice in a row, after the second time it stops giving me the error message and everything works properly. It is just annoying that error message being displayed every time I import the packages.

1 Like
2

FWIW I’ve been seeing this a lot too

1 Like
3

Weird, I can’t reproduce this but I get a different error:

[ Info: Precompiling LogExpFunctionsChainRulesCoreExt [1bf5f11d-9a0a-5d25-85d0-d1d9a28a239c]
ERROR: LoadError: UndefVarError: `logabssinh` not defined
(...)
in expression starting at stdin:2
┌ Error: Error during loading of extension LogExpFunctionsChainRulesCoreExt of LogExpFunctions, use `Base.retry_load_extensions()` to retry.
│   exception =
│    1-element ExceptionStack:
│    Failed to precompile LogExpFunctionsChainRulesCoreExt [1bf5f11d-9a0a-5d25-85d0-d1d9a28a239c] 

(...)

[ Info: Precompiling SpecialFunctionsChainRulesCoreExt [9eb7bdd4-e44c-55fc-b9cc-1a32cb715188]
ERROR: LoadError: UndefVarError: `logabssinh` not defined
(...)
in expression starting at stdin:2
┌ Error: Error during loading of extension LogExpFunctionsChainRulesCoreExt of LogExpFunctions, use `Base.retry_load_extensions()` to retry.
│   exception =
│    1-element ExceptionStack:
│    Failed to precompile LogExpFunctionsChainRulesCoreExt

(...)

[ Info: Precompiling StatsFunsChainRulesCoreExt [d41313fe-2684-5453-9fef-703f5fa397f4]
ERROR: LoadError: UndefVarError: `logabssinh` not defined

(...)

in expression starting at stdin:2
┌ Error: Error during loading of extension LogExpFunctionsChainRulesCoreExt of LogExpFunctions, use `Base.retry_load_extensions()` to retry.
│   exception =
│    1-element ExceptionStack:
│    Failed to precompile LogExpFunctionsChainRulesCoreExt

(...)

[ Info: Precompiling DistributionsChainRulesCoreExt [6db1f127-056a-568b-bd49-ae61d42389fa]
ERROR: LoadError: UndefVarError: `logabssinh` not defined

(...)

in expression starting at stdin:2
┌ Error: Error during loading of extension LogExpFunctionsChainRulesCoreExt of LogExpFunctions, use `Base.retry_load_extensions()` to retry.
│   exception =
│    1-element ExceptionStack:
│    Failed to precompile LogExpFunctionsChainRulesCoreExt [1bf5f11d-9a0a-5d25-85d0-d1d9a28a239c]

A bit hard to debug as I can’t reproduce it a second time, even doing this in a temp env and garbage collecting after probably leaves some stuff in the precompile cache.

4

Which version of Adapt.jl are you using? That should’ve been fixed by Fix StaticArrays extension. by maleadt · Pull Request #62 · JuliaGPU/Adapt.jl · GitHub 4 months ago.

1 Like
5

Adapt version is 3.6.0 and StaticArrays v0.12.6

6

What happens if you try and force Adapt 3.6.2? Also, you really should be on StaticArrays v1, that was released like 3 years ago, so that’s a bit confusing. What if you try and force those to upgrade, is there some bad package keeping them back?

7

Does the error appear in a clean environment? That is:

  1. Start julia with julia --startup-file=no

  2. Then:

julia> import Pkg; Pkg.activate(temp=true)

julia> using CSV, DataFrames, Statistics, LinearAlgebra, Plots, SharedArrays, LaTeXStrings, Measures, Interpolations, LsqFit

(it doesn’t for me)