Error while installing LinearInterpolators.

[assia855]

Thank you professor for providing such packages to the public.
Can you allow me please with the right commands to install LinearInterpolators and nterpolationKernels. It seems as there's a problem version.

(StarGenerated) pkg> add https://github.com/emmt/LinearInterpolators.jl
Updating git-repo https://github.com/emmt/LinearInterpolators.jl
Resolving package versions...
ERROR: Unsatisfiable requirements detected for package InterpolationKernels [16730964]:
InterpolationKernels [16730964] log:
├─possible versions are: 0.2.0 or uninstalled
├─InterpolationKernels [16730964] is fixed to version 0.2.0
└─restricted to versions 0.1 by LinearInterpolators [2fc109c4] — no versions left
└─LinearInterpolators [2fc109c4] log:
├─possible versions are: 0.1.0 or uninstalled
└─LinearInterpolators [2fc109c4] is fixed to version 0.1.0

[emmt]

I introduced a lot of changes in InterpolationKernels and did not correctly manage the compatibilities, this should be fixed now. I will update LinearInterpolators to work with the last version of InterpolationKernels as soon as possible.

[assia855]

We really appreciate your awesome efforts. Thank you so much.
I was wondering if there's a way to have the previous stable version. If there's a way can you share it with me please.

[emmt]

This is exactly what I attempted to do with this. Unfortunately, it seems that version restrictions only work automatically with official Julia packages which is not yet the case of InterpolationKernels and LinearInterpolators. This means that you have to specify yourself that you want to stick to a given version of InterpolationKernels that is compatible with LinearInterpolators. This can be done by using a specific environment with Julia package manager (documented here). A Julia environment is a way to specify a list of packages (possibly with version restrictions) you want to work with. You may have as many environments as you want (they will automatically share the same code if possible) but a Julia session works with a given environment at a time (it is possible to switch the environment by a activating another one but only one environment is active at a time).

I tried to figure out a concrete working example for you. Starting from scratch, you first have to create a new Julia environment for your project/work that requires LinearInterpolators. You have to decide the name and location (directory) of this environment, for instance SomeDir/SomeProject where SomeDir is a directory writable by you (the current directory if none specified) SomeProject is the name of the new environment, the environment files (Manifest.toml and Project.toml) will be stored in a directory named SomeDir/SomeProject. Start Julia and hit the ] key to switch to Julia's package manager prompt (things are really easier in that mode), then type the following commands at the ...pkg> prompt:

activate SomeDir/SomeProject
add https://github.com/emmt/InterpolationKernels.jl#v0.1
add https://github.com/emmt/LinearInterpolators.jl

The first command above activates the environment (and creates it if does not exists yet), the second command is to add the InterpolationKernels package with the restriction to use the branch v0.1 which is the one that is compatible with the LinearInterpolators package which is added to your environment by the third command. You may add any other packages. For official packages, it is sufficient to only specify the package name (no needs to give the full URL).

To go back to the Julia prompt, just hit ^C (control-C) or the backspace key. Unless you activate another environment or quit Julia, you will be working with the environment defined in SomeProject.

When you start a new Julia session, just recall the activate SomeDir/SomeProject command at the package manager prompt. You may also do that directly at Julia prompt (this is useful in scripts) by typing:

using Pkg
Pkg.activate("SomeDir/SomeProject")

Note that SomeDir/SomeProject is really a directory name, it may be specified using an absolute or relative path.

[assia855]

Thank you so much.
"#v0.1 " that's what I was looking for. It works now.