In this tutorial we'll use Julia in two or three ways:
On the interactive terminal (called REPL); and/or
On a Jupyter notebook; and
On an editor for the package development.
All of them need you to have a binary julia or julia.exe accessible.
And please download version 1.6.X
For MacOS and Linux, you could also follow the Julia Downloads page. But I also maintain an installer called Jill, which you can download using
curl -fsSL https://raw.githubusercontent.com/abelsiqueira/jill/master/jill.sh > jill.shAnd enter
sudo bash jill.sh -v 1.6-latestCheck if the installation was successful with
julia -vYour package manager might also have julia available, but make sure the version to be installed is the newest one.
You can run Julia in Jupyter in two ways:
Using your existing Python + Jupyter installation;
Installing a standalone Julia Jupyter installation.
I'm assuming that you have python and the Jupyter notebook for python working properly, so that you only need to install Julia's kernel. Open julia. Then you should see a welcome text and a prompt like julia>.
Enter ENV["PYTHON"] = "..." where ... should be the path to your python installation.
For instance, for me it's /usr/bin/python.
Enter using Pkg.
Enter Pkg.add("IJulia") (Capital i and capital j).
Open julia. You should see a welcome text and a prompt like julia>.
Press ] and the command prompt will change to (@v1.6) pkg>.
Enter add IJulia (Capital i and capital j).
This will install a minimal Python + Jupyter distribution via miniconda.
To start the notebook, enter using IJulia and then notebook().
Julia is supported by a few editors and IDEs. An incomplete list of editors is available on the main page. One of the main requirements for an editor to support Julia is the latex-unicode autocompletion feature.
Since 2019, the most used and recommended IDE is VS Code (see also the official visualstudio docs.)
On your VSCode, you just have to search for the julia extension. If your julia executable is not on the PATH, then you also have to inform its path on the "Julia: Executable Path" setting.
Alternatively, the other four common options are:
If you choose to use VSCode, you should make some tests, like:
Opening a Julia terminal
Open the "Command Palette" (Ctrl+Shift+P);
Find "Julia: Start REPL" (Alt+J Alt+O);
Sending commands to REPL
Create a new file and save as .jl;
write 2 + 3 on that file;
Ctrl+Enter to run a line;
Shift+Enter to run the whole file.
The first time you try to run commands on Julia VSCode, it will prepare the language server, and that might take a while.
When you open a Julia project folder, the environment should be automatically set. You can change it on the julia env toolbar option.
Julia comes with a built-in package manager: Pkg.
Pkg is itself a package;
it's always available because it's part of the stdlib;
it's a very robust manager, that deals with conflicts and can handle environments;
it connects to the official Julia registry to resolve package names;
it also allows for unregistered packages to be installed.
To access the package manager, open the Julia terminal and press ]. You should see a prompt like (@v1.6) pkg>. Enter help for a list of commands. Here are some commands that will be useful in this tutorial:
activate: set the primary environment that the package manager manipulates
When you activate a different environment, you can see it before the pkg prompt. E.g.:
julia> # press ]
(@v1.6) pkg> activate potato
(potato) pkg> activate .
(current-wd) pkg> activate
(@v1.6)add: add packages to project
You can add PackageName1, PackageName2 or add GitHubLink, for instance.
remove, rm: remove packages from project or manifest
status, st: summarize contents of and changes to the environment
test: run tests for packages
update, up: update packages in manifest
In this tutorial, we'll need to install several packages. You can copy and paste the following command:
pkg> add BenchmarkTools, CSV, DataFrames, JuMP, Plots, Revise, Unitful, UnitfulRecipes