Here are some tips for working with Julia efficiently.
As already elaborated in The Julia REPL, Julia's REPL provides rich functionality that facilitates an efficient interactive workflow. Here are some tips that might further enhance your experience at the command line.
The most basic Julia workflows involve using a text editor in conjunction with the
julia command line. A common pattern includes the following elements:
Put code under development in a temporary module. Create a file, say
Tmp.jl, and include within it
module Tmp export say_hello say_hello() = println("Hello!") # your other definitions here end
Put your test code in another file. Create another file, say
tst.jl, which looks like
include("Tmp.jl") import .Tmp # using .Tmp # we can use `using` to bring the exported symbols in `Tmp` into our namespace Tmp.say_hello() # say_hello() # your other test code here
and includes tests for the contents of
Tmp. Alternatively, you can wrap the contents of your test file in a module, as
module Tst include("Tmp.jl") import .Tmp #using .Tmp Tmp.say_hello() # say_hello() # your other test code here end
The advantage is that your testing code is now contained in a module and does not use the global scope in
Mainfor definitions, which is a bit more tidy.
tst.jlfile in the Julia REPL with
Lather. Rinse. Repeat. Explore ideas at the
juliacommand prompt. Save good ideas in
tst.jl. To execute
tst.jlafter it has been changed, just
There are a few ways to interact with Julia in a browser:
Whether you're at the REPL or in IJulia, you can typically improve your development experience with Revise. It is common to configure Revise to start whenever julia is started, as per the instructions in the Revise documentation. Once configured, Revise will track changes to files in any loaded modules, and to any files loaded in to the REPL with
includet (but not with plain
include); you can then edit the files and the changes take effect without restarting your julia session. A standard workflow is similar to the REPL-based workflow above, with the following modifications:
Put your code in a module somewhere on your load path. There are several options for achieving this, of which two recommended choices are:
For long-term projects, use PkgTemplates:
using PkgTemplates t = Template() t("MyPkg")
This will create a blank package,
"MyPkg", in your
.julia/devdirectory. Note that PkgTemplates allows you to control many different options through its
In step 2 below, edit
MyPkg/src/MyPkg.jlto change the source code, and
MyPkg/test/runtests.jlfor the tests.
For "throw-away" projects, you can avoid any need for cleanup by doing your work in your temporary directory (e.g.,
Navigate to your temporary directory and launch Julia, then do the following:
pkg> generate MyPkg # type ] to enter pkg mode julia> push!(LOAD_PATH, pwd()) # hit backspace to exit pkg mode
If you restart your Julia session you'll have to re-issue that command modifying
In step 2 below, edit
MyPkg/src/MyPkg.jlto change the source code, and create any test file of your choosing.
Develop your package
Before loading any code, make sure you're running Revise: say
using Reviseor follow its documentation on configuring it to run automatically.
Then navigate to the directory containing your test file (here assumed to be
"runtests.jl") and do the following:
julia> using MyPkg julia> include("runtests.jl")
You can iteratively modify the code in MyPkg in your editor and re-run the tests with
include("runtests.jl"). You generally should not need to restart your Julia session to see the changes take effect (subject to a few limitations).