electron-mocha
Mocha testing in Electron. This project has
two main value propositions:
- You can now easily test any JavaScript app in a real browser (Chromium)
without hassling with PhantomJS or Webdriver.
- You can now easily test your Electron apps!
Install
npm i -g electron-mocha
Usage
Install Electron
First, you need to install Electron. You can either run:
npm i -g electron
and then electron
will be added to your path. Or, you
can download a version from https://github.com/atom/electron/releases and
then set an environment variable ELECTRON_PATH
pointing to the binary.
Note if you're using Mac OS X, the path would be to the actual executable
and not the app directory e.g. /Applications/Electron.app/Contents/MacOS/Electron
.
You should probably just install electron-prebuilt
as it simplifies things.
Run Tests
electron-mocha
is almost a drop-in replacement for the regular mocha
command.
Here's the help output:
Usage: electron-mocha [options] [files]
Options:
-h, --help output usage information
-V, --version output the version number
-C, --no-colors force disabling of colors
-O, --reporter-options reporter-specific options, format: k=v,k2=v2,...
-R, --reporter <name> specify the reporter to use
-S, --sort sort test files
-b, --bail bail after first test failure
-g, --grep <pattern> only run tests matching <pattern>
-f, --fgrep <string> only run tests containing <string>
-i, --invert inverts --grep and --fgrep matches
-r, --require <name> require the given module
-s, --slow <ms> "slow" test threshold in milliseconds [75]
-t, --timeout <ms> set test-case timeout in milliseconds [2000]
-u, --ui <name> specify user-interface (bdd|tdd|exports)
--check-leaks check for global variable leaks
--compilers use the given module(s) to compile files
--debug enable Electron debugger on port [5858]; for --renderer tests show window and dev-tools
--debug-brk like --debug but pauses the script on the first line
--globals <names> allow the given comma-delimited global [names]
--inline-diffs display actual/expected differences inline within each string
--interactive run tests in renderer process in a visible window that can be reloaded to re-run tests
--interfaces display available interfaces
--no-timeouts disables timeouts
--opts <path> specify opts path [test/mocha.opts]
--recursive include sub directories
--renderer run tests in renderer process
--preload <name> preload the given script in renderer process
--require-main <name> load the given script in main process before executing tests
So if you run:
electron-mocha ./tests
This runs the tests in the main
process. The output that you could expect would be pretty similar to that of io.js with one exception,
it supports all of Electron libraries since it's running
in Electron! So you don't need to mock those libraries out anymore and can actually write tests to integrate with them.
If you run:
electron-mocha --renderer ./tests
This runs the tests in the renderer
.
Yes, this means that you have access to the entirety of the DOM, web storage, etc. This is because it's actually
running in a Chromium process.
Using on Travis
Your .travis.yml
will need two extra lines of configuration to run this headless on Travis:
before_script:
- export DISPLAY=:99.0; sh -e /etc/init.d/xvfb start
Debugger Support
Use the --debug
or --debug-brk
options to enable Electron's debugger. When using --renderer
this will open the test window and open the dev-tools. Note that the window will close automatically when the tests have finished, therefore this option should be used in combination with debugger
statements anywhere in your tests or code. Alternatively, you can use the --interactive
option which will keep the window open after your tests have run (you can reload the window to run the tests again) -- this gives you the opportunity to set breakpoints using the dev-tools debugger.
To debug the main process (i.e., if you run your tests without the --renderer
option), you will need to start an external debugger. For more details, see Electron's documentation.
Code Coverage
You can use electron-mocha to collect code coverage data in Electron's main and renderer processes. To do this, you will need to instrument your code, run the tests on the instrumented code, and save the coverage stats after all tests have finished. You can instrument your code on the fly using istanbul-lib or by running istanbul instrument
before you exectue your tests. Finally, use nyc to combine the results from all processes into a single coverage report.
Roadmap
- Implement a way to allow tests to run in either
main
/renderer
from within
the same test file for the purposes of integration testing.
License
MIT