Destiny-runner NPM

Destiny Runner (GUI for Karma)

This package attempts to create a GUI runner for karma. The author doesn't really like the default console runner and hence attempt to do something better.

Screen from Destiny

Warning!

This is a pre-alpha version. This package will not work (yet..)

Contributions wanted!

This project is currently in its infancy and any contribution is very much wanted & appreciated. Very soon a liberal contribution policy will be adopted. If you want to contribute, please read the list below, let me know that you start working on something & go, go, go!

Road to MVP (1.0)

The MVP is currently defined as - let Destiny be a drop-in replacement for a console reporter/logger, only more usable one.

This includes:

be able to display a list of tests executed by Karma in the hierarchical order
be able to process updates from subsequent runs
be able to handle multiple browsers in separate tabs
when Destiny is run Karma should start as well as browser window should open with the GUI
compatibility matrix with Karma versions should be made and dependencies in package.json adjusted accordingly
currently only Jasmine will be supported, Mocha is nice to have
test restart should be possible from Destiny, without resorting to Karma owned browser (refresh button)
all the logs from browser console should be available on test level & stack trace for failed tests
there should be a simple filtering for tests (name, describe, success/fail/skip)
there should be a nice way to run Destiny after it is installed i.e. npm run destiny or just destiny
Ideally you should be able to run destiny with a headless browser and work on your project.
Ideas for the future
Be able to schedule a re-run of only one test / one-suite / multiple-suites. It should render the fdescribe/fit source modification obsolete! (most likely requires changes in Karma)
Display test coverage reports with detailed information (Karma already provides this info when istanbul loader is being used)
More advanced / smarter filtering for tests
Calculating suite results based on all the test results (like any test runner does..)
Better UI/UX ?