setting up a more-or-less complete Rise dev / test environment?

  • 2
  • Question
  • Updated 4 years ago
  • Answered
I've enjoyed some pretty good success adding capabilities to the the Calendar widget in the last week, following the setup directions Donna kindly provided. Development works fine. I have been using PHPStorm from JetBrains as an editor with good results.

It's become obvious to me that the Rise system open source is worthwhile to work on; I have some other ideas I'd like to pursue (in open source, of course).

But, whether because I goofed something up or because my dev / test environment isn't complete, I haven't been able to get the end-to-end tests for the calendar widget to pass.  This has made me into a pest for Rise, because my pull requests are incomplete.

Is there anything I can do (need to do?)  to set up a more complete dev environment on my machine? 

Thanks for any wisdom you can offer.
Oliver Jones
Photo of Oliver Jones

Oliver Jones

  • 692 Points 500 badge 2x thumb

Posted 4 years ago

  • 2
Photo of Donna Peplinskie

Donna Peplinskie

  • 7,142 Points 5k badge 2x thumb

When you run 'npm install' and 'bower install', all of the dependencies you need for development and testing should be automatically installed.

For settings e2e testing, we use Protractor for selecting elements, and Chai & chai-as-promised for doing the assertions. Here's a more thorough article on e2e testing with Protractor -

For Widget e2e testing, we use CasperJS.

For unit testing, we use Chai assertions.

I would try pulling down the Calendar repository from Github without introducing any changes, and get those tests to run successfully first.

Photo of Oliver Jones

Oliver Jones

  • 692 Points 500 badge 2x thumb
Thanks for this advice. I am afraid the repo at fails the e-2-e test  (on my W7 64 machine) in precisely the same way my repo with my changes fails.

I tried running the e-2-e tests for the Rise repo on a Linux vm (Linux version 3.8.13-55.el6uek.x86_64). In that case the test system hung up trying to initialize selenium.

I'll try your suggest about copying the data from rwda to the test case.

Thanks for your help.