Dash-Industry-Forum/dash.js-browserstack

Name: dash.js-browserstack

Owner: Dash Industry Forum

Description: Spike for automated BrowserStack UATs for dash.js

Created: 2016-01-20 22:56:03.0

Updated: 2017-11-27 22:49:11.0

Pushed: 2016-07-20 18:39:42.0

Homepage: null

Size: 25

Language: JavaScript

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Latest build status

A set of tests and associated running and helper code for performing Cross-Platform tests against dash.js.

Quick Start to Viewing

These tests should be automatically ran once per day (8am GMT), and the resulting logs and video recordings uploaded to http://smp-scratch.s3-website-eu-west-1.amazonaws.com/browserstack/. Just take a look there to see the latest results.

Quick Start to Running

Prerequisites

1. Ensure you have node 0.12.x or newer, mocha, jq and BrowserStackLocal installed
2. A copy of the `dist` directory from dash.js 2.0.0 or newer to a directory named dash in the root of your local checkout
3. A valid BrowserStackLocal API user and key

Run the tests

1. Install dependencies with `npm install`
2. Start BrowserStackLocal with `BrowserStackLocal -only -f YOUR_KEY PATH_TO_WORKING_DIR -localIdentifier SESSION_IDENTIFIER`
3. Run tests in working directory with `mocha specrunner.js --user YOUR_USER --key YOUR_KEY --identifier SESSION_IDENTIFIER`

Running locally

1. Ensure you have a Selenium webdriver installed and available in your path
2. Ensure that your dash.js-browserstack is available on a webserver
3. Run the tests with, for example, `mocha specrunner.js --local chrome --url http://localhost/dash.js-browserstack --runner "Win7 Chrome"`

Advanced Running

samples/simple-runner.sh does the above, but with some bonuses such as downloading the resulting logs and creating an index to view them. It can be run both in a Jenkins CI environment and locally, with the invocation `samples/simple-runner.sh /path/to/config.json`. A sample config file would look as follows:

"secure_configuration": {
    "browserstack-user": "YOUR_USER",
    "browserstack-key": "YOUR_KEY"
}

The logs directory will contain results in the same format used for the daily runs.

Notes

This is currently a proof of concept and could do with several key improvements:

1. Parallelisation (currently only one capability suite is run at a time)
2. Far more tests (e.g. Live, MultiPeriod, additional browsers, different configurations)
3. Wiring to run against pull requests