Skip to content
A stupid simple way to run your Selenium tests locally and remotely (e.g. grid, cloud offering, etc.)
Branch: develop
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

#selenium-connect 3.7.1 (2014-04-13)

Gem Version Build Status Code Climate Coverage Status

A stupid simple way to run your Selenium tests on your computer, against a Selenium Grid, or in the cloud (e.g. SauceLabs). For a rocking implementation of this library, checkout ChemistryKit! For more usage examples check out our Friends section!

All the documentation for Selenium Connect can be found in this README, organized as follows:

Getting Started

require 'selenium-connect'

# generate a config object
config = browser: 'firefox'

# get connected
sc = SeleniumConnect.start config

# create a job
job = sc.create_job

# start the job to get a driver
@driver = job.start

# get on the road!
@driver = SeleniumConnect.start
@driver.get ""

# finish your job

# go have some fun!

Helpful Bits


If host is set to "localhost" and no jar file is specified, it will run the version of selenium-standalone-server.jar that is bundled with the library (currently 2.33.0). Or, you can specify your own jar if you have one you prefer to use. This is done with c.jar = 'path-to-jar-file'.

If no additional parameters are set, the Selenium Server will be run in the background with logging disabled. If a logging directory is provided (with c.log = 'path-to-log-dir') then the following output files will be generated:

  • Selenium Server JSON Wire Protocol output (server.log)
  • firefox.log (If using Firefox as the browser)
  • chrome.log (If using Chrome as the browser)
  • dom_0.html (Or dom_1.html etc for each open window at the time the job is finished, a dump of the html)
  • failshot.png (If failshot is marked as true, a screenshot of the end state)

This localhost functionality is driven using the Selenium Rake Server Task.


The finish command issues a quit command to the driver and stops the local server if your host is set to "localhost".



Configuration of Selenium Connect is SUPER SIMPLE if you want it to be:

config =

By default it will run a local instance of selenium server on port 4444 and launch firefox. Get going without a whole bunch of shenanigans.

If however you want to install custom settings you can use any of the following:

# Setup & Debugging
jar: # this is where my selenium server jar is
log: # the logs go to this folder

# Where to run your tests
host: 'localhost' # local, a grid ip or "saucelabs"

# Browser
browser:      'firefox'

# Saucelabs
sauce_username: 'test_user_name'
api_timeout: #how many seconds we should try to get the assets (default 10)
description: #sauce job/test description
# set any sauce options below, they will override those above
    selenium_version: #default is 2.33.0
    job_name: #sauce job/test description

You can pass parameters into the new config object like:

config = host: 'sauce labs', log: 'build'

Or you can load them up from a YAML file:

config = '/my/config.yaml'

###Additional Configuration When you create your job you can pass in parameters, right now just :name that lets you configure a job at runtime. This is helpful for using Sauce Labs where you'd want to update the description to whatever test job you are running:

job.start name: 'website should load', sauce_opts: { public: 'team' }

Note you can also pass a hash of sauce_opts to the job start function that will let you do additional run time configuration with options as detailed here:

Similarly, when you finish your job you can pass in parameters. You can use the failshot parameter to turn on the saving of the last screenshot. For SauceLabs you can mark the tests as passed or failed:

# sweet your test passed!
report = job.finish passed: true

# shucks your test failed :(
report = job.finish failed: true, failshot: true

The report is simply a container for arbitrary data. Right now we are passing back the sauce details. Here is an example of for a failed job:

{:assets=>{:server_log=>"failed_serverlog_failing_sauce_job_3ee1fddf4032476fa3f9de94298766ae.log", :job_data_log=>"failed_saucejob_failing_sauce_job_3ee1fddf4032476fa3f9de94298766ae.log"}, :sauce_data=>{:id=>"3ee1fddf4032476fa3f9de94298766ae", :"custom-data"=>nil, :owner=>"testing_arrgyle", :status=>"in progress", :error=>nil, :name=>"failing_sauce_job", :browser=>"iexplore", :browser_version=>"7.0.5730.13", :os=>"Windows 2003", :creation_time=>1373916900, :start_time=>1373916901, :end_time=>0, :video_url=>"", :log_url=>"", :public=>nil, :tags=>[], :passed=>false}}

Contribution Guidelines

This project conforms to the neverstopbuilding/craftsmanship guidelines. Specifically related to the branching model and versioning. Please see the guidelines for details.

Install Dependencies

bundle install

Run the Tests!

rake              # defaults to 'build' task, code quality, unit, and integration tests
rake unit         # unit tests
rake integration  # integration tests
rake system       # system tests

Or get your Guard On!



Will start watching the code and run the unit tests on save. Cool.


The release process is rather automated, just use one rake task with the new version number:

rake release_start['2.1.0']

And another to finish the release:

rake release_finish['A helpful tag message that will be included in the gemspec.']

This handles updating the change log, committing, and tagging the release.


Below you can find some honorable mentions of those friends that are using Selenium Connect:


We are proud that Animoto uses Selenium Connect to help them test their awesome web app.

You can’t perform that action at this time.