Selenium + Python boilerplate framework for running tests by providing external (JSON) files as the input/data source.

It decouples tests from particular environments.

##Install With tox installed run tox in the parent directory. If you don't have tox (which is very useful), run pip install -r requirements.txt

After installation you can run tests via the run.py command

##Recommendations

• Create tests in test/tests sub-directories.
• Extend the base-class Framework for your test class. The runner peeks in files ending with Test.py, so be sure to name the file accordingly.
• Create the corresponding Json file using the same name of the test file. The json should contain the contexts key with an array of environments. Each environment contains it's key & value of data, e.g.

{
  "contexts": [
    {"STAGING" : {
        "loginUser": "super@man.com"
      }
    },
    {"PROD" : {
        "loginUser": "super@woman.com"
      }
    }
  ]
}

• Script the test to utilize the context data (context map) via self.contextmap.get("loginUser").
• Add groups if necessary to the test class. Use the GROUPS list attribute in the test class.
• Run the test(s) via the runner:

$ chmod +x run.py
$ ./run.py -e STAGING -t "MyTest,MyOtherTest" # example to run 2 tests in staging
$ ./run.py -e PROD -g "smoke" # example to run 1 smoke test in prod
$ ./run.py -e STAGING # example to run all tests in staging

##Quick Notes

• The framework supports running tests (TestCase) that don't extend the Framework class.
• Default logging logs to project root test.log file with Debug or greater levels.
• The framework works best when enviornments are using different domains (rather than page names).
• When running tests from the command line, be sure to add the -e option. Python sets the environment for each execution/task only (not globablly)