---

QA OpenSource Framework

Installation

Linux:

apt-get install chromedriver libmysqlclient-dev python-pymysql python-mysqldb ffmpeg mongodb mongodb-clients mongodb-server

Install Python requirements:

pip install -r requirements.txt

To enable local Firefox testing, geckodriver is required. Place geckodriver in the same path as the Firefox binary.

Browserstack Tunnel

For some of the tests to pass, a browserstack tunnel needs to be created. The binary can be found here.

Example of binary use with tests:

./BrowserStackLocal --key {browserstack_api_key} --folder /home/User/qaopensource-framework/lib/test_data/project

Folder argument points to the test_data directory. This allows browserstack to use local files during test runs.

Example of binary use when testing against localhost:

./BrowserStackLocal --key {browserstack_api_key} --force-local

Note: While using a forced local connection with Browserstack. Test times are extremely slow since tests will be using your machines connection speeds through a vpn tunnel.

Configuration

For setting up configuration dependencies. Please refer to example configurations.

Projects and Tests

Please visit setting up your first project.

Running:

Browser Tests:

Running with suite_runner.py (preferred):

$ python suite_runner.py -h
usage: suite_runner.py [-h] [--list LIST] [--project PROJECT]
                       [--template TEMPLATE] [--env ENV]
                       [--test_type TEST_TYPE] [--runner RUNNER]
                       [--browser BROWSER] [--test_suite TEST_SUITE]
                       [--single_test SINGLE_TEST] [--report_db REPORT_DB]

optional arguments:
  -h, --help            show this help message and exit
  --list LIST           all, projects, types, templates, suites, tests,
                        browsers
  --project PROJECT
  --template TEMPLATE
  --env ENV             options: stage, qa, prod, dev
  --test_type TEST_TYPE
                        browser, api, mobile
  --runner RUNNER       options: local, local_X, grid, browserstack,
                        browserstack_mobile_ios, browserstack_mobile_droid,
                        browserstack_X
  --browser BROWSER     enter browser configs here
  --test_suite TEST_SUITE
  --single_test SINGLE_TEST
  --report_db REPORT_DB
                        options: mongodb, jenkins

Browserstack Test Example:

$ python suite_runner.py --project=project_1 --template=template_1,template_2 --env=qa --test_type=browser --runner=browserstack --browser=chrome_55,chrome_54 --test_suite=Test_Suite_1 --report_db=mongodb

Local Test Example:

$ python suite_runner.py --project=project_1 --template=template_1,template_2 --env=qa --test_type=browser --runner=local --browser=chrome --test_suite=Test_Suite_1 --report_db=mongodb

When running locally, specify which local browser you would like to run against.

API Test Example:

$ python suite_runner.py --project=project_1 --template=template_1 --env=qa --test_type=api --runner=local --test_suite=Test_Suite_1

API tests can be run against any template.

Running a single test:

$ python suite_runner.py --project=project_1 --template=template_1,template_2 --env=stage --test_type=browser --runner=browserstack --browser=chrome_55,chrome_54 --test_suite=Test_Suite_1 --single_test=test_foo.py --report_db=mongodb

You can also run a single test across multiple templates.

$ python suite_runner.py --project=project_1 --template=template_1,template_2 --env=stage --test_type=browser --runner=browserstack --test_suite=Test_Suite_1 --single=test_foo.py

Running tests with multiple instances of webdriver

python suite_runner.py --project=project_1 --template=template_1 --env=dev --test_type=browser --runner=local_X --browser=chrome --test_suite=Test_Suite_1 --single_test=test_foo.py --report_db=mongodb

local_X will allow the use of 2 webdriver instances locally. browserstack_X will trigger two webdriver instances against browserstack services. When this option is selected and a test is executed the tests will go into a queue waiting for any tests or queues in progress to finish before launching tests.

Get a list of available projects, templates, test suites and test cases options:

$ python suite_runner.py --list=all

Get list of enabled browser configs:

$ python suite_runner.py --list=browsers
chrome_54
chrome_44
chrome_55
firefox_51

Get list of all current projects:

$ python suite_runner.py --list=projects

Get list of all current test types:

$ python suite_runner.py --list=types   
mobile
browser
api

Get list of all current project templates:

$ python suite_runner.py --list=templates

Get list of available test suites:

$ python suite_runner.py --list=suites

Get list of all single test files:

$ python suite_runner.py --list=tests

Runner Options:

local Will run tests using the browser on your machine.

local_X Will run tests using the browser on your machine. However, this enables multiple driver instances.

grid Will run tests against available selenium grid instances.

saucelabs Tests will run remotely against a Saucelabs environment.

browserstack Tests will run remotely against a Browserstack environment.

browserstack_X Tests will run remotely against a Browserstack environment. However, this enables multiple driver instances.

browserstack_mobile_ios Tests will run through browserstack ios emulator.

browserstack_mobile_droid Tests will run through browserstack android emulator.

Reporting

If you would like persistent test results, there is an option to use mongodb. All that is required is a running instance of MongoDB. The mongo connect url can be found in settings.ini, for now. The framework will create a qaopensource-framework DB with a reports collection. Each test run will create a document within that collection.

An example of a test run using MongoDB:

$ python suite_runner.py --project=project_1 --template=template_1 --env=stage --test_type=browser --runner=local --test_suite=Test_Suite_1 --report_db=mongodb