DANE Client Starter Scripts

Selenium scripts for automated endless web browsing and watching videos. To be used as client behaviors in the DANE tool.

Setup

1. Set up the Python environment/dependencies.

   conda create -n selscripts
   conda activate selscripts
   conda install -y -c conda-forge pip selenium geckodriver bs4 python-dotenv
   

   Note: You may choose to use a different webdriver than Firefox (geckodriver), e.g. for Chrome you would replace geckodriver with python-chromedriver-binary. See Selenium docs and do a web search for the drivers you can install with Anaconda.

2. Specify your desired webdriver and options in scripts/driver.py.

   For options, see https://selenium-python.readthedocs.io/api.html#module-selenium.webdriver.firefox.options or https://selenium-python.readthedocs.io/api.html#module-selenium.webdriver.chrome.options.

   Note: Command line arguments can also be specified with driver_options.add_argument('--option argument').

3. Create a .env file at the root of this directory and write in your credentials.

   # Inside .env
   CURIOSITYSTREAM_EMAIL="your.email@address.com"
   CURIOSITYSTREAM_PASS="y0ur_p4ssw0rd"
   

   Note: That's a horrible password.

Running

To run a specific script, you must run it as a module like so:

python -m streaming.curiositystream

This is what causes driver.py to be run -- that way all scripts can utilize the same base driver definition (you don't need to write it multiple times if you want to make a tweak!)

Note that if you're calling this script from another directly you'll need to pass the path to the file separated by periods rather than slashes (to keep with the Python module rules). E.g. path.to.starter-scripts.streaming.curiositystream.

References

• urllib.parse Documentation -- url parsing
• Selenium Python Documentation -- browser automation
• Selenium Explicit Waits Documentation -- often we need to wait for page content to load
• BeautifulSoup4 Documentation -- page parsing (unused so far)