Automated Tests

[webnard]

Starts the requested workers with BrowserStack. Optionally starts a BrowserStack tunnel to expose a local server to the workers.
This task does not exit untli the user presses a key at which time the workers are stopped again.

Is there any way to get around the need for a keypress to stop the workers? It would be nice to be able to automatically load up the browsers through a tunnel, run a few tests, and upon their completion kill the workers.

[pghalliday]

Hmm, this was a little tricky to do due to the way grunt works. Ie. its difficult to chain tasks so that 1 starts, another runs and completes and then the first cleans up. I may look into this again though.

The other problem is that it can take a while to start the browserstack clients so in development I wanted the clients to start once at the beginning of my session and then I would stop them later when I don't need them anymore. I can see that for an integration server though this might not be ideal.

[webnard]

I've got a working solution in bash that basically runs the tunnel in the background and kills it based on its process ID when my tests are finished. I haven't done much with Grunt, but if I have time I'll try to see if I can do something similar with it and submit a pull request if I get anything viable.

[jrcryer]

👍