There is no test/validation of the scripts work

[ebrucucen]

After each change is submitted to the code repo, we need the validation of the scraping scripts are valid, and they work
so that we can have continuous integration of the codebase and resilient modelling outcomes.

[ricjhill]

Can you give me a short description, I imagine we spin up an instance of doccano and an instance (or lambada ) run to our scripts then run integration tests. I am not sure where the test data will come from. The real new sources will change all them time and creating mocks of our news sources is more work. Is there a better way?

[ricjhill]

https://github.com/marketplace/category/continuous-integration which of these tools is preferred?

[ebrucucen]

I think we should keep consistent with doccano repo, so I suggest TravisCI. What do you think Richard?