Closed brianmackenna closed 4 years ago
@brianmackenna Thanks for flagging this up. I was surprised this didn't throw an error straightaway and then I realised what's happening. It's because the dummy data generation doesn't do any kind of error checking and, because there's no database-url
defined, it never loads the real backend code which does the error checking. We'll have to think about how best to manage this, but there shouldn't be any need to write a specific test for this.
See linked issue to track progress
This is an idea for a possible test that would be nice to have but as always to be weighed up with competing priorities. Taking NSAID study def for example, I was able to write the
patients.satisfying
for the study inclusion/exclusion and includestroke
however I had not included and definedstroke
anywhere else in the study. It would be nice if there was a test to check for this so ensure we have the right inclusion/exclusion criteria.Now for studies we are reporting the number of people with a
stroke
so if was picked up pretty quickly that it was missing but I'm thinking that there may be other studies where this might not be the case and it might be missed.I've never written tests so I have no idea how much work this would be.