I’ve been in the thick of it in terms of writing tests and learning to submit the code in various ways. I was using github to more casually push my code so I could get quick feedback but have now been focusing on pushing to Gerrit so I could have the code officially reviewed.
I’ve been learning to use the mwbot node.js module which helps to automate api calls in Mediawiki. It’s been challenging to use it since it’s built on other libraries I was not too familiar with. I took a couple more courses to update myself on using javascript promises and es6 syntax, so that’s been a help in writing the code. I’ve been working on writing tests that test api functionality and have successfully written a test. The only issue is that the api itself seems to have a bug! or at least not work the way one would expect it to work. I might have to file a bug report, but as it stands my test does check that it works the way it works as it stands now.
I’ve also been automating more complex tests which involve multiple users and interactions between them. Those tests were more difficult to write and have proven to be a bit more fragile. They passed but suffer from occasional timeouts and sometimes fail inconsistently. My guess is it has to do with how long it takes to load elements on the page and I tried accounting for that in the code by using functions that make it wait for an element to load, but it still occasionally fails. I’ve also submitted it and am curious to see if it passes the automated tests in Jenkins.
