Currently it's a bit hard to know whether our completions actually work end-to-end. Whether when it's all installed, that we actually get the completions we're expecting in our API. It would be nice if we added some way that we could test this. This would've been helpful in #2846.
We might be able to use scrut to create a test for it. My team created this tool for our work and it basically allows you to easily create E2E tests. If you can make autocompletions show by running a command on the commandline, you can very easily make a scrut test out of that.
Currently it's a bit hard to know whether our completions actually work end-to-end. Whether when it's all installed, that we actually get the completions we're expecting in our API. It would be nice if we added some way that we could test this. This would've been helpful in #2846.
We might be able to use scrut to create a test for it. My team created this tool for our work and it basically allows you to easily create E2E tests. If you can make autocompletions show by running a command on the commandline, you can very easily make a scrut test out of that.