I know that right now, I am attacking a few items ...
# testing
l
I know that right now, I am attacking a few items directly: • Known bugs • Performance • Ability to do dry runs, so it can be used by IDEs and CLI • WireBox integration to support Modules • New testing UI experiences • And new VSCode extension which is partially finalized that will support jumping, highlighting, running, etc • Inline Datasets for iteration and calling • Some new matchers (toStartWith, toEndWith, and(), when(), unless() ) • Skipping function (skip (message ) ) • Localized before/afterAll by Suites
d
+1 for dry run for sure. I'm assuming that means I'd be able to browse the test suite without running them? Similar to the current test browser, but going all the way to the test level, not just file level?
Admittedly sparse on details here, but one more thing. I’d love to run tests in parallel. I know there is the asyncAll parameter in the describe block, but I’ve never been able to get it to work (but haven’t spent time digging into it). Usually the 2nd test in a suite nearly always fails for some reason or another. Is the async stuff pretty bullet proof? Is this most likely our own code’s fault? We mostly write integration tests and so our tests take forever to run. It’d be awesome if they were truly running in parallel. So yell at me if it is for sure my own fault somehow. 😄
l
running things in parallel is always tricky, because, technically we can spawn the threads no problem
the issue is tracking everything async
your app needs to support it
mocking gets in the way because youa re mocking a running app
etc
There is no easy switch here
unless you use http calls