Fix "search with a filter with arguments" spec#2581
Conversation
littleforest
left a comment
There was a problem hiding this comment.
This looks great, thanks for fixing! I don't think the second filter test is actually necessary. Would you mind fixing the linting?
|
@littleforest sounds good thank you! Should be fixed now |
The flake would happen when the second search of `kind:vip` was not applied to the URL. Running the test directly in Chrome would show it happening. As the URL parameter was never set, it was therefore blank. Whilst a `sleep(0.1)` would fix it, this instead restructures the test to avoid the problem completely. Closes thoughtbot#2523
|
Wonderful! Thank you so much for taking the time to look at this, it's been a real thorn-in-my-side for the last weeks, and I'm terrible when it comes to flakey specs! I'm really glad the RailsConf hackathon turned out so well, too! |
|
@nickcharlton glad it worked out! haha I hear you, same. Yeah really enjoyed the hackathon. I let Steffani know that I run the open source guild at my company and we have a few people who are looking for places to contribute. Would definitely love to stay in contact if you ever need a few contributors to take on similar things |
|
Oh! That's cool to hear! We've started catching up as maintainers internally, so we'll have a chat about how we can make it a bit clearer on how to jump in. In the case of Administrate, it's really just picking something that interests you — but I appreciate that can be quite hard of a suggestion! |
Hi. This PR will solve the failing flakey test listed in #2523. While debugging the issue I found a few things.
The flake would happen when the second search of
"kind:vip"was not applied to the url. I was able to run the test in headed chrome and see the failure happen. The url param was never set and was therefore blank. Adding asleep(0.1)after theclear_searchseemed to work as one pathway. CapybaraI went down the path of modifying the existing test to pass with a single param by the suggestions of a few at the RailsConf hackathon. Happy to add an additional scenario testing
"kind:standard"if that would be preferred.Thanks!