The WU bugs are very well known already, in my case however, I don’t really have to dig deep, there are some pretty obvious ones that are being ignored by bubble.
I have sent several bug reports which took almost 2 weeks to be “investigated”, only to receive a reply with an explanation completely out of context, which had nothing to do with any of the issues I reported. I am shocked!
Here are just a couple of those bugs:
As I am experiencing this throughout my app, for testing purposes, I created 2 basic repeating groups with no constraints. The first one is fetching 15 User names, the 2nd one is fetching 10 numbers. That should be 2 searches in total, 1 search per RG. Not quite, I am being charge for 4 searches, 2 searches per repeating group.
The WU is also calculated arbitrarily, for the 2nd repeating group at least - it’s double the searches but not double the WU.
My app is a SPA and only consists of Reusable Elements which are all hidden at page load. These reusables should not actively process anything until they are visible. However, there are some random searches running at page load and every time I navigate between pages; random searches belonging to random reusables. Every page load or interaction with the menu is costing me 3 WUs for nothing.
Also, are we being charged for setting custom state when the expression is not a search?
And how is WU calculated when intersecting lists? How does the search run? Let’s say I have 10 items in the 1st list and 10 items in the 2nd. Will 100 searches run at once to find the common values between these 2 lists?
This isn’t correct. Reuseable elements can still do things even when not visible. Data won’t be loaded unless it’s visible (i.e the repeating group is visible, even if you can’t see it) or referenced, but invisible elements absolutely can do stuff (you implicitly know this already if you use group variables in popups).
Setting a state doesn’t cost WU, but retrieving data to set the state would cost WU.
All of the possible results are downloaded to the client and then it is filtered client side, so avoid intersect with if you can help it.
I was under the impression that searches don’t run and data is not being retrieved unless the RE is visible. If this wasn’t the case I would be in big trouble. I have 12 REs each has 5 repeating groups (in average), plus many other searches. But that is not happening. What I don’t understand is why these 3 random searches run at page load?
As for setting states costing WU, I’m just saving the repeating group list to avoid massive WU when intersecting lists. But setting that custom state costs WU.
You can always make the datasource empty by default, and set it conditionally when ‘This Repeating Group is visible’ or ‘reuseable element X is visible’
That’s what I will end up doing. But I would still need at least an explanation from Bubble - why out of 100 searches only these particular 3 run at page load, when there’s no difference between them in how they’ve been setup.
This one is a hidden repeating group which I’m referencing 4 times in that reusable, to calculate the difference between scheduled hrs vs worked hrs. E.g: how many were scheduled vs how many clocked in, etc. That’s it. No workflows.
Par for the course when reporting bugs with WUs…sadly this won’t change. I think higher ups at Bubble have made sure no WU bugs will be addressed properly.
