Love Bubble, but 2 big time-sinks are slowing me down:
CRUD for new data types – Every time I add a new type (e.g., Client → Company), I have to rebuild all the popups, forms, dropdowns, and delete flows from scratch. Copy/paste helps, but it’s still tedious.
Saving nested JSON – I work with LLMs and deeply nested JSON (e.g., workout → group sets → sets). Building backend workflows to parse/save is slow, error-prone, and breaks if the schema changes
Question: Anyone have a plugin, pattern, or magic trick for faster CRUD setup or “JSON-to-DB” mapping?
I honestly have no idea what you’re talking about. CRUD is just CRUD. It sounds like you generally find building in Bubble, tedious.
If you’re rebuilding your UI/UX every time you add a new datatype, then you need to take a step back and take a look at your DB structure. Does it fit the goals and functions of your app?
Good Practices
Have your DB in mind when building your UI. Even better if you can map everything out before you start any UI work.
Use Reusable Elements as much as possible. Keeps your app organized and has a good side effect of increasing app performance.
Parsing JSON in Bubble
When it comes to parsing JSON, it’s always been a pain. Weird since JSON is what makes the modern Internet function. You’re going to have to use workarounds.
Data Jedi only goes one level deep in terms of nesting. In my use of API Objects, it makes things easier for me to just go only one level deep. Going deeper than one level makes it trickier to work with custom data types in the same plugin element.
This may be due to the deep nested nature. In my experience, I can change the name of the data type, the field names, and even if the field is a list or a single item via the api connector to re-initialize my api object data type and I have no issues. I can only imagine that due to the way the nested object arrays are actually creating new data types of them as well, which are attached to the main object, could be the culprit behind what you see and cause the need to update those areas.
Normally though, there would not be a need other than to maybe add some new inputs or add a single new field to the workflow actions, and not a complete rebuild from scratch.
With Data Jedi I can get a complete app data structure and ALL CRUD operations setup in a Global Data Center in less than 30 minutes if I tried to go fast. But that is not inclusive of the time to wire up the actual UI elements that the user interacts with and the communications between the UI elements and the Global Data Center.
You can save api objects as a data field on a custom data type which is the easiest way to store and retrieve the api objects later, however, without a plugin there is currently no native way in Bubble to then modify those objects values. My plugin Data Jedi makes that simple as all the actions used are built to be the same type of no-code (ie: no need to reference index value that is one off from the repeating group cell number and instead use unique IDs just like we do for Bubble custom data types).
There are definitely other free plugins or paid plugins that help you modify JSON as well, and usually that is all they do, so if all you need is something to help you modify the JSON check the plugin marketplace for some free ones and give them a test ride.
Build the best version of a complete CRUD setup and then copy and paste and repurpose for another data type.
I am guessing these will be solved by AI functions within bubble…
There is a “new page” AI function that might solve the first issue down the road. I tried a few times but I could not have enough control to actually like copy a page design and change over few critial items…. so hopefully…
The JSON use case another use case for an internal AI model. I do json manipulation all day long with various LLMs and make these connections would be helpful as well.
