I really wish I had something different to say here, but I have to agree with the awesome @basdebruin. Group IDs, like Column, Board, User, and Item IDs are all generated more or less randomly and we do not provide a way to change this through the API or in the UI.
That said, this is feedback we have been getting somewhat frequently lately and I totally agree it’s not a very convenient way of doing things. While I can’t promise anything just yet, I’ve made sure that our development team is aware of this shortcoming - hopefully, we will be able to consider a better way to do this in the future.
Last, but not least - I’d personally really appreciate it if the next time you create a new topic, it would be placed in the appropriate SubCategory within the Dev forum. It might not seem like a big deal, but it can help other users find the answer to their question more easily, so it would help keep the community nice and tidy I’ve made the changes myself now, so it’s definitely no biggie.
The id’s generated are really inconvenient. I’d almost prefer them to just be numbers, like item ids and board ids. If they are strings, it would make much more sense to be able to update them.
I am currently saving a lot of these ids a “secret” or a “parameter-in-a-store”, but it would be a bit preferable if I could just manually create a column/group in the UI, and associate that column with an expected ID, which I’d then use going forward in more automated actions.
In my case I have a board to integrate, where the board manager toon some template and renamed (100% rightfully) the groups to something meaningful to him. But he was not aware, that the groups already have some IDs.
And now, because it is impossible to alter the group IDs, I deal with something like this:
groups: [
{“id”:“emailed_items”, “title”:“To invoice”},
{“id”:“topics”, “title”:“Waiting for payment”}
] etc.
So in order to move an item to “To invoice” group I need to mutate it to “emailed_items” group id…
And I cannot do anything about it AFAIK (apart perhaps of programatically removing all the existing groups, creating new ones with the same names but corrected IDs and re-assigning the items to these new groups).
There is no correlation between the group NAME and the group ID, the same for item NAME and item ID, column NAME and column ID.
Therefore you can never rely on a group name to represent the id. This is a good approach because this allows for changing names without changing the real identifier. The side effect is what you are describing here. It would be a lot clearer if the groupId was just a number as the itemId (same for columnId).
I fully agree that the current generated ID are messy, but the same is true for columnId’s. If you create a brand new column on an empty board and name it “Home Address” the id will probably be “home_address”. When you or the user changes the name to “Business Address” you end up in the same situation where you need to access the column by (id) “home_address” to change the (name) “Business Address”
I’ve made the changes myself now, so it’s definitely no biggie.