Referencing same cached data with multiple queryKeys

[derochedesign]

Not sure if this is an anti-pattern, but I have two separate useQuery hooks that perform get requests, which will ultimately be capable of returning the exact same data.

Basically I want to be able to set a queryKey based on the results of a query, so that a different query can rely on that cache.

Some further context:

Query A: get the most recent item in a list based on oldest item

• in this case, the endpoint is given an old item (id: 1) and the backend will return the newest in that list (lets say id: 5)
• originally i set the queryKey to ['lists', 'recent', id]

Query B: get list item by id

• here we just directly get an item based on the id
• queryKey is ['lists', id]

Query A and Query B will in a lot of cases return the same data and I'd like to be able to reference the same cached data. I would expect something like:

• once Query A fetches data, I update/create the queryKey for that data to be something like ['lists', result.data.id]
• then if Query B tries to fetch, for example, ['lists', '5'], it will get that from the cache that Query A created.

Hopefully this is clear; thanks for any guidance in advance.

[TkDodo]

take a look at seeing the cache: https://tkdodo.eu/blog/seeding-the-query-cache#seeding-details-from-lists

[derochedesign]

The push approach is exactly what i was looking for, thank you!

[derochedesign]

actually, in this case there would be two records in the cache of the same data right? rather than there being two different keys that reference the same data. the reason it matters to me (other than avoiding storing duplicate data) is that I'd like to be able to invalidate either key individually and have the result be that both keys (and of course the cached data) are invalidated.

In my case, I don't have the data necessary to manually invalidate the second key when I invalidate the first one, since I don't know what args were used to fetch it originally. (for example, my first fetch could be based on an item id:2 but return id:5, and the key would be ['lists', 'recent', '2']. Then i manually create ['lists', '5'] with the result; but once i want to invalidate item id:5, I have no way to also invalidate the first ['lists', 'recent', '2']) because I can't know what the original id used was)

if its not possible to have a many-to-one relationship between keys and cache then i'll have to reconfigure how some of my data flow works