The API slice object includes various utilities that can be used for cache management, such as implementing optimistic updates, as well implementing server side rendering.
These are included as api.util
inside the API object.
info
Some of the TS types on this page are pseudocode to illustrate intent, as the actual internal types are fairly complex.
updateQueryData
A Redux thunk action creator that, when dispatched, creates and applies a set of JSON diff/patch objects to the current state. This immediately updates the Redux state with those changes.
Signatureconst updateQueryData = (
endpointName: string,
arg: any,
updateRecipe: (draft: Draft<CachedState>) => void,
updateProvided?: boolean,
) => ThunkAction<PatchCollection, PartialState, any, AnyAction>
interface PatchCollection {
patches: Patch[]
inversePatches: Patch[]
undo: () => void
}
Parameters
endpointName
: a string matching an existing endpoint namearg
: an argument matching that used for a previous query call, used to determine which cached dataset needs to be updatedupdateRecipe
: an Immer produce
callback that can apply changes to the cached stateupdateProvided
: a boolean indicating whether the endpoint's provided tags should be re-calculated based on the updated cache. Defaults to false
.The thunk action creator accepts three arguments: the name of the endpoint we are updating (such as 'getPost'
), any relevant query arguments, and a callback function. The callback receives an Immer-wrapped draft
of the current state, and may modify the draft to match the expected results after the mutation completes successfully.
The thunk returns an object containing {patches: Patch[], inversePatches: Patch[], undo: () => void}
. The patches
and inversePatches
are generated using Immer's produceWithPatches
method.
This is typically used as the first step in implementing optimistic updates. The generated inversePatches
can be used to revert the updates by calling dispatch(patchQueryData(endpointName, arg, inversePatches))
. Alternatively, the undo
method can be called directly to achieve the same effect.
Note that the first two arguments (endpointName
and arg
) are used to determine which existing cache entry to update. If no existing cache entry is found, the updateRecipe
callback will not run.
const patchCollection = dispatch(
api.util.updateQueryData('getPosts', undefined, (draftPosts) => {
draftPosts.push({ id: 1, name: 'Teddy' })
}),
)
In the example above, 'getPosts'
is provided for the endpointName
, and undefined
is provided for arg
. This will match a query cache key of 'getPosts(undefined)'
.
i.e. it will match a cache entry that may have been created via any of the following calls:
api.endpoints.getPosts.useQuery()
useGetPostsQuery()
useGetPostsQuery(undefined, { ...options })
dispatch(api.endpoints.getPosts.initiate())
dispatch(api.endpoints.getPosts.initiate(undefined, { ...options }))
Example 2
const patchCollection = dispatch(
api.util.updateQueryData('getPostById', 1, (draftPost) => {
draftPost.name = 'Lilly'
}),
)
In the example above, 'getPostById'
is provided for the endpointName
, and 1
is provided for arg
. This will match a query cache key of 'getPostById(1)'
.
i.e. it will match a cache entry that may have been created via any of the following calls:
api.endpoints.getPostById.useQuery(1)
useGetPostByIdQuery(1)
useGetPostByIdQuery(1, { ...options })
dispatch(api.endpoints.getPostById.initiate(1))
dispatch(api.endpoints.getPostById.initiate(1, { ...options }))
upsertQueryData
A Redux thunk action creator that, when dispatched, acts as an artificial API request to upsert a value into the cache.
Signatureconst upsertQueryData = <T>(endpointName: string, arg: any, newEntryData: T) =>
ThunkAction<Promise<CacheEntry<T>>, PartialState, any, UnknownAction>
Parameters
endpointName
: a string matching an existing endpoint namearg
: an argument matching that used for a previous query call, used to determine which cached dataset needs to be updatednewEntryValue
: the value to be written into the corresponding cache entry's data
fieldThe thunk action creator accepts three arguments: the name of the endpoint we are updating (such as 'getPost'
), the appropriate query arg values to construct the desired cache key, and the data to upsert.
If no cache entry for that cache key exists, a cache entry will be created and the data added. If a cache entry already exists, this will overwrite the existing cache entry data.
The thunk executes asynchronously, and returns a promise that resolves when the store has been updated. This includes executing the transformResponse
callback if defined for that endpoint.
If dispatched while an actual request is in progress, both the upsert and request will be handled as soon as they resolve, resulting in a "last result wins" update behavior.
Exampleawait dispatch(
api.util.upsertQueryData('getPost', { id: 1 }, { id: 1, text: 'Hello!' }),
)
patchQueryData
A Redux thunk action creator that, when dispatched, applies a JSON diff/patch array to the cached data for a given query result. This immediately updates the Redux state with those changes.
Signatureconst patchQueryData = (
endpointName: string,
arg: any
patches: Patch[],
updateProvided?: boolean
) => ThunkAction<void, PartialState, any, UnknownAction>;
Parameters
endpointName
: a string matching an existing endpoint namearg
: a cache key, used to determine which cached dataset needs to be updatedpatches
: an array of patches (or inverse patches) to apply to cached state. These would typically be obtained from the result of dispatching updateQueryData
updateProvided
: a boolean indicating whether the endpoint's provided tags should be re-calculated based on the updated cache. Defaults to false
.The thunk action creator accepts three arguments: the name of the endpoint we are updating (such as 'getPost'
), the appropriate query arg values to construct the desired cache key, and a JSON diff/patch array as produced by Immer's produceWithPatches
.
This is typically used as the second step in implementing optimistic updates. If a request fails, the optimistically-applied changes can be reverted by dispatching patchQueryData
with the inversePatches
that were generated by updateQueryData
earlier.
In cases where it is desired to simply revert the previous changes, it may be preferable to call the undo
method returned from dispatching updateQueryData
instead.
const patchCollection = dispatch(
api.util.updateQueryData('getPosts', undefined, (draftPosts) => {
draftPosts.push({ id: 1, name: 'Teddy' })
}),
)
dispatch(
api.util.patchQueryData(
'getPosts',
undefined,
patchCollection.inversePatches,
),
)
patchCollection.undo()
upsertQueryEntries
A standard Redux action creator that accepts an array of individual cache entry descriptions, and immediately upserts them into the store. This is designed to efficiently bulk-insert many entries at once.
Signature
export type NormalizedQueryUpsertEntry<
Definitions extends EndpointDefinitions,
EndpointName extends QueryKeys<Definitions>,
> = {
endpointName: EndpointName
arg: QueryArgFrom<Definitions[EndpointName]>
value: ResultTypeFrom<Definitions[EndpointName]>
}
const upsertQueryEntries = (entries: NormalizedQueryUpsertEntry[]) =>
PayloadAction<NormalizedQueryUpsertEntry[]>
Parameters
entries
: an array of objects that contain the data needed to upsert individual cache entries:
endpointName
: the name of the endpoint, such as "getPokemon"
arg
: the full query key argument needed to identify this cache entry, such as "pikachu"
(same as you would pass to a useQuery
hook or api.endpoints.someEndpoint.select()
)value
: the data to be upserted into this cache entry, exactly as formatted.This method is designed as a more efficient approach to bulk-inserting many entries at once than many individual calls to upsertQueryData
. As a comparison:
upsertQueryData
:
pending
and fulfilled
transformResponse
callback if defined for that endpoint, as well as the merge
callback if definedupsertQueryEntries
:
transformResponse
, so the provided value
fields must already be in the final format expected for that endpoint. However, it will still run the merge
callback if definedCurrently, this method has two main use cases. The first is prefilling the cache with data retrieved from storage on app startup. The second is to act as a "pseudo-normalization" tool. RTK Query is not a "normalized" cache. However, there are times when you may want to prefill other cache entries with the contents of another endpoint, such as taking the results of a getPosts
list endpoint response and prefilling the individual getPost(id)
endpoint cache entries.
If no cache entry for that cache key exists, a cache entry will be created and the data added. If a cache entry already exists, this will overwrite the existing cache entry data.
If dispatched while an actual request is in progress, both the upsert and request will be handled as soon as they resolve, resulting in a "last result wins" update behavior.
Exampleconst api = createApi({
endpoints: (build) => ({
getPosts: build.query<Post[], void>({
query: () => '/posts',
async onQueryStarted(_, { dispatch, queryFulfilled }) {
const res = await queryFulfilled
const posts = res.data
dispatch(
api.util.upsertQueryEntries(
posts.map((post) => ({
endpointName: 'getPost',
arg: { id: post.id },
value: post,
})),
),
)
},
}),
getPost: build.query<Post, Pick<Post, 'id'>>({
query: (post) => `post/${post.id}`,
}),
}),
})
prefetch
A Redux thunk action creator that can be used to manually trigger pre-fetching of data.
Signaturetype PrefetchOptions = { ifOlderThan?: false | number } | { force?: boolean }
const prefetch = (endpointName: string, arg: any, options: PrefetchOptions) =>
ThunkAction<void, any, any, UnknownAction>
Parameters
endpointName
: a string matching an existing endpoint nameargs
: a cache key, used to determine which cached dataset needs to be updatedoptions
: options to determine whether the request should be sent for a given situation:
ifOlderThan
: if specified, only runs the query if the difference between new Date()
and the lastfulfilledTimeStamp
is greater than the given value (in seconds)force
: if true
, it will ignore the ifOlderThan
value if it is set and the query will be run even if it exists in the cache.The thunk action creator accepts three arguments: the name of the endpoint we are updating (such as 'getPost'
), any relevant query arguments, and a set of options used to determine if the data actually should be re-fetched based on cache staleness.
React Hooks users will most likely never need to use this directly, as the usePrefetch
hook will dispatch the thunk action creator result internally as needed when you call the prefetching function supplied by the hook.
dispatch(api.util.prefetch('getPosts', undefined, { force: true }))
selectInvalidatedBy
A selector function that can select query parameters to be invalidated.
Signaturefunction selectInvalidatedBy(
state: RootState,
tags: ReadonlyArray<TagDescription<string>>,
): Array<{
endpointName: string
originalArgs: any
queryCacheKey: QueryCacheKey
}>
Parameters
state
: the root statetags
: a readonly array of invalidated tags, where the provided TagDescription
is one of the strings provided to the tagTypes
property of the api. e.g.
[TagType]
[{ type: TagType }]
[{ type: TagType, id: number | string }]
The function accepts two arguments
It returns an array that contains
const entries = api.util.selectInvalidatedBy(state, ['Post'])
const entries = api.util.selectInvalidatedBy(state, [{ type: 'Post', id: 1 }])
const entries = api.util.selectInvalidatedBy(state, [
{ type: 'Post', id: 1 },
{ type: 'Post', id: 4 },
])
invalidateTags
A Redux action creator that can be used to manually invalidate cache tags for automated re-fetching.
Signatureconst invalidateTags = (
tags: Array<TagTypes | FullTagDescription<TagTypes>>,
) => ({
type: string,
payload: tags,
})
Parameters
tags
: an array of tags to be invalidated, where the provided TagType
is one of the strings provided to the tagTypes
property of the api. e.g.
[TagType]
[{ type: TagType }]
[{ type: TagType, id: number | string }]
The action creator accepts one argument: the cache tags to be invalidated. It returns an action with those tags as a payload, and the corresponding invalidateTags
action type for the api.
Dispatching the result of this action creator will invalidate the given tags, causing queries to automatically re-fetch if they are subscribed to cache data that provides the corresponding tags.
Exampledispatch(api.util.invalidateTags(['Post']))
dispatch(api.util.invalidateTags([{ type: 'Post', id: 1 }]))
dispatch(
api.util.invalidateTags([
{ type: 'Post', id: 1 },
{ type: 'Post', id: 'LIST' },
]),
)
selectCachedArgsForQuery
A selector function that can select arguments for currently cached queries.
Signaturefunction selectCachedArgsForQuery(
state: RootState,
queryName: QueryName,
): Array<QueryArg>
Parameters
state
: the root statequeryName
: a string matching an existing query endpoint nameThe function accepts two arguments
the root state and
the name of the query
It returns an array that contains arguments used for each entry.
Exampleconst args = api.util.selectCachedArgsForQuery(state, 'getPosts')
resetApiState
Signature
const resetApiState = () => ({
type: string,
payload: undefined,
})
Description
A Redux action creator that can be dispatched to manually reset the api state completely. This will immediately remove all existing cache entries, and all queries will be considered 'uninitialized'.
Note that hooks also track state in local component state and might not fully be reset by resetApiState
.
dispatch(api.util.resetApiState())
getRunningQueriesThunk
and getRunningMutationsThunk
Signature
getRunningQueriesThunk(): ThunkWithReturnValue<Array<QueryActionCreatorResult<any>>>
getRunningMutationsThunk(): ThunkWithReturnValue<Array<MutationActionCreatorResult<any>>>
Description
Thunks that (if dispatched) return either all running queries or mutations. These returned values can be awaited like promises.
This is useful for SSR scenarios to await all queries (or mutations) triggered in any way, including via hook calls or manually dispatching initiate
actions.
Awaiting all currently running queries example
await Promise.all(dispatch(api.util.getRunningQueriesThunk()))
getRunningQueryThunk
and getRunningMutationThunk
Signature
getRunningQueryThunk<EndpointName extends QueryKeys<Definitions>>(
endpointName: EndpointName,
args: QueryArgFrom<Definitions[EndpointName]>
): ThunkWithReturnValue<
| QueryActionCreatorResult<
Definitions[EndpointName] & { type: 'query' }
>
| undefined
>
getRunningMutationThunk<EndpointName extends MutationKeys<Definitions>>(
endpointName: EndpointName,
fixedCacheKeyOrRequestId: string
): ThunkWithReturnValue<
| MutationActionCreatorResult<
Definitions[EndpointName] & { type: 'mutation' }
>
| undefined
>
Description
Thunks that (if dispatched) return a single running query (or mutation) for a given endpoint name + argument (or requestId/fixedCacheKey) combination, if it is currently running. If it is not currently running, the function returns undefined
.
These thunks are primarily added to add experimental support for suspense in the future. They enable writing custom hooks that look up if RTK Query has already got a running query/mutation for a certain endpoint/argument combination, and retrieving that to throw
it as a promise.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4