use-cache-helper
provides helper functions to easily manage and scale your redis and database caching strategies.
import { init } from 'use-cache-helper';
import { redis } from 'lib';
// your ioredis instance
init({ redis: redis });
Here's how to set the maximum number of items in the paginated list before we begin data eviction.
init({ redis: redis, maxPaginatedItems: 200 });
Upstash redis
init({
upstashRedis: redis,
});
Use getOrRefresh
with SupabaseDB
Declare query function
import { supabaseClient } from 'lib';
interface IUserProfile {
id: string;
name: string;
email: string;
}
type GetUserReturn = IUserProfile | null;
const getUserById = async (): Promise<IUserProfile> => {
const { error, data } = await supabaseClient
.from('users')
.select('*')
.eq('id', userId)
.limit(1);
if (error?.message || !data) {
return null;
}
return data[0];
};
import { getOrRefresh } from 'use-cache-helper';
const verifyUserHandler = async (userId: string) => {
////////////////////////////////////////////////
///////// Use getOrRefresh ////////
////////////////////////////////////////////////
const user = await getOrRefresh<GetUserReturn>({
parseResult: true,
key: `user:${userId}`,
cacheRefreshHandler: async (): Promise<GetUserReturn> => {
return await getUserById(userId);
},
});
if (user?.id) {
// valid user
}
};
How to manually store cache using the set
function.
import { set } from 'use-cache-helper';
const user = await getUserById(userId);
const res = await set({ key: `user:${userId}`, value: user });
if (res === 'OK') {
// success
}
This library uses Redis Sorted Sets
to implement a paginated list. We store only the unique ID from your data, which is by default sorted by the date added or modified in ascending order. We use the LRU method to evict items from the paginated list.
You can configure the maximum number of items in the paginated list. When this limit is reached, it kicks out the least recently used data, determined by score.
// 200 items limit by default
init({ redis: redis, maxPaginatedItems: 200 });
Inserting data
import { insertToPaginatedList } from 'use-cache-helper';
const handleInsertItem = async (id: string) => {
await insertToPaginatedList({
id,
key: 'myPaginatedList',
// The 'score' field is optional; if not provided, it uses the Date.now() value.
score: Date.now(),
});
};
Don't worry about duplicates as long as the ID values you're using are unique.
import { insertRecordsToPaginatedList } from 'use-cache-helper';
const list = [
{
id: 'id-xxx-1',
score: 1,
// .. more data
},
{
id: 'id-xxx-2',
score: 2,
// .. more data
},
];
await insertRecordsToPaginatedList({
listKey: `listCacheKey`,
listData: list,
cachePayload: true, // cache each record using the id
cachePayloadExpiry: 3_600, // payload cache expiry in seconds
});
See full API reference Documentation
Licensed under MIT.