In memory cache for Adonis JS, Redis cache for Adonis JS , AdonisJS, Cache for Adonis, Memcached storage
Cache for AdonisJS 5
Supported cache storages:
- Redis storage
- In memory storage
- Memcached storage
- Installation
- Sample Usage
- Custom storages
- Redis storage
- Memcached storage
- Storage toggle
- Custom context
- Cache events
- Cache tags
- Cache record TTL
npm i --save adonis5-cache
Compile your code:
node ace serve --watch
Install provider:
node ace invoke adonis5-cache
- For other configuration, please update the
config/cache.ts
.
After adding cache provider to your app, you can import CacheManager for accessing to cache.
import Cache from '@ioc:Adonis/Addons/Adonis5-Cache'
- For example you can use cache for reduce amount of requests to external API. You can storing responses to cache in such way:
import Cache from '@ioc:Adonis/Addons/Adonis5-Cache' export default class Service { constructor () { } public async loadDataFromExternalApi (userCode) { let userData = await Cache.get<UserDTO>(userCode) if (!userData) { userData = //load data from external api await Cache.put(userData) } return userData } }
You can add additional cache storages for saving cache data. You have to implement CacheStorageContract interface by your class:
import { CacheStorageContract } from '@ioc:Adonis/Addons/Adonis5-Cache'
class CustomCacheStorage implements CacheStorageContract {
get<T = any>(context: CacheContextContract, key: string): Promise<T | null> {
// method implementation
}
getMany<T = any>(context: CacheContextContract, keys: string[]): Promise<(T | null)[]> {
// method implementation
}
put(context: CacheContextContract, key: string, value, ttl: number): Promise<void> | void {
// method implementation
}
putMany(context: CacheContextContract, cacheDictionary, ttl: number): Promise<void> | void {
// method implementation
}
}
After creating custom storage you have to register your storage to cache manager:
import Cache from '@ioc:Adonis/Addons/Adonis5-Cache'
Cache.registerStorage('storage-name', storageInstance)
For using redis storage install adonis-redis package
npm i @adonisjs/redis@alpha
After it you should invoke generator for configuring package:
node ace invoke @adonisjs/redis
Configure redis in config/redis.ts
and add redis
storage to cache config in config/cache.ts
:
{
"currentCacheStorage": "redis",
"enabledCacheStorages": [
"redis"
]
}
For using memcached storage install adonis-memcached client
npm i adonis5-memcached-client
After it you should invoke generator for configuring package:
node ace invoke adonis5-memcached-client
Configure memcached client in config/memcached.ts
and add memcached
storage to cache config in config/cache.ts
:
{
"currentCacheStorage": "memcached",
"enabledCacheStorages": [
"memcached"
]
}
After registration you can use your storage in such way:
const cachedValue = await Cache.viaStorage('storage-name').get('cache-key')
Or you can enable your storage as default cache storage:
Cache.enableStorage('storage-name')
const cachedValue = await Cache.get('cache-key') // value will be received from your storage
Cache contexts responsible for serialization and deserialization your data to cache storages. For example you can add additional keys or transform you data before serialization and deserialization processes.
You can implement your context is like to custom cache storage, your custom context have to implement ** CacheContextContract**:
import { CacheContextContract } from '@ioc:Adonis/Addons/Adonis5-Cache'
const customContext: CacheContextContract = {
serialize: (data: any) => { JSON.stringify({ data: data, serializedAt: Date.now })},
deserialize: (cacheRecord:string) => ({ ...JSON.parse(cacheRecord), deserializedAt: Date.now }),
}
After implementation you have to register new context:
Cache.registerContext('custom-context-name', customContext)
And then you can using new context when you accessing to cache storage:
const cachedValue = await Cache.get<RecordDTO>('cache-key') // Reading data from cache using custom context
await Cache.put('cache-key', cachedData) // Storing data to cache using custom context
Of course, you can enable your custom context as default cache context:
Cache.enableContext('custom-context-name') // After this your cache operations will be use your custom context
You can pass fallback value when you calling get method:
const cachedValue = await Cache.get < RecordDTO > ('cache-key', fallbackValue)
Or pass async function as fallback value in the following way
const cachedValue = await Cache.get<RecordDTO>('cache-key', async () => {
return callApi();
})
If cache storage value doesn't exists fallback value will storage to cache by selected key. You can specify ttl for fallback value as function parameter:
const cachedValue = await Cache.get<RecordDTO>('cache-key', fallbackValue, customTTL)
Cache provider implements several events:
- cache-record:read
- cache-record:written
- cache-record:missed
- cache-record:forgotten
You can add listener for this events in the following ways:
import Event from '@ioc:Adonis/Core/Event'
Event.on('cache-record:missed', 'CacheEventListener.handleCacheRecordMissedEvent')
You can configure which events are emitted with cache config:
{
recordTTL: 10000,
currentCacheStorage: 'redis',
enabledCacheStorages: ['redis'],
cacheKeyPrefix: '',
enabledEvents: {
'cache-record:read': false,
'cache-record:written': false,
'cache-record:missed': false,
'cache-record:forgotten': false,
}
}
When you need partial flushing your cache you can use cache tags. This features allow you tag your cache records and then flush only tagged records. For using cache tags you should call method tags on cache manager with list of your tags as argument, then you get TaggedCacheManager, which allow you using tagged cache functionality.
Cache.tags('tag-1', 'tag-2', 'tag-3')
Actually for writing cache record with tags you can in the following way:
await Cache.tags('tag-1', 'tag-2', 'tag-3').put('key', 'value')
Or store dictionary with cache records: Actually for writing cache record with tags you can in the following way:
await Cache.tags('tag-1', 'tag-2', 'tag-3').putMany({ key: 'value' })
Then you can clear tagged records in the following way:
await Cache.tags('tag-1', 'tag-2', 'tag-3').flush()
During this operation tags and tagger records will be removed from storage, however records with another tags will remain in your storage. This is great feature for storing responses from different API's. You can tag each response by appropriate tag and flush responses only for desirable API.
You can configure record TTL using cache config:
{
recordTTL: 60000,
ttlUnits: 'ms'
}
Or you can set record ttl as function parameter:
await Cache.put(userData, 1000)
Your value will be transformed to milliseconds using time units which configured by ttlUnits parameter in your cache config.