The idea is to encapsulate the HTTP call making it easier to implement the requests by removing the need for duplicate settings. The methods are designed once all through the application and are fully accessible through the Capsule package. Another feature is that it has already been implemented with requests caching using the external "axios-extensions" package.
yarn add capsule-http
or npm instal capsule-http --save
To use this package is quite simple, after installed you need to import and register the requests with an url
import Capsule from 'capsule-http'
// Create a file called API.js and put your server requests there
// All of your apis will be accessible just importing the capsule-http
Capsule.register("https://your-global-endpoint", {
get: {
'fetch.posts': '/posts/:id',
'fetch.post': '/post',
'fetch.mongodb.post': '/posts/3',
},
post: {
'update.posts': '/posts/:id'
},
put: {
'insert.posts': '/posts/:id'
},
delete: {
'remove.post': '/posts/:id'
}
})
You can also use the register to define any object property as cache, just passing an object instead of the url:
import Capsule from 'capsule-http'
Capsule.register("https://your-global-endpoint", {
get: {
'fetch.posts': {
url: '/posts/:id',
cache: 300
},
'fetch.post': '/post',
'fetch.mongodb.post': '/posts/3',
}
})
Into your application services, import the package and execute the request
import Capsule from 'capsule-http'
Capsule.request('fetch.posts')
// 4. You can pass parameters with the request
Capsule.request('fetch.posts', { id: 3 })
You can also pass an options parameter, which you can pass info like headers, timeout, auth data or any other axios config specific
There's two configs that are not listed into axios.
- Caching data, this wrapper come with axios-extension cache adapter to handle cache requests, to take advantage of just pass the time in seconds of the cache as option parameter:
// If you want, you can also set a cache handler to optimize requests (ex: 10 minutes)
Capsule.request('fetch.posts', { id: 3 }, { cache: 600 })
The default it's configured as 5 minutes but you can pass any time that you wish
You can make a request without cache, this way the cache keeps working but your request answer without it
Capsule.request('fetch.posts', { id: 3 }, { cache: false })
You can also force the cache to be cleaned using a flag forceUpdate
to remove it or updated it's ttl
Capsule.request('fetch.posts', { id: 3 }, { forceUpdate: true, cache: false })
// or
Capsule.request('fetch.posts', { id: 3 }, { forceUpdate: true, cache: 500 })
Just run yarn test
or npm run test
and the command line will start a local server wich will be test every single route and parameters related to the js API.
The package 'axios-extesions' provides also an throttleAdapterEnhancer designed to help multiples requests return same data without any aditional check, just returning the same response did in the last second.
This adapter wasn't implemented yet because need extensive tests related on how it would perform in a node server to handle not just client side requests but also backend requests.