micro-cacheable is an utility for data caching focused on micro framework. The problem this package solves is to save the already requested data in-memory to have it available for a configurated time without processing it again.
cd my-micro-project/
npm install --save micro-cacheableand add use the package like this:
// index.js
const cache = require('micro-cacheable')
const microFn = (req, res) => {
return new Date()
}
module.exports = cache(60 * 60 * 1000, microFn)then just run your microservice normally and it will return the same result for an hour (first param as miliseconds) unless that you change the request url.
Let's say that we need a microservice that receives a name (string) and search data of a person on 3 or 4 APIs:
const { parse } = require('url')
const fetch = require('node-fetch')
module.exports = async (req, res) => {
const { searchName } = parse(req.url, true).query
const facebookData = await fetch('https://someapi1.com')
const githubData = await fetch('https://someapi2.com')
const financialData = await fetch('https://someapi3.com')
return { facebookData, githubData, financialData }
}This microservice would fetch 3 APIs every time it receives a request. Probably, in some cases, if the microservice receive the same name it will return the same data, at least for the same day, so you can just add micro-cacheable like this:
const { parse } = require('url')
const fetch = require('node-fetch')
const cache = require('micro-cacheable')
const microFn = async (req, res) => {
const { searchName } = parse(req.url, true).query
const facebookData = await fetch('https://someapi1.com')
const githubData = await fetch('https://someapi2.com')
const financialData = await fetch('https://someapi3.com')
return { facebookData, githubData, financialData }
}
module.exports = cache(24 * 60 * 60 * 1000, microFn) // One day data cachingmicro-cacheable supports Mongo (and hopefully Redis in the future) because in-memory data cache can't scale horizontally and if you work with microservices you will loose that advantage. So you can avoid that problem setting the MONGO_URL and MONGO_DB enviroment variables so all your microservice's instances use the same cache.
I worked on a project with micro using it for making web-scrapping workers that take too long the first time to get the data, and users requested often the same data so with this I can save a lot of requests, processing and time making requests of +5000ms only take 50ms.
- Fork this repository to your own GitHub account and then clone it to your local device
- Link the package to the global module directory:
npm link - Within the module you want to test your local development instance of micro-cacheable, just link it to the dependencies:
npm link micro-cacheable. Instead of the default one from npm, node will now use your clone of micro-cacheable!
Thanks to ZEIT Team for giving us micro to make our life easier!
![greenkeeper[bot] avatar](https://avatars.githubusercontent.com/in/505?v=4 "greenkeeper[bot]")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent