@harperdb/http-router

A HarperDB Component for routing requests to other components that is inspired by the Edgio router API: https://docs.edg.io/applications/v4/routing.

Installation

Go into the HarperDB application you would building and install this package and add it to the config.yaml file:

1. Install:

npm install @harperdb/http-router

2. Add to config.yaml:

'@harperdb/http-router':
  package: '@harperdb/http-router' # this can include a @version number if desired
  files: '*.*js' # Load js files so it can find the router.js file and config
# The router comes before the other main framework adapter in the pipeline
'@harperdb/nextjs':
  package: '@harperdb/nextjs'
  files: '/*'
  prebuilt: true

And then you can build a router.js file in the root of your application that looks like this:

const { Router, or, nextRoutes } = require('@harperdb/http-router');
module.exports = new Router().get('/some-path', ({ cache, proxy }) => {
	// handle the request here
})

Options

All configuration options are optional

port: number

Specify a port for the caching server. Defaults to 9926.

files: String

Used to load the necessary JS files.

config.yaml extension options

• clearRestIntervalCount - Number of records to invalidate prior to pausing when /invalidate endpoint is called
• clearRestIntervalMs - Duration of pause in milliseconds when /invalidate endpoint is called
• scheduledFullCacheClearTime - Time of day to perform a full cache clear (Expressed as hours in 24-hour format UTC time, i.e. 10.33 = 10:20 AM UTC)
• additionalCacheDatabaseGroups - Array of additional database groups to use for caching. Each group will create a new database to store cached records.

Example usage:

'@harperdb/http-router':
  package: '@harperdb/http-router'
  files: '*.*js'
  clearRestIntervalCount: 1000
  clearRestIntervalMs: 10
  scheduledFullCacheClearTime: 10.33
  additionalCacheDatabaseGroups:
    - 'api'

Multi DB Caching

By default, the cache will use a database named cache to store cached records. You can specify additional database groups to use for caching via the additionalCacheDatabaseGroups configuration option in the config.yaml file. Each additional database group will create a new database to store cached records. The database group name can optionally be passed as the cacheGroup paramater as part of the edge caching configuration within the request actions.

i.e.

cache({
    edge: {
      maxAgeSeconds: 10000,
      staleWhileRevalidateSeconds: 3600,
      cacheGroup: 'api'
    }
})

Invalidation

Cache can be invalidated via a POST request to /invalidate

This will invalidate records from the default cache database. To invalidate records from an additional cache database, use the x-cache-group request header to specify the database group name.

i.e.

POST /invalidate
HEADER: 'x-cache-group: api'