@rdf-esm/sink-map

@rdf-esm/sink-map

Map for RDFJS Sinks including shortcut methods.

Fork alert :exclamation:

This package is an ES Modules fork of @rdfjs/sink-map

It also adds the ability to register lazy sinks, which are created on first request. For example to have the respecitve modules imported dynamically. Check the example below.

Usage

The package provides Map from a string key to a Sink with a shortcut for .import. Typical it's used to store parsers or serializers for specific media types. As SinkMap extends from the ECMAScript 2015 Map and doesn't overload any standard methods, methods like set, get, has or delete can be used as defined in the specification.

Create a SinkMap

The constructor accepts arrays with key/sink pairs to fill the map:

const map = new SinkMap([
  ['text/turtle', new ParserN3()]
])

It's also possible to create an empty map and add or extend it later using the .set() method:

const map = new SinkMap()

map.set('text/turtle') = new ParserN3()

Add a lazy-loaded sink

Sinks can also be added as an async function, for example to dynamically import the module.

Here's an example from @rdf-esm/formats-common:

const map = new SinkMap()

map.set('text/turtle', async () => {
  const ParserN3 = (await import('@rdfjs/parser-n3')).default
  return new ParserN3()
})

Find a Sink

The .get method searches for the matching Sink and returns it:

const map = new SinkMap([...])

const sink = map.get('text/turtle')

if (sink) {
  // found
} else {
  // not found
}

Import shortcut

Usually you want to call the .import method of the matching Sink. The map provides a shortcut for this. It also has a .import method, but requires additionally the key as the first argument. It returns null if no matching sink was found:

const map = new SinkMap([...])

const input = fs.createReadStream('..')
const output = map.import('text/turtle', input)

if (output) {
  // found
} else {
  // not found
}

Example

Here is a complete example where the map is used to store parsers and the .import shortcut is used to parse a string input:

import { SinkMap } from '@rdfjs/sink-map'
import ParserN3 from '@rdfjs/parser-n3'
import stream from 'stream'

const map = new SinkMap([
  ['text/turtle', new ParserN3()]
])

const input = new stream.Readable({
  read: () => {
    input.push(`
      PREFIX s: <http://schema.org/>

      [] a s:Person ;
        s:jobTitle "Professor" ;
        s:name "Jane Doe" ;
        s:telephone "[phone]" ;
        s:url <http://www.janedoe.com> .
    `)
    input.push(null)
  }
})

const output = map.import('text/turtle', input)

output.on('data', quad => {
  console.log(`${quad.subject.value} - ${quad.predicate.value} - ${quad.object.value}`)
})