Skip to content

Language-Research-Technology/data-packs

 
 

Repository files navigation

A repository to host data packs for use in Describo. A data pack is simply a pre-defined array of JSON-LD objects that can be used as the value of a property.

Github: https://github.com/describo/data-packs

For users

Install the package

npm install --save @describo/data-packs

Loading the data packs into elastic search

For development you can start a local elastic search service with:

docker compose up -d ; docker compose logs -f ; docker compose stop ; docker compose rm -f

Then in your application you can trigger a load of all of the data packs:

const { IndexDataPacks } = require('@describo/data-packs')
const index = new IndexDataPacks({ elasticUrl: "http://localhost:9200", indexName: "common" });
await index.load();

Or to do it once somewhere outside of your app with something like the script ./bin/index-data-packs.js

And if you are using an external elastic server use the following example:

const { IndexDataPacks } = require('@describo/data-packs')
const index = new IndexDataPacks({
    elasticPath: "data",
    elasticUrl: "https://lookups.ldaca.edu.au/",  
    elasticAuth: { apiKey: 'base64EncodedKey' }, 
    log: true,
    indexName: "common"
});
await index.load();

Working with the package

const { DataPack } = require('@describo/data-packs')
let datapack = new DataPack({ dataPacks: ['Austlang', 'Glottolog'], indexFields: ['@id', 'name']})
await datapack.load()

let language = datapack.get({
  field: "name",
  value: "Nyaki Nyaki / Njaki Njaki",
});

Returns a JSON-LD snippet if a match is found.

If you only want a subset of the properties try:

const { DataPack } = require('@describo/data-packs')
let datapack = new DataPack({ dataPacks: ['Austlang', 'Glottolog'], indexFields: ['@id', 'name']})
await datapack.load()


let language = datapack.get({
  field: "name",
  value: "Nyaki Nyaki / Njaki Njaki",
  properties: ['name', 'languageCode']
});

And you will get only those (along with @id and @type which are always returned).

For Developers

Working on the package - writing tests

  • start elastic search docker container: docker compose up
  • Run the tests in watch mode: npm run test:watch

Updating the JSDoc documentation

  • npm run generate-docs
  • Then commit the docs and push to github

Adding a data pack to this repository

  • add a folder in the data-packs top level folder named as your data pack.
  • create an update script inside that folder named like: create-data-pack.js.
    • Your script should be self contained and retrieve the data it needs. That is, don't check a datasource into this repo if you can avoid it. Get it from the web so that the data pack is updated from the main version online. Whatever, and wherever, that is.
  • add a reference to that script in update-all-packages.sh so that your data pack is updated when the others are.
  • document the data structure by adding a file in your data pack folder called: datapack-entry-format.js. Follow the example in languages/datapack-entry-format.js.

Building and publishing an update

  • bash ./update_all_packages.sh
  • npm run generate-docs
  • Commit all of the changes
  • Bump the version: npm version minor
  • Publish npm publish

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 97.8%
  • Shell 2.0%
  • CSS 0.2%