data-repo

Basic site generator for a data repository.

Setup

Create a config.json based on the config.sample.json

npm install
npm run dev

Tipp: In order to overcome CORS problems for the geojson visualisations, the script copies the geojson files from the S3 into the static site. During development this makes everything a bit slow. Simply comment out lines 87-110 in source/_data/dataset.js. This will disable all maps, but everything else will work fine.

Overview

source/assets

Static files like images, CSS, etc. Those files are coppied to _site/assets, based on the .eleventy.json

eleventyConfig.addPassthroughCopy("source/assets", "assets");

source/_data/site.js

Basic configs for the site, like baseUrl, etc. baseUrl is the most important, for local development set this to your localhost setup, for deployment, this needs to be set to the final url (https://data.odis-berlin.de). TODO: Check if there is a way to switch between the two automatically without need of changing them manually. It also contains an array of formatters, which translate abbreviations from the meta.json's into more readable form.

source/_data/dataset.js

This is where the magic happens. The script reads out the S3 folder and creates a data object for each folder, which is turned into a dataset page. Each folder holds a meta.json, which holds the data to create the pages.

source/_includes/layouts

• dataset.liquid: individual pages for each dataset
• dcat.liquid: metadata json files for each dataset
• default.liquid: base template for the html files (head, header, footer)
• geojson.liquid: output of downloaded geojson files
• startpage.liquid: list pages

source

• 404.liquid: 404
• dataset-geojson.liquid: output of downloaded geojson files
• dataset-meta.liquid: metadata json files for each dataset
• dataset-page.liquid: individual pages for each dataset
• index.liquid: list pages (index.html, /en/index.html, /de/index.html)