STA Documentation Platform

The sta-doc project is the JS/React/Nextjs based UI representation of the STA Documentation Platform. The documentation data is fetched and parsed from a MediaWiki instance backend and represents the whole documentation in an explorative way to the enduser.

The MediaWiki instance (database) is holding documentation data of two different sources:

• GND
  • is the largest collection of cultural and research authority data in the German-speaking countries
• RDA
  • a standard for descriptive cataloging, providing instructions and guidelines on formulating bibliographic data

In the context of this application GND and RDA are namespaces, which are having a specific highlight color for distinction.

The goal of this application is to have a unified user interface to read and search these documentation standards. For example, the dataset about 'Preferred name: person or family' in MediaWiki and in this Application.

Overview

The data is pre-fetched and -parsed to compose it in a flexible way, while having no performance dependencies to MediaWiki in production.

Update data with npm run data

This code is fetching, parsing and storing the static data. It runs in the context of node and gets relevant data from a pre-defined MediaWiki instance.

Data fetching from a mediawiki instance

npm run data:fetch The fetched data gets saved in /data/raw/*.json.

Data parsing the fetched raw data

npm run data:parse Transforms all data from /data/raw/*.json and saves the result in /data/parsed/*.json. The most relevant data file is /data/parsed/entities.json, which is the last result of the composition, and it's meant to hold all relevant data for the client, pre-sorted and structured, ready for rendering.

Prepare single entities to be used to fill the solr index

npm run split-entities (Linux)

This command runs the bash script located at /bin/split-entities-solr.sh It is used to split the whole parsed entities into single representatives (like P18.json, Q2.json, ...) and stores all of them in docker/solr/data.

npm run split-entities-powershell (Windows)

This command runs the bash script located at /bin/split-entities-solr.ps1 for Windows developers.

npm run data:fetch:properties-items

This creates/updates two typescript files, each with an enum. For readable code references to Items / Properties. Be aware, this command may break the build process and enum member references may need to be adapted if the corresponding label has changed.

Running the application

The application can be run in docker containers or on the host system. The application is configured to use the basePath as to be /doc.

Developer Mode

Start developing with

• npm run install && npm run dev or
• npm run docker:dev:build; npm run docker:dev:up && sh ./docker/index-solr-dev.sh

Production Mode

Start testing the productive version with (Test local without solr search)

• npm run build && npm run start (Test with docker compose setup)
• npm run docker:build; npm run docker:up && sh ./docker/index-solr.sh

Documentation

• React
• Next JS
• Apache Solr
• Ant Design
• Emotion
• SWR
• Lodash
• react-use
• useQueryParams
• solr-client
• slugify
• react-highlight-words
• copy-to-clipboard