Corpus Browser

Elasticsearch-powered search interface to browse publicly available corpora

Try it out here: HCDS Corpus Browser

Deployment

1. Clone this repository: git clone https://github.com/uhh-lt/corpus-browser.git

2. Navigate to the docker directory: cd corpus-browser/docker

3. Adjust the environment variables

   1. Copy the .env.example file to .env: cp .env.example .env
   2. Change UID and GID to the output of id

4. Run docker compose up -d

5. Visit http://localhost:13100/ in your browser

Import new corpus to elasticsearch

1. Navigate to the importer directory: cd corpus-browser/importer
2. Install conda environment: conda env create -f environment.yaml
3. Activate conda environment: conda activate corpus-browser
4. Run python importer.py --index germanu15 --input_dir ../data/uhh/json

Developer Guide

1. Navigate to the docker directory (cd corpus-browser/docker) and remove frontend from COMPOSE_PROFILES in the .env file
2. Start the docker containers: docker compose up -d
3. Navigate to the frontend directory (cd corpus-browser/frontend) and install all dependencies (npm i)
4. Start the frontend: npm run dev
5. Visit http://localhost:5173/ in your browser