Karp backend

This in the version 6 of Karp backend, for the legacy version (v5).

Setup

This project uses poetry and MariaDB.

1. Run make install or make install-dev for a develop-install

2. Install MariaDB and create a database

3. Setup environment variables (can be placed in a .env file in the root and then ? poetry run sets those):

   export DB_DATABASE=<name of database>
   export DB_USER=<database user>
   export DB_PASSWORD=<user's password>
   export DB_HOST=localhost
   export AUTH_JWT_PUBKEY_PATH=/path/to/pubkey
   

4. Activate the virtual environment by running: poetry shell

5. Run karp-cli db up to initialize database

6. Run make serve or make serve-w-reload to start development server

   or poetry shell and then uvicorn asgi:app

7. To setup Elasticsearch, download Elasticsearch 6.x and start it

8. Install elasticsearch python libs for the right version

   1. If you use Elasticsearch 6.x, run source <VENV_NAME>/bin/activate and pip install -e .[elasticsearch6]

9. Add environment variables

export ES_ENABLED=true
export ELASTICSEARCH_HOST=localhost:9200
export SEARCH_CONTEXT=es6_search_service

Create test resources

1. poetry shell and then:
2. karp-cli entry-repo create karp/tests/data/config/places.json
3. karp-cli resource create karp/tests/data/config/places.json
4. karp-cli entries add places tests/data/places.jsonl
5. Do the same for municipalities
6. karp-cli resource publish places
7. karp-cli resource publish municipalities

Pre-processing data before publishing

** TODO: review this ** Can be used to have less downtime, because sometimes the preprocessing may be faster on another machine than the machine that talks to Elasticsearch. Do create and import on both machines, with the same data. Use machine 1 to preprocess and use result on machine 2.

1. Create resource and import data as usual.

2. Run karp-cli preprocess --resource_id places --version 2 --filename places_preprocessed

   places_preprocessed will contain a pickled dataset containing everything that is needed

3. Run karp-cli publish_preprocessed --resource_id places --version 2 --data places_preprocessed

4. Alternatively run karp-cli reindex_preprocessed --resource_id places --data places_preprocessed , if the resource was already published.

Technologies

Python

• Poetry >= 3.10
• FastAPI
• SQLAlchemy
• Typer
• Elasticsearch
• Elasticsearch DSL

Databases

• MariaDB
• Elasticsearch

Development

Version handling

Version can be bumped with bump2version.

Usage:

• Increase patch number a.b.X => a.b.(X+1): make bumpversion or bumpversion patch
• Increase minor number a.X.c => a.(X+1).0: make bumpversion-minor or bumpversion minor
• Increase major number X.b.c => (X+1).0.0: make bumpversion-major or bumpversion major
• To custom version a.b.c => X.Y.Z: bumpversion --new-version X.Y.Z

bumpversion is configured in .bumpversion.cfg.

The version is changed in the following files:

• setup.py
• src/karp/__init__.py
• .bumpversion.cfg
• doc/karp_api_spec.yaml