Finance dashboard

Tool to analyze income and expenses, similar to apps like Finanzguru.

Has four parts:

• Grafana dashboard that shows the data
• Postgres database as data source for Grafana
• Syncer script that imports data from a bank account
• Patcher service that is accessed by links in Grafana to update transactions if needed

Currently the following banks are supported:

• ING (DE)

Quickstart

docker-compose up

Then visit http://localhost:3000. You will find a Grafana dashboard with some example data.

Tip

The default login is admin/admin. You can change the password in the Grafana settings.

Loading in your own data

Make sure to have a fresh database. It's easiest to remove the complete deployment and start over:

docker-compose down -v

Now you need two things:

• A banking data importer plugin: see the example importer in plugins/example_importer
• A classifier plugin: see the example classifier in plugins/example_classifier

For a real-world example, see the ING DE CSV importer in plugins/ing_de_csv_importer. This plugin uses the CSV data that can be exported on the ING banking dashboard. You can also provide transactions by hand through the JSON importer plugin in plugins/json_importer, or adapt this plugin to fit your needs.

Tip

You can provide an additional-requirements.txt in the plugins directory to install additional dependencies that your plugin may need.

When you have your plugin(s), set the environment variables IMPORTER_PLUGINS and CLASSIFIER_PLUGIN to the Python path of the plugin. Then start the deployment. Use the --build flag to rebuild the plugins if you made changes.

export IMPORTER_PLUGINS="plugins.example_importer.importer,plugins.json_importer.importer"
export CLASSIFIER_PLUGIN="plugins.example_classifier.classifier"
docker-compose up --build

Patching transactions

Sometimes transactions may not be classified correctly. For this case you can click the Change classification link in the dashboard. This will open a separate dashboard where you can insert your custom classification:

When clicking on the link in the dashboard, a local patcher service will be called that patches this transaction in the database and create a patch json file in patcher/patches/. In the dashboard you should see an alert noting that the transaction has been patched.

If you've reset the database, you can reapply these manual changes in Grafana. You'll find a Click here to apply patched transactions link in the dashboard. This will replay all patches in the patcher/patches/ directory.

License

MIT