Skip to content

README.md

Latest commit

 

History

History
62 lines (49 loc) · 2.15 KB

README.md

File metadata and controls

62 lines (49 loc) · 2.15 KB

Reticulum OpenAPI Mustache Templates

These templates can be used with openapi-generator-cli to produce a working Reticulum LXMF service and client from an OpenAPI specification. The generated structure mirrors the EmergencyManagement example.

Installation

Install the generator with the Python CLI or run it via Docker. Using the Python package:

pip install openapi-generator-cli

Or with Docker:

docker run --rm -v "$PWD:/local" openapitools/openapi-generator-cli generate

Usage

openapi-generator-cli generate \
    -g python \
    -i path/to/spec.yaml \
    -t templates \
    -o generated

The output will contain:

  • models.py – dataclasses for all schemas
  • controllers.py – controller classes with async handlers
  • service.py – LXMFService subclass registering routes
  • server.py – entrypoint starting the service
  • client.py – simple client invoking the first operation
  • database.py – example async database setup

Adjust the generated code as needed for your specification.

Post-generation adjustments

  • Set auth_token on the generated service and client if your deployment requires message authentication.
  • Extend the generated server.py with the same runtime CLI options used by the Emergency Management example (--config-path, --storage-path, --display-name, --auth-token, --database-path, --database-url, and --link-keepalive-interval). The scaffolded file only sleeps for 30 seconds; update it to wait on a signal-aware shutdown event so the service keeps running until interrupted, prints its identity hashes, and retries LXMF link establishment when necessary.
  • Call configure_database() with a caller-supplied override before init_db to honour environment variables such as EMERGENCY_DATABASE_URL or CLI flags.
  • Add additional schema dataclasses if your API defines objects outside the supplied OpenAPI spec.
  • Mirror the gateway helpers if you need FastAPI adapters: load shared LXMF client configuration from NORTH_API_CONFIG_PATH/NORTH_API_CONFIG_JSON, expose /notifications/stream, and surface interface or link status in startup logs.