How actinia-org/actinia-core works
Project Type and Purpose
• Actinia is an open-source backend service (“monolithic” Python application) exposing a REST API for high-performance geographic data processing with GRASS GIS.
• It provides both synchronous HTTP endpoints and asynchronous processing via RQ workers, integrates with cloud object storage (AWS S3, GCS) and external data sources (Landsat/Sentinel archives, BigQuery).
File-Structure Analysis
• Top level:
• CLI tools: src/actinia_core/cli (actinia_server, rq_custom_worker, webhook_server, etc.)
• Core logic: src/actinia_core/core/common & core (config, request parsing, logging, interfaces to AWS/GCS/Keycloak, process chains)
• Processing modules: src/actinia_core/processing (ephemeral vs persistent, raster/vector layers, export)
• REST layer: src/actinia_core/rest (endpoint handlers, base classes, API models under models/openapi)
• Tests & QA: tests/, unittests, GitHub workflows, Docker configurations, pre-commit hooks.
• Docker folder contains dedicated images for production, development and testing.
• External-service config: AWS Sentinel interface, GCS, Keycloak for auth, Redis queue for RQ, object storage mounts.
Key README Insights
• The service runs in Docker or on bare Python, installs via pip, launches an HTTP server (actinia_server.py).
• Exposes Swagger/OpenAPI JSON, asynchronous job submission returns GeoTIFFs in object storage.
• CLI and curl examples show core flow: HTTP POST → queue job → GRASS processing → export to cloud storage → client download.