dev-denish/platform

Purpose and Runtime

dMRV is a Phase-1 land-use, forestry, and carbon-project monitoring platform that ingests satellite/geospatial data, runs analyses, renders map layers, and produces reports. It runs as a FastAPI modular monolith plus a React/Vite frontend served by nginx. Container orchestration is Docker Compose: deploy/docker-compose.yml.

Entry Points and Delivery

The API application factory owns lifecycle, middleware, error handling, and router registration in backend/app/main.py; the production ASGI entry point is backend/app/asgi.py. The SPA starts at frontend/src/main.jsx and composes routes/application state through frontend/src/App.jsx. Docker images are defined by backend/Dockerfile and frontend/Dockerfile.

Versioned API and Internal Boundaries

HTTP endpoints are grouped under a v1 router, including auth, projects, datasets, layers, analyses, jobs, reports, tile delivery, and WMS/external layers: backend/app/api/v1/router.py, backend/app/api/v1/analyses.py, backend/app/api/v1/tiles.py. Dependency injection and request authorization form the API boundary in backend/app/api/deps.py. The intended flow is API → service orchestration → repository persistence, with typed contracts and policy concepts in backend/app/domain/dtos.py and backend/app/domain/authz.py.

Persistence, Access Control, and Auditing

PostGIS is the material datastore, initialized and versioned through Alembic migrations: backend/migrations/env.py, backend/migrations/versions/0001_initial.py. Repositories isolate parameterized SQL for core aggregates and geospatial records, e.g. backend/app/repositories/projects.py, backend/app/repositories/datasets.py, and backend/app/repositories/vector_layers.py. Authentication, token handling, memberships, permission grants, and project-scoped access are separated into backend/app/services/auth_service.py, backend/app/services/permission_service.py, and backend/app/services/project_access.py.

Geospatial Ingestion and Layer Serving

Dataset ingestion is a bounded raster/vector pipeline that detects and processes COGs, rasters, vectors, and storage concerns: backend/app/services/ingestion/service.py, backend/app/services/ingestion/raster.py, backend/app/services/ingestion/vector.py. Ingested and external/WMS data become map-facing layers through backend/app/services/tile_service.py, backend/app/services/tile_renderer.py, and backend/app/services/wms_service.py. This is the main boundary between geospatial source formats/services and browser-consumable tiles/layer metadata.

Analysis and Background Execution

Analysis configuration/catalog definitions drive service-level Google Earth Engine (GEE) execution: backend/app/domain/analysis_catalog.py, backend/app/services/gee_analysis_service.py, backend/app/services/gee_client.py. Jobs decouple long-running GEE analyses and report generation from request handling: backend/app/services/jobs_service.py, backend/app/workers/arq_worker.py, backend/app/workers/gee_analysis_jobs.py. Redis/ARQ is implied by the queue-worker abstraction; the README specifies a threadpool-now/queue-later evolution boundary.

Frontend Map and Project Workspaces

The authenticated SPA organizes project lifecycle, upload, forest-definition, user, and WMS administration pages: frontend/src/pages/ProjectDetailPage.jsx, frontend/src/pages/UploadPage.jsx, frontend/src/pages/ForestDefinitionPage.jsx. Interactive map composition, draw/measure tools, layer controls, and analysis UI reside in frontend/src/components/ProjectMap.jsx, frontend/src/components/LayersPanel.jsx, and frontend/src/components/AnalysisPanel.jsx. Auth state and route enforcement are client-side boundaries at frontend/src/context/AuthContext.jsx and frontend/src/routes/ProtectedRoute.jsx.

Reporting and External Dependencies

Reports combine calculated content, charts, map images, and PDF rendering in backend/app/services/report_service.py, backend/app/services/report_map_image.py, and backend/app/services/report_pdf.py. Key external integrations are Google Earth Engine for satellite analysis, CDSE ingestion (backend/app/services/cdse_ingestion.py), external layer retrieval, and WMS servers. Operational security, rate limiting, metrics, and structured logging are centralized under backend/app/core/security.py, backend/app/core/ratelimit.py, and backend/app/core/metrics.py.