Files
TJWaterServerBinary/.github/copilot-instructions.md
2026-02-05 10:47:54 +08:00

201 lines
6.9 KiB
Markdown

# TJWater Server - Copilot Instructions
This is a FastAPI-based water network management system (供水管网智能管理系统) that provides hydraulic simulation, SCADA data integration, network element management, and risk analysis capabilities.
## Running the Server
```bash
# Install dependencies
pip install -r requirements.txt
# Start the server (default: http://0.0.0.0:8000 with 2 workers)
python scripts/run_server.py
# Note: On Windows, the script automatically sets WindowsSelectorEventLoopPolicy
```
## Running Tests
```bash
# Run all tests
pytest
# Run a specific test file with verbose output
pytest tests/unit/test_pipeline_health_analyzer.py -v
# Run from conftest helper
python tests/conftest.py
```
## Architecture Overview
### Core Components
1. **Native Modules** (`app/native/`): Platform-specific compiled extensions (`.so` for Linux, `.pyd` for Windows) providing performance-critical functionality including:
- SCADA device integration
- Water distribution analysis (WDA)
- Pipe risk probability calculations
- Wrapped through `app.services.tjnetwork` interface
2. **Services Layer** (`app/services/`):
- `tjnetwork.py`: Main network API wrapper around native modules
- `simulation.py`: Hydraulic simulation orchestration (EPANET integration)
- `project_info.py`: Project configuration management
- `epanet/`: EPANET hydraulic engine integration
3. **API Layer** (`app/api/v1/`):
- **Network Elements**: Separate endpoint modules for junctions, reservoirs, tanks, pipes, pumps, valves
- **Components**: Curves, patterns, controls, options, quality, visuals
- **Network Features**: Tags, demands, geometry, regions/DMAs
- **Core Services**: Auth, project, simulation, SCADA, data query, snapshots
4. **Database Infrastructure** (`app/infra/db/`):
- **PostgreSQL**: Primary relational database (users, audit logs, project metadata)
- **TimescaleDB**: Time-series extension for historical data
- **InfluxDB**: Optional time-series database for high-frequency SCADA data
- Connection pools initialized in `main.py` lifespan context
- Database instance stored in `app.state.db` for dependency injection
5. **Domain Layer** (`app/domain/`):
- `models/`: Enums and domain objects (e.g., `UserRole`)
- `schemas/`: Pydantic models for request/response validation
6. **Algorithms** (`app/algorithms/`):
- `api_ex/`: Analysis algorithms (k-means sensor placement, sensitivity analysis, pipeline health)
- `data_cleaning.py`: Data preprocessing utilities
- `simulations.py`: Simulation helpers
### Security & Authentication
- **Authentication**: JWT-based with access tokens (30 min) and refresh tokens (7 days)
- **Authorization**: Role-based access control (RBAC) with 4 roles:
- `VIEWER`: Read-only access
- `USER`: Read-write access
- `OPERATOR`: Modify data
- `ADMIN`: Full permissions
- **Audit Logging**: `AuditMiddleware` automatically logs POST/PUT/DELETE requests
- **Encryption**: Fernet symmetric encryption for sensitive data (`app.core.encryption`)
Default admin accounts:
- `admin` / `admin123`
- `tjwater` / `tjwater@123`
### Key Files
- `app/main.py`: FastAPI app initialization, lifespan (DB pools), CORS, middleware, router mounting
- `app/api/v1/router.py`: Central router aggregating all endpoint modules
- `app/core/config.py`: Settings management using `pydantic-settings`
- `app/auth/dependencies.py`: Auth dependencies (`get_current_active_user`, `get_db`)
- `app/auth/permissions.py`: Permission decorators (`require_role`, `get_current_admin`)
- `configs/project_info.yml`: Default project configuration (auto-loaded on startup)
- `.env`: Environment configuration (database credentials, JWT secret, encryption key)
## Important Conventions
### Database Connections
- Database instances are initialized in `main.py` lifespan and stored in `app.state.db`
- Access via dependency injection:
```python
from app.auth.dependencies import get_db
async def endpoint(db = Depends(get_db)):
# Use db connection
```
### Authentication in Endpoints
Use dependency injection for auth requirements:
```python
from app.auth.dependencies import get_current_active_user
from app.auth.permissions import require_role, get_current_admin
from app.domain.models.role import UserRole
# Require any authenticated user
@router.get("/data")
async def get_data(current_user = Depends(get_current_active_user)):
return data
# Require specific role (USER or higher)
@router.post("/data")
async def create_data(current_user = Depends(require_role(UserRole.USER))):
return result
# Admin-only access
@router.delete("/data/{id}")
async def delete_data(id: int, current_user = Depends(get_current_admin)):
return result
```
### API Routing Structure
- All v1 APIs are mounted under `/api/v1` prefix via `api_router`
- Legacy routes without version prefix are also mounted for backward compatibility
- Group related endpoints in separate router modules under `app/api/v1/endpoints/`
- Use descriptive tags in `router.py` for OpenAPI documentation grouping
### Native Module Integration
- Native modules are pre-compiled for specific platforms
- Always import through `app.native.api` or `app.services.tjnetwork`
- The `tjnetwork` service wraps native APIs with constants like:
- Element types: `JUNCTION`, `RESERVOIR`, `TANK`, `PIPE`, `PUMP`, `VALVE`
- Operations: `API_ADD`, `API_UPDATE`, `API_DELETE`
- `ChangeSet` for batch operations
### Project Initialization
- On startup, `main.py` automatically loads project from `project_info.name` if set
- Projects are opened via `open_project(name)` from `tjnetwork` service
- Initial project config comes from `configs/project_info.yml`
### Audit Logging
Manual audit logging for critical operations:
```python
from app.core.audit import log_audit_event, AuditAction
await log_audit_event(
action=AuditAction.UPDATE,
user_id=current_user.id,
username=current_user.username,
resource_type="resource_name",
resource_id=str(resource_id),
ip_address=request.client.host,
request_data=data
)
```
### Environment Configuration
- Copy `.env.example` to `.env` before first run
- Required environment variables:
- `SECRET_KEY`: JWT signing (generate with `openssl rand -hex 32`)
- `ENCRYPTION_KEY`: Data encryption (generate with Fernet)
- Database credentials for PostgreSQL, TimescaleDB, and optionally InfluxDB
### Database Migrations
SQL migration scripts are in `migrations/`:
- `001_create_users_table.sql`: User authentication tables
- `002_create_audit_logs_table.sql`: Audit logging tables
Apply with:
```bash
psql -U postgres -d tjwater -f migrations/001_create_users_table.sql
```
## API Documentation
- Swagger UI: http://localhost:8000/docs
- ReDoc: http://localhost:8000/redoc
- OpenAPI schema: http://localhost:8000/openapi.json
## Additional Resources
- `SECURITY_README.md`: Comprehensive security feature documentation
- `DEPLOYMENT.md`: Integration guide for security features
- `readme.md`: Project overview and directory structure (in Chinese)