Add WebSocket streaming endpoint and deterministic mocked tests

[Copilot]

Adds WebSocket support for real-time market data streaming and mocked tests for CI reliability. The existing REST API and indicator calculations remain unchanged.

Changes

WebSocket Endpoint (src/app/api/v1/ws_live_data.py)

• WS /api/v1/live_data/ws streams OHLCV + indicators at configurable intervals (1-60s)
• Same query parameters as REST endpoint (symbol, timeframe, limit, exchange)
• Error handling: retries on NetworkError/ExchangeError, closes on ValueError

# Client usage
ws = new WebSocket("ws://localhost:8000/api/v1/live_data/ws?symbol=BTC/USDT&interval=5")
ws.onmessage = (event) => {
  const data = JSON.parse(event.data);
  // Same structure as GET /api/v1/live_data response
}

Mocked Tests (tests/test_live_data_mocked.py)

• 6 tests using monkeypatch to stub ExchangeClient.fetch_ohlcv with synthetic data
• Eliminates network dependency for CI (deterministic, no rate limits, no API keys)
• Coverage: success cases, edge cases (insufficient data for indicators), error handling, validation

Type Safety

• Added pandas-stubs to fix mypy strict mode errors
• Type alias OHLCVData = list[list[int | float]] for candle structure

Documentation

• WebSocket API usage and error behavior documented in README
• Project structure updated to reflect new files

Testing

All 9 tests pass (3 existing + 6 new mocked). CI runs reliably without network calls.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• api.binance.com
  • Triggering command: /home/REDACTED/.local/bin/pytest pytest -v (dns block)
  • Triggering command: /usr/bin/python python /tmp/test_ws_testclient.py (dns block)
  • Triggering command: `/usr/bin/python3 python3 -c
    from fastapi.testclient import TestClient
    from app.main import app

client = TestClient(app)

print('Testing WebSocket endpoint...')
with client.websocket_connect('/api/v1/live_data/ws?interval=1&symbol=ETH/USDT') as websocket:
print('��� WebSocket c` (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

Create a complete FastAPI service for fetching OHLCV candles and computing technical indicators and open a PR against main in the repository Nfectious/crypto-trade-analysis.

Requirements (deliverables)

1. Project metadata and tooling (top-level files)

• pyproject.toml with dependencies: fastapi, uvicorn[standard], pydantic>=2.6, pydantic-settings, ccxt, pandas, ta; dev-deps: pytest, httpx, ruff, mypy, types-requests. Configure ruff, mypy (strict), and pytest testpaths.
• README.md with quickstart (local + Docker), API examples, note about indicator null handling.
• .env.example containing template config (APP_NAME, DEFAULT_EXCHANGE=binance, API_PREFIX=/api/v1, etc.).
• .gitignore (ignore .venv, pycache, .pytest_cache, .mypy_cache, .ruff_cache, dist, build, .env).
• Dockerfile (python:3.11-slim, set PYTHONDONTWRITEBYTECODE & PYTHONUNBUFFERED, install deps, install package in editable mode including dev deps, EXPOSE 8000, CMD uvicorn app.main:app --host 0.0.0.0 --port 8000).
• docker-compose.yml (service "api" mapping 8000:8000 and env_file .env).
• .github/workflows/ci.yml (on: [push,pull_request]; runs on python 3.11; steps: checkout, setup-python, pip install -e '.[dev]', run ruff, mypy, pytest).

2. Python package layout (src/app)

• src/app/init.py
• src/app/main.py: create FastAPI app, include router at settings.api_prefix, health endpoint GET / returning {'status':'ok','message':'service running'}, configure logging on startup.

3. Core

• src/app/core/config.py: pydantic-settings Settings class with defaults: app_name, api_prefix, default_exchange='binance', default_market_type='spot', enable_rate_limit=True. Load from .env.
• src/app/core/logging.py: basic logging setup (timestamp, level, name, message).

4. API routes

• src/app/api/v1/routes_live_data.py: GET {settings.api_prefix}/live_data with query params: symbol (str default 'BTC/USDT', min_length=3,max_length=30), timeframe (str default '1h', min_length=1,max_length=10), limit (int default 250, ge=20, le=2000), exchange (str default settings.default_exchange, min_length=2,max_length=40). Response model LiveDataResponse. Map ccxt.NetworkError->HTTP 503, ccxt.ExchangeError->HTTP 400, other exceptions->HTTP 500.
• src/app/api/v1/ws_live_data.py: WebSocket endpoint at {settings.api_prefix}/live_data/ws that streams recent candle payloads every N seconds (query param interval default 5). Gracefully handle disconnects.

5. Integrations

• src/app/integrations/exchanges.py: ExchangeClient wrapper around ccxt exchange instances, and build_exchange(name, settings) helper. Provide fetch_ohlcv method for testability.

6. Schemas

• src/app/schemas/live_data.py: Pydantic models
  • Candle: timestamp (datetime), open, high, low, close, volume (float), ema_20/ema_50/ema_200/rsi_14/atr_14 (Optional[float])
  • LatestIndicators: ema_20/ema_50/ema_200/rsi_14/atr_14 (Optional[float])
  • LiveDataResponse: symbol, timeframe, exchange, last_price (float), last_timestamp (datetime), candles_count (int), recent_candles: List[Candle], latest_indicators: LatestIndicators, meta: dict

7. Services

• src/app/services/indicators.py: functions to compute EMA(20/50/200), RSI(14), ATR(14) using pandas + ta. Ensure NaN/inf converted to None; round floats to 6 decimals.
• src/app/services/market_data.py: get_live_data(exchange_name, symbol, timeframe, limit) that uses ExchangeClient.fetch_ohlcv, builds a pandas DataFrame, converts ms timestamps to UTC datetimes, computes indicators, prepares recent_candles (last 20), computes latest_indicators, last_price, last_timestamp, meta (including exchange.options.get('defaultType')), and returns a LiveDataResponse.

8. Tests

• tests/test_health.py: test GET / returns 200 and contains 'running' or 'service running'.
• tests/test_live_data.py: optional network smoke test (marked or skipped in CI); verifies endpoint returns expected schema when using a live exchange.
• tests/test_live_data_mocked.py: deterministic tests using pytest and monkeypatch to stub ExchangeClient.fetch_ohlcv with synthetic OHLCV data so CI is reliable. Ensure pytest uses mocked tests in CI.

Behavioral details

• Default exchange: 'binance' unless overridden via .env or query param.
• Use ta for indicators; pandas for DataFrame manipulation.
• Return only last 20 candles in recent_candles.
• Convert NaN/infinite values to None prior to JSON serialization so nulls appear in response.
• Validate inputs via FastAPI Query validators; return clear HTTPException messages on validation errors.
• Error mapping: ccxt.NetworkError -> 503, ccxt.ExchangeError -> 400, other exceptions -> 500.
• Logging configured via core.logging.

Branch & PR details

• Branch name: add/fastapi-live-data (from main)
• Open a PR from add/fastapi-live-data -> main
• PR title: "Add professional FastAPI crypto data service with OHLCV + indicators"
• PR description: include this full problem statement and list of files/behavior (so reviewers can see scope)
• Preserve multiple logical comm...

This pull request was created from Copilot chat.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[Nfectious]

reviewed