Add professional FastAPI crypto data service with OHLCV + indicators

[Copilot]

Implements a production-ready FastAPI service for fetching cryptocurrency OHLCV data with technical indicators (EMA 20/50/200, RSI 14, ATR 14) from multiple exchanges via ccxt.

Core Implementation

API Endpoints

• REST: GET /api/v1/live_data - Fetch OHLCV with indicators, returns last 20 candles
• WebSocket: WS /api/v1/live_data/ws - Stream updates at configurable intervals (5-300s)
• Query params validated via Pydantic: symbol, timeframe, limit (20-2000), exchange

Data Pipeline

• Exchange client wraps ccxt with rate limiting and market type configuration
• Indicators calculated using ta library on pandas DataFrame
• NaN/infinite values converted to None for JSON serialization
• Metadata includes exchange defaultType when available

Error Handling

• Network errors → 503
• Exchange errors → 400
• Generic errors → 500
• WebSocket implements circuit breaker (max 5 consecutive errors)

Testing Strategy

Mocked tests use pytest monkeypatch to stub exchange.fetch_ohlcv, ensuring deterministic CI:

class MockExchange:
    def __init__(self, config: dict[str, bool] | None = None) -> None:
        self.options = {"defaultType": "spot"}
    
    def fetch_ohlcv(self, symbol: str, timeframe: str = "1h", limit: int = 250) -> list[list[float]]:
        return SAMPLE_OHLCV_DATA[:limit]

def test_live_data_endpoint_mocked(monkeypatch: pytest.MonkeyPatch) -> None:
    monkeypatch.setattr("ccxt.binance", lambda config: MockExchange(config))
    response = client.get("/api/v1/live_data")
    assert response.status_code == 200

Network smoke tests retained but not required for CI to pass.

Infrastructure

• Docker: Multi-stage build with editable install
• CI: GitHub Actions runs ruff, mypy (strict mode), pytest
• Type safety: Full coverage with mypy, pandas-stubs included
• Config: pydantic-settings with .env support

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 -m uvicorn app.main:app --host 0.0.0.0 --port 8000 (dns block)
  • Triggering command: /home/REDACTED/.local/bin/pytest pytest -v /home/REDACTED/work/crypto-trade-analysis/crypto-trade-analysis git conf�� --global credential.username /usr/sbin/git (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 new branch add/fastapi-live-data and open a pull request against main that adds a complete FastAPI service for fetching OHLCV candles and computing indicators (EMA/RSI/ATR) with Docker, CI, typing, and tests.

Files and structure to add (create under repo root):

Project files

• pyproject.toml (dependencies: fastapi, uvicorn[standard], pydantic, pydantic-settings, ccxt, pandas, ta; dev: pytest, httpx, ruff, mypy, types-requests)
• README.md (quickstart, Docker, notes)
• .env.example
• .gitignore
• Dockerfile
• docker-compose.yml
• .github/workflows/ci.yml

Application package (src/app)

• src/app/init.py
• src/app/main.py (FastAPI app, health endpoint, include router)

Core

• src/app/core/init.py
• src/app/core/config.py (pydantic-settings Settings; load .env)
• src/app/core/logging.py (basic logging config)

API routes

• src/app/api/init.py
• src/app/api/v1/init.py
• src/app/api/v1/routes_live_data.py (GET /api/v1/live_data; Query validation; response_model)
• src/app/api/v1/ws_live_data.py (WebSocket endpoint /api/v1/live_data/ws that streams recent candle payloads every N seconds; interval query param)

Integrations

• src/app/integrations/init.py
• src/app/integrations/exchanges.py (ExchangeClient wrapper; build_exchange to instantiate ccxt exchange with enableRateLimit and options.defaultType)

Schemas

• src/app/schemas/init.py
• src/app/schemas/live_data.py (Pydantic models: Candle, LatestIndicators, LiveDataResponse)

Services

• src/app/services/init.py
• src/app/services/indicators.py (add_indicators using ta: EMA20/50/200, RSI14, ATR14; helper to convert NaN->None)
• src/app/services/market_data.py (get_live_data: fetch OHLCV via ccxt.fetch_ohlcv, build DataFrame, add indicators, prepare recent_candles (last 20), latest indicators, meta)

Tests

• tests/init.py
• tests/test_health.py (health endpoint)
• tests/test_live_data.py (network smoke test)
• tests/test_live_data_mocked.py (mocked ccxt tests using pytest + monkeypatch to stub exchange.fetch_ohlcv)

Behavior and implementation details

• Default exchange: "binance" (configurable via .env)
• Indicators: EMA 20/50/200, RSI 14, ATR 14 using ta and pandas
• Convert NaN/infinite to None before JSON serialization
• Validate query params: symbol (3-30 chars), timeframe (1-10), limit (20-2000), exchange (2-40)
• Error mapping: ccxt.NetworkError -> HTTP 503, ccxt.ExchangeError -> HTTP 400, other exceptions -> HTTP 500
• recent_candles: return last 20 candles
• meta: include exchange.options.get("defaultType") when available
• Logging via core.logging
• Dockerfile installs package in editable mode with dev deps and runs uvicorn
• GitHub Actions CI: checkout, setup python 3.11, install dev deps, run ruff, mypy, pytest

Testing/CI

• Include mocked ccxt tests so CI is deterministic; keep the network smoke test but it won't be required for CI

PR details

• Branch: add/fastapi-live-data
• Base: main
• PR title: "Add professional FastAPI crypto data service with OHLCV + indicators"
• Include this problem statement text in the PR description
• Preserve multiple logical commits (do not squash)
• Do not add assignees, reviewers, or labels by default

Repository: Nfectious/crypto-trade-analysis

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.