backend/mcp_router.py

"""MCP Router for Portfolio Intelligence Platform.

Compatibility layer that routes to unified mcp_tools module.
This file maintains backwards compatibility with existing code that uses
mcp_router.call_*_mcp() methods.

All actual implementation is now in backend.mcp_tools module.
"""

import json
import logging
from typing import Any, Dict, List

from backend import mcp_tools
from backend.utils.serialisation import dumps_str

logger = logging.getLogger(__name__)


class MCPRouter:
    """Router for orchestrating MCP tool calls.

    This is a compatibility layer - all actual implementation
    is in backend.mcp_tools module with namespaced functions.
    """

    def __init__(self):
        """Initialise MCP router (compatibility layer)."""
        logger.info("Initialising MCP router (compatibility layer)")

    # Yahoo Finance MCP methods
    async def call_yahoo_finance_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Yahoo Finance MCP tool (delegates to market_* functions).

        Args:
            tool: Tool name (get_quote, get_historical_data, get_fundamentals)
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Yahoo Finance MCP call: {tool}")

        if tool == "get_quote":
            tickers = params.get("tickers", [])
            return await mcp_tools.market_get_quote(dumps_str(tickers))

        elif tool == "get_historical_data":
            return await mcp_tools.market_get_historical_data(
                ticker=params["ticker"],
                period=params.get("period", "1y"),
                interval=params.get("interval", "1d"),
            )

        elif tool == "get_fundamentals":
            return await mcp_tools.market_get_fundamentals(ticker=params["ticker"])

        else:
            raise ValueError(f"Unknown Yahoo Finance tool: {tool}")

    # FMP MCP methods
    async def call_fmp_mcp(self, tool: str, params: Dict[str, Any]) -> Dict[str, Any]:
        """Call Financial Modeling Prep MCP tool (delegates to market_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing FMP MCP call: {tool}")

        if tool == "get_company_profile":
            return await mcp_tools.market_get_company_profile(ticker=params["ticker"])

        elif tool == "get_income_statement":
            return await mcp_tools.market_get_income_statement(
                ticker=params["ticker"],
                period=params.get("period", "annual"),
                limit=str(params.get("limit", 5)),
            )

        elif tool == "get_balance_sheet":
            return await mcp_tools.market_get_balance_sheet(
                ticker=params["ticker"],
                period=params.get("period", "annual"),
                limit=str(params.get("limit", 5)),
            )

        elif tool == "get_cash_flow_statement":
            return await mcp_tools.market_get_cash_flow_statement(
                ticker=params["ticker"],
                period=params.get("period", "annual"),
                limit=str(params.get("limit", 5)),
            )

        elif tool == "get_financial_ratios":
            return await mcp_tools.market_get_financial_ratios(
                ticker=params["ticker"],
                ttm=str(params.get("ttm", True)).lower(),
            )

        elif tool == "get_key_metrics":
            return await mcp_tools.market_get_key_metrics(
                ticker=params["ticker"],
                ttm=str(params.get("ttm", True)).lower(),
            )

        else:
            raise ValueError(f"Unknown FMP tool: {tool}")

    # Trading MCP methods
    async def call_trading_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Trading MCP tool (delegates to technical_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Trading MCP call: {tool}")

        if tool == "get_technical_indicators":
            return await mcp_tools.technical_get_indicators(
                ticker=params["ticker"],
                period=params.get("period", "3mo"),
            )

        else:
            raise ValueError(f"Unknown Trading MCP tool: {tool}")

    # FRED MCP methods
    async def call_fred_mcp(self, tool: str, params: Dict[str, Any]) -> Dict[str, Any]:
        """Call FRED MCP tool (delegates to market_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing FRED MCP call: {tool}")

        if tool == "get_economic_series":
            return await mcp_tools.market_get_economic_series(
                series_id=params["series_id"],
                observation_start=params.get("observation_start"),
                observation_end=params.get("observation_end"),
            )

        else:
            raise ValueError(f"Unknown FRED tool: {tool}")

    # Portfolio Optimizer MCP methods
    async def call_portfolio_optimizer_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Portfolio Optimizer MCP tool (delegates to portfolio_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Portfolio Optimizer MCP call: {tool}")

        market_data = params.get("market_data", [])
        market_data_json = dumps_str(market_data)
        risk_tolerance = params.get("risk_tolerance", "moderate")

        if tool == "optimize_hrp":
            return await mcp_tools.portfolio_optimize_hrp(
                market_data_json=market_data_json,
                risk_tolerance=risk_tolerance,
            )

        elif tool == "optimize_black_litterman":
            return await mcp_tools.portfolio_optimize_black_litterman(
                market_data_json=market_data_json,
                risk_tolerance=risk_tolerance,
            )

        elif tool == "optimize_mean_variance":
            return await mcp_tools.portfolio_optimize_mean_variance(
                market_data_json=market_data_json,
                risk_tolerance=risk_tolerance,
            )

        else:
            raise ValueError(f"Unknown Portfolio Optimizer tool: {tool}")

    # Risk Analyzer MCP methods
    async def call_risk_analyzer_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Risk Analyzer MCP tool (delegates to risk_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Risk Analyzer MCP call: {tool}")

        if tool == "analyze_risk":
            portfolio = params.get("portfolio", [])
            benchmark = params.get("benchmark")

            return await mcp_tools.risk_analyze(
                portfolio_json=dumps_str(portfolio),
                portfolio_value=str(params.get("portfolio_value", 100000)),
                confidence_level=str(params.get("confidence_level", 0.95)),
                time_horizon=str(params.get("time_horizon", 1)),
                method=params.get("method", "historical"),
                num_simulations=str(params.get("num_simulations", 10000)),
                benchmark_json=dumps_str(benchmark) if benchmark else None,
            )

        elif tool == "forecast_volatility_garch":
            returns = params.get("returns", [])

            return await mcp_tools.risk_forecast_volatility_garch(
                ticker=params["ticker"],
                returns_json=dumps_str([float(r) for r in returns]),
                forecast_horizon=str(params.get("forecast_horizon", 30)),
                garch_p=str(params.get("garch_p", 1)),
                garch_q=str(params.get("garch_q", 1)),
            )

        else:
            raise ValueError(f"Unknown Risk Analyzer tool: {tool}")

    # Ensemble Predictor MCP methods
    async def call_ensemble_predictor_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Ensemble Predictor MCP tool (delegates to ml_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Ensemble Predictor MCP call: {tool}")

        if tool == "forecast_ensemble":
            prices = params.get("prices", [])
            dates = params.get("dates")

            return await mcp_tools.ml_forecast_ensemble(
                ticker=params["ticker"],
                prices_json=dumps_str([float(p) for p in prices]),
                dates_json=dumps_str(dates) if dates else None,
                forecast_horizon=str(params.get("forecast_horizon", 30)),
                confidence_level=str(params.get("confidence_level", 0.95)),
                use_returns=str(params.get("use_returns", True)).lower(),
                ensemble_method=params.get("ensemble_method", "mean"),
            )

        else:
            raise ValueError(f"Unknown Ensemble Predictor tool: {tool}")

    # News Sentiment MCP methods
    async def call_news_sentiment_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call News Sentiment MCP tool (delegates to sentiment_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing News Sentiment MCP call: {tool}")

        if tool == "get_news_with_sentiment":
            return await mcp_tools.sentiment_get_news(
                ticker=params["ticker"],
                days_back=str(params.get("days_back", 7)),
            )

        else:
            raise ValueError(f"Unknown News Sentiment tool: {tool}")

    # Feature Extraction MCP methods
    async def call_feature_extraction_mcp(
        self, tool: str, params: Dict[str, Any]
    ) -> Dict[str, Any]:
        """Call Feature Extraction MCP tool (delegates to technical_* functions).

        Args:
            tool: Tool name
            params: Tool parameters

        Returns:
            Tool result
        """
        logger.debug(f"Routing Feature Extraction MCP call: {tool}")

        if tool == "extract_technical_features":
            return await mcp_tools.technical_extract_features(
                ticker=params["ticker"],
                prices=dumps_str(params.get("prices", [])),
                volumes=dumps_str(params.get("volumes", [])),
                include_momentum=str(params.get("include_momentum", True)).lower(),
                include_volatility=str(params.get("include_volatility", True)).lower(),
                include_trend=str(params.get("include_trend", True)).lower(),
            )

        elif tool == "normalise_features":
            return await mcp_tools.technical_normalise_features(
                ticker=params["ticker"],
                features=dumps_str(params.get("features", {})),
                historical_features=dumps_str(params.get("historical_features", [])),
                window_size=str(params.get("window_size", 100)),
                method=params.get("method", "ewm"),
            )

        elif tool == "select_features":
            return await mcp_tools.technical_select_features(
                ticker=params["ticker"],
                feature_vector=dumps_str(params.get("feature_vector", {})),
                max_features=str(params.get("max_features", 15)),
                variance_threshold=str(params.get("variance_threshold", 0.95)),
            )

        elif tool == "compute_feature_vector":
            return await mcp_tools.technical_compute_feature_vector(
                ticker=params["ticker"],
                technical_features=dumps_str(params.get("technical_features", {})),
                fundamental_features=dumps_str(params.get("fundamental_features", {})),
                sentiment_features=dumps_str(params.get("sentiment_features", {})),
                max_features=str(params.get("max_features", 30)),
                selection_method=params.get("selection_method", "pca"),
            )

        else:
            raise ValueError(f"Unknown Feature Extraction tool: {tool}")

    # High-level helper methods
    async def fetch_market_data(self, tickers: List[str]) -> Dict[str, Any]:
        """Fetch market data for given tickers.

        Args:
            tickers: List of stock/asset tickers

        Returns:
            Market data from Yahoo Finance
        """
        logger.info(f"Fetching market data for {len(tickers)} tickers")
        results = await mcp_tools.market_get_quote(dumps_str(tickers))
        return {r.get("ticker", r.get("symbol")): r for r in results}

    async def fetch_fundamentals(self, tickers: List[str]) -> Dict[str, Any]:
        """Fetch fundamental data for tickers.

        Args:
            tickers: List of stock tickers

        Returns:
            Fundamental data per ticker
        """
        logger.info(f"Fetching fundamentals for {len(tickers)} tickers")
        results = {}
        for ticker in tickers:
            results[ticker] = await mcp_tools.market_get_company_profile(ticker)
        return results

    async def fetch_technical_indicators(self, tickers: List[str]) -> Dict[str, Any]:
        """Fetch technical indicators for tickers.

        Args:
            tickers: List of stock tickers

        Returns:
            Technical indicators per ticker
        """
        logger.info(f"Fetching technical indicators for {len(tickers)} tickers")
        results = {}
        for ticker in tickers:
            results[ticker] = await mcp_tools.technical_get_indicators(ticker)
        return results

    async def fetch_macro_data(self) -> Dict[str, Any]:
        """Fetch macroeconomic data from FRED.

        Returns:
            Macroeconomic indicators
        """
        logger.info("Fetching macroeconomic data")
        results = {}
        for series_id in ["GDP", "UNRATE", "DFF"]:
            results[series_id] = await mcp_tools.market_get_economic_series(series_id)
        return results


# Global MCP router instance
mcp_router = MCPRouter()