ftdt-quant-lab/cli.py

"""
FTDT Quant Lab — unified CLI.

Subcommands:
  collect   — Run data collector (streams to Parquet)
  analyze   — Run analytics on stored data (Phase 2)
  simulate  — Run market-making simulator on stored data (Phase 3)
  run       — Start production trading node (Phase 4)
  backtest  — Run VectorBT backtest (existing)

Usage:
  python -m cli collect --coins BTC,ETH --data-dir data/raw
  python -m cli analyze --data-dir data/raw --start 2026-08-01 --end 2026-08-07
  python -m cli simulate --data-dir data/raw --coin BTC --hours 24
  python -m cli run --coins BTC,ETH --mode paper
  python -m cli backtest --strategy pairs --interval 1h
"""

from __future__ import annotations

import asyncio
import logging
import os
import sys
from pathlib import Path

sys.path.insert(0, str(Path(__file__).resolve().parent))


def cmd_collect(args):
    """Run the Hyperliquid data collector."""
    from data.collectors.hyperliquid import HyperliquidCollector
    from data.store import RawMessageStore

    store = RawMessageStore(
        data_dir=args.data_dir,
        flush_interval_sec=args.flush_interval,
    )

    collector = HyperliquidCollector(
        store=store,
        coins=args.coins,
        testnet=not args.mainnet,
        poll_interval_sec=args.poll_interval,
    )

    asyncio.run(collector.run())


def cmd_analyze(args):
    """Run microstructure analytics on stored data."""
    from data.store import read_range

    print(f"Reading {args.channel}/{args.coin} from {args.start_date} to {args.end_date}...")
    messages = read_range(
        args.data_dir,
        channel=args.channel,
        coin=args.coin.upper(),
        start_date=args.start_date,
        end_date=args.end_date,
    )
    print(f"Loaded {len(messages)} messages")

    if args.channel == "l2book":
        from microstructure.book import batch_book_stats
        snapshots = []
        for msg in messages:
            payload = msg["payload"]
            levels = payload.get("levels", [])
            if levels and isinstance(levels, list) and len(levels) >= 2:
                bids = {}
                asks = {}
                for bid in levels[0]:
                    if float(bid.get("sz", 0)) > 0:
                        bids[float(bid["px"])] = float(bid["sz"])
                for ask in levels[1]:
                    if float(ask.get("sz", 0)) > 0:
                        asks[float(ask["px"])] = float(ask["sz"])
                snapshots.append({"bids": bids, "asks": asks})

        stats = batch_book_stats(snapshots)
        print(json.dumps(stats, indent=2, default=str))

    elif args.channel == "trades":
        from microstructure.trades import classify_bulk_lee_ready, trade_arrival_rate, trade_volume_profile

        trades = [msg["payload"] for msg in messages]
        mids = [float(msg["payload"].get("px", 0)) for msg in messages]
        times = [msg["exchange_ts"] for msg in messages]
        sides = classify_bulk_lee_ready(trades, mids)
        buys = sum(1 for s in sides if s == "buy")
        sells = sum(1 for s in sides if s == "sell")
        arrival = trade_arrival_rate(times)
        vol = trade_volume_profile(trades)

        print(f"Trades: {len(trades)} total ({buys} buy, {sells} sell)")
        print(f"Arrival rate: {json.dumps(arrival, indent=2, default=str)}")
        print(f"Volume profile: {json.dumps(vol, indent=2, default=str)}")

    elif args.channel == "funding":
        from microstructure.funding import funding_regime, basis_spread

        rates = [float(msg["payload"].get("funding", 0)) for msg in messages]
        marks = [float(msg["payload"].get("mark_px", 0)) for msg in messages]
        regime = funding_regime(rates, window_hours=24, n_samples_per_hour=1)
        print(f"Funding regime: {json.dumps(regime, indent=2, default=str)}")

    else:
        print(f"Channel '{args.channel}' — raw dump:")
        for msg in messages[:5]:
            print(json.dumps(msg, indent=2, default=str))
        if len(messages) > 5:
            print(f"... and {len(messages) - 5} more")


def cmd_simulate(args):
    """Run market-making simulator on stored data with L2 events."""
    from data.store import read_range
    from sim.engine import SimulationEngine, SimConfig
    from sim.maker import MakerConfig

    print(f"Loading L2 book data for {args.coin} from {args.start_date} to {args.end_date}...")
    l2_messages = read_range(
        args.data_dir,
        channel="l2book",
        coin=args.coin.upper(),
        start_date=args.start_date,
        end_date=args.end_date,
    )
    print(f"Loaded {len(l2_messages)} L2 updates")

    trade_messages = read_range(
        args.data_dir,
        channel="trades",
        coin=args.coin.upper(),
        start_date=args.start_date,
        end_date=args.end_date,
    )
    print(f"Loaded {len(trade_messages)} trades")

    events = []
    for msg in l2_messages:
        payload = msg["payload"]
        levels = payload.get("levels", [])
        bids = {}
        asks = {}
        if levels and isinstance(levels, list) and len(levels) >= 2:
            for bid in levels[0]:
                if float(bid.get("sz", 0)) > 0:
                    bids[float(bid["px"])] = float(bid["sz"])
            for ask in levels[1]:
                if float(ask.get("sz", 0)) > 0:
                    asks[float(ask["px"])] = float(ask["sz"])
        events.append({
            "type": "l2",
            "data": {"bids": bids, "asks": asks},
            "time": msg["local_ts"],
            "coin": args.coin.upper(),
        })

    for msg in trade_messages:
        payload = msg["payload"]
        events.append({
            "type": "trade",
            "data": payload,
            "time": msg["local_ts"],
            "coin": args.coin.upper(),
        })

    events.sort(key=lambda e: e["time"])
    print(f"Total events: {len(events)}")

    config = SimConfig(
        maker=MakerConfig(
            base_size=args.base_size,
            max_inventory=args.max_inventory,
            gamma=args.gamma,
        ),
        max_inventory=args.max_inventory,
        cancel_after_ms=args.cancel_after_ms,
        quote_refresh_ms=args.quote_refresh_ms,
        seed=args.seed,
    )

    engine = SimulationEngine(config=config, seed=args.seed)
    engine.run(events)

    stats = engine.stats()
    breakdown = engine.breakdown()

    print("\n=== Simulation Results ===")
    print(f"Duration: {events[-1]['time'] - events[0]['time']:.0f}s" if events else "0s")
    print(f"Trades: {stats.total_trades} ({stats.bid_fills} bid, {stats.ask_fills} ask)")
    print(f"Toxic fills: {stats.toxic_fills} ({stats.adverse_rate:.1%})")
    print(f"Cancels: {stats.cancels}")
    print(f"Avg spread: {stats.avg_spread_bps} bps")
    print(f"Max inventory: {stats.max_inventory}")
    print(f"Max drawdown: {stats.max_drawdown}%")
    print(f"Sharpe: {stats.sharpe}  Sortino: {stats.sortino}")
    print(f"Uptime: {stats.uptime_pct}%")
    print(f"\nPnL Breakdown:")
    print(f"  Spread capture:  ${breakdown.spread_capture:.4f}")
    print(f"  Inventory PnL:    ${breakdown.inventory_pnl:.4f}")
    print(f"  Maker fees:       ${breakdown.maker_fees:.4f}")
    print(f"  Taker fees:       ${breakdown.taker_fees:.4f}")
    print(f"  Funding PnL:      ${breakdown.funding_pnl:.4f}")
    print(f"  Adverse selection: ${breakdown.adverse_selection_cost:.4f}")
    print(f"  ─────────────────────────────")
    print(f"  Gross PnL:        ${breakdown.gross_pnl:.4f}")
    print(f"  Net PnL:          ${breakdown.net_pnl:.4f}")


def cmd_run(args):
    """Start the production trading node."""
    import asyncio
    from live.node_v2 import ProductionNode

    coins = [c.strip().upper() for c in args.coins.split(",") if c.strip()]

    node = ProductionNode(
        coins=coins,
        testnet=not args.mainnet,
        mode=args.mode,
        max_position_per_coin=args.max_position,
        base_quote_size=args.base_size,
        initial_equity=args.equity,
        tick_interval_sec=args.tick_interval,
        metrics_file=args.metrics_file,
    )

    asyncio.run(node.run())


def cmd_backtest(args):
    """Run a VBT backtest (existing functionality)."""
    from backtests.vbt_runner import VBTBacktestRunner
    runner = VBTBacktestRunner()
    result = runner.run_strategy(strategy=args.strategy, interval=args.interval, limit=args.limit)
    import json as _json
    print(_json.dumps({k: v for k, v in (result or {}).items()
                       if k not in ("trades", "equity_curve")}, indent=2, default=str))
    if result and result.get("trades"):
        print(f"\n{len(result['trades'])} trades")


def main():
    import argparse
    p = argparse.ArgumentParser(description="FTDT Quant Lab CLI")
    sp = p.add_subparsers(dest="command", required=True)

    # collect
    pc = sp.add_parser("collect", help="Run data collector")
    pc.add_argument("--coins", nargs="+", default=["BTC", "ETH"])
    pc.add_argument("--mainnet", action="store_true")
    pc.add_argument("--data-dir", default="data/raw")
    pc.add_argument("--poll-interval", type=float, default=60.0)
    pc.add_argument("--flush-interval", type=float, default=5.0)

    # analyze
    pa = sp.add_parser("analyze", help="Run microstructure analytics")
    pa.add_argument("--data-dir", default="data/raw")
    pa.add_argument("--channel", default="l2book", choices=["l2book", "trades", "funding", "mark", "open_interest", "liquidation"])
    pa.add_argument("--coin", default="BTC")
    pa.add_argument("--start-date", default="2026-08-01")
    pa.add_argument("--end-date", default="2026-08-07")

    # simulate
    ps = sp.add_parser("simulate", help="Run market-making simulator")
    ps.add_argument("--data-dir", default="data/raw")
    ps.add_argument("--coin", default="BTC")
    ps.add_argument("--start-date", default="2026-08-01")
    ps.add_argument("--end-date", default="2026-08-07")
    ps.add_argument("--gamma", type=float, default=0.1)
    ps.add_argument("--base-size", type=float, default=0.001)
    ps.add_argument("--max-inventory", type=float, default=0.005)
    ps.add_argument("--cancel-after-ms", type=float, default=5000.0)
    ps.add_argument("--quote-refresh-ms", type=float, default=2000.0)
    ps.add_argument("--seed", type=int, default=42)

    # run
    pr = sp.add_parser("run", help="Start production node")
    pr.add_argument("--coins", default="BTC,ETH", help="Comma-separated coin list")
    pr.add_argument("--mainnet", action="store_true")
    pr.add_argument("--mode", default="paper", choices=["paper", "live"])
    pr.add_argument("--max-position", type=float, default=0.003)
    pr.add_argument("--base-size", type=float, default=0.0002)
    pr.add_argument("--equity", type=float, default=10000.0)
    pr.add_argument("--tick-interval", type=float, default=2.0)
    pr.add_argument("--metrics-file", default="/tmp/ftdt-metrics-v2.json")

    # backtest
    pb = sp.add_parser("backtest", help="Run VBT backtest")
    pb.add_argument("--strategy", default="pairs")
    pb.add_argument("--interval", default="1h")
    pb.add_argument("--limit", type=int, default=500)

    args = p.parse_args()

    import json as _json
    import json

    if args.command == "collect":
        cmd_collect(args)
    elif args.command == "analyze":
        cmd_analyze(args)
    elif args.command == "simulate":
        cmd_simulate(args)
    elif args.command == "run":
        cmd_run(args)
    elif args.command == "backtest":
        cmd_backtest(args)


if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO, format="%(asctime)s %(message)s", datefmt="%H:%M:%S")
    main()