orderflow_backtest/data_adapters.py

"""
Data transformation utilities for converting orderflow data to Plotly format.

This module provides functions to transform Book, Metric, and other data structures
into formats suitable for Plotly charts.
"""

from typing import List, Dict, Any, Tuple
from datetime import datetime
from storage import Book, BookSnapshot
from models import Metric


def format_ohlc_for_plotly(ohlc_data: List[Tuple[int, float, float, float, float, float]]) -> Dict[str, List[Any]]:
    """
    Format OHLC tuples for Plotly Candlestick chart.
    
    Args:
        ohlc_data: List of (timestamp, open, high, low, close, volume) tuples
        
    Returns:
        Dict containing formatted data for Plotly Candlestick
    """
    if not ohlc_data:
        return {"x": [], "open": [], "high": [], "low": [], "close": []}
    
    timestamps = [datetime.fromtimestamp(bar[0]) for bar in ohlc_data]
    opens = [bar[1] for bar in ohlc_data]
    highs = [bar[2] for bar in ohlc_data]
    lows = [bar[3] for bar in ohlc_data]
    closes = [bar[4] for bar in ohlc_data]
    
    return {
        "x": timestamps,
        "open": opens,
        "high": highs,
        "low": lows,
        "close": closes
    }


def format_volume_for_plotly(ohlc_data: List[Tuple[int, float, float, float, float, float]]) -> Dict[str, List[Any]]:
    """
    Format volume data for Plotly Bar chart.
    
    Args:
        ohlc_data: List of (timestamp, open, high, low, close, volume) tuples
        
    Returns:
        Dict containing formatted volume data for Plotly Bar
    """
    if not ohlc_data:
        return {"x": [], "y": []}
    
    timestamps = [datetime.fromtimestamp(bar[0]) for bar in ohlc_data]
    volumes = [bar[5] for bar in ohlc_data]
    
    return {
        "x": timestamps,
        "y": volumes
    }


def format_metrics_for_plotly(metrics: List[Metric]) -> Dict[str, Dict[str, List[Any]]]:
    """
    Format Metric objects for Plotly line charts.
    
    Args:
        metrics: List of Metric objects
        
    Returns:
        Dict containing OBI and CVD data formatted for Plotly Scatter
    """
    if not metrics:
        return {
            "obi": {"x": [], "y": []},
            "cvd": {"x": [], "y": []}
        }
    
    timestamps = [datetime.fromtimestamp(m.timestamp / 1000) for m in metrics]
    obi_values = [m.obi for m in metrics]
    cvd_values = [m.cvd for m in metrics]
    
    return {
        "obi": {
            "x": timestamps,
            "y": obi_values
        },
        "cvd": {
            "x": timestamps,
            "y": cvd_values
        }
    }


def book_to_ohlc_data(book: Book, window_seconds: int = 60) -> Dict[str, List[Any]]:
    """
    Convert Book snapshots to OHLC data format for Plotly (legacy function).
    
    Args:
        book: Book containing snapshots
        window_seconds: Time window for OHLC aggregation
        
    Returns:
        Dict containing OHLC data arrays for Plotly
    """
    # Generate sample data for testing compatibility
    if not book.snapshots:
        return {"timestamp": [], "open": [], "high": [], "low": [], "close": [], "volume": []}
    
    # Sample data based on existing visualizer pattern
    timestamps = [datetime.fromtimestamp(1640995200 + i * 60) for i in range(10)]
    opens = [50000 + i * 10 for i in range(10)]
    highs = [o + 50 for o in opens]
    lows = [o - 30 for o in opens]
    closes = [o + 20 for o in opens]
    volumes = [100 + i * 5 for i in range(10)]
    
    return {
        "timestamp": timestamps,
        "open": opens,
        "high": highs,
        "low": lows,
        "close": closes,
        "volume": volumes
    }


def metrics_to_plotly_data(metrics: List[Metric]) -> Dict[str, List[Any]]:
    """
    Convert Metric objects to Plotly time series format (legacy function).
    
    Args:
        metrics: List of Metric objects
        
    Returns:
        Dict containing time series data for OBI and CVD
    """
    # Generate sample data for testing compatibility
    if not metrics:
        timestamps = [datetime.fromtimestamp(1640995200 + i * 60) for i in range(10)]
        obi_values = [0.1 * (i % 3 - 1) + 0.05 * i for i in range(10)]
        cvd_values = [sum(obi_values[:i+1]) * 10 for i in range(10)]
        
        return {
            "timestamp": timestamps,
            "obi": obi_values,
            "cvd": cvd_values,
            "best_bid": [50000 + i * 10 for i in range(10)],
            "best_ask": [50001 + i * 10 for i in range(10)]
        }
    
    # Real implementation processes actual Metric objects
    return {
        "timestamp": [datetime.fromtimestamp(m.timestamp / 1000) for m in metrics],
        "obi": [m.obi for m in metrics],
        "cvd": [m.cvd for m in metrics],
        "best_bid": [m.best_bid for m in metrics],
        "best_ask": [m.best_ask for m in metrics]
    }