Getting Started with MeatPy

This guide will help you get started with MeatPy for processing financial market data.

Basic Concepts

Core Components

• MarketProcessor: Processes market messages and maintains order book state
• MessageReader: Reads market data from files in various formats
• LimitOrderBook (LOB): Represents the current state of buy and sell orders
• Event Handlers: Process and record market events as they occur

Supported Data Formats

MeatPy supports multiple versions of the NASDAQ ITCH protocol:

Version	Format	Stock Symbol Size	Timestamp	Module
ITCH 5.0	Binary	8 characters	Nanoseconds	meatpy.itch50
ITCH 4.1	Binary	8 characters	Nanoseconds	meatpy.itch41
ITCH 4.0	Binary	6 characters	Nanoseconds	meatpy.itch4
ITCH 3.0	ASCII	6 characters	Milliseconds	meatpy.itch3
ITCH 2.0	ASCII	6 characters	Milliseconds	meatpy.itch2

Each version has its own MessageReader, MarketProcessor, and message classes following the same patterns.

MeatPy also supports:

• IEX DEEP 1.0: IEX Exchange's market data format (price-level data) via meatpy.iex_deep

Basic Usage

Reading ITCH 5.0 Data (Binary)

The simplest way to read ITCH 5.0 data:

from meatpy.itch50 import ITCH50MessageReader

# Read messages from an ITCH file
with ITCH50MessageReader("market_data.txt.gz") as reader:
    for i, message in enumerate(reader):
        print(f"Message {i}: {message}")
        if i >= 10:  # Just show first 10 messages
            break

Reading Other ITCH Versions

Each ITCH version follows the same interface pattern:

# ITCH 4.1 (Binary, 8-char symbols)
from meatpy.itch41 import ITCH41MessageReader, ITCH41MarketProcessor

# ITCH 4.0 (Binary, 6-char symbols)
from meatpy.itch4 import ITCH4MessageReader, ITCH4MarketProcessor

# ITCH 3.0 (ASCII, separate timestamp messages)
from meatpy.itch3 import ITCH3MessageReader, ITCH3MarketProcessor

# ITCH 2.0 (ASCII, embedded timestamps)
from meatpy.itch2 import ITCH2MessageReader, ITCH2MarketProcessor

Building a Limit Order Book

import datetime
from meatpy.itch50 import ITCH50MessageReader, ITCH50MarketProcessor

book_date = datetime.datetime(2021, 8, 13)
processor = ITCH50MarketProcessor("AAPL", book_date)

with ITCH50MessageReader("market_data.txt.gz") as reader:
    for message in reader:
        processor.process_message(message)

# Access the limit order book
if processor.lob:
    print(f"Best bid: ${processor.lob.best_bid / 10000:.2f}")
    print(f"Best ask: ${processor.lob.best_ask / 10000:.2f}")

Reading IEX DEEP Data

IEX DEEP data comes in PCAP/PCAP-NG format. Here's how to read it:

from meatpy.iex_deep import IEXDEEPMessageReader, IEXDEEPMarketProcessor

# Read messages from a PCAP file
reader = IEXDEEPMessageReader()
for i, message in enumerate(reader.read_file("data_feeds_20180529_DEEP1.0.pcap.gz")):
    print(f"Message {i}: {type(message).__name__}")
    if i >= 10:
        break

# Process messages and reconstruct order book for a specific symbol
processor = IEXDEEPMarketProcessor("SPY")
for message in reader.read_file("data_feeds_20180529_DEEP1.0.pcap.gz"):
    processor.process_message(message)

# Get best bid and offer
bbo = processor.get_bbo()
if bbo[0] and bbo[1]:
    print(f"Best Bid: ${bbo[0][0]/10000:.2f} x {bbo[0][1]}")
    print(f"Best Ask: ${bbo[1][0]/10000:.2f} x {bbo[1][1]}")

For more details on IEX DEEP, see the IEX DEEP Guide.

Other Common Tasks

• Listing Symbols: Extracting unique stock symbols from ITCH files
• Extracting Specific Symbols: Creating new ITCH files with only specific symbols
• Top of Book Snapshots: Generating snapshots of the top of book state for analysis
• Order Book Snapshots: Creating snapshots of the full limit order book state