shm-reader 0.1.0

A general-purpose SDK for reading and monitoring shared memory segments

SHM Reader

A general-purpose Python SDK for reading and monitoring shared memory segments.

Features

• List available shared memory segments
• Read data from shared memory segments
• Monitor shared memory segments for changes
• Register custom parsers for different data formats
• Command-line interface for quick access

Installation

pip install shm-reader

Usage

As a Library

from shm_reader import ShmReader
import struct

# Create a reader instance
reader = ShmReader()

# List available shared memory segments
segments = reader.list_segments()
for segment in segments:
    print(f"{segment.name}: {segment.size} bytes")

# Register a custom parser for a specific segment type
def market_data_parser(data):
    """Parse market data in format: symbol(12s), price(d), volume(q), timestamp(d)"""
    if len(data) < 28:  # 12 + 8 + 8 + 8 bytes
        raise ValueError("Not enough data for market format")
        
    symbol_bytes, price, volume, timestamp = struct.unpack("12sdqd", data[:28])
    symbol = symbol_bytes.decode('utf-8').strip('\x00')
    
    return {
        "symbol": symbol,
        "price": price,
        "volume": volume,
        "timestamp": timestamp
    }

# Register the parser with the SHM Reader
ShmReader.register_parser("market", market_data_parser)

# Read a specific segment
segment = reader.read_segment("market:btcusdt")
if segment.parsed_data:
    print(f"Symbol: {segment.parsed_data['symbol']}")
    print(f"Price: {segment.parsed_data['price']}")

# Monitor a segment for changes, checking the 'timestamp' field for updates
for updated_segment in reader.monitor_segment("market:btcusdt", interval=0.5, update_field="timestamp"):
    print(f"Update: {updated_segment.parsed_data['symbol']} @ {updated_segment.parsed_data['price']}")

Using the Command Line

List available segments:

shm-reader --list

Read a segment:

shm-reader --read market:btcusdt

Monitor a segment:

shm-reader --monitor market:btcusdt --update-field timestamp --interval 0.5

Clear segments:

shm-reader --clear

Register custom parsers from a JSON file:

shm-reader --register-parsers parsers.json

Example parser definition file (parsers.json):

[
  {
    "type": "market",
    "format": "12sdqd",
    "fields": ["symbol", "price", "volume", "timestamp"]
  },
  {
    "type": "order",
    "format": "12sddq",
    "fields": ["symbol", "price", "quantity", "order_id"]
  }
]

Segment Type Convention

The SDK uses a naming convention for segments to identify their type:

• Segment names can be prefixed with a type identifier: type:name
• Example: market:btcusdt, order:book, volatility:btcusdt
• The segment type is used to select the appropriate parser

Built-in Parsers

The SDK includes a built-in parser for "volatility" type segments with the following format:

• timestamp (double, 8 bytes)
• price (double, 8 bytes)
• volatility_1s (double, 8 bytes)
• high_1s (double, 8 bytes)
• low_1s (double, 8 bytes)
• volume_1s (long long, 8 bytes)
• update_count (long long, 8 bytes)

License

MIT License

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.