reth-db-py

Python package allowing you to interact with the Reth DB via Python. Written with Rust and Pyo3.

This python wrapper can access node data 3x-5x faster than local web3.py calls. Using this package, a block can be retrieved in ~10ms on a local reth DB.

Installation

This package has been published to PyPi and can be installed using pip:

pip install reth-db-py

Assets

This package only has a single python class made available: PyDatabaseHandler.

PyDatabaseHandler is a class used to interact with the Reth DB. It's a wrapper around the Rust DatabaseHandler struct. It has a few methods which all return json strings:

• get_header_by_block_number: get a single header by block number
• get_headers_by_block_number_range: get multiple headers by block number range
• get_transaction_by_id: get a single transaction by transaction id
• get_transactions_by_id_range: get multiple transactions by transaction id range
• get_transactions_by_block_number_range: get multiple transactions by block number range
• get_block_by_number: get a single block by block number
• get_uncles_by_block_number: get uncles by block number
• get_receipts_by_transaction_id: get receipts by transaction id
• get_receipts_by_block_number: get receipts by block number

impl PyDatabaseHandler {
    pub fn get_header_by_block_number(&self, number: u64) -> PyResult<String>
    pub fn get_headers_by_block_number_range(&self, start: u64, end: u64) -> PyResult<String>
    pub fn get_transaction_by_id(&self, id: u64) -> PyResult<String>
    pub fn get_transactions_by_id_range(&self, start: u64, end: u64) -> PyResult<String>
    pub fn get_transactions_by_block_number_range(&self, start: u64, end: u64) -> PyResult<String>
    pub fn get_block_by_number(&self, number: u64) -> PyResult<String>
    pub fn get_uncles_by_block_number(&self, number: u64) -> PyResult<String>
    pub fn get_receipts_by_transaction_id(&self, id: u64) -> PyResult<String>
    pub fn get_receipts_by_block_number(&self, number: u64) -> PyResult<String>
}

class PyDatabaseHandler:
    def get_header_by_block_number(self, number: int) -> str
    def get_headers_by_block_number_range(self, start: int, end: int) -> str
    def get_transaction_by_id(self, id: int) -> str
    def get_transactions_by_id_range(self, start: int, end: int) -> str
    def get_transactions_by_block_number_range(self, start: int, end: int) -> str
    def get_block_by_number(self, number: int) -> str
    def get_uncles_by_block_number(self, number: int) -> str
    def get_receipts_by_transaction_id(self, id: int) -> str
    def get_receipts_by_block_number(self, number: int) -> str

Usage

Import reth-db-py assets:

from reth_db_py import PyDatabaseHandler

Create a PyDatabaseHandler instance:

handler = PyDatabaseHandler("/path/to/db/mdbx.dat")

Get the header by block number

header = handler.get_header_by_block_number(17_000_000)

Get the headers by block number range

headers = handler.get_headers_by_block_number_range(17_000_000, 17_000_005)

Get transaction by id

transaction = handler.get_transaction_by_id(1000)

Get transactions by id range

transactions = handler.get_transactions_by_id_range(1000, 1005)

Get transactions by block number range

transactions = handler.get_transactions_by_block_number_range(17_000_000, 17_000_005)

Get block by number

block = handler.get_block_by_number(17_000_000)

Get uncles by block number

uncles = handler.get_uncles_by_block_number(17_000_000)

Get receipts by transaction id

receipts = handler.get_receipts_by_transaction_id(1000)

Get receipts by block number

receipts = handler.get_receipts_by_block_number(17_000_000)

Tests

Coming soon.

Benchmarks

Speed tests were conducted by retrieving 1,000 block bodies (including transactions) using:

• reth-db-py (local reth db)
• web3.py (local reth JSON RPC)
• http requests (remote alchemy API)

See benchmark tests here.