solarchive 0.0.1

Python SDK for accessing Solana blockchain data from solarchive.org - free Parquet datasets of transactions, accounts, and tokens

solarchive

Python SDK for accessing Solana blockchain data from solarchive.org.

solarchive is a project to archive Solana's public transaction data and make it freely accessible in ergonomic formats (Apache Parquet) for developers, researchers, and the entire Solana community.

Features

• Download Solana transaction, account, and token data in Parquet format
• Access historical data from 2020 to present
• No API keys or rate limits - direct HTTP access to data files
• Query data with DuckDB, pandas, Spark, or any Parquet-compatible tool
• Browse available datasets via index files
• Licensed under CC-BY-4.0

Installation

pip install solarchive

Usage

from solarchive import SolArchive

# Initialize the client
client = SolArchive()

# List available transaction dates
dates = client.list_transaction_dates()
print(f"Available dates: {dates[:5]}...")

# Get index for a specific date
index = client.get_transaction_index("2025-11-01")
print(f"Files available: {len(index['files'])}")

# Download transaction data for a date
# Downloads all Parquet files for that date
client.download_transactions("2025-11-01", output_dir="./data/txs")

# Get account snapshots for a month
client.download_accounts("2025-12", output_dir="./data/accounts")

# Get token snapshots for a month
client.download_tokens("2024-09", output_dir="./data/tokens")

Data Schema

The archive contains three main datasets:

1. Transactions - All non-vote transactions with signatures, fees, account changes, and token balances

   • Schema: https://data.solarchive.org/schemas/solana/transactions.json
   • Partitioned by day: txs/YYYY-MM-DD/*.parquet

2. Accounts - Periodic snapshots of account states including balances and program data

   • Schema: https://data.solarchive.org/schemas/solana/accounts.json
   • Partitioned by month: accounts/YYYY-MM/*.parquet

3. Tokens - Metadata snapshots for fungible and non-fungible tokens

   • Schema: https://data.solarchive.org/schemas/solana/tokens.json
   • Partitioned by month: tokens/YYYY-MM/*.parquet

Development

This package requires Python 3.11 or higher.

# Install dependencies
pip install -e .

# Run the example
python main.py

About solarchive.org

Visit solarchive.org to explore the data directly in your browser using DuckDB-WASM, or to support the project. Hosting hundreds of terabytes costs nearly $10,000/year!

License

MIT