How to Find All Transactions for a Wallet

One of the most common questions when working with blockchain data is: "What transactions has this wallet address been involved in?" This guide shows you how to query SolArchive data to find all transactions for a specific wallet.

Understanding Account Keys

Every Solana transaction includes an account_keys array containing all wallet addresses involved in the transaction. This includes:

• The fee payer (usually index 0)
• All accounts that signed the transaction
• All accounts read from or written to
• Program accounts that were invoked

To find all transactions for a wallet, we search for transactions where the wallet address appears anywhere in the account_keys array.

Basic Wallet Query

Here's how to find all transactions involving a specific wallet address:

-- Find all transactions for a wallet
SELECT 
    block_timestamp,
    signature,
    fee / 1e9 as fee_sol,
    status
FROM read_parquet('data/txs/**/*.parquet')
WHERE list_contains(account_keys, 'YOUR_WALLET_ADDRESS_HERE')
ORDER BY block_timestamp DESC
LIMIT 100;

This query uses list_contains() to check if your wallet address appears in the account_keys array. The results show:

• block_timestamp - When the transaction occurred
• signature - Unique transaction ID
• fee_sol - Transaction fee in SOL (converted from lamports)
• status - Whether the transaction succeeded or failed

Finding Token Transfers

To specifically find token transfers (sends and receives), we need to check the pre_token_balances and post_token_balances arrays:

-- Find transactions with token transfers for a wallet
SELECT 
    t.block_timestamp,
    t.signature,
    tok.symbol,
    tok.name,
    (post.amount - pre.amount) / POW(10, pre.decimals) as amount_change,
    CASE 
        WHEN post.amount > pre.amount THEN 'received'
        ELSE 'sent'
    END as direction
FROM read_parquet('data/txs/**/*.parquet') as t,
    UNNEST(t.pre_token_balances) as pre,
    UNNEST(t.post_token_balances) as post
LEFT JOIN read_parquet('data/tokens/**/*.parquet') as tok
    ON pre.mint = tok.mint
WHERE list_contains(t.account_keys, 'YOUR_WALLET_ADDRESS_HERE')
  AND pre.mint = post.mint
  AND pre.account_index = post.account_index
  AND pre.amount != post.amount
  AND t.status = 'Success'
  AND tok.symbol IS NOT NULL
ORDER BY t.block_timestamp DESC
LIMIT 50;

This query shows:

• Token symbol and name (USDC, BONK, etc)
• Amount sent or received
• Direction (sent vs received)
• Only successful transactions with known tokens

Wallet Statistics

Want to see overall statistics for a wallet? This query calculates totals:

-- Transaction statistics for a wallet
SELECT 
    COUNT(*) as total_transactions,
    COUNT(CASE WHEN status = 'Success' THEN 1 END) as successful,
    COUNT(CASE WHEN status != 'Success' THEN 1 END) as failed,
    SUM(fee) / 1e9 as total_fees_sol,
    MIN(block_timestamp) as first_transaction,
    MAX(block_timestamp) as latest_transaction
FROM read_parquet('data/txs/**/*.parquet')
WHERE list_contains(account_keys, 'YOUR_WALLET_ADDRESS_HERE');

This gives you:

• Total number of transactions
• Success vs failed transaction count
• Total fees paid (in SOL)
• Date of first and most recent transaction

Complete Python Script

Here's a complete Python script that ties it all together:

import duckdb

# Your wallet address
WALLET_ADDRESS = "YOUR_WALLET_ADDRESS_HERE"

con = duckdb.connect()

# Find all transactions for the wallet
query = f"""
SELECT 
    block_timestamp,
    signature,
    fee / 1e9 as fee_sol,
    status,
    list_count(account_keys) as num_accounts
FROM read_parquet('data/txs/**/*.parquet')
WHERE list_contains(account_keys, '{WALLET_ADDRESS}')
ORDER BY block_timestamp DESC
LIMIT 100
"""

result = con.execute(query).fetchdf()
print(f"\nFound {len(result)} transactions for wallet {WALLET_ADDRESS[:8]}...\n")
print(result.to_string())

# Get statistics
stats_query = f"""
SELECT 
    COUNT(*) as total_transactions,
    COUNT(CASE WHEN status = 'Success' THEN 1 END) as successful,
    COUNT(CASE WHEN status != 'Success' THEN 1 END) as failed,
    SUM(fee) / 1e9 as total_fees_sol,
    MIN(block_timestamp) as first_transaction,
    MAX(block_timestamp) as latest_transaction
FROM read_parquet('data/txs/**/*.parquet')
WHERE list_contains(account_keys, '{WALLET_ADDRESS}')
"""

stats = con.execute(stats_query).fetchdf()
print("\n=== Wallet Statistics ===")
print(stats.to_string())

con.close()

Run it with:

uv run find_wallet_txs.py

Example Output

You'll see output like:

Found 1,247 transactions for wallet 8NN9ZVVa...

   block_timestamp          signature                    fee_sol    status
0  2025-12-01 23:59:58     5J9Kf2h8Ld3m9pQ7sR...       0.000005   Success
1  2025-12-01 18:32:15     3nF4sK8Lm3pQ9tY2v...       0.000005   Success
2  2025-12-01 12:08:42     7kP3mN8Lp2sR9wX5t...       0.000005   Success
...

=== Wallet Statistics ===
   total_transactions  successful  failed  total_fees_sol  first_transaction    latest_transaction
0  1247               1243        4       0.006235        2024-03-15 08:22:11  2025-12-01 23:59:58

Next Steps

Now that you can find wallet transactions, you might want to:

• Analyze spending patterns over time
• Track interactions with specific programs
• Export wallet history to CSV for reporting
• Build a wallet activity dashboard