Index ERC-20 transfers

This tutorial builds an event table of USDC transfers on Ethereum mainnet, recording the sender, recipient, amount, and transaction hash.

chain.event.eth → Filter USDC Transfer events → Extract transfer fields → Transfer event table

Before you begin

Sign in to the Chaintable console. If you don’t have a Space yet, click New Space on the Spaces page, enter an Id, and click Create space.

New Space dialog with the Id field and Create space button

Open your new or existing Space. This tutorial uses demo; replace it with your Space ID.

1. Create a transfer event table

In Tables, create a table named usdc_transfers, select Block Event, and choose Ethereum. The resulting table ID is demo.usdc_transfers.eth.

Create an Ethereum Block Event Table in Tables

Add the following fields in Schema, keeping the system fields and other defaults:

Field	Type	Content
`tx_id`	`TXID`	Transaction hash
`from_address`	`ADDRESS`	Sender address
`to_address`	`ADDRESS`	Recipient address
`amount`	`UINT256`	Raw integer transfer amount

Transfer table schema with default configuration retained

2. Create a Function

In Functions, create parse_usdc_transfer and enter the following code:

def _(event):
    sender, recipient, amount = event['params']
    return {
        'id': event['id'],
        'tx_id': event['tx_id'],
        'from_address': sender,
        'to_address': recipient,
        'amount': amount,
        'block_id': event['block_id'],
        'block_height': event['block_height'],
        'block_timestamp': event['block_timestamp'],
    }

Add an event parameter below the editor, select JSON as its type, and save the Function. For the steps, see Create a Function. The Function extracts the sender, recipient, and amount from the event’s params and returns a record for the transfer table. The record also carries the event’s block fields, which larger backfills require; see Result handlers.

The parse_usdc_transfer Function with its event parameter

3. Run the Pipeline

Configure the Pipeline

In Notebooks, create backfill_usdc_transfers and enter the following code. Replace SPACE with your Space ID:

from blockdb import Column
from blockdb.operator import filter
from blockx import Pipeline, ROW

SPACE = 'demo'

pipeline = Pipeline(
    sources=[{
        'table_id': 'chain.event.eth',
        'func_id': f'{SPACE}.parse_usdc_transfer',
        'args_tmpl': [ROW],
        'operator': filter(
            (Column('contract_id') == '0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')
            & (Column('name') == 'Transfer')
        ),
    }],
    target=f'{SPACE}.usdc_transfers.eth',
)

pipeline.backfill(start=25939925, end=25939934)

sources specifies the source table and saved Function. operator filters events, and ROW passes each source record as an argument. The Pipeline writes the Function’s returned records to target.

Run the backfill

Save and click Run. View progress and results in Output or Instance detail. backfill() processes the specified 10 blocks, including both boundary heights, then exits.

Block processing logs and completed backfill_usdc_transfers instance

4. View transfer data

Return to usdc_transfers.eth and check the processed blocks in Overview. This run backfills only a limited range; earlier history is still missing.

Transfer table progress with default settings and the event subtable entry

Open event under Subtables to view transfer records.

Actual USDC transfer records in the event subtable

Each record uses the source event’s id. The Pipeline fills in block fields automatically. amount retains the raw integer value. Divide it by 10^6 to get the USDC amount.

Optional: keep the table updated

Create a Notebook named update_usdc_transfers, copy the code above, and replace the final pipeline.backfill(...) line with the following call. Save and run:

pipeline.update()

By default, update() fills history from the earliest unfinished block, then keeps updating after reaching the latest block. If you only need data from a specific height onward, set the start height first. For a large backlog, you can also run backfill() in another Notebook at the same time to speed up historical processing.

Continuous update call and historical processing logs in update_usdc_transfers

Select Switch to Background to run in the background. Cancel the instance in the account’s Instances page to stop it. Data already written is retained.

Overview

Quickstart

Guides

Reference