Skip to main content
Version: Old ArrowSquid docs

Block data for Substrate

SubstrateBatchProcessor follows the common squid processor architecture, in which data processing happens within the batch handler, a function repeatedly called on batches of on-chain data. The function takes a single argument called "batch context". Its structure follows the common batch context layout, with ctx.blocks being an array of Block objects containing the data to be processed, aligned at the block level.

For SubstrateBatchProcessor the Block interface is defined as follows:

export type Block<F extends FieldSelection = {}> = {
header: BlockHeader<F>
extrinsics: Extrinsic<F>[]
calls: Call<F>[]
events: Event<F>[]

F here is the type of the argument of the setFields() processor method.

Block.header contains the block header data. The rest of the fields are iterables containing four kinds of blockchain data. The canonical ordering within each iterable is the same as it is within the blocks.

The exact fields available in each data item type are inferred from the setFields() call argument. They are documented on the Field selection page:


The handler below simply outputs all the Balances.transfer_all calls on Kusama in real time:

import {SubstrateBatchProcessor} from '@subsquid/substrate-processor'
import {TypeormDatabase} from '@subsquid/typeorm-store'

const processor = new SubstrateBatchProcessor()
chain: '',
archive: lookupArchive('kusama', {release: 'ArrowSquid'})
.setBlockRange({from: 19_600_000})
name: ['Balances.transfer_all'],
call: {
origin: true,
success: true
}) TypeormDatabase(), async ctx => {
for (let block of ctx.blocks) {
for (let call of block.calls) {, `Call:`)

One can experiment with the setFields() argument and see how the output changes.