Skip to content

Latest commit

 

History

History
58 lines (40 loc) · 2.57 KB

README.md

File metadata and controls

58 lines (40 loc) · 2.57 KB
Raw

TX indexer

This is a micro-indexer built on top of NEAR Lake Framework.

The purpose of this indexer is to watch for every single Transaction in NEAR Protocol and collect all the children entities related to the Transaction into a single structure TransactionDetails.

TransactionDetails definition

TODO: Update the README once the structure is finalized.

ref: struct TransactionDetails

How it works

The tx-indexer works like any other Lake-based indexer watching the network.

In order to collect a Transactions with all related entities it does the following:

  1. Converts Transaction into a tx-indexer-defined structure TransactionDetails
  2. Stores the data into cache
  3. Watched for ExecutionOutcomes and its statuses. Depending on the status:
    • SuccessValue means the Receipt "branch" has finished successfully
    • SuccessReceiptId means a new Receipt branch has started and we need to follow it
    • Other error/failure statuses means the Receipt branch has finished with an error
  4. Whenever we observe a new Receipt branch we increment the Transaction's anticipated ExecutionOutcomes
  5. Whenever a Receipt branch is finishes we decrement the Transaction's anticipated ExecutionOutcomes
  6. We add related Receipts and ExecutionOutcomes to the corresponding fields of the TransactionDetails during the indexing
  7. Once the anticipated ExecutionOutcomes counter becomes zero we consider the TransactionDetails collecting as over and can serialize and store in the Storage

How to run

Prerequisites

An example .env file:

INDEXER_ID=tx-indexer
DATABASE_URL=127.0.0.1:9042
DATABASE_USER=admin
DATABASE_PASSWORD=password

AWS_ACCESS_KEY_ID=AKIAIOSFODNN7EXAMPLE
AWS_SECRET_ACCESS_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
AWS_DEFAULT_REGION=eu-central-1

Features

Command to run

cargo run --release -- <start_options>
  • start_options:
    • from-latest fetches the final block height from the RPC and starts indexing from that block
    • from-interruption <N?> is used to retrieve the last_processed_block_height from the database. This value is used as the starting point for processing blocks. If a specific value <N?> is provided, it will be used as the fallback option. If <N?> is not provided or if the database does not have a record (for example, in the case of a fresh start with an empty storage), the fallback option will be from-latest.
    • from-block <N> starts indexing from the block height <N>