Skip to content

aptos-labs/aptos-indexer-processors

Repository files navigation

codecov

Aptos Indexer Client Guide

This guide will get you started with creating an Aptos indexer with custom parsing. We have several endpoints that provided a streaming RPC of transaction data.

GRPC Data Stream Endpoints

Request

  • config.yaml
    • chain_id: ID of the chain used for validation purposes.
    • grpc_data_stream_endpoint: Replace with the grpc data stream endpoints for mainnet, devnet, testnet, or previewnet.
    • grpc_data_stream_api_key: Replace YOUR_TOKEN with your auth token.
    • db_connection_uri: The DB connection used to write the processed data
    • (optional) starting-version
      • If starting-version is set, the processor will begin indexing from transaction version = starting_version.
      • To auto restart the client in case of an error, you can cache the latest processed transaction version. In the example, the processor restarts from cached transaction version that is stored in a table, and if neither starting_version nor cached version are set, the processor defaults starting version to 0.

Response

  • The response is a stream of RawDatastreamResponse objects.
  • To learn more about the protos and the code generated from those protos see protos/ in aptos-core.

Warning

For production-grade indexers, we recommend the Rust processors. The Python implementation is known to have a grpc deserialization recursion limit. The issue is with the GRPC library and we haven't had a chance to look into this. Please proceed with caution. The typescript implementation is known to get stuck when there are lots of data to process. The issue is with the GRPC client and we haven't had a chance to optimize. Please proceed with caution.