Asset MetaData Service

Store and access information about assets

This is an interim version used primarily for testing and demonstrations

Run

This will run the indexer RPC server and an Arweave testnet node

node dist/src/index.js

OR

make run

You can assess the service on port 8081

Ethereum

For simplicty it also provides an instance of Ganache to run for testing

Run make eth in a separate terminal to start ganache

API

Create a new asset

POST /asset with an object in the format of an AssetMetaData as the body of the message:

curl --header "Content-Type: application/json" 
     --request POST --data {"did":"1234""name": "example", ...}
     localhost:8081/asset

The body of the response will contain an object with the Arweave transaction id:

{
    txid: '1Lv13UEyuX9pw710TqSDBJ1gLU5zVdndwgFyItTFRFY'
}

Get asset details

GET /asset/{did} given the did return the asset's details from storage

curl --header "Content-Type: application/json" --request GET localhost:8081/asset/1234

On success, the body of the response will contain and object in the format of AssetMetaData. If not found, a status code of 404 will be returned.

Get a list of assets

GET /assets returns a list of information on assets. See db/assets.ts AssetInfo for the format.

curl --header "Content-Type: application/json" --request GET localhost:8081/assets

This will always return a 200 status code. If no assets are found, it will return an empty list []

How it works

The indexer provides a small REST API to allow a client to store and fetch metadata associated with BioAssets. The full metadata is stored on permanent decentralized storage (Arweave). The API server maintains an in-memory database to serve as a cache for the information. A subset of the metadata is stored and returned from the cache, while the full metadata is returned from Arweave.

Because the entire history of metadata is stored on Arweave, it can be rebuilt from scratch if needed. So the reliance on a central server such as the indexer is primarily for convenience. In otherwords, if the indexer goes offline, the metadata is still available via Arweave.

Each asset is assigned a globally unique identity using a decentralized identitifer (DID) format. The identifier is calculated from the Keccak256 hash of the contract address and ethereum chain id. The DID serves as a unique pointer to the actual transaction hash of the content stored on Arweave.

We plan to expand the role of the DID and it's associated document (outlined in the w3 specification) in the future.

TODO

• Authentication via Ethereum signatures