Skip to content

KenCloud-Tech/prolly-tree-oracle

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

47 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Prolly Tree Oracle

What

  • EVM oracle
  • Based on ipld prolly tree indexer
  • Enable contracts to create/read/update/delete to prolly trees.
  • Index over collections to enable faster queries
  • charge gas or eth for transactions (e.g. if the field you query isn't indexed, charge more)
  • maybe gas costs per lPLD block loaded? That way people deploying oracles can set theiown costs.
  • allow querying by existing cid
  • advertise prolly tree root on lPFS network to query offchain
  • oracle publishes tree under ipns
  • can we get read/write access gating based on wallets or multisigs?

API:

  • create(owner wallet, prolly config params)=>id

  • allowWrite(wallet)

  • put(id,collection,JSON document, primaryKey?) => primaryKey

  • index(id, collection, fields)

  • get(id, collection, primaryKey)->JSON

  • search(id, collection, query {equals, sort, limit, skip})

Architecture

Blockchain and decentralized applications (DApps) are increasingly important for establishing trust and transparency in data storage and computing. However, on-chain transactions are often costly and slow. To overcome this challenge, off-chain nodes can be used to store and compute data. Blockchain oracles are introduced and implemented in the form of application programming interfaces (APIs) to connect the real world to the blockchain. Prolly Tree Oracle connects the IPLD Prolly Tree storage service to the blockchain, allowing users to store large amounts of data off-chain at a fraction of the cost while enjoying the transparency and security brought by the blockchain.

Constructor:

constructor(uint basefee, uint bytefee) payable {
        owner = payable(msg.sender);
        baseGasFee = basefee;
        gasPerByte = bytefee;
        CurrentReqID = 1;
    }

When deploying the oracle contract, you can set the basic service fee for each service and the service fee per byte of data when importing data using the Put method.
So users should make sure to carry enough fees to pay the service fee when calling the API.

API Implementation:

  • function Create(string calldata dbName, string calldata colName, string calldata primaryKey)
    Creates a new collection named colName in the database pointed to by the provided dbName. If two users register the same dbName successively, the owner of this database is the first user. When the second user registers, it is necessary to determine whether it has the permission to the database.

  • function AllowWrite(address to)
    Grant user to read and write permissions on your database.
    service fee = baseGasFee

  • function Put(string calldata dbName, string calldata colName, bytes calldata data)
    Save the data data to the colName collection in the database pointed to by dbName.
    service fee = baseGasFee + gasPerByte * datasize

  • Import_from_url(string calldata dbName, string calldata colName, string calldata url, string calldata format)
    If you use the Put method to import a large amount of data, you will pay expensive service fees, so we provide off-chain data import services.
    The off-chain storage service provider will initiate a Get request to obtain data through the URL you provide.
    You need to specify the data format through format. format = {ndjson or csv}
    service fee = baseGasFee

  • function Get(string calldata dbName, string calldata colName, bytes calldata recordID, string calldata callBack)
    Get data from the colName collection in the database pointed to by dbName according to recordID. You need to provide a callback function to receive the data. For example, the callback function I defined in the consumer contract is:function CallBackFunc(bytes calldata data). So, the callBack parameter of string type is CallBackFunc(bytes).
    service fee = baseGasFee

  • function Index(string calldata dbName, string calldata colName, string calldata idx)
    Add index idx to the colName collection in the database pointed to by dbName
    service fee = baseGasFee

  • function Search(string calldata dbName, string calldata colName, bytes calldata querys, string calldata callBack)
    Use index.Query to query data from the colName collection in the database pointed to by dbName. When using it, you can build one or more index.Query objects, and serialize the index.Query object array into json as the parameter queries. Query methods:{equal, compare, sort, limit, skip}
    Similarly, a callback function needs to be provided to receive data
    service fee = baseGasFee

  • function GetRootCid(string calldata dbName, string calldata callBack) Get the RootCid of the database of the method caller
    Similarly, a callback function needs to be provided to receive an array of collection indexes
    service fee = baseGasFee

  • function GetIndex(string calldata dbName, string calldata colName, string calldata callBack)

    Get the index of the colName collection in the database pointed to by dbName
    Similarly, a callback function needs to be provided to receive an array of collection indexes
    service fee = baseGasFee

  • function GetCol(string calldata dbName, string calldata callBack) Get all collections in the database pointed to by dbName Similarly, a callback function needs to be provided to receive an array of collection names
    service fee = baseGasFee

Usage

Consumer Contract

The Consumer Contract used for testing is located at contracts/test.sol, and there is an example of a WebApp project that interacts directly with the oracle: ./webApp

Compile the Consumer Contract to obtain the ABI file, and use the abigen to generate Golang package corresponding to the contract: abigen --abi={ Path to abi.json } --pkg={ Package name } --out={ The path to the generated .go file }

Golang Client

Use a Golang test program golangServer/test/Oracle_test.go to call the request function in the Consumer Contract for testing the oracle.