hillstreetlabs / parr Goto Github PK

Example:

[
    {
      "constant": true,
      "inputs": [],
      "name": "rate",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "weiRaised",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "wallet",
      "outputs": [
        {
          "name": "",
          "type": "address"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "token",
      "outputs": [
        {
          "name": "",
          "type": "address"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "inputs": [
        {
          "name": "_rate",
          "type": "uint256"
        },
        {
          "name": "_wallet",
          "type": "address"
        },
        {
          "name": "_token",
          "type": "address"
        }
      ],
      "payable": false,
      "stateMutability": "nonpayable",
      "type": "constructor"
    }]

Example output:

Failed to download transaction 0x752924da840fbe610484c3d510de165bc42c0ddc4cd35971727f65e3b9d97b4f TypeError: Cannot read property 'to' of null
    at /Users/pfh/Projects/ferrari/src/lib/TransactionDownloader.js:77:50
    at Generator.next (<anonymous>)
    at step (/Users/pfh/Projects/ferrari/dist/lib/TransactionDownloader.js:20:191)
    at /Users/pfh/Projects/ferrari/dist/lib/TransactionDownloader.js:20:361
    at <anonymous>
    at process._tickCallback (internal/process/next_tick.js:118:7)

Update Importer to only download a block and persist the block and its included transactions to postgres. It should not call getTransactionReceipt for the transactions.

Should include:

• Module to take a block and index it, its transactions and related accounts
• CLI to call the index method for all blocks or a specific block

Includes:

• Hook up Ethmoji contract
• Index all logs and internal transactions

• Histogram showing percentage of fully-index blocks, chunked into 100 pieces
• total number of blocks, transactions

Included:

• Reference data stored in postgres when indexing in Elasticsearch versus web3 directly

Index 1: Blocks and transactions

• Index containing documents with type field that can be either block or transaction
• All transactions with from, to, logs and internal_transactions all embedded
• Blocks will be parents, and transactions will be children

Index 2: Addresses

• Should include addresses as parents and transactions (with embedded block, log and internal transaction data) as children
• Will have to store each transaction twice, since a child can only have one parent via join field
• Index will have a type field that can be address, fromTransaction and toTransaction

Should include:

• Name
• Ticker

Not sure if this is intended. But seems to make bunch of transactions fail

Includes:

• Express API with /blocks, /transactions, and /accounts routes

The goal of this issue is to list out all use-cases for parr, so that we build only useful things. This is meant to serve as a living document of interesting use-cases.

1. Make a dashboard of personal ethmoji earnings, both over time and cumulative.
2. See a list of all ethmojis my ethmoji items contributed to.
3. See a list of all accounts with more than 1000 ETH

Not sure what to call this, but it should take an arbitrary, un-imported transaction from postgres, call getTransactionReceipt, get internal transactions, decode logs, and write all the data back to postgres for indexing.

Included items:

• Add postgres database for storing all downloaded data
• Create blocks, transactions, internal_transactions and logs tables
• Add status fields (or another name) to indicate parsing status (i.e. downloaded transaction but not transaction receipt or added block ID but not block data)

Allow people to save and share queries.

POST /queries:

• { query: "[THE QUERY]", hash: "[hash]" }

GET /queries/[hash]

• { query: "[THE QUERY]", hash: "[hash]" }

Data should include:

• start time
• end time
• token (address)

All of this is in the Crowdsale ABI:

{
  "contractName": "Crowdsale",
  "abi": [
    {
      "constant": true,
      "inputs": [],
      "name": "rate",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "endTime",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "weiRaised",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "wallet",
      "outputs": [
        {
          "name": "",
          "type": "address"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "startTime",
      "outputs": [
        {
          "name": "",
          "type": "uint256"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "token",
      "outputs": [
        {
          "name": "",
          "type": "address"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    },
    {
      "anonymous": false,
      "inputs": [
        {
          "indexed": true,
          "name": "purchaser",
          "type": "address"
        },
        {
          "indexed": true,
          "name": "beneficiary",
          "type": "address"
        },
        {
          "indexed": false,
          "name": "value",
          "type": "uint256"
        },
        {
          "indexed": false,
          "name": "amount",
          "type": "uint256"
        }
      ],
      "name": "TokenPurchase",
      "type": "event"
    },
    {
      "constant": false,
      "inputs": [
        {
          "name": "beneficiary",
          "type": "address"
        }
      ],
      "name": "buyTokens",
      "outputs": [],
      "payable": true,
      "stateMutability": "payable",
      "type": "function"
    },
    {
      "constant": true,
      "inputs": [],
      "name": "hasEnded",
      "outputs": [
        {
          "name": "",
          "type": "bool"
        }
      ],
      "payable": false,
      "stateMutability": "view",
      "type": "function"
    }
  ]
}

Includes:

• API for searching Elasticsearch index with /query or /search POST method

TODO:

• Need to work out format of Elasticsearch indices

The simplest version of this website will be just a textarea and a submit button. A user can input an ES json query into the textarea and send to /search by submitting the form.

Because encodeSignature requires its argument to have both a name and a type, just checking that all inputs have those works for now. However, it might make sense to tell the user that their ABI is invalid, if they pass an ABI with an entry that is of type function but lacks a name. It seems only constructor and fallback types are allowed to miss name.

Also, possible add a stats CLI command

When adding a new ABI to check every address against, we shouldn't have to re-download transactions. We should have a separate script for re-checking bytecode against ABIs and updating the implements jsonb column.

ES errors just cause a timeout right now 👎

• Find generic ERC 20 and ERC 721 ABIs and import into contracts table
• Make addresses optional, so the decoder can check generic standards like ERC 721 as well as specific address ABIs

We should update Indexer to index blocks, transactions, logs, internal transactions, and addresses from postgres. The script should watch postgres and index fully imported rows once the importers update them.

The parr_blocks_transactions index embeds addresses inside transactions (i.e. {type: "transaction", from: {address: "...", is_contract: false}}). When the embedded address changes, each transaction should be re-indexed.

The parr_addresses index embeds addresses as well, even though it also has a parent/child relationship between transactions and addresses.

When addresses are being indexed, it fetches all the transactions available and indexes those. If new transactions that involve already indexed addresses gets added, that should trigger a re-index.

We currently only index blocks, transactions and logs.

We should import and index internal transactions as well.

Includes:

• Script to poll Infura and index all new blocks (can use index method here: #1)

We need a way for community members to submit contract ABIs and specify their addresses

Not sure what the best setup will be, but we should start working on this

A JSON route that a user can hit that gives stats about the import/download/index status of the whole system.