Contract Name Types in Truffle Solidity Tests

Question

The example in the docs for truffle .sol tests has these lines:

MetaCoin meta = MetaCoin(DeployedAddresses.MetaCoin());

and

MetaCoin meta = new MetaCoin();

I've had a good read through the docs and don't understand why the MetaCoin type shows up so often. Wouldn't it be easier to access the deployed address with:

var meta = DeployedAddresses.MetaCoin(); ?

Answer 1

You're creating a state variable meta cast as Type MetaCoin. The compiler understands this Type by seeing the .sol code for MetaCoin, so meta gets all the methods of a MetaCoin contract.

You would accomplish that much with:

MetaCoin meta;

To be meaningful, it needs an address. Maybe it's already published, so a contract that wants to talk to it could get situated using an address passed to the constuctor.

   function MyContract(address metaAddress) {
      meta = MetaCoin(metaAddress);
   }

This is allowed, because meta is a MetaCoin - both sides are cast as the same Type.

Or, maybe you want to deploy a new MetaCoin each time:

function newMetaCoin() returns(address newContract)
    meta = new MetaCoin();
    return(meta);
}

The return works because contracts are directly convertable to Type address. It's shorthand for return(address(meta));

Hope it helps.

Answer 2

MetaCoin is simply what the truffle framework uses for demo purposes. Note how the contract is defined under the truffle-webpack-demo repository.

In order for you to interact with any of the functions defined in that contract (i.e. sendCoin), you need to instantiate a MetaCoin "object" that points to the deployed address by doing MetaCoin(DeployedAddresses.MetaCoin()). Simply doing DeployedAddresses.MetaCoin() will give you the address the contract is deployed to but won't give you the ability to make any calls or transactions.