corda/docs/source/hello-world-flow.rst

Writing the flow

A flow describes the sequence of steps for agreeing a specific ledger update. By installing new flows on our node, we allow the node to handle new business processes. Our flow will allow a node to issue an IOUState onto the ledger.

Flow outline

Our flow needs to take the following steps for a borrower to issue a new IOU onto the ledger:

1. Create a valid transaction proposal for the creation of a new IOU
2. Verify the transaction
3. Sign the transaction ourselves
4. Record the transaction in our vault
5. Send the transaction to the IOU's lender so that they can record it too

Subflows

Although our flow requirements look complex, we can delegate to existing flows to handle many of these tasks. A flow that is invoked within the context of a larger flow to handle a repeatable task is called a subflow.

In our initiator flow, we can automate steps 4 and 5 using FinalityFlow.

All we need to do is write the steps to handle the creation and signing of the proposed transaction.

FlowLogic

Flows are implemented as FlowLogic subclasses. You define the steps taken by the flow by overriding FlowLogic.call.

We'll write our flow in either TemplateFlow.java or App.kt. Delete both the existing flows in the template, and replace them with the following:

example-code/src/main/kotlin/net/corda/docs/tutorial/helloworld/flow.kt

example-code/src/main/java/net/corda/docs/java/tutorial/helloworld/IOUFlow.java

If you're following along in Java, you'll also need to rename TemplateFlow.java to IOUFlow.java.

We now have our own FlowLogic subclass that overrides FlowLogic.call. There's a few things to note:

• FlowLogic.call has a return type that matches the type parameter passed to FlowLogic - this is type returned by running the flow
• FlowLogic subclasses can have constructor parameters, which can be used as arguments to FlowLogic.call
• FlowLogic.call is annotated @Suspendable - this means that the flow will be check-pointed and serialised to disk when it encounters a long-running operation, allowing your node to move on to running other flows. Forgetting this annotation out will lead to some very weird error messages
• There are also a few more annotations, on the FlowLogic subclass itself:
  • @InitiatingFlow means that this flow can be started directly by the node
  • @StartableByRPC allows the node owner to start this flow via an RPC call
• We override the progress tracker, even though we are not providing any progress tracker steps yet. The progress tracker is required for the node shell to establish when the flow has ended

Let's walk through the steps of FlowLogic.call one-by-one:

Retrieving participant information

The identity of our counterparty is passed in as a constructor argument. However, we need to use the ServiceHub to retrieve our identity, as well as the identity of the notary we'll be using for our transaction.

You can see that the notary's identity is being retrieved from the node's ServiceHub. Whenever we need information within a flow - whether it's about our own node, its contents, or the rest of the network - we use the node's ServiceHub. In particular, ServiceHub.networkMapCache provides information about the other nodes on the network and the services that they offer.

Building the transaction

We'll build our transaction proposal in two steps:

• Creating a transaction builder
• Adding the desired items to the builder

Creating a transaction builder

To start building the proposed transaction, we need a TransactionBuilder. This is a mutable transaction class to which we can add inputs, outputs, commands, and any other items the transaction needs. We create a TransactionBuilder that uses the notary we retrieved earlier.

Transaction items

Now that we have our TransactionBuilder, we need to add the desired items. Remember that we're trying to build the following transaction:

So we'll need the following:

• The output IOUState and its associated contract
• A Create command listing the IOU's lender as a signer

The command we use pairs the IOUContract.Create command defined earlier with our public key. Including this command in the transaction makes us one of the transaction's required signers.

We add these items to the transaction using the TransactionBuilder.withItems method, which takes a vararg of:

• StateAndContract or TransactionState objects, which are added to the builder as output states
• StateAndRef objects (references to the outputs of previous transactions), which are added to the builder as input state references
• Command objects, which are added to the builder as commands
• SecureHash objects, which are added to the builder as attachments
• TimeWindow objects, which set the time-window of the transaction

It will modify the TransactionBuilder in-place to add these components to it.

Verifying the transaction

We've now built our proposed transaction. Before we sign it, we should check that it represents a valid ledger update proposal by verifying the transaction, which will execute each of the transaction's contracts.

If the verification fails, we have built an invalid transaction. Our flow will then end, throwing a TransactionVerificationException.

Signing the transaction

Now that we have a valid transaction proposal, we need to sign it. Once the transaction is signed, no-one will be able to modify the transaction without invalidating our signature, effectively making the transaction immutable.

The call to ServiceHub.toSignedTransaction returns a SignedTransaction - an object that pairs the transaction itself with a list of signatures over that transaction.

Finalising the transaction

Now that we have a valid signed transaction, all that's left to do is to have it notarised and recorded by all the relevant parties. By doing so, it will become a permanent part of the ledger. As discussed, we'll handle this process automatically using a built-in flow called FinalityFlow:

FinalityFlow completely automates the process of:

• Notarising the transaction if required (i.e. if the transaction contains inputs and/or a time-window)
• Recording it in our vault
• Sending it to the other participants (i.e. the lender) for them to record as well

Our flow, and our CorDapp, are now ready!

Progress so far

We have now defined a flow that we can start on our node to completely automate the process of issuing an IOU onto the ledger. The final step is to spin up some nodes and test our CorDapp.