Testing with Hardhat and Typechain

Testing smart contracts with Hardhat and Typechain.

Testing with Hardhat and Typechain

In this article, you’ll learn how to test smart contracts with Hardhat and Typechain.

---

Objectives

By the end of this lesson, you should be able to:

• Set up TypeChain to enable testing
• Write unit tests for smart contracts using Mocha, Chai, and the Hardhat Toolkit
• Set up multiple signers and call smart contract functions with different signers

---

Overview

Testing is an important aspect of software development and developing smart contracts is no different. In fact, you need to be more careful because smart contracts usually manage money and live in an adversarial environment, where anyone can see the code and interact with your smart contract. This means you can expect bad actors to try to exploit your smart contracts.

Setup Typechain

In the previous guide, you created a new project using the init command that by default installs @nomicfoundation/hardhat-toolbox. This package already contains Typechain, which is a plugin that generates static types for your smart contracts. This means you can interact with your contracts and get immediate feedback about the parameters received by a particular function and the functions of a smart contract.

The best way to see its true potential is to start writing tests.

After compiling the hardhat project in the previous lesson, a new folder called typechain-types was created, which Typechain is already installed and running.

Writing your first unit test with Typechain

Hardhat includes a sample smart contract named Lock.sol and a sample test inside the test folder named Lock.ts.

In the following, you reuse this smart contract but rewrite the test using Typechain.

To remove the body of the Lock.ts file:

import { expect } from 'chai';
import { ethers } from 'hardhat';

describe('Lock', function () {});

Then, import two files from typechain-types, Lock, and Lock__Factory.

Typechain always creates two files per contract. The first one Lock refers to the type and functions of a particular contract. Lock__Factory is used to deploy the Lock contract or to create instances of a particular contract.

The Lock.sol contract allows the creator to lock Ether until an unlock time has passed.

Notice the constructor has a payable keyword:

constructor(uint _unlockTime) payable {
        require(
            block.timestamp < _unlockTime,
            "Unlock time should be in the future"
        );

        unlockTime = _unlockTime;
        owner = payable(msg.sender);
    }

This means the contract is expecting to receive an amount of ether.

Next, test the following:

• The unlock time value
• The value locked during creation
• The owner address
• The withdraw function

Testing unlockTime

Next, you include test cases after the before function.

The first test case should verify that the unlockTime variable is correct.

Testing Ether balance

In order to get the balance of your Lock contract, you simply call ethers.provider.getBalance.

Create a new test case:

Then, run npx hardhat test and you should get:

  Lock
    ✔ should get the unlockTime value
    ✔ should have the right ether balance

  2 passing (1s)

Testing owner

Similar to the previous test cases, you can verify that the owner is correct.

Then, run npx hardhat test and you should get:

  Lock
    ✔ should get the unlockTime value
    ✔ should have the right ether balance
    ✔ should have the right owner

  3 passing (1s)

Testing withdraw

Testing withdrawal is more complex because you need to assert certain conditions, such as:

• The owner cannot withdraw before the unlock time.
• Only the owner can withdraw.
• The withdraw function works as expected.

Hardhat allow you to test reverts with a set of custom matchers.

For example, the following code checks that an attempt to call the function withdraw reverts with a particular message:

it('shouldn"t allow to withdraw before unlock time', async () => {
  await expect(lockInstance.withdraw()).to.be.revertedWith("You can't withdraw yet");
});

In addition, Hardhat also allows you to manipulate the time of the environment where the tests are executed. You can think of it as a Blockchain that is running before the tests and then the tests are executed against it.

You can modify the block.timestamp by using the time helper:

it('shouldn"t allow to withdraw a non owner', async () => {
  const newLastBlockTimeStamp = await time.latest();

  // We set the next block time stamp using this helper.
  // We assign a value further in the future.
  await time.setNextBlockTimestamp(newLastBlockTimeStamp + UNLOCK_TIME);

  // Then we try to withdraw using other user signer. Notice the .connect function that is useful
  //  to create and instance but have the msg.sender as the new signer.
  const newInstanceUsingAnotherSigner = lockInstance.connect(otherUserSigner);

  // We attempt to withdraw, but since the sender is not the owner, it will revert.
  await expect(newInstanceUsingAnotherSigner.withdraw()).to.be.revertedWith("You aren't the owner");
});

Finally, test that the owner can withdraw. You can manipulate the time similarly to the previous test case but you won’t change the signer and will assert the new balances.

You can then run npx hardhat test and you should get:

  Lock
    ✔ should get the unlockTime value
    ✔ should have the right ether balance
    ✔ should have the right owner
    ✔ shouldn"t allow to withdraw before unlock time (51ms)
    ✔ shouldn"t allow to withdraw a non owner
    ✔ should allow to withdraw an owner

  6 passing (2s)

Conclusion

In this lesson, you’ve learned how to test smart contracts using Hardhat and Typechain.

---

See also

Solidity Docs [Remix Project]: https://remix-project.org/ [Hardhat]: https://hardhat.org/