Skip to main content

Build

As usual, let's install app dependencies and then build the app.

Install app dependencies​

note

The following instructions assume that you have installed Cubist.

Let's start by installing the Cubist Node.js SDK and related dependencies. cd to the application directory of your chosen example and language. Then:

npm install

Build instructions​

We run cubist build from the top-level token bridge app directory. From the top-level token bridge app directory, we (as usual) run:

cubist build

ERC20Bridged is currently configured to run on Polygon and TokenSender on Ethereum, but you can change that easily from the configuration file.

What Cubist generates​

When you build, Cubist generates one directory per target chain within the build output directory:

build
├── orm              // Bindings for interacting with the contracts
├── ethereum
│   ├── artifacts    // Compiled contracts that will run on ethereum
│   │   ├── ...   
│   └── contracts    // Source files of all original contracts (or ethereum shims)
│       └── ... 
└── polygon
    ├── artifacts    // Compiled contracts that will run on polygon
    │   ├── ...
    └── contracts    // Source files of all original contracts (or polygon shims)

Recall that the orm directory contains the Cubist-generated binding code that allows us to interface with our contracts.

In build, Cubist saves:

  1. The ABIs and other data produced when compiling the contracts with solc (within each target's artifacts directory)
  2. Original and shim source files (within each target's contracts directory). Cubist copies each contract in your contracts directory and puts them here (or replaces them with their shim, if applicable)
  3. The information that lets the off-chain relayer do its job (in ${ContractName}.bridge.json files).

The polygon and ethreum artifacts directories both contain an ERC20Bridged file and a TokenSender file; that's because both contracts are used cross-chain, so they have shim on the chain that's not their original target. Within the artifacts directory, we also see a number of other directories corresponding to code that we didn't write. For example, in the polygon/artifacts directory, we see directories for ERC20, IERC20, and more. This extra data is generated during compilation because our token bridge contracts require it; for example, ERC20Bridged inherits from ERC20 (which it imports from OpenZepplin via npm).

danger

In general, you should not modify any of the files in the build directory. These files are automatically generated, so your changes may get overwritten.