Getting Started with Anchor

The Anchor framework is a tool that simplifies the process of building Solana programs. Whether you're new to blockchain development or an experienced programmer, Anchor simplifies the process of writing, testing, and deploying Solana programs.

In this section, we'll walk through:

  • Creating a new Anchor project
  • Building and testing your program
  • Deploying to Solana clusters
  • Understanding the project file structure

前提条件 #

For detailed installation instructions, visit the installation page.

Before you begin, ensure you have the following installed:

  • Rust: The programming language for building Solana programs.
  • Solana CLI: Command-line tool for Solana development.
  • Anchor CLI: Command-line tool for the Anchor framework.

To verify Anchor CLI installation, open your terminal and run:

Terminal
anchor --version

Expected output:

Terminal
anchor-cli 0.30.1

Getting Started #

This section covers the basic steps to create, build, and test your first local Anchor program.

Create a new Project #

To start a new project, use the anchor init command followed by your project's name. This command creates a new directory with the specified name and sets up a default program and test file.

Terminal
anchor init my-program

Navigate to the new project directory and open it in your code editor.

Terminal
cd my-project

The default Anchor program is located at /programs/my-project/src/lib.rs.

The default Typescript test file is located at /tests/my-project.ts.

If you prefer Rust for testing, initialize your project with the --test-template rust flag.

anchor init --test-template rust my-program

The Rust test file will be at /tests/src/test_initialize.rs.

Build the Program #

Build the program by running anchor build.

Terminal
anchor build

The compiled program will be at /target/deploy/my_project.so. The content of this file is what gets stored on the Solana network (as an executable account) when you deploy your program.

Test the Program #

To test the program, run anchor test.

Terminal
anchor test

By default, the Anchor.toml config file specifies the localnet cluster. When developing on localnet, anchor test will automatically:

  1. Start a local Solana validator
  2. Build and deploy your program to the local cluster
  3. Run the tests in the tests folder
  4. Stop the local Solana validator

Alternatively, you can manually start a local Solana validator and run tests against it. This is useful if you want to keep the validator running while you iterate on your program. It allows you to inspect accounts and transaction logs on the Solana Explorer while developing locally.

Open a new terminal and start a local Solana validator by running the solana-test-validator command.

Terminal
solana-test-validator

In a separate terminal, run the tests against the local cluster. Use the --skip-local-validator flag to skip starting the local validator since it's already running.

Terminal
anchor test --skip-local-validator

Deploy to Devnet #

By default, the Anchor.toml config file in an Anchor project specifies the localnet cluster.

Anchor.toml
[toolchain]
 
[features]
resolution = true
skip-lint = false
 
[programs.localnet]
my_program = "3ynNB373Q3VAzKp7m4x238po36hjAGFXFJB4ybN2iTyg"
 
[registry]
url = "https://api.apr.dev"
 
[provider]
cluster = "Localnet"
wallet = "~/.config/solana/id.json"
 
[scripts]
test = "yarn run ts-mocha -p ./tsconfig.json -t 1000000 tests/**/*.ts"

To deploy your program to devnet, change the cluster value to Devnet. Note that this requires your wallet to have enough SOL on Devnet to cover deployment cost.

-cluster = "Localnet"
+cluster = "Devnet"
Anchor.toml
[provider]
cluster = "Devnet"
wallet = "~/.config/solana/id.json"

Now when you run anchor deploy, your program will be deployed to the devnet cluster. The anchor test command will also use the cluster specified in the Anchor.toml file.

anchor deploy

To deploy to mainnet, simply update the Anchor.toml file to specify the mainnet cluster.

Anchor.toml
[provider]
cluster = "Mainnet"
wallet = "~/.config/solana/id.json"

Update the Program #

Solana programs can be updated by redeploying the program to the same program ID.

To update a program, simply make changes to your program's code and run the anchor build command to generated an updated .so file.

anchor build

Then run the anchor deploy command to redeploy the updated program.

anchor deploy

Close the Program #

To reclaim the SOL allocated to a program account, you can close your Solana program.

To close a program, use the solana program close <PROGRAM_ID> command. For example:

solana program close 3ynNB373Q3VAzKp7m4x238po36hjAGFXFJB4ybN2iTyg --bypass-warning

Note that once a program is closed, the program ID cannot be reused to deploy a new program.

Project File Structure #

Below is an overview of default file structure in an Anchor workspace:

.
├── .anchor
│   └── program-logs
├── app
├── migrations
├── programs
│   └── [project-name]
│       └── src
│           ├── lib.rs
│           ├── Cargo.toml
│           └── Xargo.toml
├── target
│   ├── deploy
│   │   └── [project-name]-keypair.json
│   ├── idl
│   │   └── [project-name].json
│   └── types
│       └── [project-name].ts
├── tests
│   └── [project-name].ts
├── Anchor.toml
├── Cargo.toml
└── package.json

Programs Folder #

The /programs folder contains your project's Anchor programs. A single workspace can contain multiple programs.

Tests Folder #

The /tests folder contains test files for your project. A default test file is created for you when you create your project.

Target Folder #

The /target folder contains build outputs. The main subfolders include:

  • /deploy: Contains the keypair and program binary for your programs.
  • /idl: Contains the JSON IDL for your programs.
  • /types: Contains the TypeScript type for the IDL.

Anchor.toml File #

The Anchor.toml file configures workspace settings for your project.

.anchor Folder #

Includes a program-logs file that contains transaction logs from the last run of test files.

App Folder #

The /app folder is an empty folder that can be optionally used for your frontend code.