Dead simple rust CLI to ease workflows management inside monorepos.
You can install hawk via cargo (homebrew coming soon) or by downloading the latest binary
cargo install hawk-cliRun hawk init to initialize an empty config file. With the --read-from-env flag hawk will try to
retrive your workspaces from pnpm-workspace.yaml or pacakge.json workspaces key.
You can also pass --json if you want to save the config file as json.
hawk 0.1.4
USAGE:
hawk [OPTIONS] [SUBCOMMAND]
OPTIONS:
-c, --config <CONFIG> Specify the config file path
-h, --help Print help information
--scope <SCOPE> Specify which workspaces files copy / watch Usage: --scope
<workspace-name>
-V, --version Print version information
SUBCOMMANDS:
clean Delete generated files
copy Copy files to the `target` directory
help Print this message or the help of the given subcommand(s)
init Initialize a repository
list List workflows in the `target` directoryCheck out the example folder.
Below an example monorepo situation:
example
├── hawk-config.yaml
├── .github
│ └── workflows
│ ├── my-second-app--deploy.yml # name generated by folder
│ └── the-app--deploy.yml # name is read from package.json
└── packages
├── my-app
│ ├── .DS_Store
│ ├── package.json // reads workspace name from package.json (the-app)
│ └── .github/workflows
│ └── deploy.yml
└── my-second-app
└── .github/workflows
└── deploy.yml $ cd example
$ hawk --watch
... let the magic happenGithub actions don't yet support workflows inside subfolders, neither in your .github/workflows/ folder or project custom folders.
So I made hawk to solve this problem without using custom commands. It lets you copy workflows from custom paths and paste them with a prefix, handling most of the pain.
With 10 lines config you have a working monorepo setup.
Download the latest release and move in your $PATH
make sure to have your rust environment ready, then:
- Clone the repo
- Run
cargo build -rormake build - Copy
target/release/hawkto your path or usesudo make install(it will copy the bin into/usr/local/bin) - Enjoy
To setup a new project just run hawk init. If you're in a node environment you can pass the --read-from-env flag to generate config based on the monorepo configuration.
- uses: rawnly/hawk@main
with:
config: hawk-config.yaml- File watching
- Cleanup
workflowsfolder from generated files. - Custom configuration
- Generate config from
pnpm-workspace.yamland yarnspackage.json:workspaces - Create an action to automate this process. (so the user can update a workflow, push and get the generated one updated automatically)
