pandahook

PandaHook

The Ultimate Tool to Manage and Deploy Githook Scripts.

---

Githooks are powerful tools. Whatever you can script can be triggered from nothing but a git command. PandaHook helps you manage this magic.

PandaHook is designed to not only assist you in generating githook scripts, but also setting them up in a remote server. PandaHook is meant to be your Swiss Army knife, so it features a modular design with multiple sub-commands to keep the codebase manageable and future-friendly.

Installation

PandaHook is easily installed via npm. However, how you install PandaHook depends on how you'd like to use it. You may either use PandaHook as a command-line tool or install it as a dependency into your project. In either case, you will need CoffeeScript installed.

Command-Line Tool

If you'd like to use PandaHook's command-line tool on your local machine, install it globally.

npm install -g coffee-script
npm install -g pandahook

This gives you a symlinked executable to invoke on your command-line. See Command-Line Guide below for more information on this executable.

Node Library

If you would like to install PandaCluster as a library and programmatically access its methods, install it locally to your project.

npm install -g coffee-script
npm install --save pandahook

This places the PandaCluster Node module into your project and in your package.json file as a dependency. See API Guide below for more information on programatic access.

Command-Line Guide

The command-line tool is accessed via several sub-commands. Information is available at any time by placing "help" or "-h" after most commands. Here is a list of currently available sub-commands.

--------------------------------------------
Usage: pandahook COMMAND [arg...]
--------------------------------------------
Follow any command with "help" for more information.

A tool to manage githook scripts and deploy them to your hook-server.

Commands:
build     Generates a githook script
create    Clones a remote, bare repo on the hook-server
destroy   Deletes a remote repo from the hook-server
push      Adds the specified githook script to remote repo on the hook-server
rm        Deletes the specifed githook script from the remote repo on the hook-server

In particular, the build sub-command is meant to be flexible and powerful. Please see build_subcommand.md for more information.

Configuration Dotfile

Reusable configuration data is stored in the dotfile .pandahook.cson. This keeps you from having to re-type the same data repeatedly into commands. This data must be provided in your code if you plan to access the library programmatically. Here is a sample file layout:

# Required Hook-Server Stanza
hook_server:
  address: "[e-mail]"   # This is an SSH connection, not HTTPS

  # Optional Stanzas for Target Services.  We will focus on CoreOS here.
  coreos:
    address: "myCoreOSCluster.com"

API Guide

To keep this ReadMe short, the API documentation has been placed into a separate file. See the file API Guide.md for complete information.

Usage Example

To see how PandaHook is used, you'll need a couple things:

1. The hook-server, a remote server to host your server-side githooks.
2. The configuration dotfile, .pandahook.cson, completed for your use-case and placed into your local $HOME directory (ie, at ~/.pandahook.cson). See below for more details.
3. A git repository on your local machine, from which you will push to the hook-server.
4. The service targeted by your githook, running on a second remote server out there somewhere on the Internet. Our initial examples will target services running inside CoreOS clusters.

You can see a Hello World example of PandaHook usage in this repository.