Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Builds JavaScript projects using PackR and ERB

tree: 6df0645d16

Fetching latest commit…

Octocat-spinner-32-eaf2f5

Cannot retrieve the latest commit at this time

Octocat-spinner-32 bin
Octocat-spinner-32 lib
Octocat-spinner-32 test
Octocat-spinner-32 .gitignore Migrate to gemspec+Bunlder. March 24, 2012
Octocat-spinner-32 .gitmodules Install packr as a submodule while we hack on source maps. March 24, 2012
Octocat-spinner-32 Gemfile Remove packr submodule. April 07, 2012
Octocat-spinner-32 History.txt
Octocat-spinner-32 README.rdoc
Octocat-spinner-32 jake.gemspec
README.rdoc

Jake

Jake is a command line tool for building JavaScript packages from source code. Using simple YAML config files, you can specify any number of build files to be generated by concatenating and minifying groups of source files. It allows ERB to be used inside source files to generate code, and provides event hooks into the build process so you can extend it for your own needs.

Features

  • Concatenate and minify groups of source files for distribution

  • Configured using simple YAML files

  • Easily generate multiple distros with different compression settings

  • Use ERB templating to generate source code

  • Extensible using Ruby/ERB and event hooks during the build process

  • Tested on Ruby 1.8.6, 1.8.7, 1.9.1

Installation

sudo gem install jake

Usage

To begin with, create a file called jake.yml in the root directory of your project; you will run the jake command from here. A basic config looks like this:

---
source_directory:     source
build_directory:      build

layout:               together

header:               COPYRIGHT

builds:
  src:
    packer:           false
  min:
    shrink_vars:      true
    private:          true

packages:
  [ DESCRIBED BELOW ]
  • source_directory is the directory relative to jake.yml where your source files are, and build_directory is where all the generated build files will be placed.

  • layout describes whether files from separate builds should go in separate directories. For example if you have a package called foo, with the above config the together layout will generate build/foo-src.js and build/foo-min.js, whereas a layout value of apart will generate build/src/foo.js and build/min/foo.js.

  • header specifies a file whose content should appear at the top of all generated build files. The content of this file will typically be JavaScript comments containing copyright and license information. This content is never minified. The header option may be omitted.

Build listing

The build listing, given by the builds option in the config file, lists all the builds you want to produce for distribution, and what minification settings each build should use. JavaScript projects typically distribute both compressed and uncompressed copies of their code to suit both production and development environments.

You can have as many builds as you like and the names are up to you. I'm using src and min as readily understood examples. Each build may specify some combination of the following options:

  • packer: false - disables minification for this build. This precludes use of further minification options.

  • shrink_vars: true - tells the minifier to compress local variable names inside functions.

  • private: true - tells the minifier to obfuscate 'private' variables with numeric replacements. JavaScript convention is that any name beginning with an underscore, e.g. _foo or obj._bar should be considered private. They are replaced with _0, _1, etc.

  • base62: true - produces base-62 encoded minification.

  • suffix: false - files from this build should not have a suffix if using the together layout, so you get build/foo.js rather than build/foo-src.js, for example. Only one build may use this option, otherwise file name clashes will occur.

Package listing

The package listing, given under the packages config option, describes the packages you want to produce and which source files are used to generate them. A package is named using the path under build_directory where it should be generated, e.g. foo or ext/awesome (you may omit the .js extension). Each package lists one or more source files used to build it, and may optionally list some extra options as described below.

For the examples, assume the source directory is src and the build directory is dist. This package uses a single source file src/foo.js and generates dist/foo_dist.js:

foo_dist:     foo

This package generates dist/bar.js from src/bar1.js and src/bar2.js

bar:
  - bar1
  - bar2

This generates a package at dist/sub/dir.js from src/path/file.js and src/path/baz.js:

sub/dir:
  - path/file
  - path/baz

If all the source files for a package live in the same subdirectory, you can tidy things up using the directory option. If you use any package-level options, you must list the files under the files option (the above examples are just syntactic shorthands for this):

sub/dir:
  directory:  path
  files:
    - file
    - baz

The full list of package options is as follows:

  • files - lists the source files used to build the package. Shorthand may be used as above if no further options are used.

  • extends - name of another package from which to inherit configuration. Useful for making a package that includes all the files from another, plus a few extras.

  • directory - the directory under source_directory in which to find source files. May be omitted.

  • header - a custom header file to use on this package. Overrides the root header option. May be omitted.

  • packer - lists minification settings that override settings being used for the current build. If a build listed above uses packer: false, this takes precedence over package-specific instructions. Typically used to override options for the minified build.

  • meta - should be a YAML dictionary containing arbitrary data useful to user-defined build events. May be omitted. See 'Event hooks' below.

For example, here's a package listing that uses all the options:

packages:
  foo_dist:         foo

  bar:
    - bar1
    - bar2

  sub/whizz:
    extends:        foo_dist
    directory:      path
    header:         CUSTOM_HEADER
    files:
      - file1
      - file2

  last:
    packer:
      private:      false
    meta:
      requires:
        - jQuery
        - GMap2
    files:
      - one_file
      - another_file

In conjunction with the build options listed above, this matches the following project layout (omitting build name suffixes for brevity):

- build/
    - sub/
        - whizz.js
    - bar.js
    - foo_dist.js
    - last.js
- source/
    - path/
        - CUSTOM_HEADER
        - file1.js
        - file2.js
    - another_file.js
    - bar1.js
    - bar2.js
    - foo.js
    - one_file.js
- COPYRIGHT
- jake.yml

Using ERB in source files

Jake lets you use Ruby's ERB templating system within your source code so you can insert values generated from Ruby functions. To use this feature, you need to create a file called Jakefile in the root of your project. This contains helper functions that are called in your source code to inject data.

For example, say you want to extract a version number from your version control system and inject it into your code along with the build name. Your source code should contain something like this:

MyJavaScriptLib.VERSION = "<%= version %>-<%= build %>";

And your Jakefile should contain a helper called version:

jake_helper :version do
  # extract version number from svn, git, whatever
  # e.g. return '1.0'
end

Jake has a built-in helper called build that returns the current build name. When built, the output would contain the following:

MyJavaScriptLib.VERSION = "1.0-src";    // or "1.0-min" for the 'min' build

Event hooks

The Jakefile may also define event hooks that are fired during a build when interesting things happen. This allows you to extend your build process using configuration data from Jake. All event callbacks are passed a Build object as the first argument, and may receive additional arguments depending on the event type. We currently have two events:

file_created is fired whenever a new build file is created. The callback is passed the Buildable package object, the current build type (src or min using the above examples), and the full path to the newly created file. The package object may contain metadata (set using the meta option, see above) which you can use for further code generation.

build_complete is fired after a build has finished running, that is after all sets of minification options have been run. At this point you can use any metadata you've gathered to generate more code, copy files to your distribution directory, etc.

$register = {}

jake_hook :file_created do |build, pkg, build_type, path|
  $register[path] = pkg.meta
end

jake_hook :build_complete do |build|
  FileUtils.cp 'README', build.build_directory + '/README'
  # generate code from $register
end

License

(The MIT License)

Copyright © 2008-2012 James Coglan

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Something went wrong with that request. Please try again.