Sign Up for Free

RunKit +

Try any Node.js package right in your browser

This is a playground to test code. It runs a full Node.js environment and already has all of npm’s 400,000 packages pre-installed, including blake with all npm packages installed. Try it out:

var blake = require("blake")

This service is provided by RunKit and is not affiliated with npm, Inc or the package authors.

blake v1.0.1

Generate anything

blake - generate anything

The blake Node package provides a file generation pipeline. Originally, I wrote it to generate my blog. Separating IO from data transformation—by using an intermediate representation—blake takes care of IO, and lets you get on with generating your stuff. Of course, gulp puts itself forward as a streaming build system, but if you—like me—experience slight framework fatigue, and prefer plain Node, you might want to give blake a shot.

Build Status


Alas, I don't have a silly example yet, however, after installing blake, you could generate my blog, to see how it works:

git clone
cd troubled
npm install
blake /tmp/troubled

Start an HTTP server in the target directory:

cd /tmp/troubled
python -m SimpleHTTPServer 8081

Point your browser to http://localhost:8081 to inspect the result.


blake [source_directory] target_directory [source_file ...]

If you skip passing the source directory, and give the target only, like in the example, blake will use the current working directory as source.

To generate specific files exclusively, you can pass arbitary source file paths.



Optional String type, which can be String(), null, or undefined.


Optional error type: Error(), null, or undefined.


An object to configure paths.

  • resources str() An optional path for static resources.
  • data String() The path to the content directory.
  • templates String() The folder to load templates from.
  • posts str() An optional posts directory for bloglike sites.
  • target String() The target directory.


Just a bag of things, passed to your view function, you can put anything you want in here; just remember that blake uses following properties internally:

  • template String() The filename of the template.
  • title str() The title of the item or null.
  • date str() This optional date overrides the current date.
  • path str() If this optional target path is not provided, blake will mirror the path of the source file.
  • name str() A target filename to override the default source filename.


The item object, an intermediate representation of each file to generate, forms the core of blake; it is constructed internally and passed to the view function, in which you use its properties to produce the final output.

  • header header() The original header.
  • paths paths() The paths object populated by blake.
  • body str() Everything after the header in the source file.
  • title str() The title of the item.
  • name String() The target file name.
  • date Date() The current date or the date set in header().
  • templatePath String() The absolute path to the template file.
  • path The absolute target file path.
  • link The relative target file path applicable as local link on the site.
  • template Buffer() The template data.
  • read read()

read(path, cb)

Reads all source files in the given path recursively to apply the callback with an optional error and the resulting items—handy to create archives, feeds, etc.

  • path String()
  • cb Function(er, items)
    • er err() An optional Error.
    • items [item()] The resulting items.


An object that maps view functions by template filename.


The blake module exports the constructor of the Blake Transform stream. To use it do require('blake'). A Blake stream has two additional getters providing access to parts of the configuration:

  • resources str() The optional path to the resources directory.
  • data String() The path to the data source directory.

This constructor—the solely exported function by the module—is decorated with two stateless functions, given var blake = require('blake'):


Returns a readable stream of all filenames in the given directory. The filenames are read recursively, directory names are skipped.

  • path String() The path of the directory to read.

blake.copy(source, target)

Recursively copies all files from the source directory to the target directory; returns a readable stream of the copied target filenames.

  • source String() The source directory.
  • target String() The target directory.

Configuring a site

The source directory has to contain a source module, which has to export paths() and views():

exports.paths = {
  data: 'data',
  templates: 'templates',
  resources: 'resources',
  posts: 'data/posts'

// Associate your view functions with template names.
exports.views = {
  'rss.jade': rss,
  'article.jade': article,
  'home.jade': home,
  'about.jade': about,
  'error.jade': about,
  'archive.jade': archive,
  'likes.jade': likes,
  'tweet.jade': tweet

Writing a view

view (item, cb)

  • item item()
  • cb Function(error, result)
    • error err() Pass an error if something went wrong.
    • result Buffer() The resulting artifact generated by this view.

Each item() is associated with a view function by a template name in the configuration. This function—implemented by you—is responsible to generate the artifact. It is in this function where the actual work is done. Here you use values from the item and your template, also provided via the item, to generate the final output, which you apply to the callback once you are done—so it can be written to disk by blake.

Creating a new instance

blake(source, target)

  • source str() The source directory.
  • target String() The target directory.
var blake = require('blake')
var b = blake('./test/data', '/tmp/blake-example')

Copying static resources

blake.copy(resources, target)

  • resources String() A directory containing static resources.
  • target String() The target directory.
var b = blake(source, target)
blake.copy(b.resources, target)

Generating a site

For a complete build, you would typically generate the site after copying static resources:

var b = blake(source, target)
blake.copy(b.resources, target).on('end', function () {

Generating specific files

Since blake is a Transform stream, you can easily generate only specific files:

var b = blake(source, target)


With npm do:

npm install blake

To use the command-line interface:

npm install -g blake


MIT License


RunKit is a free, in-browser JavaScript dev environment for prototyping Node.js code, with every npm package installed. Sign up to share your code.
Sign Up for Free