Glue is a simple static site generator for Ruby that uses Haml, Sass and Textile/Markdown.

You can create static website so easily that you’ll imagine how could you live without it!
Glue has only what you’ll need:

Getting started

To install Glue, you’ll need a working Ruby installation with Rubygems. If you’re cool with it, just run the following command to install it. 

sudo gem install glue-site

After installing Glue and its dependencies, you can create your site so easily! Just run the following command from your terminal window. 

$ glue mysite
creating mysite directory... done!
creating public directory... done!
creating public/javascripts directory... done!
creating public/images directory... done!
creating views/layouts directory... done!
creating config directory... done!
 copying config/glue.yml... done!
 copying views/stylesheets/style.sass... done!
 copying Rakefile... done!
 copying config/helper.rb... done!
 copying views/layouts/main.haml... done!
 copying views/index.haml... done!
 copying views/404.haml... done!
 copying views/_footer.haml... done!
 copying public/.htaccess... done!
 copying public/robots.txt... done!

This command will create a mysite directory with all the Glue structure. Now you're ready to configure your Glue site.

Configuring

All your site settings are defined in config/glue.yml. There are 3 main blocks: site, sitemap and feed. 

site:
  # Set default page title.
  title: "The awesome Glue!"

  # Set default page description
  description: "Glue is a simple static site generator for Ruby that uses Haml, Sass and Textile/Markdown."

  # Set default keywords
  keywords: "glue, ruby, haml, sass, static site, textile, markdown"

  # Set your base url.
  # Include domain and path (if needed).
  # Domain is required for sitemap generation.
  base_url: http://fnando.github.com/

  # If you're using Apache you can enable `MultiViews`
  # or mod_rewrite and access your URL like
  # http://example.com/about instead of http://example.com/about.html
  friendly_url: true

  # Set your language for feeds and meta tags
  language: en-us

sitemap:
  # Do generate sitemap
  enabled: true

  # Filters will be applied against full filename.
  # Use .*? to match all. Use regular expressions.
  filters:
    - .*?

feed:
  # Do generate feed
  enabled: true

  # Filters will be applied against full filename.
  # Use .*? to match all. Use regular expressions.
  filters:
    - .*?

Views

Views are placed in the "views" directory. All .haml files will be exported as .html, respecting the structure you define.

So, if you create a file views/index.haml, a correspondent public/index.html will be exported; if you create views/articles/using-glue.haml, Glue will export it as public/articles/using-glue.html, and so on.

Meta data

To specify layouts, titles, descriptions and any other value you want, your document must have meta data. Meta data is nothing but Haml comments. Here's the meta data from this page you're viewing: 

/ title: The awesome Glue!
/ description: Glue is a simple static site generator for Ruby that uses Haml, Sass and Textile/Markdown.
/ keywords: Glue, Ruby, Static Site, Haml, Sass, Markdown, Textile
/ layout: main
/ id: glue
/ tagline: Static site generator
/ version: 0.0.5

The are some meta data that is Glue-reserved.

layout
Specify the layout your page is going to use. Layouts must be placed on views/layouts. If you don't want to use a layout, set it to none.
index
When set to yes, define if the current page should also exported as public/index.html. The default is no.
sitemap
When set to no, the page is not added to public/sitemap.xml. The default is yes.
feed
When set to no, the page is not added to public/feed.xml. The default is yes.

Using helpers

There are some helpers that comes with Glue:

markdown(text)
Parse a given text as Markdown code. 
%p
  :markdown
    This text will be converted to HTML using
    [RDiscount](http://github.com/rtomayko/rdiscount/tree/master).
You can call it directly: 
%p= markdown("*Markdown* rules!")
textile(text)
Parse a given text as Textile code. 
%p
  :textile
    This text will be converted to HTML using
    "RedCloth":http://redcloth.org/.
You can call it directly: 
%p= textile("*Textile* rules!")
haml(text)
Parse a given text as Haml code. You can call it directly: 
%p= haml("%strong Haml\nis so nice!")
url_for(path, options={})
Return an URL considering the base_url configuration option. If you want to retrieve a full URL (that include the host), set the option :full to true. 
%p= url_for("feed.xml", :full => true)
feed_tag
Return the link tag pointing to your feed.xml. 
= feed_tag
render(path)
Render will lookup for the path your provide, according to the file extension. The example below will search for footer.haml, footer.markdown and then footer.textile, rendering the first it founds. 
= render("footer")

To add your own helpers, use the file config/helper. Let's create a helper that reverses a given text. 

module Helper
  def reversed(text)
    text.reverse
  end
end

Just call it from your view like this: 

%p
  = reversed("Some cool text!")

Notice that you need to return a string. DO NOT output directly; use the equal sign like in = reversed("Some cool text!")

If you want to output directly, use the haml_concat method in your helper. 

module Helper
  def reversed(text)
    haml_concat text.reverse
  end
end

Then you can just call the helper. 

%p
  - reversed("Some cool text!")

Capistrano & Dreamhost

Here's a simple deploy task for Capistrano & Dreamhost. First, create a new domain/subdomain pointing to the public directory.

Then run capify from the root directory. Edit the config/deploy.rb file, pasting the sample below. 

set :application, "howto"
set :scm, :git
set :repository,  "git@github.com:USER/some-repo.git"
set :deploy_to, "/home/USER/example.com"
set :user, "USER"
set :use_sudo, false

role :web, "example.com"

Now, run cap deploy:setup to prepare your server. To update your site, just run cap deploy:update.

Could not be easier!

Maintainer

This project was created by Nando Vieira.

Check it out another interesting projects at http://fnando.github.com.