henrik-jekyll 0.5.0

data/lib/jekyll/convertible.rb

@@ -0,0 +1,82 @@
+# Convertible provides methods for converting a pagelike item
+# from a certain type of markup into actual content
+#
+# Requires
+#   self.site -> Jekyll::Site
+module Jekyll
+  module Convertible
+    # Return the contents as a string
+    def to_s
+      self.content || ''
+    end
+
+    # Read the YAML frontmatter
+    #   +base+ is the String path to the dir containing the file
+    #   +name+ is the String filename of the file
+    #
+    # Returns nothing
+    def read_yaml(base, name)
+      self.content = File.read(File.join(base, name))
+
+      if self.content =~ /^(---\s*\n.*?)\n---\s*\n/m
+        self.content = self.content[($1.size + 5)..-1]
+
+        self.data = YAML.load($1)
+      end
+    end
+
+    # Transform the contents based on the file extension.
+    #
+    # Returns nothing
+    def transform
+      case self.content_type
+      when 'textile'
+        self.ext = ".html"
+        self.content = self.site.textile(self.content)
+      when 'markdown'
+        self.ext = ".html"
+        self.content = self.site.markdown(self.content)
+      end
+    end
+
+    # Determine which formatting engine to use based on this convertible's
+    # extension
+    #
+    # Returns one of :textile, :markdown or :unknown
+    def content_type
+      case self.ext[1..-1]
+      when /textile/i
+        return 'textile'
+      when /markdown/i, /mkdn/i, /md/i
+        return 'markdown'
+      end
+      return 'unknown'
+    end
+
+    # Add any necessary layouts to this convertible document
+    #   +layouts+ is a Hash of {"name" => "layout"}
+    #   +site_payload+ is the site payload hash
+    #
+    # Returns nothing
+    def do_layout(payload, layouts)
+      info = { :filters => [Jekyll::Filters], :registers => { :site => self.site } }
+
+      # render and transform content (this becomes the final content of the object)
+      payload["content_type"] = self.content_type
+      self.content = Liquid::Template.parse(self.content).render(payload, info)
+      self.transform
+
+      # output keeps track of what will finally be written
+      self.output = self.content
+
+      # recursively render layouts
+      layout = layouts[self.data["layout"]]
+      while layout
+        payload = payload.deep_merge({"content" => self.output, "page" => layout.data})
+        self.output = Liquid::Template.parse(layout.content).render(payload, info)
+
+        layout = layouts[layout.data["layout"]]
+      end
+    end
+  end
+end

data/lib/jekyll/core_ext.rb

@@ -0,0 +1,22 @@
+class Hash
+  # Merges self with another hash, recursively.
+  #
+  # This code was lovingly stolen from some random gem:
+  # http://gemjack.com/gems/tartan-0.1.1/classes/Hash.html
+  #
+  # Thanks to whoever made it.
+  def deep_merge(hash)
+    target = dup
+
+    hash.keys.each do |key|
+      if hash[key].is_a? Hash and self[key].is_a? Hash
+        target[key] = target[key].deep_merge(hash[key])
+        next
+      end
+
+      target[key] = hash[key]
+    end
+
+    target
+  end
+end

data/lib/jekyll/filters.rb

@@ -0,0 +1,43 @@
+module Jekyll
+
+  module Filters
+    def textilize(input)
+      RedCloth.new(input).to_html
+    end
+
+    def date_to_string(date)
+      date.strftime("%d %b %Y")
+    end
+
+    def date_to_long_string(date)
+      date.strftime("%d %B %Y")
+    end
+
+    def date_to_xmlschema(date)
+      date.xmlschema
+    end
+
+    def xml_escape(input)
+      input.gsub("&", "&amp;").gsub("<", "&lt;").gsub(">", "&gt;")
+    end
+
+    def number_of_words(input)
+      input.split.length
+    end
+
+    def array_to_sentence_string(array)
+      connector = "and"
+      case array.length
+      when 0
+        ""
+      when 1
+        array[0].to_s
+      when 2
+        "#{array[0]} #{connector} #{array[1]}"
+      else
+        "#{array[0...-1].join(', ')}, #{connector} #{array[-1]}"
+      end
+    end
+
+  end
+end

data/lib/jekyll/layout.rb

@@ -0,0 +1,36 @@
+module Jekyll
+
+  class Layout
+    include Convertible
+
+    attr_accessor :site
+    attr_accessor :ext
+    attr_accessor :data, :content
+
+    # Initialize a new Layout.
+    #   +site+ is the Site
+    #   +base+ is the String path to the <source>
+    #   +name+ is the String filename of the post file
+    #
+    # Returns <Page>
+    def initialize(site, base, name)
+      @site = site
+      @base = base
+      @name = name
+
+      self.data = {}
+
+      self.process(name)
+      self.read_yaml(base, name)
+    end
+
+    # Extract information from the layout filename
+    #   +name+ is the String filename of the layout file
+    #
+    # Returns nothing
+    def process(name)
+      self.ext = File.extname(name)
+    end
+  end
+
+end

data/lib/jekyll/page.rb

@@ -0,0 +1,67 @@
+module Jekyll
+
+  class Page
+    include Convertible
+
+    attr_accessor :site
+    attr_accessor :ext
+    attr_accessor :data, :content, :output
+
+    # Initialize a new Page.
+    #   +site+ is the Site
+    #   +base+ is the String path to the <source>
+    #   +dir+ is the String path between <source> and the file
+    #   +name+ is the String filename of the file
+    #
+    # Returns <Page>
+    def initialize(site, base, dir, name)
+      @site = site
+      @base = base
+      @dir = dir
+      @name = name
+
+      self.data = {}
+
+      self.process(name)
+      self.read_yaml(File.join(base, dir), name)
+      #self.transform
+    end
+
+    # Extract information from the page filename
+    #   +name+ is the String filename of the page file
+    #
+    # Returns nothing
+    def process(name)
+      self.ext = File.extname(name)
+    end
+
+    # Add any necessary layouts to this post
+    #   +layouts+ is a Hash of {"name" => "layout"}
+    #   +site_payload+ is the site payload hash
+    #
+    # Returns nothing
+    def render(layouts, site_payload)
+      payload = {"page" => self.data}.deep_merge(site_payload)
+      do_layout(payload, layouts)
+    end
+
+    # Write the generated page file to the destination directory.
+    #   +dest+ is the String path to the destination dir
+    #
+    # Returns nothing
+    def write(dest)
+      FileUtils.mkdir_p(File.join(dest, @dir))
+
+      name = @name
+      if self.ext != ""
+        name = @name.split(".")[0..-2].join('.') + self.ext
+      end
+
+      path = File.join(dest, @dir, name)
+      File.open(path, 'w') do |f|
+        f.write(self.output)
+      end
+    end
+  end
+
+end

data/lib/jekyll/post.rb

@@ -0,0 +1,205 @@
+module Jekyll
+
+  class Post
+    include Comparable
+    include Convertible
+
+    class << self
+      attr_accessor :lsi
+    end
+
+    MATCHER = /^(.+\/)*(\d+-\d+-\d+)-(.*)(\.[^.]+)$/
+
+    # Post name validator. Post filenames must be like:
+    #   2008-11-05-my-awesome-post.textile
+    #
+    # Returns <Bool>
+    def self.valid?(name)
+      name =~ MATCHER
+    end
+
+    attr_accessor :site
+    attr_accessor :date, :slug, :ext, :categories, :topics, :published
+    attr_accessor :data, :content, :output
+
+    # Initialize this Post instance.
+    #   +site+ is the Site
+    #   +base+ is the String path to the dir containing the post file
+    #   +name+ is the String filename of the post file
+    #   +categories+ is an Array of Strings for the categories for this post
+    #
+    # Returns <Post>
+    def initialize(site, source, dir, name)
+      @site = site
+      @base = File.join(source, dir, '_posts')
+      @name = name
+
+      self.categories = dir.split('/').reject { |x| x.empty? }
+
+      parts = name.split('/')
+      self.topics = parts.size > 1 ? parts[0..-2] : []
+
+      self.process(name)
+      self.read_yaml(@base, name)
+
+      if self.data.has_key?('published') && self.data['published'] == false
+        self.published = false
+      else
+        self.published = true
+      end
+
+      if self.categories.empty?
+        if self.data.has_key?('category')
+          self.categories << self.data['category']
+        elsif self.data.has_key?('categories')
+          # Look for categories in the YAML-header, either specified as
+          # an array or a string.
+          if self.data['categories'].kind_of? String
+            self.categories = self.data['categories'].split
+          else
+            self.categories = self.data['categories']
+          end
+        end
+      end
+    end
+
+    # Spaceship is based on Post#date
+    #
+    # Returns -1, 0, 1
+    def <=>(other)
+      self.date <=> other.date
+    end
+
+    # Extract information from the post filename
+    #   +name+ is the String filename of the post file
+    #
+    # Returns nothing
+    def process(name)
+      m, cats, date, slug, ext = *name.match(MATCHER)
+      self.date = Time.parse(date)
+      self.slug = slug
+      self.ext = ext
+    end
+
+    # The generated directory into which the post will be placed
+    # upon generation. This is derived from the permalink or, if
+    # permalink is absent, set to the default date
+    # e.g. "/2008/11/05/" if the permalink style is :date, otherwise nothing
+    #
+    # Returns <String>
+    def dir
+      if permalink
+        permalink.to_s.split("/")[0..-2].join("/") + '/'
+      else
+        prefix = self.categories.empty? ? '' : '/' + self.categories.join('/')
+        if [:date, :pretty].include?(self.site.permalink_style)
+          prefix + date.strftime("/%Y/%m/%d/")
+        else
+          prefix + '/'
+        end
+      end
+    end
+
+    # The full path and filename of the post.
+    # Defined in the YAML of the post body
+    # (Optional)
+    #
+    # Returns <String>
+    def permalink
+      self.data && self.data['permalink']
+    end
+
+    # The generated relative url of this post
+    # e.g. /2008/11/05/my-awesome-post.html
+    #
+    # Returns <String>
+    def url
+      ext = self.site.permalink_style == :pretty ? '' : '.html'
+      permalink || self.id + ext
+    end
+
+    # The UID for this post (useful in feeds)
+    # e.g. /2008/11/05/my-awesome-post
+    #
+    # Returns <String>
+    def id
+      self.dir + self.slug
+    end
+
+    # Calculate related posts.
+    #
+    # Returns [<Post>]
+    def related_posts(posts)
+      return [] unless posts.size > 1
+
+      if self.site.lsi
+        self.class.lsi ||= begin
+          puts "Running the classifier... this could take a while."
+          lsi = Classifier::LSI.new
+          posts.each { |x| $stdout.print(".");$stdout.flush;lsi.add_item(x) }
+          puts ""
+          lsi
+        end
+
+        related = self.class.lsi.find_related(self.content, 11)
+        related - [self]
+      else
+        (posts - [self])[0..9]
+      end
+    end
+
+    # Add any necessary layouts to this post
+    #   +layouts+ is a Hash of {"name" => "layout"}
+    #   +site_payload+ is the site payload hash
+    #
+    # Returns nothing
+    def render(layouts, site_payload)
+      # construct payload
+      payload =
+      {
+        "site" => { "related_posts" => related_posts(site_payload["site"]["posts"]) },
+        "page" => self.to_liquid
+      }
+      payload = payload.deep_merge(site_payload)
+
+      do_layout(payload, layouts)
+    end
+
+    # Write the generated post file to the destination directory.
+    #   +dest+ is the String path to the destination dir
+    #
+    # Returns nothing
+    def write(dest)
+      FileUtils.mkdir_p(File.join(dest, dir))
+
+      path = File.join(dest, self.url)
+
+      if self.site.permalink_style == :pretty
+        FileUtils.mkdir_p(path)
+        path = File.join(path, "index.html")
+      end
+
+      File.open(path, 'w') do |f|
+        f.write(self.output)
+      end
+    end
+
+    # Convert this post into a Hash for use in Liquid templates.
+    #
+    # Returns <Hash>
+    def to_liquid
+      { "title" => self.data["title"] || self.slug.split('-').select {|w| w.capitalize! || w }.join(' '),
+        "url" => self.url,
+        "date" => self.date,
+        "id" => self.id,
+        "topics" => self.topics,
+        "categories" => self.categories,
+        "content" => self.content }.deep_merge(self.data)
+    end
+
+    def inspect
+      "<Post: #{self.id}>"
+    end
+  end
+
+end

data/lib/jekyll/site.rb

@@ -0,0 +1,238 @@
+module Jekyll
+
+  class Site
+    attr_accessor :config, :layouts, :posts, :categories
+    attr_accessor :source, :dest, :lsi, :pygments, :permalink_style
+
+    # Initialize the site
+    #   +config+ is a Hash containing site configurations details
+    #
+    # Returns <Site>
+    def initialize(config)
+      self.config          = config.clone
+
+      self.source          = config['source']
+      self.dest            = config['destination']
+      self.lsi             = config['lsi']
+      self.pygments        = config['pygments']
+      self.permalink_style = config['permalink'].to_sym
+
+      self.reset
+      self.setup
+    end
+
+    def reset
+      self.layouts         = {}
+      self.posts           = []
+      self.categories      = Hash.new { |hash, key| hash[key] = Array.new }
+    end
+
+    def setup
+      # Check to see if LSI is enabled.
+      require 'classifier' if self.lsi
+
+      # Set the Markdown interpreter (and Maruku self.config, if necessary)
+      case self.config['markdown']
+        when 'rdiscount'
+          begin
+            require 'rdiscount'
+
+            def markdown(content)
+              RDiscount.new(content).to_html
+            end
+
+            puts 'Using rdiscount for Markdown'
+          rescue LoadError
+            puts 'You must have the rdiscount gem installed first'
+          end
+        when 'maruku'
+          begin
+            require 'maruku'
+
+            def markdown(content)
+              Maruku.new(content).to_html
+            end
+
+            if self.config['maruku']['use_divs']
+              require 'maruku/ext/div'
+              puts 'Maruku: Using extended syntax for div elements.'
+            end
+
+            if self.config['maruku']['use_tex']
+              require 'maruku/ext/math'
+              puts "Maruku: Using LaTeX extension. Images in `#{self.config['maruku']['png_dir']}`."
+
+              # Switch off MathML output
+              MaRuKu::Globals[:html_math_output_mathml] = false
+              MaRuKu::Globals[:html_math_engine] = 'none'
+
+              # Turn on math to PNG support with blahtex
+              # Resulting PNGs stored in `images/latex`
+              MaRuKu::Globals[:html_math_output_png] = true
+              MaRuKu::Globals[:html_png_engine] =  self.config['maruku']['png_engine']
+              MaRuKu::Globals[:html_png_dir] = self.config['maruku']['png_dir']
+              MaRuKu::Globals[:html_png_url] = self.config['maruku']['png_url']
+            end
+          rescue LoadError
+            puts "The maruku gem is required for markdown support!"
+          end
+      end
+    end
+
+    def textile(content)
+      RedCloth.new(content).to_html
+    end
+
+    # Do the actual work of processing the site and generating the
+    # real deal.
+    #
+    # Returns nothing
+    def process
+      self.reset
+      self.read_layouts
+      self.transform_pages
+      self.write_posts
+    end
+
+    # Read all the files in <source>/_layouts into memory for later use.
+    #
+    # Returns nothing
+    def read_layouts
+      base = File.join(self.source, "_layouts")
+      entries = []
+      Dir.chdir(base) { entries = filter_entries(Dir['*.*']) }
+
+      entries.each do |f|
+        name = f.split(".")[0..-2].join(".")
+        self.layouts[name] = Layout.new(self, base, f)
+      end
+    rescue Errno::ENOENT => e
+      # ignore missing layout dir
+    end
+
+    # Read all the files in <base>/_posts and create a new Post object with each one.
+    #
+    # Returns nothing
+    def read_posts(dir)
+      base = File.join(self.source, dir, '_posts')
+      entries = []
+      Dir.chdir(base) { entries = filter_entries(Dir['**/*']) }
+
+      # first pass processes, but does not yet render post content
+      entries.each do |f|
+        if Post.valid?(f)
+          post = Post.new(self, self.source, dir, f)
+
+          if post.published
+            self.posts << post
+            post.categories.each { |c| self.categories[c] << post }
+          end
+        end
+      end
+
+      # second pass renders each post now that full site payload is available
+      self.posts.each do |post|
+        post.render(self.layouts, site_payload)
+      end
+
+      self.posts.sort!
+      self.categories.values.map { |cats| cats.sort! { |a, b| b <=> a} }
+    rescue Errno::ENOENT => e
+      # ignore missing layout dir
+    end
+
+    # Write each post to <dest>/<year>/<month>/<day>/<slug>
+    #
+    # Returns nothing
+    def write_posts
+      self.posts.each do |post|
+        post.write(self.dest)
+      end
+    end
+
+    # Copy all regular files from <source> to <dest>/ ignoring
+    # any files/directories that are hidden or backup files (start
+    # with "." or "#" or end with "~") or contain site content (start with "_")
+    # unless they are "_posts" directories or web server files such as
+    # '.htaccess'
+    #   The +dir+ String is a relative path used to call this method
+    #            recursively as it descends through directories
+    #
+    # Returns nothing
+    def transform_pages(dir = '')
+      base = File.join(self.source, dir)
+      entries = filter_entries(Dir.entries(base))
+      directories = entries.select { |e| File.directory?(File.join(base, e)) }
+      files = entries.reject { |e| File.directory?(File.join(base, e)) }
+
+      # we need to make sure to process _posts *first* otherwise they
+      # might not be available yet to other templates as {{ site.posts }}
+      if directories.include?('_posts')
+        directories.delete('_posts')
+        read_posts(dir)
+      end
+      [directories, files].each do |entries|
+        entries.each do |f|
+          if File.directory?(File.join(base, f))
+            next if self.dest.sub(/\/$/, '') == File.join(base, f)
+            transform_pages(File.join(dir, f))
+          else
+            first3 = File.open(File.join(self.source, dir, f)) { |fd| fd.read(3) }
+
+            if first3 == "---"
+              # file appears to have a YAML header so process it as a page
+              page = Page.new(self, self.source, dir, f)
+              page.render(self.layouts, site_payload)
+              page.write(self.dest)
+            else
+              # otherwise copy the file without transforming it
+              FileUtils.mkdir_p(File.join(self.dest, dir))
+              FileUtils.cp(File.join(self.source, dir, f), File.join(self.dest, dir, f))
+            end
+          end
+        end
+      end
+    end
+
+    # Constructs a hash map of Posts indexed by the specified Post attribute
+    #
+    # Returns {post_attr => [<Post>]}
+    def post_attr_hash(post_attr)
+      # Build a hash map based on the specified post attribute ( post attr => array of posts )
+      # then sort each array in reverse order
+      hash = Hash.new { |hash, key| hash[key] = Array.new }
+      self.posts.each { |p| p.send(post_attr.to_sym).each { |t| hash[t] << p } }
+      hash.values.map { |sortme| sortme.sort! { |a, b| b <=> a} }
+      return hash
+    end
+
+    # The Hash payload containing site-wide data
+    #
+    # Returns {"site" => {"time" => <Time>,
+    #                     "posts" => [<Post>],
+    #                     "categories" => [<Post>],
+    #                     "topics" => [<Post>] }}
+    def site_payload
+      {"site" => {
+        "time" => Time.now,
+        "posts" => self.posts.sort { |a,b| b <=> a },
+        "categories" => post_attr_hash('categories'),
+        "topics" => post_attr_hash('topics')
+      }}
+    end
+
+    # Filter out any files/directories that are hidden or backup files (start
+    # with "." or "#" or end with "~") or contain site content (start with "_")
+    # unless they are "_posts" directories or web server files such as
+    # '.htaccess'
+    def filter_entries(entries)
+      entries = entries.reject do |e|
+        unless ['_posts', '.htaccess'].include?(e)
+          # Reject backup/hidden
+          ['.', '_', '#'].include?(e[0..0]) or e[-1..-1] == '~'
+        end
+      end
+    end
+
+  end
+end