provide customizable sanitization

lib/gollum.rb

@@ -18,54 +18,13 @@ require 'gollum/page'
 require 'gollum/file'
 require 'gollum/markup'
 require 'gollum/albino'
+require 'gollum/sanitization'
 
 module Gollum
   VERSION = '1.0.1'
 
-  SANITIZATION_OPTIONS = {
-    :elements => [
-      'a', 'abbr', 'acronym', 'address', 'area', 'b', 'big',
-      'blockquote', 'br', 'button', 'caption', 'center', 'cite',
-      'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir',
-      'div', 'dl', 'dt', 'em', 'fieldset', 'font', 'form', 'h1',
-      'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'img', 'input',
-      'ins', 'kbd', 'label', 'legend', 'li', 'map', 'menu',
-      'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp',
-      'select', 'small', 'span', 'strike', 'strong', 'sub',
-      'sup', 'table', 'tbody', 'td', 'textarea', 'tfoot', 'th',
-      'thead', 'tr', 'tt', 'u', 'ul', 'var'
-    ],
-    :attributes => {
-      :all => ['abbr', 'accept', 'accept-charset',
-               'accesskey', 'action', 'align', 'alt', 'axis',
-               'border', 'cellpadding', 'cellspacing', 'char',
-               'charoff', 'charset', 'checked', 'cite',
-               'class', 'clear', 'cols', 'colspan', 'color',
-               'compact', 'coords', 'datetime', 'dir',
-               'disabled', 'enctype', 'for', 'frame',
-               'headers', 'height', 'href', 'hreflang',
-               'hspace', 'id', 'ismap', 'label', 'lang',
-               'longdesc', 'maxlength', 'media', 'method',
-               'multiple', 'name', 'nohref', 'noshade',
-               'nowrap', 'prompt', 'readonly', 'rel', 'rev',
-               'rows', 'rowspan', 'rules', 'scope',
-               'selected', 'shape', 'size', 'span', 'src',
-               'start', 'summary', 'tabindex', 'target',
-               'title', 'type', 'usemap', 'valign', 'value',
-               'vspace', 'width']
-    },
-    :protocols => {
-      'a' => {'href' => ['http', 'https', 'mailto', :relative]},
-      'img' => {'href' => ['http', 'https', :relative]}
-    }
-  }
-  HISTORY_SANITIZATION_OPTIONS = SANITIZATION_OPTIONS.merge(
-    :add_attributes => {
-      'a' => {'rel' => 'nofollow'}
-    }
-  )
-
   class Error < StandardError; end
+
   class DuplicatePageError < Error
     attr_accessor :dir
     attr_accessor :existing_path

lib/gollum/markup.rb

@@ -27,9 +27,9 @@ module Gollum
     #
     # Returns the formatted String content.
     def render(no_follow = false)
-      sanitize_options = no_follow   ? 
+      sanitize_options = (
-        HISTORY_SANITIZATION_OPTIONS : 
+        no_follow ? @wiki.history_sanitization : @wiki.sanitization
-        SANITIZATION_OPTIONS
+      ).to_h
       data = extract_tex(@data)
       data = extract_code(data)
       data = extract_tags(data)

lib/gollum/sanitization.rb

@@ -0,0 +1,150 @@
+module Gollum
+
+  # Encapsulate sanitization options.
+  #
+  # This class does not yet support all options of Sanitize library.
+  # See http://github.com/rgrove/sanitize/.
+  class Sanitization
+
+    #
+    ELEMENTS = [
+      'a', 'abbr', 'acronym', 'address', 'area', 'b', 'big',
+      'blockquote', 'br', 'button', 'caption', 'center', 'cite',
+      'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir',
+      'div', 'dl', 'dt', 'em', 'fieldset', 'font', 'form', 'h1',
+      'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'img', 'input',
+      'ins', 'kbd', 'label', 'legend', 'li', 'map', 'menu',
+      'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp',
+      'select', 'small', 'span', 'strike', 'strong', 'sub',
+      'sup', 'table', 'tbody', 'td', 'textarea', 'tfoot', 'th',
+      'thead', 'tr', 'tt', 'u', 'ul', 'var'
+    ]
+
+    #
+    ATTRIBUTES = {
+      :all => [
+        'abbr', 'accept', 'accept-charset',
+        'accesskey', 'action', 'align', 'alt', 'axis',
+        'border', 'cellpadding', 'cellspacing', 'char',
+        'charoff', 'charset', 'checked', 'cite',
+        'class', 'clear', 'cols', 'colspan', 'color',
+        'compact', 'coords', 'datetime', 'dir',
+        'disabled', 'enctype', 'for', 'frame',
+        'headers', 'height', 'href', 'hreflang',
+        'hspace', 'id', 'ismap', 'label', 'lang',
+        'longdesc', 'maxlength', 'media', 'method',
+        'multiple', 'name', 'nohref', 'noshade',
+        'nowrap', 'prompt', 'readonly', 'rel', 'rev',
+        'rows', 'rowspan', 'rules', 'scope',
+        'selected', 'shape', 'size', 'span', 'src',
+        'start', 'summary', 'tabindex', 'target',
+        'title', 'type', 'usemap', 'valign', 'value',
+        'vspace', 'width'
+      ]
+    }
+
+    #
+    PROTOCOLS = {
+      'a' => {'href' => ['http', 'https', 'mailto', :relative]},
+      'img' => {'href' => ['http', 'https', :relative]}
+    }
+
+    #
+    ADD_ATTRIBUTES = {
+      'a' => {'rel' => 'nofollow'}
+    }
+
+    # Additional options specifically for histories.
+    HISTORY_OPTIONS = {
+      :add_attributes => {
+        'a' => {'rel' => 'nofollow'}
+      }
+    }
+
+    #
+    def initialize(*settings)
+      if settings.empty?
+        @elements   = ELEMENTS
+        @attributes = ATTRIBUTES
+        @protocols  = PROTOCOLS
+        @add_attributes = {}
+        @allow_comments = false
+      else
+        @elements   = []
+        @attributes = {}
+        @protocols  = {}
+        @add_attributes = {}
+        @allow_comments = false
+        settings.each{ |s| merge!(s) }
+      end
+    end
+
+    def elements
+      @elements
+    end
+
+    def attributes
+      @attributes
+    end
+
+    def protocols
+      @protocols
+    end
+
+    def add_attributes
+      @add_attributes
+    end
+
+    def allow_comments
+      !!@allow_comments
+    end
+
+    #
+    def to_h
+      { :elements       => elements,
+        :attributes     => attributes,
+        :protocols      => protocols,
+        :add_attributes => add_attributes,
+        :allow_comments => allow_comments
+      }
+    end
+
+    #
+    def merge(settings)
+      self.class.new(self, settings)
+    end
+
+    #
+    def merge!(settings)
+      case settings
+      when Sanitization
+        other_elements       = settings.elements
+        other_attributes     = settings.attributes
+        other_protocols      = settings.protocols
+        other_add_attributes = settings.add_attributes
+      else
+        other_elements       = settings[:elements]       || []
+        other_attributes     = settings[:attributes]     || {}
+        other_protocols      = settings[:protocols]      || {}
+        other_add_attributes = settings[:add_attributes] || {}
+      end
+
+      elements.concat(other_elements)
+      other_attributes.each do |k,v|
+        attributes[k] ||= []
+        attributes[k].concat(v)
+      end
+      other_protocols.each do |k,v|
+        protocols[k] ||= {}
+        protocols[k].merge!(v)
+      end
+      other_add_attributes.each do |k,v|
+        add_attributes[k] ||= {}
+        add_attributes[k].merge!(v)
+      end
+    end
+
+  end
+
+end
+

lib/gollum/wiki.rb

@@ -15,6 +15,12 @@ module Gollum
       # Sets the default email for commits.
       attr_accessor :default_committer_email
 
+      #
+      attr_writer :sanitization_options
+
+      #
+      attr_writer :history_sanitization_options
+
       # Gets the page class used by all instances of this Wiki.
       # Default: Gollum::Page.
       def page_class
@@ -36,6 +42,12 @@ module Gollum
             ::Gollum::File
           end
       end
+
+      #
+      def sanitization
+        @sanitization ||= Sanitization.new
+      end
+
     end
 
     self.default_committer_name  = 'Anonymous'
@@ -51,21 +63,35 @@ module Gollum
     # repo    - The String path to the Git repository that holds the Gollum
     #           site.
     # options - Optional Hash:
-    #           :base_path  - String base path for all Wiki links.
+    #           :base_path    - String base path for all Wiki links.
-    #                         Default: "/"
+    #                           Default: "/"
-    #           :page_class - The page Class. Default: Gollum::Page
+    #           :page_class   - The page Class. Default: Gollum::Page
-    #           :file_class - The file Class. Default: Gollum::File
+    #           :file_class   - The file Class. Default: Gollum::File
+    #           :sanitization - An instance of Santitization.
     #
     # Returns a fresh Gollum::Repo.
     def initialize(path, options = {})
-      @path       = path
+      @path         = path
-      @repo       = Grit::Repo.new(path)
+      @repo         = Grit::Repo.new(path)
-      @base_path  = options[:base_path]  || "/"
+      @base_path    = options[:base_path]    || "/"
-      @page_class = options[:page_class] || self.class.page_class
+      @page_class   = options[:page_class]   || self.class.page_class
-      @file_class = options[:file_class] || self.class.file_class
+      @file_class   = options[:file_class]   || self.class.file_class
+      @sanitization = options[:sanitization] || self.class.sanitization
+
       clear_cache
     end
 
+    # Public: Instance of Sanitization class.
+    attr :sanitization
+
+    # Public: Return a new sanitization instance combining #sanitiazation
+    # and additional history sanitization options.
+    #
+    # Returns a Sanitization instance.
+    def history_sanitization
+      @history_sanitiazation ||= sanitization.merge(Sanitization::HISTORY_OPTIONS)
+    end
+
     # Public: check whether the wiki's git repo exists on the filesystem.
     #
     # Returns true if the repo exists, and false if it does not.