tiegz-kadoku 0.1.1 → 0.1.2

data/README

@@ -0,0 +1,40 @@
+kadoku
+    by Tieg Zaharia
+    http://github.com/tiegz/kadoku
+ 
+== DESCRIPTION:
+ 
+Kadoku is a Ruby library that uses Hpricot to clean up your unreadable 
+HTML (ie ERB-generated html) [use with caution]. 
+
+== REQUIREMENTS:
+ 
+* Hpricot
+ 
+== INSTALL:
+ 
+  $ gem sources -a http://gems.github.com/ (you only need to do this once)
+  $ gem install tiegz-kadoku
+ 
+== USAGE:
+ 
+Require the gem...
+ 
+  require 'tiegz-kadoku'
+ 
+To load the doc,
+ 
+  a = open('my.site.com/index.html')
+  b = Kadoku::Markup.new(a)
+ 
+To get the cleaned-up markup,
+ 
+  b.to_clean_html
+ 
+To get the original html (rebuilt by Hpricot),
+
+  b.to_html
+
+== RELEVANT LINKS:
+
+* http://blog.solid1pxred.com/articles/2008/12/18/kadoku-the-markup-apologist

data/kadoku.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name     = "kadoku"
-  s.version  = "0.1.1"
+  s.version  = "0.1.2"
   s.date     = "2008-12-17"
   s.summary  = "Uses Hpricot to clean up your unreadable HTML (ie ERB-generated html) [use with caution]"
   s.email    = "tieg.zaharia+kadoku@gmail.com"
@@ -13,9 +13,9 @@ Gem::Specification.new do |s|
     "Rakefile", 
     "README", 
     "TODO",
-    "lib/kadoku/kadoku.rb", 
-    "lib/kadoku/kadoku/markup.rb", 
-    "lib/kadoku/kadoku/markup_after_filter.rb"]
+    "lib/kadoku.rb", 
+    "lib/kadoku/markup.rb", 
+    "lib/kadoku/markup_after_filter.rb"]
   s.test_files = ["test/test_helper.rb",
     "test/kadoku_markup_test.rb"]
   s.rdoc_options = ["--main", "README"]

data/lib/kadoku.rb

@@ -0,0 +1,7 @@
+require 'hpricot'
+require 'kadoku/markup'
+require 'kadoku/markup_after_filter'
+
+# ( かどく or Kadoku).is_for("making readable markup").with("Hpricot")
+module Kadoku
+end

data/lib/kadoku/markup.rb

@@ -0,0 +1,76 @@
+require 'hpricot'
+
+module Kadoku
+  class Markup
+    attr_accessor :hpricot
+
+    def initialize(content, options={})
+      @clean_newline = "#{options[:clean_newline] || "\n"}"
+      @clean_indent  = "#{options[:clean_indent] || "\t"}"
+      @hpricot = Hpricot(content)
+      @indent_counter = 0
+    end
+  
+    # Returns the original document parsed by Hpricot without cleanup
+    def to_html
+      @hpricot.to_html
+    end
+
+    # Returns the original markup parsed by Hpricot with extra cleanup
+    def to_clean_html
+      str = ''
+
+      # A recursive lambda that rejects any empty (includes newlines/spaces/tabs) nodes
+      empty_elements = lambda { |el| el.respond_to?(:children) ? (el.children.reject!(&empty_elements); false) : (el.to_html.strip.empty?) }
+
+      # A recursive lambda that re-maps the nodes with all newlines and extra spaces stripped out
+      stripped_down = lambda do |el|
+        if el.respond_to?(:children)
+          el.children.map(&stripped_down) # iterate over children if possible
+        else
+          #unless el.parent.respond_to?(:name) && el.parent.name == 'script' # preserve JS newlines
+            el.content.gsub!(/\n/, '') if el.is_a?(Hpricot::Text) || el.is_a?(Hpricot::Comment) # remove newlines from text
+            el.content.gsub!(/\s+/, ' ') if el.is_a?(Hpricot::Text) || el.is_a?(Hpricot::Comment) # convert all groups of spaces into one space
+          #end
+        end
+        el
+      end
+
+      # A recursive lambda that iterates over each node and rebuilds the html
+      pretty_print = lambda do |el, i|
+        if el.respond_to?(:children)
+          if el.children.empty? 
+            # isn't there a less hackish way to turn a tag into a self-ending tag?
+            _el = Hpricot::Elem.new(Hpricot::STag.parse(el.name, el.attributes, nil, nil))
+            str << indented(_el) + @clean_newline
+          else
+            str << indented(el.stag) + @clean_newline # add start tag
+            unless el.children.empty?
+              @indent_counter += 1
+              el.children.each_with_index(&pretty_print)
+              @indent_counter -= 1
+              str << indented(el.etag) + @clean_newline if el.etag # add end tag
+            end
+          end
+        else
+          case el
+          when Hpricot::DocType # need to do anything special to the doc type?
+            str << indented(el) + @clean_newline
+          else
+            str << indented(el) + @clean_newline
+          end
+        end
+      end
+
+      @hpricot.children.reject!(&empty_elements).map!(&stripped_down).each_with_index(&pretty_print)
+      @indent_counter = 0 # reset
+
+      str
+    end
+
+    protected
+    def indented(el)
+      @clean_indent * @indent_counter + el.output('')
+    end
+  end
+end

data/lib/kadoku/markup_after_filter.rb

@@ -0,0 +1,24 @@
+module Kadoku
+  # An ActionController after_filter.
+  #
+  # Note: Kadoku is still experimental and isn't guaranteed to not break your html. 
+  #       You might need to restyle some things, so I wouldn't use it untested in Production env.
+  #
+  # Example:
+  #
+  #   class Application < ActionController::Base
+  #     ... all your filters here
+  #     after_filter ERBApologist::Filters::ReadableHTML
+  #   end
+  #
+  # -or-
+  #   class Application < ActoinController::Base
+  #     ... all your filters here
+  #     cleanup_html
+  #   end
+  class MarkupAfterFilter
+    def self.filter(controller)
+      controller.response.body = Markup.new(controller.response.body)
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: tiegz-kadoku
 version: !ruby/object:Gem::Version 
-  version: 0.1.1
+  version: 0.1.2
 platform: ruby
 authors: 
 - Tieg Zaharia
@@ -29,9 +29,9 @@ files:
 - Rakefile
 - README
 - TODO
-- lib/kadoku/kadoku.rb
-- lib/kadoku/kadoku/markup.rb
-- lib/kadoku/kadoku/markup_after_filter.rb
+- lib/kadoku.rb
+- lib/kadoku/markup.rb
+- lib/kadoku/markup_after_filter.rb
 has_rdoc: true
 homepage: http://github.com/tiegz/kadoku
 post_install_message: