reverse_markdown 0.0.1 → 0.1.0

data/lib/reverse_markdown.rb

@@ -1,4 +1,4 @@
-require 'rexml/document'
+require 'nokogiri'
 
 # reverse markdown for ruby
 # author: JO
@@ -13,6 +13,7 @@ require 'rexml/document'
 # -
 
 class ReverseMarkdown
+  attr_reader :errors
 
   # set basic variables:
   # - @li_counter: numbering list item (li) tags in an ordered list (ol)
@@ -33,9 +34,8 @@ class ReverseMarkdown
   # To garantuee well-formed xml for REXML a <root> element will be added, but has no effect.
   # After parsing all elements, the 'reference style'-links will be inserted.
   def parse_string(string)
-    doc = REXML::Document.new("<root>\n"+string+"\n</root>")
-    parse_element(doc.root, :none)
-    insert_links()
+    parse_element(Nokogiri::HTML.fragment(string))
+    insert_links
     @output
   end
 
@@ -45,7 +45,7 @@ class ReverseMarkdown
   # 3a. if element only contains text, handle it like a text node
   # 3b. if element is a container handle its children, which may be text- or element nodes
   # 4. finally add the markdown ending tag for this element
-  def parse_element(element, parent)
+  def parse_element(element, parent = nil)
     name = element.name.to_sym
     # 1.
     @output << ("  " * @indent) if name.eql?(:li)
@@ -53,28 +53,27 @@ class ReverseMarkdown
     @output << opening(element, parent)
 
     # 3a.
-    if (element.has_text? and element.children.size < 2)
+    if element.children.size == 1 && element.children.first.text?
       @output << text_node(element, parent)
-    end
-
-    # 3b.
-    if element.has_elements?
+    else
+      # 3b.
       element.children.each do |child|
         # increase indent if nested list
-        @indent += 1 if element.name=~/(ul|ol)/ and parent.eql?(:li)
+        @indent += 1 if nested_list?(element, parent)
 
-        if child.node_type.eql?(:element)
+        if child.element?
           parse_element(child, element.name.to_sym)
         else
-          if parent.eql?(:blockquote)
-            @output << child.to_s.gsub("\n ", "\n>")
-          else
-            @output << child.to_s
-          end
+          @output <<
+            if parent.eql?(:blockquote)
+              child.inner_text.gsub("\n ", "\n>")
+            else
+              child.inner_text
+            end
         end
 
         # decrease indent if end of nested list
-        @indent -= 1 if element.name=~/(ul|ol)/ and parent.eql?(:li)
+        @indent -= 1 if nested_list?(element, parent)
       end
     end
 
@@ -82,13 +81,19 @@ class ReverseMarkdown
     @output << ending(element, parent)
   end
 
+  private
+
+  def nested_list?(element, parent)
+    element.name=~/(ul|ol)/ and parent.eql?(:li)
+  end
+
   # Returns opening markdown tag for the element. Its parent matters sometimes!
-  def opening(type, parent)
+  def opening(type, parent_name)
     case type.name.to_sym
       when :h1
         "# "
       when :li
-        parent.eql?(:ul) ? " - " : " "+(@li_counter+=1).to_s+". "
+        parent_name == :ul ? " - " : " "+(@li_counter+=1).to_s+". "
       when :ol
         @li_counter = 0
         ""
@@ -110,18 +115,16 @@ class ReverseMarkdown
         "**"
       when :blockquote
         # remove leading newline
-        type.children.first.value = ""
+        type.children.first.content = ""
         "> "
       when :code
-        parent.eql?(:pre) ? "    " : "`"
+        parent_name == :pre ? "    " : "`"
       when :a
         "["
       when :img
         "!["
       when :hr
         "----------\n\n"
-      when :root
-        ""
       else
         @errors << "unknown start tag: "+type.name.to_s
         ""
@@ -166,8 +169,6 @@ class ReverseMarkdown
         @links << type.attribute('src').to_s
         "" + type.attribute('alt').to_s + "][" + @links.size.to_s + "] "
         "#{type.attribute('alt')}][#{@links.size}] "
-      when :root
-        ""
       else
         @errors << "  unknown end tag: "+type.name.to_s
         ""
@@ -178,12 +179,16 @@ class ReverseMarkdown
   # If its a code block to indent of 4 spaces.
   # For block quotation add a leading '>'
   def text_node(element, parent)
-    if element.name.to_sym.eql?(:code) and parent.eql?(:pre)
-      element.text.gsub("\n","\n    ") << "\n"
-    elsif parent.eql?(:blockquote)
-      element.text.gsub!("\n ","\n>")
-    else
-      element.text
+    text_node = element.children.first
+    if text_node.text?
+      text = text_node.text
+      if element.name.to_sym.eql?(:code) and parent.eql?(:pre)
+        text.gsub("\n","\n    ") << "\n"
+      elsif parent.eql?(:blockquote)
+        text.gsub!("\n ","\n>")
+      else
+        text
+      end
     end
   end
 
@@ -195,13 +200,6 @@ class ReverseMarkdown
     end.join
   end
 
-  # Print out all errors, that occured and have been written to @errors.
-  def print_errors
-    @errors.each do |error|
-      puts error
-    end
-  end
-
   # Perform a benchmark on a given string n-times.
   def speed_benchmark(string, n)
     require 'benchmark'

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: reverse_markdown
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.1.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,8 +9,24 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-09 00:00:00.000000000 Z
-dependencies: []
+date: 2012-07-11 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 description: 
 email: 
 executables: []