Ensure interior empty list items and some nbsp are not stripped away

Gemfile

@@ -0,0 +1,6 @@
+source 'https://rubygems.org'
+
+gemspec
+
+gem 'pry'; require 'pry'
+gem 'clipboard'; require 'clipboard'

Gemfile.lock

@@ -0,0 +1,46 @@
+PATH
+  remote: .
+  specs:
+    html2confluence (1.3.23)
+      nokogiri
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    clipboard (1.1.1)
+    coderay (1.1.1)
+    diff-lcs (1.3)
+    method_source (0.8.2)
+    mini_portile2 (2.2.0)
+    nokogiri (1.8.0)
+      mini_portile2 (~> 2.2.0)
+    pry (0.10.4)
+      coderay (~> 1.1.0)
+      method_source (~> 0.8.1)
+      slop (~> 3.4)
+    rspec (3.6.0)
+      rspec-core (~> 3.6.0)
+      rspec-expectations (~> 3.6.0)
+      rspec-mocks (~> 3.6.0)
+    rspec-core (3.6.0)
+      rspec-support (~> 3.6.0)
+    rspec-expectations (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-mocks (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-support (3.6.0)
+    slop (3.6.0)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  clipboard
+  html2confluence!
+  pry
+  rspec
+
+BUNDLED WITH
+   1.13.7

Rakefile

@@ -0,0 +1,23 @@
+require 'bundler/gem_tasks'
+require 'rake/testtask'
+
+Rake::TestTask.new do |t|
+  t.libs << 'test'
+  t.test_files = Dir['test/**/*_test.rb']
+end
+task default: :test
+
+desc "Open a pry console preloaded with this library"
+task console: 'console:pry'
+
+namespace :console do
+
+  task :pry do
+    sh "bundle exec pry -I lib -r html2confluence.rb"
+  end
+
+  task :irb do
+    sh "bundle exec irb -I lib -r html2confluence.rb"
+  end
+
+end

html2confluence.gemspec

@@ -1,7 +1,7 @@
 Gem::Specification.new do |s|
   s.platform    = Gem::Platform::RUBY
   s.name        = 'html2confluence'
-  s.version     = "1.3.22"
+  s.version     = "1.3.23"
   s.summary     = 'Converter from HTML to Confluence Wiki Markup'
   s.description = 'Provides an SGML parser to convert HTML into the Wiki Markup format'
 
@@ -13,6 +13,7 @@ Gem::Specification.new do |s|
 
   s.require_path = 'lib'
   s.files        = Dir.glob("{lib,spec}/**/*") + %w(example.rb README.mdown)
-
+  
   s.add_dependency "nokogiri"
+  s.add_development_dependency "rspec"
 end

lib/html2confluence.rb

@@ -1,5 +1,7 @@
 require 'rexml/document'
 
+require 'walker'
+
 require 'nokogiri' # For validating html from our editor
 
 # A class to convert HTML to confluence markup. Based on the python parser
@@ -74,7 +76,6 @@ def make_quicktag_start_pair(tag, wrapchar, attributes)
 
   def make_quicktag_end_pair(wrapchar)
     content = stop_capture
-
     # Don't make quicktags with empty content. 
     if content.join("").strip.empty?
       write(content)
@@ -85,15 +86,25 @@ def make_quicktag_end_pair(wrapchar)
       unless in_nested_quicktag?
         #write([" "]) 
       end
-      write(["#{wrapchar}"])
+      write_quicktag_wrapchar(wrapchar)
     end
-    write(content.collect(&:strip))
-    write([wrapchar]) unless @skip_quicktag
+    write(content)
+    write_quicktag_wrapchar(wrapchar) unless @skip_quicktag
     unless in_nested_quicktag?
       #write([" "]) 
     end
   end
 
+  def write_quicktag_wrapchar(wrapchar)
+    write([
+      if @last_write[-1] =~ /\s/
+        "#{wrapchar}"
+      else
+        "{#{wrapchar}}"
+      end
+    ])
+  end
+
   def in_nested_quicktag?
     @quicktags ||= QUICKTAGS.keys
     @stack.size > 1 && @quicktags.include?(@stack[@stack.size-1]) && @quicktags.include?(@stack[@stack.size-2])
@@ -147,25 +158,25 @@ def handle_data(data)
   PAIRS = { 'bq' => 'bq', 'p' => 'p' }
   QUICKTAGS = { 'b' => '*', 'strong' => '*', 'del' => '-',  'strike' => '-',
     'i' => '_', 'ins' => '+', 'u' => '+', 'em' => '_', 'cite' => '??', 
-    'sup' => '^', 'sub' => '~', 'code' => '@'}
+    'sup' => '^', 'sub' => '~'}
 
-  PAIRS.each do |key, value|
-    define_method "start_#{key}" do |attributes|
-      make_block_start_pair(value, attributes)
+  PAIRS.each do |node, markup|
+    define_method "start_#{node}" do |attributes|
+      make_block_start_pair(markup, attributes)
     end
 
-    define_method "end_#{key}" do
+    define_method "end_#{node}" do
       make_block_end_pair
     end
   end
 
-  QUICKTAGS.each do |key, value|
-    define_method "start_#{key}" do |attributes|
-      make_quicktag_start_pair(key, value, attributes)
+  QUICKTAGS.each do |node, wrapchar|
+    define_method "start_#{node}" do |attributes|
+      make_quicktag_start_pair(node, wrapchar, attributes)
     end
 
-    define_method "end_#{key}" do
-      make_quicktag_end_pair(value)
+    define_method "end_#{node}" do
+      make_quicktag_end_pair(wrapchar)
     end
   end
 
@@ -363,6 +374,17 @@ def end_blockquote
       write(s)
     end
   end
+
+  def start_code(attrs)
+    @preserveWhitespace = true
+    write("{code}")
+  end
+
+  def end_code
+    stop_capture_and_write
+    write("{code}")
+    @preserveWhitespace = false
+  end
 
   def start_pre(attrs)
     @preserveWhitespace = true
@@ -392,7 +414,7 @@ def preprocess(data)
     data.gsub!(/&(mdash|#8212);/,'---')
     data.gsub!(/&(ndash|#8211);/,'--')
 
-    # remove empty blockquotes and list items (other empty elements are easy enough to deal with)
+    # remove empty blockquotes (other empty elements are easy enough to deal with)
     data.gsub!(/<blockquote>\s*(<br[^>]*>)?\s*<\/blockquote>/x,' ')
 
     # Fix unclosed <br>
@@ -443,16 +465,15 @@ def preprocess(data)
 
   # Return the textile after processing
   def to_wiki_markup
-    fix_textile_whitespace!(result.join).gsub(/\n(\*|#)+\s*\n(\*|#)+/) do |match|
-      "\n#{match.split("\n").last.squeeze(' ')}"
-    end
+    fix_textile_whitespace!(result.join)
   end
 
   def fix_textile_whitespace!(output)
-    # fixes multiple blank lines, blockquote indicator followed by blank lines, and trailing whitespace after quicktags
-    # modifies input string and also returns it
+    # fixes multiple blank lines
     output.gsub!(/(\n\s*){2,}/,"\n\n")
+    # fixes blockquote indicator followed by blank lines
     output.gsub!(/bq. \n+(\w)/,'bq. \1')
+    # fixes trailing whitespace after quicktags
     QUICKTAGS.values.uniq.each do |t|
       output.gsub!(/ #{Regexp.escape(t)}[ \t]+#{Regexp.escape(t)} /,' ') # removes empty quicktags
       #output.gsub!(/(\[?#{Regexp.escape(t)})(\w+)([^#{Regexp.escape(t)}]+)(\s+)(#{Regexp.escape(t)}\]?)/,'\1\2\3\5\4') # fixes trailing whitespace before closing quicktags
@@ -461,6 +482,14 @@ def fix_textile_whitespace!(output)
     #output.gsub!(/^[ \t]/,'') # leading whitespace
     #output.gsub!(/[ \t]$/,'') # trailing whitespace
     output.strip!
+    # fixes extra bullets generated when nesting list items
+    output.gsub!(/\n([\*|#]+)\s*\n([\*|#]+)/) do |match|
+      if $1 == $2
+        match
+      else
+        "\n#{match.split("\n").last}"
+      end.squeeze(' ')
+    end
     return output
   end