Two bugfixes: display output when there are no comments, allow non-UTF-8 encodings

.gitignore

@@ -1,3 +1,4 @@
+/docs
 *.gem
 *.rbc
 .bundle

lib/rocco.rb

@@ -73,26 +73,30 @@
 #
 # * `:stylesheet`, which specifies the css stylesheet to use for each
 #   rendered template.  _Defaults to `http://jashkenas.github.com/docco/resources/docco.css`
-#   (the original docco stylesheet)
+#   (the original docco stylesheet)._
+#
+# * `:encoding`: specifies the encoding that input files are written in.
+#   _Defaults to `UTF-8`_.
 class Rocco
   VERSION = '0.8.2'
 
   def initialize(filename, sources=[], options={})
     @file       = filename
     @sources    = sources
 
-    # When `block` is given, it must read the contents of the file using
-    # whatever means necessary and return it as a string. With no `block`,
-    # the file is read to retrieve data.
-    @data = if block_given? then yield else File.read(filename) end
-
     @options =  {
       :language      => 'ruby',
       :comment_chars => '#',
       :template_file => nil,
-      :stylesheet    => 'http://jashkenas.github.com/docco/resources/docco.css'
+      :stylesheet    => 'http://jashkenas.github.com/docco/resources/docco.css',
+      :encoding => 'UTF-8'
     }.merge(options)
 
+    # When `block` is given, it must read the contents of the file using
+    # whatever means necessary and return it as a string. With no `block`,
+    # the file is read to retrieve data.
+    @data = if block_given? then yield else read_with_encoding(filename) end
+
     # If we detect a language
     if "text" != detect_language
       # then assign the detected language to `:language`, and look for
@@ -148,6 +152,20 @@ def to_html
   # Helper Functions
   # ----------------
 
+  # Read *file* encoded `@options[:encoding]` into a string encoded in UTF-8.
+  def read_with_encoding filename
+    # This works differently in Ruby 1.8 and Ruby 1.9, which are
+    # distinguished by checking if `IO#external_encoding` exists.
+    if IO.method_defined?("external_encoding")
+      File.read(filename, :external_encoding => @options[:encoding],
+                :internal_encoding => "UTF-8")
+    else
+      require 'iconv'
+      data = File.read(filename)
+      Iconv.conv("UTF-8", @options[:encoding], data)
+    end
+  end
+
   # Returns `true` if `pygmentize` is available locally, `false` otherwise.
   def pygmentize?
     @_pygmentize ||= ENV['PATH'].split(':').
@@ -380,6 +398,7 @@ def highlight(blocks)
     # into separate sections.
     markdown = docs_blocks.join("\n\n##### DIVIDER\n\n")
     docs_html = process_markdown(markdown).split(/\n*<h5>DIVIDER<\/h5>\n*/m)
+    docs_html << "" if docs_html.empty?
 
     # Combine all code blocks into a single big stream with section dividers and
     # run through either `pygmentize(1)` or <http://pygments.appspot.com>

test/test_reported_issues.rb

@@ -28,11 +28,9 @@ def test_issue10_utf8_processing
       "UTF-8 input files ought behave correctly."
     )
     # and, just for grins, ensure that iso-8859-1 works too.
-    # @TODO:    Is this really the correct behavior?  Converting text
-    #           to UTF-8 on the way out is probably preferable.
-    r = Rocco.new( File.dirname(__FILE__) + "/fixtures/issue10.iso-8859-1.rb" )
+    r = Rocco.new( File.dirname(__FILE__) + "/fixtures/issue10.iso-8859-1.rb", [], :encoding => 'ISO-8859-1' )
     assert_equal(
-      "<p>hello w\366rld</p>\n",
+      "<p>hello wörld</p>\n",
       r.sections[0][0],
       "ISO-8859-1 input should probably also behave correctly."
     )
@@ -83,4 +81,12 @@ def test_issue15_extra_space_after_comment_character_remains
     )
     assert_equal("<h2>Comment 1</h2>\n", r.sections[0][0])
   end
+
+  def test_issue71_empty_comment
+    r = roccoize( "filename.rb", "def codeblock\nend\n" )
+    assert_equal 1, r.sections.length
+    assert_equal 2, r.sections[ 0 ].length
+    assert_match /\s*/, r.sections[ 0 ][ 0 ]
+    assert_equal "<span class=\"k\">def</span> <span class=\"nf\">codeblock</span>\n<span class=\"k\">end</span>", r.sections[ 0 ][ 1 ]
+  end
 end