1 files changed, 27 insertions, 21 deletions
diff --git a/lib/rdoc/encoding.rb b/lib/rdoc/encoding.rb
index 54ecd89816..cf60badd24 100644
--- a/lib/rdoc/encoding.rb
+++ b/lib/rdoc/encoding.rb
@@ -7,6 +7,18 @@
 
 module RDoc::Encoding
 
+  HEADER_REGEXP = /^
+    (?:
+      \A\#!.*\n
+      |
+      ^\#\s+frozen[-_]string[-_]literal[=:].+\n
+      |
+      ^\#[^\n]+\b(?:en)?coding[=:]\s*(?<name>[^\s;]+).*\n
+      |
+      <\?xml[^?]*encoding=(?<quote>["'])(?<name>.*?)\k<quote>.*\n
+    )+
+  /xi # :nodoc:
+
   ##
   # Reads the contents of +filename+ and handles any encoding directives in
   # the file.
@@ -18,12 +30,13 @@ module RDoc::Encoding
   # unknown character in the target encoding will be replaced with '?'
 
   def self.read_file filename, encoding, force_transcode = false
-    content = open filename, "rb" do |f| f.read end
+    content = File.open filename, "rb" do |f| f.read end
     content.gsub!("\r\n", "\n") if RUBY_PLATFORM =~ /mswin|mingw/
 
     utf8 = content.sub!(/\A\xef\xbb\xbf/, '')
 
-    content = RDoc::Encoding.set_encoding content
+    enc = RDoc::Encoding.detect_encoding content
+    content = RDoc::Encoding.change_encoding content, enc if enc
 
     begin
       encoding ||= Encoding.default_external
@@ -85,29 +98,22 @@ module RDoc::Encoding
   end
 
   ##
-  # Sets the encoding of +string+ based on the magic comment
-
-  def self.set_encoding string
-    string = remove_frozen_string_literal string
-
-    string =~ /\A(?:#!.*\n)?(.*\n)/
-
-    first_line = $1
+  # Detects the encoding of +string+ based on the magic comment
 
-    name = case first_line
-           when /^<\?xml[^?]*encoding=(["'])(.*?)\1/ then $2
-           when /\b(?:en)?coding[=:]\s*([^\s;]+)/i   then $1
-           else                                           return string
-           end
+  def self.detect_encoding string
+    result = HEADER_REGEXP.match string
+    name = result && result[:name]
 
-    string = string.sub first_line, ''
-
-    string = remove_frozen_string_literal string
+    name ? Encoding.find(name) : nil
+  end
 
-    enc = Encoding.find name
-    string = RDoc::Encoding.change_encoding string, enc if enc
+  ##
+  # Removes magic comments and shebang
 
-    string
+  def self.remove_magic_comment string
+    string.sub HEADER_REGEXP do |s|
+      s.gsub(/[^\n]/, '')
+    end
   end
 
   ##