From 46ff009168cfb29b473bc37d38b64719ad9e6526 Mon Sep 17 00:00:00 2001
From: naruse <naruse@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>
Date: Wed, 14 Apr 2010 13:11:11 +0000
Subject: * lib/rexml/source.rb: force_encoding("UTF-8") when the input   is
 already UTF-8. patched by Kouhei Sutou [ruby-core:23404]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@27342 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
---
 lib/rexml/source.rb | 39 +++++++++++++++++++++++++--------------
 1 file changed, 25 insertions(+), 14 deletions(-)

(limited to 'lib/rexml')

diff --git a/lib/rexml/source.rb b/lib/rexml/source.rb
index 3f6d4ffa26..1206150b16 100644
--- a/lib/rexml/source.rb
+++ b/lib/rexml/source.rb
@@ -162,6 +162,15 @@ module REXML
         @line_break = ">"
       end
       super( @source.eof? ? str : str+@source.readline( @line_break ) )
+
+      if !@to_utf and
+          @buffer.respond_to?(:force_encoding) and
+          @source.respond_to?(:external_encoding) and
+          @source.external_encoding != ::Encoding::UTF_8
+        @force_utf8 = true
+      else
+        @force_utf8 = false
+      end
     end
 
     def scan(pattern, cons=false)
@@ -174,11 +183,7 @@ module REXML
       if rv.size == 0
         until @buffer =~ pattern or @source.nil?
           begin
-            # READLINE OPT
-            #str = @source.read(@block_size)
-            str = @source.readline(@line_break)
-            str = decode(str) if @to_utf and str
-            @buffer << str
+            @buffer << readline
           rescue Iconv::IllegalSequence
             raise
           rescue
@@ -193,12 +198,7 @@ module REXML
 
     def read
       begin
-        str = @source.readline(@line_break)
-        str = decode(str) if @to_utf and str
-        @buffer << str
-        if not @to_utf and @buffer.respond_to? :force_encoding
-          @buffer.force_encoding Encoding::UTF_8
-        end
+        @buffer << readline
       rescue Exception, NameError
         @source = nil
       end
@@ -213,9 +213,7 @@ module REXML
       @buffer = $' if cons and rv
       while !rv and @source
         begin
-          str = @source.readline(@line_break)
-          str = decode(str) if @to_utf and str
-          @buffer << str
+          @buffer << readline
           rv = pattern.match(@buffer)
           @buffer = $' if cons and rv
         rescue
@@ -254,5 +252,18 @@ module REXML
       end
       [pos, lineno, line]
     end
+
+    private
+    def readline
+      str = @source.readline(@line_break)
+      return nil if str.nil?
+
+      if @to_utf
+        decode(str)
+      else
+        str.force_encoding(::Encoding::UTF_8) if @force_utf8
+        str
+      end
+    end
   end
 end
-- 
cgit v1.2.3