nokogiri 1.4.6-java → 1.4.7-java

data/CHANGELOG.ja.rdoc

@@ -1,3 +1,9 @@
+=== 1.4.7 / 2011年7月1日
+
+* バグの修正
+
+  * エンコーディング宣言のないHTMLファイルで部分的に重複したドキュメントが生成される問題を修正した. #478
+
 === 1.4.6 / 2011年6月19日
 
 * ノート

data/CHANGELOG.rdoc

@@ -1,3 +1,11 @@
+=== 1.4.7 / 2011-07-01
+
+* Bugfixes
+
+  * Fix a bug in advanced encoding detection that leads to partially
+    duplicated document when parsing an HTML file with unknown
+    encoding. Thanks, Timothy Elliott (@ender672)! #478
+
 === 1.4.6 / 2011-06-19
 
 * Notes

data/lib/nokogiri/html/document.rb

@@ -92,7 +92,7 @@ module Nokogiri
           if string_or_io.respond_to?(:read)
             url ||= string_or_io.respond_to?(:path) ? string_or_io.path : nil
             if !encoding
-              # Perform further encoding detection that libxml2 does
+              # Perform advanced encoding detection that libxml2 does
               # not do.
               string_or_io = EncodingReader.new(string_or_io)
               begin
@@ -181,16 +181,13 @@ module Nokogiri
           if !@firstchunk
             @firstchunk = @io.read(len) or return nil
 
-            # This implementation expects and assumes that the first
-            # call from htmlReadIO() is made with a length long enough
-            # (~1KB) to achieve further encoding detection that
-            # libxml2 does not do.
+            # This implementation expects that the first call from
+            # htmlReadIO() is made with a length long enough (~1KB) to
+            # achieve advanced encoding detection.
             if encoding = EncodingReader.detect_encoding(@firstchunk)
+              # The first chunk is stored for the next read in retry.
               raise EncodingFoundException, encoding
             end
-
-            # This chunk is stored for the next read in retry.
-            return @firstchunk
           end
 
           ret = @firstchunk.slice!(0, len)

data/lib/nokogiri/version.rb

@@ -1,6 +1,6 @@
 module Nokogiri
   # The version of Nokogiri you are using
-  VERSION = '1.4.6'
+  VERSION = '1.4.7'
 
   # More complete version information about libxml
   VERSION_INFO = {}

data/test/helper.rb

@@ -22,6 +22,7 @@ module Nokogiri
     SHIFT_JIS_HTML  = File.join(ASSETS_DIR, 'shift_jis.html')
     ENCODING_XHTML_FILE = File.join(ASSETS_DIR, 'encoding.xhtml')
     ENCODING_HTML_FILE  = File.join(ASSETS_DIR, 'encoding.html')
+    NOENCODING_FILE  = File.join(ASSETS_DIR, 'noencoding.html')
     PO_XML_FILE     = File.join(ASSETS_DIR, 'po.xml')
     PO_SCHEMA_FILE  = File.join(ASSETS_DIR, 'po.xsd')
     ADDRESS_SCHEMA_FILE = File.join(ASSETS_DIR, 'address_book.rlx')

data/test/html/test_document_encoding.rb

@@ -89,6 +89,13 @@ module Nokogiri
         File.open(file, 'rb')
       end
 
+      def test_document_html_noencoding
+        from_stream = Nokogiri::HTML(binopen(NOENCODING_FILE))
+        from_string = Nokogiri::HTML(binread(NOENCODING_FILE))
+
+        assert_equal from_string.to_s.size, from_stream.to_s.size
+      end
+
       def test_document_xhtml_enc
         [ENCODING_XHTML_FILE, ENCODING_HTML_FILE].each { |file|
           doc_from_string_enc = Nokogiri::HTML(binread(file), nil, 'Shift_JIS')

metadata

@@ -2,7 +2,7 @@
 name: nokogiri
 version: !ruby/object:Gem::Version 
   prerelease: 
-  version: 1.4.6
+  version: 1.4.7
 platform: java
 authors: 
   - Aaron Patterson
@@ -11,7 +11,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-06-19 00:00:00 -04:00
+date: 2011-07-01 00:00:00 -04:00
 default_executable: nokogiri
 dependencies: 
   - !ruby/object:Gem::Dependency