xmlhash 1.3.3 → 1.3.4

data/History.txt

@@ -1,3 +1,7 @@
+=== 1.3.4 / 2012-12-04
+
+* "upgrade" latin1 and us-ascii to utf-8 automatically
+
 === 1.3.3 / 2012-11-06
 * <files/> should return [] not [{}]
 

data/ext/xmlhash/xmlhash.c

@@ -199,6 +199,8 @@ static VALUE parse_xml_hash(VALUE self, VALUE rb_xml)
   Check_Type(rb_xml, T_STRING);
 #ifdef HAVE_RUBY_ENCODING_H
   m_current_encoding = rb_enc_get(rb_xml);
+  if (m_current_encoding == rb_ascii8bit_encoding() || m_current_encoding == rb_usascii_encoding())
+      m_current_encoding = rb_utf8_encoding();
 #endif
 
   data = (char*)calloc(RSTRING_LEN(rb_xml), sizeof(char));

data/lib/xmlhash.rb

@@ -1,7 +1,7 @@
 require 'xmlhash/xmlhash'
 
 module Xmlhash
-  VERSION = '1.3.3'
+  VERSION = '1.3.4'
 
   class XMLHash < Hash
     

data/test/test_xmlhash.rb

@@ -124,4 +124,35 @@ eos
     ret = Xmlhash.parse(xml)
     assert_equal ret.elements('files'), []
   end
+
+  def test_utf8
+     xml = '<package name="libconfig" project="home:coolo">
+  <title>libconfig &#8211; C/C++ Configuration File Library</title>
+  <description>Libconfig is a simple library for processing structured configuration files, like this one: test.cfg. This file format is more compact and more readable than XML. And unlike XML, it is type-aware, so it is not necessary to do string parsing in application code.
+
+Libconfig is very compact &#8212; just 38K for the stripped C shared library (less than one-fourth the size of the expat XML parser library) and 66K for the stripped C++ shared library. This makes it well-suited for memory-constrained systems like handheld devices.
+
+The library includes bindings for both the C and C++ languages. It works on POSIX-compliant UNIX systems (GNU/Linux, Mac OS X, Solaris, FreeBSD) and Windows (2000, XP and later).</description>
+  </package>'
+    xh = Xmlhash.parse(xml)
+    assert_equal "UTF-8", xh['title'].encoding.to_s
+
+    # now try with different input encoding
+    xml.encode!('US-ASCII')
+    xh = Xmlhash.parse(xml)
+    assert_equal "UTF-8", xh['title'].encoding.to_s
+
+    xml = '<?xml version="1.0" encoding="ISO-8859-1"?>
+    <package><title>Äöß</title></package>'
+    xml.encode!('ISO-8859-1')
+    xh = Xmlhash.parse(xml)
+    assert_equal "ISO-8859-1", xh['title'].encoding.to_s
+    
+    xml = '<?xml version="1.0" encoding="ISO-8859-1"?>
+    <package><title>&#228;&#211;&#254;</title></package>'
+    xml.encode!('US-ASCII')
+    xh = Xmlhash.parse(xml)
+    assert_equal "UTF-8", xh['title'].encoding.to_s
+
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: xmlhash
 version: !ruby/object:Gem::Version
-  version: 1.3.3
+  version: 1.3.4
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-11-06 00:00:00.000000000 Z
+date: 2012-12-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: pkg-config
@@ -66,7 +66,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '3.1'
+        version: '3.3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -74,7 +74,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '3.1'
+        version: '3.3'
 description: ! 'A small C module that wraps libxml2''s xmlreader to parse a XML
 
   string into a ruby hash'