字符串序列化时碰到的小问题

最近在使用.NET当中的序列化方法序列化字符串时碰到一个小问题。序列化时一切正常，反序列化时却

碰到了问题。
先看代码:

//对象定义 [Serializable] public class TestObject { public string Value { get; set; } } #region XmlFormatter 序列化类 /// <summary> /// 序列化类 /// </summary> public class XmlFormatter { public static string ToXml(object obj) { if (obj == null) return string.Empty; StringBuilder sb = new StringBuilder(); StringWriter sw = new StringWriter(sb); try { XmlSerializer serializer = new XmlSerializer(obj.GetType(), string.Empty); serializer.Serialize(sw, obj); sw.Close(); return sb.ToString(); } catch (Exception ex) { Debug.WriteLine(ex); return string.Empty; } } public static object FromXml(string xml, Type type) { try { StringReader reader = new StringReader(xml); XmlSerializer serializer = new XmlSerializer(type); object obj = serializer.Deserialize(reader); reader.Close(); return obj; } catch (Exception ex) { Debug.WriteLine(ex); } return null; } } #endregion //测试代码 //普通的字符串 static void TestMethod1() { TestObject expected = new TestObject() { Value = "Test 1234" }; string xmlString = XmlFormatter.ToXml(expected); TestObject actual = new TestObject(); actual = (TestObject)XmlFormatter.FromXml(xmlString, actual.GetType()); if (actual != null && expected.Value == actual.Value) Console.WriteLine("TestClass1.TestMethod1 passed!"); else Console.WriteLine("TestClass1.TestMethod1 failed!"); } //包含特殊字符的字符串 static void TestMethod2() { TestObject expected = new TestObject() { Value = "Test1234" }; string xmlString = XmlFormatter.ToXml(expected); TestObject actual = new TestObject(); actual = (TestObject)XmlFormatter.FromXml(xmlString, actual.GetType()); if (actual != null && expected.Value == actual.Value) Console.WriteLine("TestClass1.TestMethod2 passed!"); else Console.WriteLine("TestClass1.TestMethod2 failed!"); }

 

输出结果：
TestClass1.TestMethod1 passed!
TestClass1.TestMethod2 failed!

 

Debug发现TestMethod1的序列化结果是：

<?xml version="1.0" encoding="utf-16"?> <TestObject xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"> <Value>Test 1234</Value> </TestObject>

TestMethod2的序列化结果是：

<?xml version="1.0" encoding="utf-16"?> <TestObject xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"> <Value>Test&#x1;&#x2;&#x3;&#x4;1234</Value> </TestObject>

 

如果我们把TestMethod2的这段Xml存为一个文件，然后用Visual Studio打开，会发现有以下的错误提示：

Error 3 Character ' ', hexadecimal value 0x1 is illegal in XML documents. XMLFile2.xml 3 14 Miscellaneous Files
Error 4 Character ' ', hexadecimal value 0x2 is illegal in XML documents. XMLFile2.xml 3 19 Miscellaneous Files
Error 5 Character ' ', hexadecimal value 0x3 is illegal in XML documents. XMLFile2.xml 3 24 Miscellaneous Files
Error 6 Character ' ', hexadecimal value 0x4 is illegal in XML documents. XMLFile2.xml 3 29 Miscellaneous Files

 

这下我们明白了，原来就是：...这些个特殊字符在作怪。网上搜了一把，这个过程比较郁闷，不知道该用什么关键字去搜，后来直接在Google里试了一下“&#x1”发现前两个就是和我相似的问题，其中第一个里面讲到了MSDN里面的原文

参考这里吧：

http://bytes.com/groups/net-xml/176249-system-xml-xmldocument-parses-well-formed

http://www.codeguru.com/forum/archive/index.php/t-232348.html

 

特别是MSDN的原文：

http://msdn.microsoft.com/zh-cn/library/system.xml.xmltextreader.normalization(VS.80).aspx

 

这下就明白了，我反序列化的时候没有直接用到XmlTextReader，而Normalization默认是true，而是用的StringReader把字符串读取为流，只要把反序列化的方法稍微改改就行：

 TextReader tr = new StringReader(xml); XmlTextReader reader = new XmlTextReader(tr); reader.Normalization = false; XmlSerializer serializer = new XmlSerializer(type); object obj = serializer.Deserialize(reader); tr.Close(); reader.Close(); return obj ;

 

好了，完事大吉！

最后奉上测试代码：测试代码