当我尝试使用sax在套接字上解析xml时,我遇到了一个奇怪的现象.
经过分析,我注意到DataOutputStream在我的数据前添加了2个字节.

DataOutputStream发送的消息：

0020  50 18 00 20 0f df 00 00  00 9d 3c 3f 78 6d 6c 20   P.. .... ..<?xml 
0030  76 65 72 73 69 6f 6e 3d  22 31 2e 30 22 3f 3e 3c   version= "1.0"?><
0040  63 6f 6d 70 61 6e 79 3e  3c 73 74 61 66 66 3e 3c   company> <staff><
0050  66 69 72 73 74 6e 61 6d  65 3e 79 6f 6e 67 3c 2f   firstnam e>yong</
0060  66 69 72 73 74 6e 61 6d  65 3e 3c 6c 61 73 74 6e   firstnam e><lastn
0070  61 6d 65 3e 6d 6f 6f 6b  20 6b 69 6d 3c 2f 6c 61   ame>mook  kim</la
0080  73 74 6e 61 6d 65 3e 3c  6e 69 63 6b 6e 61 6d 65   stname>< nickname
0090  3e c2 a7 3c 2f 6e 69 63  6b 6e 61 6d 65 3e 3c 73   >..</nic kname><s
00a0  61 6c 61 72 79 3e 31 30  30 30 30 30 3c 2f 73 61   alary>10 0000</sa
00b0  6c 61 72 79 3e 3c 2f 73  74 61 66 66 3e 3c 2f 63   lary></s taff></c
00c0  6f 6d 70 61 6e 79 3e                               ompany>

使用Transformer发送消息：

0020  50 18 00 20 b6 b1 00 00  3c 3f 78 6d 6c 20 76 65   P.. .... <?xml ve
0030  72 73 69 6f 6e 3d 22 31  2e 30 22 20 65 6e 63 6f   rsion="1 .0" enco
0040  64 69 6e 67 3d 22 75 74  66 2d 38 22 3f 3e 3c 63   ding="ut f-8"?><c
0050  6f 6d 70 61 6e 79 3e 3c  73 74 61 66 66 3e 3c 66   ompany>< staff><f
0060  69 72 73 74 6e 61 6d 65  3e 79 6f 6e 67 3c 2f 66   irstname >yong</f
0070  69 72 73 74 6e 61 6d 65  3e 3c 6c 61 73 74 6e 61   irstname ><lastna
0080  6d 65 3e 6d 6f 6f 6b 20  6b 69 6d 3c 2f 6c 61 73   me>mook  kim</las
0090  74 6e 61 6d 65 3e 3c 6e  69 63 6b 6e 61 6d 65 3e   tname><n ickname>
00a0  c2 a7 3c 2f 6e 69 63 6b  6e 61 6d 65 3e 3c 73 61   ..</nick name><sa
00b0  6c 61 72 79 3e 31 30 30  30 30 30 3c 2f 73 61 6c   lary>100 000</sal
00c0  61 72 79 3e 3c 2f 73 74  61 66 66 3e 3c 2f 63 6f   ary></st aff></co
00d0  6d 70 61 6e 79 3e                                  mpany>  

正如人们可能会注意到DataOutputStream在消息前面添加了两个字节.因此,sax解析器抛出异常“org.xml.sax.SAXParseException：prolog中不允许使用内容.”但是,当我跳过这两个字节时,sax解析器工作得很好.
另外我注意到DataInputStream无法读取Transformer消息.

我的问题是：为什么DataOutputStream会添加这些字节,为什么不使用Transformer？

对于那些有兴趣复制问题的人,这里有一些代码：

使用DataInputStream的服务器：

String data = "<?xml version=\"1.0\"?><company><staff><firstname>yong</firstname><lastname>mook kim</lastname><nickname>§</nickname><salary>100000</salary></staff></company>";
ServerSocket server = new ServerSocket(60000);
Socket socket = server.accept();
DataOutputStream os = new DataOutputStream(socket.getOutputStream());
os.writeUTF(data);
os.close();
socket.close();

使用Transformer的服务器：

ServerSocket server = new ServerSocket(60000);
Socket socket = server.accept();
Document doc = createDocument();
printXML(doc, os);
os.close();
socket.close();

public synchronized static void printXML(Document document, OutputStream stream) throws TransformerException
{
    DOMSource domSource = new DOMSource(document);
    StreamResult streamResult = new StreamResult(stream);
    TransformerFactory tf = TransformerFactory.newInstance();
    Transformer serializer = tf.newTransformer();
    serializer.setOutputProperty(OutputKeys.ENCODING, "utf-8");
    serializer.setOutputProperty(OutputKeys.INDENT, "no");
    serializer.transform(domSource, streamResult);
}

private static Document createDocument() throws ParserConfigurationException
{
    Document document = DocumentBuilderFactory.newInstance().newDocumentBuilder().newDocument();
    Element company = document.createElement("company");
    Element staff = document.createElement("staff");
    Element firstname = document.createElement("firstname");
    Element lastname = document.createElement("lastname");
    Element nickname = document.createElement("nickname");
    Element salary = document.createElement("salary");
    Text firstnameText = document.createTextNode("yong");
    Text lastnameText = document.createTextNode("mook kim");
    Text nicknameText = document.createTextNode("§");
    Text salaryText = document.createTextNode("100000");
    document.appendChild(company);
    company.appendChild(staff);
    staff.appendChild(firstname);
    staff.appendChild(lastname);
    staff.appendChild(nickname);
    staff.appendChild(salary);
    firstname.appendChild(firstnameText);
    lastname.appendChild(lastnameText);
    nickname.appendChild(nicknameText);
    salary.appendChild(salaryText);
    return document;
}

使用SAX Parser的客户：

SAXParserFactory factory = SAXParserFactory.newInstance();
SAXParser saxParser = factory.newSAXParser();
DefaultHandler handler = new MyHandler();
Socket socket = new Socket("localhost", 60000);
InputSource is = new InputSource(new InputStreamReader(socket.getInputStream()));
is.setEncoding("UTF-8");
//socket.getInputStream().skip(2); // skip over the 2 bytes from the DataInputStream
saxParser.parse(is, handler);

使用DataInputStream的客户端：

Socket socket = new Socket("localhost", 60000);
DataInputStream os = new DataInputStream(socket.getInputStream());
while(true) {
    String data = os.readUTF();
    System.out.println("Data: " + data);
}

解决方法:

DataOutputStream.writeUTF()的输出是一种自定义格式,旨在由DataInputStream.readUTF()读取.

你正在调用的writeUTF方法的javadoc说：

Writes a string to the underlying output stream using modified UTF-8 encoding in a machine-independent manner.

First, two bytes are written to the output stream as if by the writeShort method giving the number of bytes to follow. This value is the number of bytes actually written out, not the length of the string. Following the length, each character of the string is output, in sequence, using the modified UTF-8 encoding for the character. If no exception is thrown, the counter written is incremented by the total number of bytes written to the output stream. This will be at least two plus the length of str, and at most two plus thrice the length of str.