c＃解析包含HTML特殊字符的字符串XElement
问题描述：
在客户端的服务器+ GWT上使用.NET c＃，我有一个Web窗体，它接受用户输入，然后构建一个XML字符串并将其存储在数据库中。然后我需要从数据库中读取它，通过tcp将它发送到手持设备，并将其解析为XElement。一切运作良好，直到您从Word中复制和粘贴文本或在这种情况下脱颖而出，当我尝试这样做：c＃解析包含HTML特殊字符的字符串XElement
XElement.parse(str);
它抛出一个异常：
'.', hexadecimal value 0x00, is an invalid character. Line 132, position 111.
例字符，将导致此问题是正确的撇号字符（0x2019）。现在可能会有一大堆特殊字符可能从excel/word复制粘贴等。处理此问题的最佳方法是什么？下面是我如何构建从流串：
protected CallResult callUsingSocketClass(string methodName, params Action<CallParameters>[] addParameters)
{ WebServicesClient.Debug.DebugMessage（WebServicesClient.Debug.MASK_ENTRY_EXIT， “++调用（{0}，...）”，方法名）;
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
CallParameters parameters = new CallParameters(this, methodName);
foreach (var addParameter in addParameters)
{
addParameter(parameters);
}
string post = this.CreatePost(parameters);
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC(Method={0}, host={1}, port={2}, post='{3}')", methodName, this.Host, this.Port, post);
byte[] postBytes = Encoding.UTF8.GetBytes(post);
//
// Send the request and wait for the reply.
//
char[] replyContentChars = null;
for (int attempts = 0; attempts < 3; attempts++)
{
Socket socket = null;
try
{
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - creating socket for RPC call...");
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
string hostID = this.Host;
if (this.HostIPAddress != null)
{
hostID = this.HostIPAddress;
}
using (socket = this.connectToServer(hostID, this.Port))
{
if (socket == null)
{
return null;
}
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - socket created!");
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - communicating with server...");
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - writing post...");
this.sendDataToServer(socket, postBytes);
int replyLength = -1;
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - reading reply...");
using (var reader = this.receiveStreamFromServer(socket))
{
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
socket.Close();
socket = null;
return null;
}
}
for (; ;)
{
string lineRaw = reader.ReadLine().Trim();
string line = lineRaw.ToLowerInvariant();
if (line.StartsWith("content-length:"))
{
replyLength = int.Parse(line.Substring(15));
}
else if (line == "")
{
if (replyLength < 0)
{
throw new InvalidOperationException("Reply hasn't specified content-length");
}
break;
}
else
{
if (this.CookieJar != null)
{
this.CookieJar.ProcessFromServer(lineRaw);
}
}
}
// Content starts here
replyContentChars = new char[replyLength];
int replyRecv = 0;
do
{
int charsRecv = reader.Read(replyContentChars, replyRecv, replyLength - replyRecv);
if (charsRecv <= 0)
{
break;
}
replyRecv += charsRecv;
} while (replyRecv < replyLength);
//int charsRecv = reader.Read(replyContentChars, 0, replyLength);
if (replyRecv != replyLength)
{
untime.Logger.Logger.Error("Web Service call '{0}' received {1} bytes, header indicated {2} bytes", methodName, replyRecv, replyLength);
throw new InvalidOperationException(String.Format("Have not received all of reply data - received {0} bytes, expected {1}", replyRecv, replyLength));
}
}
socket.Close();
socket = null;
}
}
catch (Exception e)
{
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC - exception thrown - {0} [{1},{2}]", e.Message, e.Source, e.StackTrace);
}
finally
{
if (socket != null)
{
socket.Close();
socket = null;
}
}
if (replyContentChars != null)
{
break;
}
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
}
//
// Verify that data has been received.
//
if (replyContentChars == null)
{
return null;
}
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
//
// Process the received data.
//
string replyContent = new string(replyContentChars);
WebServicesClient.Debug.DebugMessage(WebServicesClient.Debug.MASK_RPC_CALL, "RPC(Method={0}, replyContent='{1}')", methodName, replyContent);
XElement xReplyContent = XElement.Parse(replyContent);
var xReplyBody = xReplyContent.Element(nsSoap + "Body");
var xFault = xReplyBody.Element(nsSoap + "Fault");
if (xFault != null)
{
// Something has gone wrong on the server
var xFaultCode = xFault.Element(nsSoap + "Code");
var xFaultReason = xFault.Element(nsSoap + "Reason");
untime.Logger.Logger.Error("Web Service call to method '{0}' failed: Code='{1}', Reason='{2}'", methodName, (string)xFaultCode, (string)xFaultReason);
string faultCode = (string)xFaultCode;
var codeParts = faultCode.Split(':');
XmlQualifiedName xmlQualifiedName;
if (codeParts.Length == 2)
{
xmlQualifiedName = new XmlQualifiedName(codeParts[1], codeParts[0]);
}
else
{
xmlQualifiedName = new XmlQualifiedName(faultCode);
}
throw new SoapException((string)xFaultReason, xmlQualifiedName);
}
var xResponse = xReplyBody.Element(this.nsArgs + (methodName + "Response"));
var xResult = xResponse.Element(this.nsArgs + (methodName + "Result"));
if (this.OnTransferring != null)
{
if (!this.OnTransferring())
{
return null;
}
}
var result = new CallResult(xResult);
return result;
}
答
我已经找到了答案，这里的问题：
的想法是要么剥离/从输入替换Unicode字符或XML使用Unicode编码构建时：
<?xml version="1.0" encoding="UTF-16"?>
在上面的帖子中，接受的答案建议使用下面的方法（在此粘贴），但是如果您需要替换某些字符，则需要添加一些代码来执行此操作：
public string getRidOfUnprintablesAndUnicode (string inpString)
{
string outputs = String.Empty;
for (int jj = 0; jj < inpString.Length; jj++)
{
char ch = inpString[ jj ];
if (((int)(byte)ch) >= 32 & ((int)(byte)ch) <= 128)
{
outputs += ch;
}
}
return outputs;
}
代码不完整。 'replyLength'在哪里定义？在这个例子中你有一个无尽的'for'循环。 – jgauffin 2011-03-29 10:03:41
好的，我已经粘贴了现在在这里完成这项工作的方法。从第70行开始，你可以看到我如何建立字符串。 – Shahid 2011-03-29 10:17:21
你从WebServicesClient.Debug.DebugMessage（WebServicesClient.Debug.MASK_RPC_CALL，“RPC（Method = {0}，replyContent ='{1}'）”，methodName，replyContent）获得了什么输出结果？' – jgauffin 2011-03-29 10:28:29