Hi there,

I am running QHtml to get a Website with German chars.
For some reason they are not encoded (e.g. &auml), but come raw ('ä').

After getting it, I want to to run it through HTML tidy (to make it valid XML) and then parse it and search for text, but the German chars are not processed correctly.

main.cpp:
Qt Code: 
  1. #include <QtCore/QCoreApplication>
  2. #include <QProcess>
  3. #include <QBuffer>
  4. #include <QDomDocument>
  5. #include <iostream>
  6. #include "synchttp.h"
  7. #include <QFile>
  8.  
  9. using namespace std;
  10.  
  11. int main(int argc, char **argv )
  12. {
  13. 	QCoreApplication a(argc, argv);
  14. 	QBuffer buffer;
  15. 	SyncHTTP http("back2hack.cc");
  16. 	http.syncGet("/", &buffer);
  17. 	QByteArray ar = buffer.data();
  18. 	ar.replace("ä", "ae");
  19. 	QProcess process;
  20. 	QStringList arguments;
  21. 	arguments << "-q" << "-asxml";
  22. 	process.start("./tidy", arguments );
  23. 	process.waitForStarted();
  24. 	process.write(ar);
  25. 	process.waitForBytesWritten();
  26. 	process.closeWriteChannel();
  27.  
  28. 	process.waitForFinished();
  29. 	QByteArray array = process.readAllStandardOutput();
  30. 	//std::cout << array.data();
  32. 	QString error;
  33. 	int row, column;
  34. 	doc.setContent(array, false, &error, &row, &column );
  35. 	std::cout << error.toLatin1().data();
  36. 	//std::cout << doc.firstChild().toText().data().toLatin1().data();
  37. 	QString text;
  38. 	QList<QDomNode> l;
  39. 	QDomElement element = doc.documentElement();
  40. 	QDomNode c = element.firstChild();
  41. 	while(!c.isNull()) {
  42. 		QDomNode n = c.nextSibling();
  43. 		while(!c.isNull()) {
  44. 			if(c.toElement().text() == "Blogsystem: Einträge absteigend sortieren" )
  45. 				l.append(c);
  46. 			c = c.nextSibling();
  47. 		}
  48. 		n = n.firstChild();
  49. 	}
  50. 	foreach(QDomNode n, l) {
  51. 		cout << qPrintable(n.toElement().text());
  52. 		cout << qPrintable(n.toElement().tagName());
  53. 		cout << l.length();
  54. 	}
  55. 	QFile file("pwn");
  56. 	file.open(QFile::WriteOnly);
  57. 	file.write(buffer.data());
  58. 	return 0;
  59. }
To copy to clipboard, switch view to plain text mode 

synchttp.h:
Qt Code: 
  1. /***************************************************************************
  2.  *   Copyright (C) 2005 by Iulian M                                        *
  3.  *   eti@erata.net                                                         *
  4.  ***************************************************************************/
  5. #ifndef ETKSYNCHTTP_H
  6. #define ETKSYNCHTTP_H
  7.  
  8. #include <QHttp>
  9. #include <QEventLoop>
  10. #include <QBuffer>
  11.  
  12. /**
  13.  * Provide a synchronous api over QHttp
  14.  * Uses a QEventLoop to block until the request is completed
  15.  * @author Iulian M <eti@erata.net>
  16. */
  17. class SyncHTTP: public QHttp
  18. {
  19. 	Q_OBJECT
  20. 	public:
  21. 		/// structors
  22. 		SyncHTTP( QObject * parent = 0 )
  23. 		:QHttp(parent),requestID(-1),status(false){}
  24.  
  25. 		SyncHTTP( const QString & hostName, quint16 port = 80, QObject * parent = 0 )
  26. 		:QHttp(hostName,port,parent),requestID(-1),status(false){}
  27.  
  28. 		virtual ~SyncHTTP(){}
  29.  
  30. 		/// send GET request and wait until finished
  31. 		bool syncGet ( const QString & path, QIODevice * to = 0 )
  32. 		{
  33. 			///connect the requestFinished signal to our finished slot
  34. 			connect(this,SIGNAL(requestFinished(int,bool)),SLOT(finished(int,bool)));
  35. 			/// start the request and store the requestID
  36. 			requestID = get(path, to );
  37. 			/// block until the request is finished
  38. 			loop.exec();
  39. 			/// return the request status
  40. 			return status;
  41. 		}
  42.  
  43. 		/// send POST request and wait until finished
  44. 		bool syncPost ( const QString & path, QIODevice * data, QIODevice * to = 0 )
  45. 		{
  46. 			///connect the requestFinished signal to our finished slot
  47. 			connect(this,SIGNAL(requestFinished(int,bool)),SLOT(finished(int,bool)));
  48. 			/// start the request and store the requestID
  49. 			requestID = post(path, data , to );
  50. 			/// block until the request is finished
  51. 			loop.exec();
  52. 			/// return the request status
  53. 			return status;
  54. 		}
  55.  
  56. 		bool syncPost ( const QString & path, const QByteArray& data, QIODevice * to = 0 )
  57. 		{
  58. 			/// create io device from QByteArray
  59. 			QBuffer buffer;
  60. 			buffer.setData(data);
  61. 			return syncPost(path,&buffer,to);
  62. 		}
  63.  
  64. 	protected slots:
  65. 		virtual void finished(int idx, bool err)
  66. 		{
  67. 			/// check to see if it's the request we made
  68. 			if(idx!=requestID)
  69. 				return;
  70. 			/// set status of the request
  71. 			status = !err;
  72. 			/// end the loop
  73. 			loop.exit();
  74. 		}
  75.  
  76. 	private:
  77. 		/// id of current request
  78. 		int requestID;
  79. 		/// error status of current request
  80. 		bool status;
  81. 		/// event loop used to block until request finished
  82. 		QEventLoop loop;
  83. };
  84.  
  85. #endif
To copy to clipboard, switch view to plain text mode 

What could be the reason?