The best way is Tidy the html or xml code

http://fop-miniscribus.googlecode.co...dy/bestqtidy.h
class QTidy
qt4 compatible lib...
http://fop-miniscribus.googlecode.co.../modules/tidy/

after you can parse as xml and read/write all your own chunk html...

or if you search only image on html code to load.....


Qt Code: 
  1. /* read remote image from html code */
  2.  void Load_Image_Connector()
  3. {
  4.     ///////////////////qDebug() << "### Load_Image_Connector";
  5.  
  6.     QRegExp expression( "src=[\"\'](.*)[\"\']", Qt::CaseInsensitive );
  7.     expression.setMinimal(true);
  8.     int iPosition = 0;
  9.     int canna = 0;
  10.     while( (iPosition = expression.indexIn( html , iPosition )) != -1 ) {
  11.         QString semi1 = expression.cap( 1 );
  12.         canna++;
  13.         dimage.append(semi1);    /* image lista 1 */
  14.         AppendImage(semi1);     /* register qmap  images & name & replace qurl remove css file */  
  15.         iPosition += expression.matchedLength();
  16.         ////////////////////qDebug() << "### iPosition " << iPosition;
  17.     }
  18.     QTimer::singleShot(1, this, SLOT(GetRemoteFile()));
  19.  
  20. }
To copy to clipboard, switch view to plain text mode