Qt-interest Archive, March 2002
QDom and XML
Message 1 in thread
Hello,
I have an XML file beginning with
<?xml version="1.0" encoding="ISO-8859-1"?>
No I created a QDomDocument and used setContents to set its contents
to the contents of the file. How can I retrieve the encoding, so that I can
use QTextCodec to decode the text contained in nodes? especially for
other encodings that iso-8859-1?
Regards,
Volker augustin
Message 2 in thread
Hi,
> I have an XML file beginning with
> <?xml version="1.0" encoding="ISO-8859-1"?>
> No I created a QDomDocument and used setContents to set its contents
> to the contents of the file. How can I retrieve the encoding, so that I can
> use QTextCodec to decode the text contained in nodes? especially for
> other encodings that iso-8859-1?
The text in the nodes (and all other strings) are QStrings -- i.e.
already unicode. The parsing that via the setContent() should already
take care of the right encoding. There is no need to use a QTextCodec to
get the strings.
Rainer
--
[ signature omitted ]
Message 3 in thread
> > The text in the nodes (and all other strings) are QStrings -- i.e.
> > already unicode. The parsing that via the setContent() should already
> > take care of the right encoding. There is no need to use a QTextCodec to
> > get the strings.
>
> This problem came up in the KNewsticker applet of KDE 3.0. In
> kdenetwork/knewsticker/common/xmlnewsaccess.cpp there is the following
> (shortened) code:
>
> QDomDocument domDoc;
> domDoc.setContent(QCString(data));
> QDomNodeList items = domDoc.elementsByTagName(QString::fromLatin1("item"));
> QDomNode itemNode;
> QString headline;
> for (unsigned int i = 0; i < items.count(); i++) {
> itemNode = items.item(i);
> headline =
> itemNode.namedItem(QString::fromLatin1("title")).toElement().text();
> }
One problem in your code is that you use the QDomDocument::setContent()
overload that takes a QCString. In this case the input is considered to
be UTF8 (which it isn't!) -- in general, one should not use this
function, since the XML document could be in an encoding that is not
representable as a C string anyway (e.g. in UTF16).
> The initial data here is a QByteArray. [...]
So why do you convert it into a QCString instead of using the QByteArray
directly?
> Replacing the QCString in line #2 with QString [...]
This is in your case as bad as using a QCString, since the encoding is
ignored as well.
> [...]
> For japanese http://slashdot.jp/slashdot.rdf. [...]
The file I get from this URL does not seem to be an XML file at all.
Rainer
--
[ signature omitted ]
