Reading XML file

JonB

@russjohn834
Well I'm not a C++-er, I thought you'd know from the doc page syntax, you're already managing to pass &file for QIODevice *dev.

QDomDocument::setContent(QIODevice *dev, QString *errorMsg = nullptr, int *errorLine = nullptr, int *errorColumn = nullptr)

so something like

QFile file;
QString errorMsg;
int errorLine, errorColumn;
if (!document.setContent(&file, &errorMsg, &errorLine, &errorColumn))
{
    qDebug () << errorMsg << errorLine << errorColumn
}

and then look at your XML in that light.

russjohn834

Thank you @JonB, @jsulm

I'm getting following error (AND error line and column) while loading the xml file:

"unexpected character" 2 18

I'm trying to read xml file, which looks like this:

<!DOCTYPE detailsentry>
<subject details catagory="run session">
 <Name>Rob</Name>
 <Surname>John</Surname>
 <Patient ID>PID12HL</Patient ID>
 <Date>06/11/2019 20:00</Date>
 <Clinician Note>Some notes to test this section</Clinician Note>
</subject details>

Any thoughts where I'm getting wrong!?

JonB

@russjohn834
Yes, on line #2 at character #18! That's what the error info is telling you!

<subject details catagory="run session">

Count to character #18. What is that space in details catagory? (BTW, if you care, category is the correct spelling.) You can't have a space there. Nor for that matter in your <Patient ID> and others. Where did you get this XML from? You need to test your code with well-formed XML before you do anything.

russjohn834

Thank you @JonB !

Now I can load the xml file:

<!DOCTYPE detailsentry>
<subject_details>
 <Name>Pete</Name>
 <Surname>Batty</Surname>
 <Patient_ID>GH34TRM</Patient_ID>
 <Date>06/11/2019 16:00</Date>
 <Clinician_Note>test note to display</Clinician_Note>
</subject_details>

I'm trying to parse data ( Name, Surname, Patient_ID, Date and Clinician_Note ) from this xml file.

 domDocument.setContent(&file);
 QDomElement topElement = domDocument.documentElement();
 QDomNode domNode = topElement.firstChild();

 while (!domNode.isNull())
 {
    QDomElement domElement = domNode.toElement();
     if (!domElement.isNull())
        {
         qDebug()<<"Tag name"<<topElement.tagName();
        }
  domNode = domNode.nextSibling();
 }

This does output Tag name "subject_details" for 5 times. How can I parse content in each line?

JonB

@russjohn834 said in Reading XML file:

qDebug()<<"Tag name"<<topElement.tagName();

This does output Tag name "subject_details" for 5 times. How can I parse content in each line?

Now come on :) Nicely, if it prints the same thing 5 times, look at what it's printing! topElement doesn't change in the loop, does it? But domElement does ...

russjohn834

This post is deleted!

JonB

@russjohn834
I'm not going to answer every question, you have to do some work yourself! Did you try looking at the documentation for QDomElement since that is what you are asking about? Did you like the look of https://doc.qt.io/qt-5/qdomelement.html#text?

russjohn834

Thanks @JonB . Sorry I just found what I need. I can directly parse content of eachdomElement by following

QDomElement topElement = domDocument.documentElement();
topElement.elementsByTagName("Name").at(0).firstChild().nodeValue();
topElement.elementsByTagName("Surname").at(0).firstChild().nodeValue(); ....

JonB

@russjohn834
You can, but why change over to that way of doing it when you are already using QDomElement calls elsewhere which is more convenient?

russjohn834

I was too much confused about different terminologies of xml and methods of QDomDocument. I'm fiddling with my very basic knowledge to parse data from xml :) Yes, the documentation makes it more clear.

@JonB said in Reading XML file:

@russjohn834
You can, but why change over to that way of doing it when you are already using QDomElement calls elsewhere which is more convenient?

Did not get your point. You recommend topElement.elementsByTagName("Name").at(0).firstChild().nodeValue();
than using loop to fetch content?

JonB

@russjohn834
My thought/inclination is to make the reading code look like the code you already show, which uses QDomElement. So something like:

QDomNode domElement = topElement.elementsByTagName("Name").at(0).firstChild();
QDomElement domElement = domNode.toElement();
if (!domElement.isNull())
    return domElement.text();

It's just a matter of style. If I access things in one place via QDomElement calls, with its dedicated text() method, I prefer not to go via QDomNode and its nodeValue() method somewhere else. Just makes code easier to search/maintain/debug if you keep consistency where possible.

russjohn834

Thanks for your feedback @JonB , @jsulm