[Moved] Document/View application

giesbert

Hi mlong,

I think you meant:

MainWindow.cpp:
@
MainWindow::someMethod() {
userEntry->userdata(&info); // Pass the address of info.
// or
userEntry->userdata(info); // using reference
}
@

as userEntry is a pointer to the UserEntry class...

andre

Note that you might want to use forward declares in headers instead of includes. That means:

@
//MainWindow.h
struct Information;

//...
private:
Information* m_information;
@

and then only in the MainWindow.cpp do the #include of Information.h.

It makes compilation faster.

CapnMikey

No luck - below is the MainWindow and Information code. I left out the UserData code to see if this would work before making it any more complex. This generates a compile error code:

         request for member 'a' in 'info', which is of non-class type 'Information*'

Is this caused by a syntax error that I am missing?

MainWindow.h
@
#ifndef MAINWINDOW_H
#define MAINWINDOW_H

#include <QMainWindow>
#include "Information.h"

namespace Ui {
class MainWindow;
}

class MainWindow : public QMainWindow
{
Q_OBJECT

public:
explicit MainWindow(QWidget *parent = 0);
~MainWindow();

private:
struct information *info;

private:
Ui::MainWindow *ui;
};
#endif // MAINWINDOW_H
@

MainWindow.cpp
@
#include "mainwindow.h"
#include "ui_mainwindow.h"

MainWindow::MainWindow(QWidget *parent) :
QMainWindow(parent),
ui(new Ui::MainWindow)
{

ui->setupUi(this);

info = new information;

info.a = "A";
info.b = "B";
info.c = "C";

}

MainWindow::~MainWindow()
{
delete ui;
}
@

Information.h
@
#ifndef INFORMATION_H
#define INFORMATION_H

#include <QtGui>

struct information
{
QString a;
QString b;
QString c;
};

#endif // INFORMATION_H

@

Added @ tags around code sections. Please do that yourself next time; Andre

andre

There seems to be some weird mismatch of #ifdefs and #endif's in your code. Perhaps you should fix these first?

Oh, and in your MainWindow, you have a pointer to an Information instance, while you do not seem to be creating an instance and trying to assign to it as if it is a normal member variable.

CapnMikey

That seems to have come from copying and pasting. The definitions are okay in my files.

andre

[quote author="CapnMikey" date="1315495836"]That seems to have come from copying and pasting. The definitions are okay in my files.[/quote]
Then please update your previous post to reflect the actual state of your code.

CapnMikey

Line 15 in Information.h should be line 25 in MainWindow.h. Sorry about the mixup.

andre

[quote author="CapnMikey" date="1315496093"]Line 15 in Information.h should be line 25 in MainWindow.h. Sorry about the mixup.[/quote]
Please just use the Edit button next to your post to modify it and fix the mixup.

CapnMikey

Okay, the code above has been modified.

Thanks for pointing out the edit button.

giesbert

Your struct is named information, but the definition in mainwindow.h is Information. C++ is case sensitive...

CapnMikey

I corrected the case error as well (see post above). I still get the same error message.

andre

Did you also note the remark I posted above?

[quote author="Andre" date="1315495737"]Oh, and in your MainWindow, you have a pointer to an Information instance, while you do not seem to be creating an instance and trying to assign to it as if it is a normal member variable. [/quote]

CapnMikey

I added line 13 to the MainWindow.cpp above, but I still get the same error message. Is this what you are referring to?

andre

Either, change line 20 in MainWindow.h to
@
Information info;
@
and remove line 13 from MainWindow.cpp, or

change line 20 in MainWindow.h to
@
Information* info;
@
and change lines 15-17 in MainWindow.cpp into:
@
info->a = "A";
info->b = "B";
info->c = "C";
@
and insert this line as line 24:
@
delete info;
@

In either case, get yourself a good book on C++ to get yourself a solid introduction into the language you are working with. Qt does not shield you from having to learn C++.

CapnMikey

Andre (and everyone else)-

Thanks for your help. The first option compiled correctly. Having seen something that works I hope I can figure it out from here.

Your last suggestion is appropriate. However, I have a large stack of books on C++ programming. The problem is the definition of the word "good". I learned chemical thermodynamics by using a large stack of textbooks. None of them were particulary good. But, where one authors explanation failed another author provided a more thorough explanation. I was able to piece it together that way. I haven't found that approach to work well for understanding C++.