Question

I’m using Ubuntu 11.10, Qt 4, Qwt 6.0.1

the problem is that in general everything works, examples from Qwt compile without ANY problems, but when I try to convert from QPointF.x to double I get error. The funnies thing is that (on x86) qreal is supposed to be redefinition of double…

here is what doesn’t work (

QPointF pt;
pt.setX(1.0);
pt.setY(2.0);
double px=pt.x;

the compiler returns error:

error: argument of type ‘qreal (QPointF::)()const {aka double (QPointF::)()const}’ does not match ‘double’

Am I doing something stupid?

In short:

I need to convert it to use in class inheriting from QwtSeriesData to make custom interpolation between points (in this case it’s not gonna be simple linear interpolation). For that I need method

double y(double x) const

which will return value of function of given x

when I simplified it to use QwtSeriesData it compiles. For this to work it was just necessary to implement

QPointF sample(size_t i) const
size_t size() const

but as I said, I need custom interpolation, so seems that implementing interface inheriting from QwtSyntheticPointData is the best option.

Answer 1

QPointF is a class (not a struct) and QPointF::x() is a method, not a public member. Try:

double px = pt.x();