something wrong when using ccGLCameraParameters

[kasper]

I want to get the ccGLCameraParameters data.

Code: Select all

 ccGLCameraParameters camera;
 glWindow->getGLCameraParameters(camera);

but there's something wrong.

Code: Select all

"QCC_DB_LIB.lib(QCC_DB_LIB.dll):-1: error:
 LNK2005: "public: virtual void __cdecl ccGLMatrixTpl<double>::toIdentity(void)" (?toIdentity@?$ccGLMatrixTpl@N@@UEAAXXZ) 已在 mainwindow.obj 中定義過了"   

it already defined in mainwindow.obj

so I remove all code .

Only left with

Code: Select all

CreateGLWindow(glWindow, widget);
ccGLCameraParameters camera;

it still error.

I found that the problem is "ccGLMatrixd"

Code: Select all

#include <ccGLMatrixTpl.h>
ccGLMatrixd m;

then the same error.

but no error when using "ccGLMatrix"


I also can not find "ccGLCameraParameters" in document?

How to fix it? Thanks.

[kasper]

I don't know how , LNK2005 disappeared Linux OS.

But now I got a new problem,

when I'm using

Code: Select all

ccGenericPointCloud::VisibilityTableType *visibilityArray = cloud->getTheVisibilityArray();

it was error.
I click into getTheVisibilityArray() function, and in header file,
its return type is ccGenericPointCloud::VisibilityTableType not ccGenericPointCloud::VisibilityTableType *

but I serach in document-api online it return ccGenericPointCloud::VisibilityTableType *
why they are different?

Second, after I get the visibility array, visibilityArray is not a null object but its size is 0.

should I do any setting before it?

it's my code.

Code: Select all

    ccGLCameraParameters camera;
    m_glWindow_->getGLCameraParameters(camera);

    const double half_w = camera.viewport[2] / 2.0;
    const double half_h = camera.viewport[3] / 2.0;

    ccGenericPointCloud::VisibilityTableType *visibilityArray = &c->getTheVisibilityArray();

    assert(visibilityArray);

    unsigned cloudSize = c->size();


    qDebug() << "visibilityArray" <<  visibilityArray->size();   // it return 0

    for (int i = 0; i < static_cast<int>(cloudSize); ++i)
    {
        qDebug() << "cloudSize"<< cloudSize;
        if (visibilityArray->at(i) == POINT_VISIBLE)            // it will runtime error because of its size.
        {
            const CCVector3* P3D = c->getPoint(i);

            CCVector3d Q2D;
            camera.project(*P3D, Q2D);

            CCVector2 P2D(	static_cast<PointCoordinateType>(Q2D.x-half_w),
                            static_cast<PointCoordinateType>(Q2D.y-half_h) );

            bool pointInside = CCLib::ManualSegmentationTools::isPointInsidePoly(P2D, m_segmentationPoly_);
            visibilityArray->at(i) = ( false != pointInside ? POINT_HIDDEN : POINT_VISIBLE );
        }
    }
    

[daniel]

The online code documentation is a little bit old (well, actually it's very old). The new API indeed returns a reference on the 'visibility table':
https://github.com/CloudCompare/CloudCo ... oud.h#L171

And before using it, you have to actually allocate it (with resetVisibilityArray) and then populate it (not sure what you want to do with it).