Bresenham line drawer crashes after certain loop before rendering anything?

Question

I am trying to draw histogram using Bresenham Line drawing algorithm using OpenGL in CodeBlocks. When arguments to drawLine are changed (manually/ hardcoded) the program crashes. The code is not complete and I am just trying to experiment around. Why is the program crashing after certain loop before rendering anything?

I have tried different argument within the drawLine function inside drawHistogram function. Also there seems problem with loop itself. I can't work my head around there. Though the logic seems OK to me, the result is not quite as expected.

#include <iostream>
#include <GL/glut.h>

using namespace std;

void drawLine();
void inputData();
void CoordinateAxes();
void plot(float x, float y);

void drawBar(float );
void drawHistogram();

float x, y, dx, dy, D1, D2, xInc, yInc;

//int frequency[5] = {100, 200, 150, 300, 500} ;

int main(int argc, char** argv)
{
    //inputData();

    glutInit(&argc, argv);
    glutInitDisplayMode(GLUT_RGB);
    glutInitWindowPosition(100, 100);
    glutInitWindowSize(500,500);

    glutCreateWindow("DDA Algorithm");
    glutDisplayFunc(drawHistogram);

    glutMainLoop();
    return 0;
}

void drawLine(float x1, float y1, float x2, float y2)
{
    glClear(GL_COLOR_BUFFER_BIT);
    CoordinateAxes();
    glColor3ub(255 ,255,255);
    glBegin(GL_POINTS);

        dx = x2 - x1;
        dy = y2 - y1;

        float P[int(dx)];

        if (dy >= dx)
        {
            D1 = dx;
            D2 = dy;
            xInc = 0;
            yInc = 1;
        }
        else
        {
            D1 = dy;
            D2 = dx;
            xInc = 1;
            yInc = 0;
        }

        P[0] = 2*D1 - D2;


        x = x1;
        y = y1;
        //plot(x, y);
        //cout << "(" << x << "," << y << ") ";
        //cout << D2;



        for(int i = 0; i < D2 ; i++)
        {
            if (P[i] <= 0)
            {
                plot(x , y);
                P[i+1] = P[i] + 2*D1;
                x += xInc;
                y += yInc;
                //cout << "(" << x << "," << y << ")";
            }
            else
            {
                plot(x , y );
                P[i+1] = P[i] + 2*D1 - 2*D2;
                x += 1;
                y += 1;
                //cout << "(" << x << "," << y << ")";
            }
        }
    glEnd();
    glFlush();
}

//void inputData()
//{
//    cout << "Enter frequencies of 5 data: ";
//
//    for (int i = 0; i < 5; i++)
//    {
//        cin >> freq[i];
//    }
//}

void CoordinateAxes()
{
    glColor3ub(0,0,255);
    glBegin(GL_LINES);
        glVertex2f(0,1);
        glVertex2f(0,-1);
        glVertex2f(-1,0);
        glVertex2f(1,0);
    glEnd();
}

void plot(float x, float y)
{
    glVertex2f(x/500, y/500);
    cout << "(" << x << "," << y << ")";
}

//void drawBar(float freq)
//{
//    drawLine(0, 0, 0, freq );
//    drawLine(0, freq, 25, freq);
//    drawLine(25, freq, 25, 0);
//}

void drawHistogram()
{
    drawLine(10, 10, 10, 100 );
}

I expect a straight line to be rendered on OpenGL window.

In which line does it crash? Which error do you get? Also: when you are using OpenGL, whats the point of implementing your own line-drawing? OpenGL can already draw lines. — BDL, Jun 11 '19 at 08:53
When you call `drawLine(10, 10, 10, 100 )` then `x1=10` and `x2=10` so `dx=0`because of `dx = x2 - x1;` and finally the array `P` has size 0 `float P[int(dx)];` Probably it has to be `float P[int(std::max(abs(dx)+0.5, abs(dy)+0.5))+1]` — Rabbid76, Jun 11 '19 at 09:18

score 1 · Answer 1 · answered Jun 11 '19 at 09:29

The issue start at the line

drawLine(10, 10, 10, 100 )

This call causes that x1=10 and x2=10, so dx=0 because of dx = x2 - x1;
Finally the array P has size 0,because of float P[int(dx)];

void drawLine(float x1, float y1, float x2, float y2)
{
    // [...]

    dx = x2 - x1;
    dy = y2 - y1;

    float P[int(dx)];

You've to create an array with the size of the absolut maximum difference of x2-x1 and y2-y1:

#include <algorithm>

float P[int(std::max(abs(dx), abs(dy))+0.5)+1];

In c++ I recommend to use std::vector:

#include <vector>

std::vector<float> P(int(std::max(abs(dx), abs(dy))+0.5)+1, 0.0f);

Thanks. Yes the problem was with the array. Now it works fine. — iheathers, Jun 11 '19 at 10:14

Bresenham line drawer crashes after certain loop before rendering anything?

1 Answers1