How to draw a square from point data with a geometry shader

Question

I want to draw a square from point data with the geometry shader.

In the vertex shader, I emit a single point.

#version 330 core
void main() {
    gl_Position = vec4(0, 0, 0, 1.0);
}

In the geometry shader, I want to create a triangle strip forming a square.

The size is irrelevant at the moment so the model should have a size of 1 (ranging from (-0.5, -0.5) from the initial point position to (+0.5, +0.5).

I need help to calculate the position of the emitted vertices, as visible in the code:

#version 330 core

layout(points) in;
layout(triangle_strip, max_vertices=4) out;

out vec2 tex_coord;

uniform mat4x4 model;
uniform mat4x4 view;
uniform mat4x4 projection;

void main() {
    int i;
    for(i = 0; i < gl_in.length(); ++i) {
        mat4x4 trans;

        trans = //???
        gl_Position = projection * view * model * trans * gl_in[i].gl_Position;
        tex_coord = vec2(0, 0);
        EmitVertex();

        trans = //???
        gl_Position = projection * view * model * trans * gl_in[i].gl_Position;
        tex_coord = vec2(1, 0);
        EmitVertex();

        trans = //???
        gl_Position = projection * view * model * trans * gl_in[i].gl_Position;
        tex_coord = vec2(1, 1);
        EmitVertex();

        trans = //???
        gl_Position = projection * view * model * trans * gl_in[i].gl_Position;
        tex_coord = vec2(0, 1));
        EmitVertex();
    }
    EndPrimitive();
}

I thought to use trans to translate the initial point to the desired coordinates. How would I realize this?

Edit for clarity

I want to generate from a single point what else would be given by the vertex buffer; a single plane:

float vertex[] {
        -0.5,  0.5, 0.0,
         0.5,  0.5, 0.0,
         0.5, -0.5, 0.0,    
        -0.5, -0.5, 0.0
}

Instead I give only a point in the middle of these points and want to generate the real points by subtracting and adding the differences to the center (0.5 and -0.5 etc.). All I need is to know how to apply this transformation in the code (where the ??? are).

Answer 1

Judging by your updated question, I think this pseudo-code should get you pointed in the right direction. It seems to me that all you want to do is offset the x and y coordinates of your point by a constant amount, so an array is the perfect way to do this.

const vec3 offset [4] =
  vec3 [] ( vec3 (-0.5,  0.5, 0.0),
            vec3 ( 0.5,  0.5, 0.0),
            vec3 ( 0.5, -0.5, 0.0),
            vec3 (-0.5, -0.5, 0.0) );

const vec2 tc     [4] =
  vec2 [] ( vec2 (0.0, 0.0),
            vec2 (1.0, 0.0),
            vec2 (1.0, 1.0),
            vec2 (0.0, 1.0) );

void
main (void)
{
  int i;
  for (i = 0; i < gl_in.length (); ++i) {
    gl_Position = projection * view * model * (gl_in [i].gl_Position + offset [0]);
    tex_coord   = tc [0];
    EmitVertex   ();

    gl_Position = projection * view * model * (gl_in [i].gl_Position + offset [1]);
    tex_coord   = tc [1];
    EmitVertex   ();

    gl_Position = projection * view * model * (gl_in [i].gl_Position + offset [2]);
    tex_coord   = tc [2];
    EmitVertex   ();

    gl_Position = projection * view * model * (gl_in [i].gl_Position + offset [3]);
    tex_coord   = tc [3];
    EmitVertex   ();
  }
  EndPrimitive ();
}