public void setViewport(int i, int j, int width, int height) {
SLog.d(did, String.format("Viewport set: %d/%d", width, height));
viewport[0] = i;
viewport[1] = j;
viewport[2] = width;
viewport[3] = height;
}
@Override
public void onSurfaceChanged(GL10 gl, int width, int height) {
gl.glViewport(0, 0, width, height);
camera.setViewport(0, 0, width, height);
camera.changePerspective(gl, Projection.PERSPECTIVE);
SLog.d(did, "Surface created.");
}
public void changePerspective(GL10 gl, Projection p) {
SLog.d(did, "Changing perspective.");
gl.glMatrixMode(GL10.GL_PROJECTION); // select projection
gl.glLoadIdentity(); // reset
float aspect = ((float) viewport[2] / viewport[3]);
switch (p) {
case PERSPECTIVE:
SLog.d(did, "Setting perspective mode width aspect = " + aspect);
gl.glFrustumf(-aspect, aspect, -1f, 1f, near, far);
break;
case ORTHOGRAPHIC:
SLog.d(did, "Setting orthographic mode.");
gl.glOrthof(-1, 1, -1, 1, -1f, 10.0f);
break;
default:
break;
}
hasChanged = true;
}
static {
SLog.setTag(did, "GL Renderer.");
}
public class GLRenderer implements GameRenderer {
private static final int did = SLog.register(GLRenderer.class);
static {
SLog.setTag(did, "GL Renderer.");
}
private GLSurface surface;
private RenderScheduler renderScheduler;
private GLCamera camera;
public GLRenderer(RenderScheduler scheduler, GLCamera camera) {
this.camera = camera;
this.renderScheduler = scheduler;
}
public void reset(GL10 gl) {}
@Override
public void onSurfaceChanged(GL10 gl, int width, int height) {
gl.glViewport(0, 0, width, height);
camera.setViewport(0, 0, width, height);
camera.changePerspective(gl, Projection.PERSPECTIVE);
SLog.d(did, "Surface created.");
}
public void setCamera(GLCamera camera) {
this.camera = camera;
}
/** Makes renderer work. */
public void renderFrame() {
surface.requestRender();
}
@Override
public void onSurfaceCreated(GL10 gl, EGLConfig config) {
camera.reset(gl);
}
@Override
public void onDrawFrame(GL10 gl) {
gl.glClear(GL10.GL_COLOR_BUFFER_BIT | GL10.GL_DEPTH_BUFFER_BIT);
camera.update(gl);
camera.focusOnObjects(gl);
for (Renderable r : renderScheduler.getRenderList(gl)) {
gl.glPushMatrix();
r.draw(gl);
gl.glPopMatrix();
}
}
@Override
public void setSurface(GLSurface surface) {
this.surface = surface;
}
@Override
public GLSurface getSurface() {
return surface;
}
@Override
public Camera getCamera() {
return camera;
}
}
static {
SLog.setTag(did, "Camera.");
SLog.setLevel(did, Level.VERBOSE);
}
/**
* Motion observer musi być świadomy gdzie znajduje się kamera! Inaczej współrzędne nic nie znaczą,
* bo nie wiadomo gdzie jesteśmy w przestrzeni.
*
* <p>http://stackoverflow.com/questions/2093096/implementing-ray-picking
* http://ovcharov.me/2011/01/14/android-opengl-es-ray-picking/
*
* <p>Może macierze powinienem przetrzymywać jako zwykle floaty i je przestawiać po prostu sobie?
* http://www.learnopengles.com/android-lesson-one-getting-started/ a na koniec podawać do opengl
*
* <p>Faktycznie, samemu powinno się trzymać te macierze! !!!!!!!!!!!!!!!!!!!!!!!!!!! When working
* with opengles 2 in Android you keep track of your matrices yourself, usually as a couple of
* float[]. If you calculate them directly in the shader, I don't think that you can get them.
* !!!!!!!!!!!!!!!!!!!!!!!!!!!
*
* <p>Opengl 2.0 nie ma już nawet funkcji translate czy scale, trzeba robić to samemu! A więc każdy
* obiekt będzie pewnie musiał mieć dostęp jakoś do kamery i zlecać przestawienia modelview...
*
* <p>dobre http://stackoverflow.com/questions/6261867/question-about-3d-picking-in-
* android-with-opengl-es-2
*
* <p>Wszelkie glulookat itd przestają mieć rację bytu, wszystkie operacje na macierzach robimy
* systemowo a dostarczamy je po prostu do renderowania
*
* <p>Znikają funkcje jak gl.matrixMode itd.
*
* <p>Na razie oldschoolowo PICK RAY (przez wczytywanie macierzy z opengl na siłę przy użyciu
* glGetFloat - które oczywiście znika razem z translate itp itd. w następnej wersji opengl). Na
* razie mogę sobie korzystać ze wszystkich GLU.cośtam Pojawia się pytanie najważniejsze - czy przy
* obliczaniu ray musimy znać aktaulną dla danego obiektu macierz model?
*
* <p>Jeśli tak by było to każdy renderer musi sobie trzymać kopię modelview matrix. Projection się
* nie zmienia więc może siedzieć tutaj.
*
* @author kboom
*/
public class GLCamera implements Camera {
private static final int did = SLog.register(GLCamera.class);
static {
SLog.setTag(did, "Camera.");
SLog.setLevel(did, Level.VERBOSE);
}
private boolean hasChanged = true;
// Position the eye behind the origin.
private float xEye = 0f;
private float yEye = 0f; // put somewhere here to get pretty much good view
// over things
private float zEye = 3f;
// We are looking toward the distance
private float xLook = 0f;
private float yLook = 0f;
private float zLook = 0f;
// This is where our head (not eyes!) would be pointing were we holding the
// camera.
final float xUp = 0.0f;
final float yUp = 1.0f;
final float zUp = 0.0f;
private float near = 1f;
private float far = 100f;
/*
*
* Most important matrices. (chyba do wywalenia...)
*/
/** Screen dimensions. */
private final int[] viewport = new int[4];
/**
* This matrix can be thought of as our camera. This matrix transforms world space to eye space;
* it positions things relative to our eye.
*/
private final float[] modelview = new float[16];
/**
* How the objects are rendered in the space. Probably will be used by every object to draw itself
* onto the screen.
*/
private final float[] projection = new float[16];
// dummy one
private final float[] dummy = new float[16];
public GLCamera() {
SLog.v(did, "+++ constructor +++");
}
public void setViewport(int i, int j, int width, int height) {
SLog.d(did, String.format("Viewport set: %d/%d", width, height));
viewport[0] = i;
viewport[1] = j;
viewport[2] = width;
viewport[3] = height;
}
public void reset(GL10 gl) {
SLog.v(did, "Reseting..");
hasChanged = true;
}
public enum Projection {
PERSPECTIVE,
ORTHOGRAPHIC
}
public void changePerspective(GL10 gl, Projection p) {
SLog.d(did, "Changing perspective.");
gl.glMatrixMode(GL10.GL_PROJECTION); // select projection
gl.glLoadIdentity(); // reset
float aspect = ((float) viewport[2] / viewport[3]);
switch (p) {
case PERSPECTIVE:
SLog.d(did, "Setting perspective mode width aspect = " + aspect);
gl.glFrustumf(-aspect, aspect, -1f, 1f, near, far);
break;
case ORTHOGRAPHIC:
SLog.d(did, "Setting orthographic mode.");
gl.glOrthof(-1, 1, -1, 1, -1f, 10.0f);
break;
default:
break;
}
hasChanged = true;
}
public void focusOnObjects(GL10 gl) {
gl.glMatrixMode(GL10.GL_MODELVIEW);
gl.glEnable(GL10.GL_DEPTH_TEST);
gl.glClearColor(0f, 0f, 0f, 1.0f);
}
public void update(GL10 gl) {
gl.glMatrixMode(GL10.GL_PROJECTION);
// dopisane
gl.glMatrixMode(GL10.GL_MODELVIEW);
// gl.glLoadIdentity();
// koniec
if (hasChanged) {
GLU.gluLookAt(gl, xEye, yEye, zEye, xLook, yLook, zLook, xUp, yUp, zUp);
hasChanged = false;
}
// GL11 gl11 = (GL11) gl;
// gl11.glGetFloatv(GL10.GL_MODELVIEW, modelview, 0); // wystarczyło
// dopisać 0 i już
// traktuje jako
// float
// gl11.glGetFloatv(GL10.GL_PROJECTION, projection, 0);
// NIEPOTRZEBNE BO DZIAła i bez tego
/*
* gl.glLoadIdentity(); gl.glFrustumf(-aspect, aspect, -1f, 1f, near,
* far); gl.glRotatef(xAngle * 10, 1f, 0, 0); gl.glRotatef(yAngle * 10,
* 0, 1f, 0); gl.glRotatef(zAngle, 0f, 0f, 1f); gl.glTranslatef(xPos,
* yPos, zPos);
*/
}
@Override
public void accomodateTo(float near, float far) {
this.near = near;
this.far = far;
hasChanged = true;
}
@Override
public void move(float x, float y, float z) {
xEye = x;
yEye = y;
zEye = z;
hasChanged = true;
}
@Override
public void rotate(float xAngle, float yAngle, float zAngle) {
// just rotate the coordinates
hasChanged = true;
}
@Override
public void getRay(float x, float y, float z, float[] ray) {
/*
*
* normalised_x = 2 * mouse_x / win_width - 1 normalised_y = 1 - 2 *
* mouse_y / win_height // note the y pos is inverted, so +y is at the
* top of the screen
*
* unviewMat = (projectionMat * modelViewMat).inverse()
*
* near_point = unviewMat * Vec(normalised_x, normalised_y, 0, 1)
* camera_pos = ray_origin = modelViewMat.inverse().col(4) ray_dir =
* near_point - camera_pos
*/
// GLU.gluUnProject(winX, winY, winZ, model, modelOffset, project,
// projectOffset, view, viewOffset, obj, objOffset)
GLU.gluUnProject(x, viewport[3] - y, z, modelview, 0, projection, 0, viewport, 0, ray, 0);
// fix
if (ray[3] != 0) {
ray[0] = ray[0] / ray[3];
ray[1] = ray[1] / ray[3];
ray[2] = ray[2] / ray[3];
}
ray[0] = ray[0] - xEye;
ray[1] = ray[1] - yEye;
ray[2] = ray[2] - zEye;
// 0f jako 4 ta?
}
@Override
public void lookAt(float x, float y, float z) {
xLook = x;
yLook = y;
zLook = z;
}
/*
* OBOWIAZUJACE PODEJSCIE normalizedPoint[0] = (x * 2 / screenW) -1;
* normalizedPoint[1] = 1 - (y * 2 / screenH); normalizedPoint[2] = ?
* normalizedPoint[3] = ? matrix = perspective_matrix x model_matrix
* inv_matrix = inverse(matrix) outpoint = inv_matrix x normalizedPoint
*/
// albo to?
/*
* public void unproject(float [] in, float [] out) {
*
* // multiply to get mvp float [] mvp = dummy; Matrix.multiplyMM(mvp, 0,
* projection, 0, modelview, 0); // invert it Matrix.invertM(mvp, 0, mvp,
* 0);
*
* // screen coordinates are already normalized // compute it
* Matrix.multiplyMV(out, 0, mvp, 0, in, 0);
*
* if (out[3] == 0) return null;
*
* out[3] = 1 / out[3];
*
* float o1 = out[1]; float o2 = out[2]; float o3 = out[3]; float o4 =
* out[4];
*
* out[0] = o1 * o3; out[1] = o1 * o3; out[2] = o2 * o3; out[3] = 0; // not
* really needed...
*
* // potem
*
* Vec3 near = unProject(30, 50, 0, mvpMatrix, 800, 480); Vec3 far =
* unProject(30, 50, 1, mvpMatrix, 800, 480); // 1 for winz means projected
* on the far plane Vec3 pickingRay = Subtract(far, near); // Vector
* subtraction
*
* }
*/
}
public void reset(GL10 gl) {
SLog.v(did, "Reseting..");
hasChanged = true;
}
public GLCamera() {
SLog.v(did, "+++ constructor +++");
}