Develop Reference

lwjgl 3 macOS issue with creating windows

I created some code to work with lwjgl3, and ran into an issue with the macOS bindings since I am working on a mac.
The issue is that I cannot seem to get my program to create any windows. I know this for sure as the program just stops at that point. There are no errors.
The line the code gets stuck on is:
long window = glfwCreateWindow(300, 300, "Hello World!", MemoryUtil.NULL, MemoryUtil.NULL);
Since I am developing on IntelliJ, whenever I disconnect the process, it crashes with :
Exception Type: EXC_BAD_ACCESS (SIGABRT)
Exception Codes: EXC_I386_GPFLT
Exception Note: EXC_CORPSE_NOTIFY
I feel that that is not the issue, as I tried running another code dependant on lwjgl3, and it also got stuck at the part creating the window.
I know that the code is stuck at the part to create the window as I have put print statements every line.
I really hope that this is not a compatibility issue with either macOS Big Sur or the mac book pro 2015 model.
while glfwInit() return true, when I run System.out.println(glGetString(GL_VERSION) );, I get this error Fatal error: Thread[main,5,main]: No context is current or a function that is not available in the current context was called. The JVM will abort execution.
Here is the complete code that I am using for testing:
public static void main(String[] args) throws Exception{
GLFWErrorCallback.createPrint(System.err).set();
// Initialize GLFW. Most GLFW functions will not work before doing this.
if ( !glfwInit() )
throw new IllegalStateException("Unable to initialize GLFW");
glfwWindowHint(GLFW_CONTEXT_VERSION_MAJOR, 2);
glfwWindowHint(GLFW_CONTEXT_VERSION_MINOR, 0);
// Configure GLFW
glfwDefaultWindowHints(); // optional, the current window hints are already the default
glfwWindowHint(GLFW_VISIBLE, GLFW_FALSE); // the window will stay hidden after creation
glfwWindowHint(GLFW_RESIZABLE, GLFW_TRUE); // the window will be resizable
glfwWindowHint(GLFW_CONTEXT_VERSION_MAJOR, 3);
glfwWindowHint(GLFW_CONTEXT_VERSION_MINOR, 3);
// Should be true for macOS, according to GLFW docs, to get core profile.
glfwWindowHint(GLFW_OPENGL_FORWARD_COMPAT, GL_TRUE);
// According to Apple docs, non-core profiles are limited to version 2.1.
glfwWindowHint(GLFW_OPENGL_PROFILE, GLFW_OPENGL_CORE_PROFILE);
// Create the window
long window = glfwCreateWindow(300, 300, "Hello World!", MemoryUtil.NULL, MemoryUtil.NULL);
//glClearColor(1.0f, 0.0f, 0.0f, 0.0f);
System.out.println('0');
glfwDestroyWindow(window);
}
I have googled this multiple times:
cannot seem to find a relevant answer. One person suggested that macOS has a hidden popup that had to be closed for the window to be assigned, but I did everything to disable it (as I could not find it) and still nothing happened. I think the GL_VERSION indicates an issue with how openGL is being bound to Java.
[Edit]
Ensured that the app was running on the main thread by passing in the JVM option -XstartOnFirstThread.
[EDIT2]
After leaving the project for a few hours, I came back and reran it.
I added one extra line on the top:
System.out.println("Hello LWJGL " + Version.getVersion() + "!");
The createWindow(...) function started working, which is weird as I do not know why. It also started working if I removed the line, so if anyone knows what was happening please let me know.
I am leaving the question unanswered as I still do not know the reason for the issue, I just happened to be lucky that it fixed itself.
[EDIT3]
I force rebuilt the entire project on intellij and it stopped working again. Am really confused as to why it was working in between

I had the -XstartOnFirstThread on the gradle.properties. Try set it in the jvmargs of the run configuration you currently have. That fixed it for me

Android native SDL glContext messed up picture

I'm trying to setup SDL2 native application with use of custom glContext.
Java part: this has by default Absolute layout which is deprecated. And i wonder if i need to use some kind of SurfaceView:
mSurface = new SDLSurface(getApplication());
mLayout = new RelativeLayout(this);
mLayout.setLeft(0);
mLayout.setTop(0);
mLayout.addView(mSurface);
setContentView(mLayout);
Next, i'm creating opengl context in my native code:
SDL_SetHint(SDL_HINT_RENDER_DRIVER, "opengles2");
SDL_SetHint(SDL_HINT_RENDER_OPENGL_SHADERS, "1");
SDL_GL_SetAttribute(SDL_GL_CONTEXT_PROFILE_MASK, SDL_GL_CONTEXT_PROFILE_ES);
SDL_GL_SetAttribute(SDL_GL_CONTEXT_MINOR_VERSION, 2);
SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 2);
SDL_GL_SetAttribute( SDL_GL_DOUBLEBUFFER, 1 );
SDL_GL_SetAttribute( SDL_GL_DEPTH_SIZE, 0 );
SDL_GL_SetAttribute( SDL_GL_RED_SIZE, 8 );
SDL_GL_SetAttribute( SDL_GL_GREEN_SIZE, 8 );
SDL_GL_SetAttribute( SDL_GL_BLUE_SIZE, 8 );
SDL_GL_SetAttribute( SDL_GL_ALPHA_SIZE, 8 );
m_glContext = SDL_GL_CreateContext(_window);
SDL_GL_MakeCurrent(_window, m_glContext);
Window is created like this:
m_mainWindow = SDL_CreateWindow("clovo", SDL_WINDOWPOS_CENTERED, SDL_WINDOWPOS_CENTERED, 2048, 1536,
SDL_WINDOW_SHOWN | SDL_WINDOW_OPENGL | SDL_WINDOW_RESIZABLE | SDL_WINDOW_ALLOW_HIGHDPI | SDL_WINDOW_FULLSCREEN);
I thought the case may be in size of window, but after creation the size is fits the screen. Also, this code works fine in iOS.
I can't find a solution for a problem. Can't even find a problem. I'm new to Android, so maybe there something about glContext what i should know?

Found the problem.
It's occur when i'm using my own framebuffers and then setting it back to default.
In order to not set wrong buffer at my renderer initialization i'm storing default framebuffer
GLint default_buffer;
glGetIntegerv(GL_FRAMEBUFFER_BINDING, &default_buffer);
And then i can restore default buffer to draw to screen.

Exporting Eclipse project causes incorrect texturing and crashes,

This issue is now fixed. My shader attributes were not bound correctly.
I have got a game, which when ran from the IDE looks like this:
However, when I export it from Eclipse with these settings,
The texturing is completely incorrect. The textures are still loaded, but they are not wrapping correctly onto the object.
The code is exactly the same, as I only just exported it, and I am currently running both windows of the game at the same time(One working fine from the IDE, one looking weird from an exported JAR).
I have also copied all the resources from the IDE directory to the folder with the external JAR. IDE directory:
Directory I run the external JAR from:
Also, I know that the textures are actually loading - They are not wrapping correctly. I know this because:
If you look at the plane, you can see that it still has elements of the texture - They are just all stretched and messed up.
Same thing with the Skybox. If you look at that, parts are still there, but again, it's incorrectly wrapped around the OBJ model.
If I hit the Q key to render the terrain with a DisplayList (Wrapping the terrain texture multiple times), it shows the texture. Can't get a screenshot of this because I can't hit Q and take a screenshot.
I have checked inside the JAR file, and the fragment and the vertex shader are still there. The correct libraries also appear to be there (Besides, if they weren't, the game would not even start).
Update:
As I was restarting the game multiple times to check for more information, I noticed that as soon as the display shows (so I can see the incorrectly-textured terrain and plane), the game freezes about 90% of the time. No Stacktrace, just a "This window is not responding and windows is closing it".
The game still works perfectly without crashing when I run it in the IDE.
The server for the game exports and runs perfectly. Only the client is the issue.
What about exporting could make the game any different than running it in the IDE, and how can I solve it?
Update:
So here is my texture loading code:
loader.loadTexture("PlaneTexture", 1);
//loadTexture() method:
public int loadTexture(String fileName, int mipmap) {
Texture texture = null;
try {
try{
texture = TextureLoader.getTexture("PNG", new FileInputStream("res/" + fileName + ".png")); //TextureLoader is a Slick-Util class.
}catch (Exception e){
e.printStackTrace();
}
if (texture == null){
throw new Exception("Null texture!");
}
//texture = TextureLoader.getTexture("GIF", new FileInputStream("res/" + fileName + ".png"));
if (mipmap > -10){
GL30.glGenerateMipmap(GL11.GL_TEXTURE_2D);
GL11.glTexParameteri(GL11.GL_TEXTURE_2D, GL11.GL_TEXTURE_MIN_FILTER, GL11.GL_LINEAR_MIPMAP_LINEAR);
GL11.glTexParameterf(GL11.GL_TEXTURE_2D, GL14.GL_TEXTURE_LOD_BIAS, mipmap);
}
} catch (Exception e) {
e.printStackTrace();
System.err.println("Tried to load texture " + fileName + ".png , didn't work");
System.exit(1);
return -1;
}
textures.add(texture.getTextureID());
return texture.getTextureID();
}
I now have the texture ID of the texture. I then render the object (in this case the plane) like this:
Plane you = Main.TerrainDemo.shipsID.get(Main.TerrainDemo.UID);
Main.TerrainDemo.shader.start();
TexturedModel texturedModel = TerrainDemo.shipModel; //The plane model
RawModel model = texturedModel.getRawModel();
GL30.glBindVertexArray(model.getVaoID());
GL20.glEnableVertexAttribArray(0);
GL20.glEnableVertexAttribArray(1);
GL20.glEnableVertexAttribArray(2);
GL13.glActiveTexture(GL13.GL_TEXTURE0);
GL11.glBindTexture(GL11.GL_TEXTURE_2D, TerrainDemo.shipModel.getTexture().getID()); //The ID of the texture.
glDrawElements(GL_TRIANGLES, model.getVertexCount(), GL11.GL_UNSIGNED_INT, 0);
And, although I don't think they're important, my vertex and fragment shaders:
//Vertex shader
#version 130
in vec3 position;
in vec2 textureCoords;
in vec3 normal;
out vec2 pass_textureCoords;
out vec3 surfaceNormal;
out vec3 toLightVector;
uniform mat4 transformationMatrix;
uniform vec3 lightPosition;
void main(void){
gl_Position = ftransform();
pass_textureCoords = textureCoords;
surfaceNormal = (transformationMatrix * vec4(normal, 0.0)).xyz;
toLightVector = (vec3(500, 50000, 500)) - (transformationMatrix * vec4(position, 1.0)).xyz;
}
//Fragment shader
#version 130
in vec2 pass_textureCoords;
in vec3 surfaceNormal;
in vec3 toLightVector;
out vec4 out_Color;
uniform sampler2D textureSampler;
uniform vec3 lightColour;
void main(void){
vec3 unitNormal = normalize(surfaceNormal);
vec3 unitLightVector = normalize(toLightVector);
float nDot1 = dot(unitNormal, unitLightVector);
float brightness = max(nDot1, 0.2);
brightness = brightness + 0.5;
vec3 diffuse = brightness * vec3(1, 1, 1);
vec4 textureColor = vec4(diffuse, 1.0) * texture(textureSampler, pass_textureCoords);
if(textureColor.a<1){
discard;
}
out_Color = vec4(textureColor.r, textureColor.g, textureColor.b, 1);
}
Again, I will stress that all of this is working perfectly if the game is running from the IDE. It is just if it runs from an external JAR that the issue occurs.
I will be experimenting with different texture loading techniques and methods (e.g. packing textures into the JAR) and seeing if anything different happens.
Yet another update:
So, I sent the game to another person (They also use windows 8), and the game worked perfectly! No texturing errors whatsoever! So now I'm unsure if the problem is with my PC specifically or something else.
For those who wish to try, you can download the game at http://endcraft.net/PlaneGame and see it yourself (Please read the instructions.txt - Also, you'll need a program to decompress .rar files).
I will be getting as many people as I know to give the game a go and see if they have the same issue or if the texturing is correct.
It is completely baffling me that it works fine when I run it from the IDE, but does not work when I export into an external jar, but does work when I export it into an external jar and send it to someone else!
(another) Update:
I have sent the game to multiple people, some of them are coming across this crash:
# A fatal error has been detected by the Java Runtime Environment:
#
# EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x1cd6b57f, pid=36828, tid=35556
#
# JRE version: Java(TM) SE Runtime Environment (8.0_31-b13) (build 1.8.0_31-b13)
# Java VM: Java HotSpot(TM) Client VM (25.31-b07 mixed mode windows-x86 )
# Problematic frame:
# C [atioglxx.dll+0xc0b57f]
#
In the log file, I see this:
Stack: [0x011d0000,0x01220000], sp=0x0121f1e8, free space=316k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
C [atioglxx.dll+0xc0b57f]
C [atioglxx.dll+0xbe8b95]
C [atioglxx.dll+0x641852]
C [lwjgl.dll+0x9a28]
j org.lwjgl.opengl.GL20.glUniformMatrix4(IZLjava/nio/FloatBuffer;)V+33
j TMLoading.ShaderProgram.loadMatrix(ILorg/lwjgl/util/vector/Matrix4f;)V+23j
TMLoading.StaticShader.loadTransformationMatrix(Lorg/lwjgl/util/vector/Matrix4f;) V+6
j Joehot200.Plane.render()V+407
j Joehot200.TerrainDemo.render()V+4045
j Joehot200.TerrainDemo.enterGameLoop()V+356
j Joehot200.TerrainDemo.startGame()V+320
j StartScreenExperiments.Test2.resartTDemo()V+128
j StartScreenExperiments.Test2.main([Ljava/lang/String;)V+27
v ~StubRoutines::call_stub
V [jvm.dll+0x1473e5]
In other words, the entire issue (the texturing and the crashes) are beginning to look a lot like they are related to the shaders (either parsing information to them, or the actual shader code itself).
I have also done more testing, and the texturing works fine without shaders using a DisplayList.
Here is the code up to the glUniformMatrix() call:
//Some unnecessary code has been removed. For example, you do not care about what colour I make the plane depending on what team it is on.
//Plane class. (Referring to a jet plane which is the main object the player controls)
public void render(){
twodcoords = TextDemo.getScreenCoords(sx, sy + 30, sz);
glPushAttrib(GL_ENABLE_BIT);
glBindBuffer(GL_ARRAY_BUFFER, 0);
glPushMatrix();
glTranslatef(sx, sy, sz);
GL30.glBindVertexArray(Main.TerrainDemo.vaoID);
glEnableClientState(GL_VERTEX_ARRAY);
glEnableClientState(GL_NORMAL_ARRAY);
glRotatef(srot, 0f, 1f, 0f);
glRotatef(pitch, -1f, 0f, 0f);
Main.TerrainDemo.shader.start();
glPushMatrix();
glDisable(GL_LIGHTING);
TexturedModel texturedModel = TerrainDemo.shipModel;
RawModel model = texturedModel.getRawModel();
GL30.glBindVertexArray(model.getVaoID());
GL20.glEnableVertexAttribArray(0);
GL20.glEnableVertexAttribArray(1);
GL20.glEnableVertexAttribArray(2);
GL13.glActiveTexture(GL13.GL_TEXTURE0);
GL11.glBindTexture(GL11.GL_TEXTURE_2D, TerrainDemo.shipModel.getTexture().getID());
org.lwjgl.util.vector.Matrix4f m = Assist.createTransformationMatrix(new Vector3f(sx, sy, sz), new Vector3f(pitch, rot, roll), new Vector3f(5, 5, 5)); //Creates a transformation matrix based on the X, Y, Z, Pitch, yaw, roll, and scale of the plane.
Main.TerrainDemo.shader.loadTransformationMatrix(m);
glDrawElements(GL_TRIANGLES, model.getVertexCount(), GL11.GL_UNSIGNED_INT, 0);
GL20.glDisableVertexAttribArray(0);
GL20.glDisableVertexAttribArray(1);
GL20.glDisableVertexAttribArray(2);
GL30.glBindVertexArray(0);
GL30.glBindVertexArray(0);
glPopMatrix();
glPopAttrib();
}
//StaticShader class. This method literally just passes it to the loadMatrix class.
public void loadTransformationMatrix(Matrix4f matrix){
super.loadMatrix(location_transformationMatrix, matrix);
}
//ShaderProgram class.
FloatBuffer buf = BufferUtils.createFloatBuffer(4 * 4);
public void loadMatrix(int location, Matrix4f matrix){
matrix.store(buf);
buf.flip();
GL20.glUniformMatrix4(location, false, buf);
}
Update - So, with one hour left on the Bounty, I thought I'd add a few more details:
As I probably said somewhere above, the game works for some when exported, but not for others. I've noticed that the game has always worked when ran on java 7, but with me only me and one other tester on java 7, this really isn't conclusive.
The texturing renders correctly in a DisplayList. The textures are loading. However, they are not being displayed correctly.
Even if you don't know the problem, trying out the game (ignore the start screen, I need to make a new one) and telling me the results, as well as your OS/Java details, etc, would be really appreciated.
Yes, the mipmapping is correct. I know someone in the comments mentioned it possibly wasn't, but I've tried setting it stupidly high and I do indeed get a very blurred texture.
I've already tried "package libraries into external jar". I appreciate the time taken for that answer, but I did say in the comments that I've already tried it.
The issue may be the fragment shader (as someone suggested in the comments), but I am currently unsure how to test it, nor do I understand why it would work inside the IDE but not outside of it.

I can't test this, so I'm not sure if this will help, but some implementations of OpenGL don't save element buffers in VAOs. So try binding the element buffer before the call to glDrawElements, e.g.
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, someBuffer);
glDrawElements(GL_TRIANGLES, model.getVertexCount(), GL11.GL_UNSIGNED_INT, 0);
If you don't do this, you may be using the last bound buffer for all draw calls. Probably not the answer, but it couldn't hurt to test it. I also noticed that you are using an older version of Slick, as you still have the slick-util, which is not part of the latest release, so you could also try updating that.

So try this instead of Extracting the required libraries try Package option instead, Because i personally believe there is an issue while finding those library files or linking some file then simply run the package.

Trying to set a texture from Java plugin in Unity 4.6

What i'm trying to do:
Pull an image from sd-card on phone using Java Plugin.
Unity passes a texture ID to plugin.
Plugin uses opengl to assign the image to the texture in Unity through the ID.
Will (eventually) be used to play a video clip from the phone in Unity, for now, it's just trying to change a texture outside of unity.
My issue:
When i call the method in the plugin, passing texture.GetNativeTextureID() into it, the texture does not change. I'm currently only using a simple black 50x50 texture for testing, and the original texture is a flat white.
I'm worried that i've missed something significant, as this is my first time working with Gl calls in java. Much of the answers to similar problems involve using native C++ instead of Java, but I can't find a concrete answer saying that C++ must be used. I'd like to do my best to avoid writing another set of plugins and plugin handlers for C++, but if it's the most efficient/only way to get this working, i'll do it as i'm not unfamiliar with OpenGL and C++
Code:
The plugin method is called from OnPreRender() in a script attached to the main camera:
if (grabTex) {
int texPtr = m_VideoTex.GetNativeTextureID();
Debug.Log( "texPtr = " + texPtr );
m_JVInterface.SetTex( texPtr );
}
m_VideoTex is a basic Texture2D( 50, 50 ) with all pixels set to white, attached to the diffuse shader on the quad in the scene.
The Java plugin code is as follows:
public void SetTexture(Context cont, int _texPointer) {
if (_texPointer != 0) {
final BitmapFactory.Options options = new BitmapFactory.Options();
options.inScaled = false;
options.inJustDecodeBounds = false;
final Bitmap bitmap = BitmapFactory.decodeFile("/storage/emulated/0/Pictures/black.jpg", options);
GLES20.glBindTexture(GLES20.GL_TEXTURE_2D, _texPointer);
GLUtils.texImage2D(GLES20.GL_TEXTURE_2D, 0, bitmap, 0);
Log.i("VideoHandler", "Recieved ID: " + _texPointer);
bitmap.recycle();
}
}

This is most likely a problem with the OpenGL Context. The easiest way would be to send the texture as raw bytes to Unity and then upload as texture inside Unity.

How do I get rid of the mouse cursor in full-screen exclusive mode?

I'm working on a simple 2D game engine in Java, and having no trouble with FSEM, buffer strategies, and so on; my issue is with the mouse cursor. In windowed mode, I can hide the mouse cursor, no problem, by using setCursor() from my JFrame to set a wholly-transparent cursor. However, after a call to device.setFullScreenWindow(this) to go into FSEM, the mouse cursor comes back, and subsequent calls to setCursor() to set it back to my blank cursor have no effect. Calling device.setFullScreenWindow(null) allows me to get rid of the cursor again - it's only while I'm in FSEM that I can't get rid of it.
I'm working under JDK 6, target platform is JDK 5+.
UPDATE: I've done some more testing, and it looks like this issue occurs under MacOS X 10.5 w/Java 6u7, but not under Windows XP SP3 with Java 6u7. So, it could possibly be a bug in the Mac version of the JVM.

Try Creating a custom invisible cursor:
Toolkit toolkit = Toolkit.getDefaultToolkit();
Point hotSpot = new Point(0,0);
BufferedImage cursorImage = new BufferedImage(1, 1, BufferedImage.TRANSLUCENT);
Cursor invisibleCursor = toolkit.createCustomCursor(cursorImage, hotSpot, "InvisibleCursor");
setCursor(invisibleCursor);

One developer found a way around it by creating a one pixel cursor out of a transparent GIF.
http://sevensoft.livejournal.com/23460.html
I know you tried that, but his is specifically addressing the issue of full-screen mode, exactly as you say, so perhaps there's something he's done that you haven't.

I think I've finally found the solution:
System.setProperty("apple.awt.fullscreenhidecursor","true");
This is an Apple-proprietary system property that hides the mouse cursor when an application is in full-screen mode. It's the only way I've found to fix it.

Here's what has been working for me:
Toolkit toolkit = Toolkit.getDefaultToolkit();
// get the smallest valid cursor size
Dimension dim = toolkit.getBestCursorSize(1, 1);
// create a new image of that size with an alpha channel
BufferedImage cursorImg = new BufferedImage(dim.width, dim.height, BufferedImage.TYPE_INT_ARGB);
// get a Graphics2D object to draw to the image
Graphics2D g2d = cursorImg.createGraphics();
// set the background 'color' with 0 alpha and clear the image
g2d.setBackground(new Color(0.0f, 0.0f, 0.0f, 0.0f));
g2d.clearRect(0, 0, dim.width, dim.height);
// dispose the Graphics2D object
g2d.dispose();
// now create your cursor using that transparent image
hiddenCursor = toolkit.createCustomCursor(cursorImg, new Point(0,0), "hiddenCursor");
Granted, I haven't tested it on Mac (yet), only Windows. But when I used the common methods I was getting the cursor as black box, so I use the code above the create a transparent box and set it as the cursor instead. Of course you have to use the setCursor method on an AWT object (such as your app's Frame) to set this hiddenCursor. Here is my hideMouse method ('fr' is my Frame):
public void hideMouse(boolean hide) {
if(hide) {
fr.setCursor(hiddenCursor);
} else {
fr.setCursor(Cursor.getDefaultCursor());
}
}

I don't know if this knowledge applies but in a old VB6 app I had the same problem and I got rid of it moving the cursor out of the screen giving it some very large values.
Hope it helps.

If you're running only on Windows, it looks like you'll need to call ShowCursor(FALSE) through JNI. At least, to make the cursor hide complete.
Here's some code which creates the 1x1 cursor. It works for me, though I still get a 1x1 cursor.
Toolkit toolkit = Toolkit.getDefaultToolkit();
Dimension dim = toolkit.getBestCursorSize(1,1);
transCursor = toolkit.createCustomCursor(gc.createCompatibleImage(dim.width, dim.height),
new Point(0, 0), "transCursor");
((Component)mainFrame).setCursor(transCursor);

Specifically for your Mac problem, through JNI you could use the following:
Quartz Display Services Reference - CGDisplayHideCursor