Related
I want to make a loading animation in After Effects using scripts
I get no arc only white screen with shape layer there
and this is what gets rendered in the app
This is my code that doesn't work
app.project.close(CloseOptions.DO_NOT_SAVE_CHANGES);
app.newProject();
app.beginUndoGroup("Create Comp");
// Create a new composition with a solid layer
var comp = app.project.items.addComp("My Composition", 1920, 1080, 1, 10, 24);
comp.openInViewer();
var solid = comp.layers.addSolid([1, 1, 1], "My Solid", 1920, 1080, 1, 10);
// Create a new shape layer
var shapeLayer = comp.layers.addShape();
// shapeLayer.moveToStart();
shapeLayer.enabled= true;
// Set position of the shape layer
var shapePosition = shapeLayer.property("ADBE Transform Group").property("ADBE Position");
shapePosition.setValue([960,540]);
var path = shapeLayer.property("ADBE Root Vectors Group").addProperty("ADBE Vector Shape - Group");
// make a arc
path.property("ADBE Vector Shape").setValue(new Shape());
path.property("ADBE Vector Shape").value.vertices = [[0, 0], [100, 100], [200, 0]];
path.property("ADBE Vector Shape").value.inTangents = [[-50, -50], [0, 0]];
path.property("ADBE Vector Shape").value.outTangents = [[0, 0], [50, 50]];
path.property("ADBE Vector Shape").value.closed = true;
path.enabled = true;
var stroke = shapeLayer.property("ADBE Root Vectors Group").addProperty("ADBE Vector Graphic - Stroke");
stroke.property("ADBE Vector Stroke Width").setValue(5);
stroke.property("ADBE Vector Stroke Color").setValue([1, 0, 0]);
stroke.enabled = true;
var fill = shapeLayer.property("ADBE Root Vectors Group").addProperty("ADBE Vector Graphic - Fill");
fill.property("ADBE Vector Fill Color").setValue([1, 0, 0]);
fill.enabled = true;
// Create an animation for the rotation property
var rotation = shapeLayer.property("ADBE Transform Group").property("ADBE Rotate Z");
rotation.setValueAtTime(0, 0);
rotation.setValueAtTime(5, 180);
// Set the composition duration to 10 seconds
comp.duration = 10;
app.endUndoGroup();
Properties in AE can't be set by assigning values to them. You need to use the setValue() method. Additionally in this instance you want to set the values of the shape object before assigning it to the path property.
var theShape = new Shape();
theShape.vertices = [[0, 0], [100, 100], [200, 0]];
theShape.inTangents = [[-50, -50], [0, 0]];
theShape.outTangents = [[0, 0], [50, 50]];
theShape.closed = true;
path.property("ADBE Vector Shape").setValue(theShape);
Also, as an AE developer:
In my main application I have some SVGPaths that I add to an XYChart. Sometimes they have an ImagePattern fill which now needs to have a LinearGradient fill. The ImagePattern fill is a crosshatch and this needs to be colored with the LinearGradient the same as if it was a solid Rectangle with a LinearGradient applied. The SVGPath also has a dotted outline and the LinearGradient should fill the dotted outline and the ImagePattern fill as it they were part of the same shape.
I've written some sample code to show where I'm at. This colors the crosshatch as it's created and looks ok but isn't the effect I describe above as each cross in the ImagePattern has the LinearGradient applied individually. Ideally the LinearGradient would just be applied to the final SVGPath once the ImagePattern fill has been applied.
I've also tried some effects using Blend and ColorInput but haven't managed to get any closer to the solution.
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import javafx.application.Application;
import javafx.scene.Scene;
import javafx.scene.SnapshotParameters;
import javafx.scene.image.Image;
import javafx.scene.layout.BorderPane;
import javafx.scene.layout.Pane;
import javafx.scene.paint.Color;
import javafx.scene.paint.CycleMethod;
import javafx.scene.paint.ImagePattern;
import javafx.scene.paint.LinearGradient;
import javafx.scene.paint.Paint;
import javafx.scene.paint.Stop;
import javafx.scene.shape.Line;
import javafx.scene.shape.SVGPath;
import javafx.stage.Stage;
public class Main extends Application {
#Override
public void start(Stage primaryStage) {
try {
List<Color> colors = Arrays.asList(Color.RED, Color.YELLOW, Color.GREEN);
ArrayList<Stop> stops = new ArrayList<>(colors.size() * 2);
for (int i = 0; i < colors.size(); i++) {
stops.add(new Stop(getOffset(i, colors.size()), colors.get(i)));
stops.add(new Stop(getOffset(i + 1, colors.size()), colors.get(i)));
}
LinearGradient lg = new LinearGradient(0, 0, 20, 20, false, CycleMethod.REPEAT, stops);
SVGPath svgPath = new SVGPath();
svgPath.setContent("M-84.1487,-15.8513 a22.4171,22.4171 0 1 0 0,31.7026 h168.2974 a22.4171,22.4171 0 1 0 0,-31.7026 Z");
Image hatch = createCrossHatch(lg);
ImagePattern pattern = new ImagePattern(hatch, 0, 0, 10, 10, false);
svgPath.setFill(pattern);
svgPath.setStroke(lg);
BorderPane root = new BorderPane();
root.setCenter(svgPath);
Scene scene = new Scene(root, 400, 400);
primaryStage.setScene(scene);
primaryStage.show();
}
catch (Exception e) {
e.printStackTrace();
}
}
protected static Image createCrossHatch(Paint paint) {
Pane pane = new Pane();
pane.setPrefSize(20, 20);
pane.setStyle("-fx-background-color: transparent;");
Line fw = new Line(-5, -5, 25, 25);
Line bw = new Line(-5, 25, 25, -5);
fw.setStroke(paint);
bw.setStroke(paint);
fw.setStrokeWidth(3);
bw.setStrokeWidth(3);
pane.getChildren().addAll(fw, bw);
new Scene(pane);
SnapshotParameters sp = new SnapshotParameters();
sp.setFill(Color.TRANSPARENT);
return pane.snapshot(sp, null);
}
private double getOffset(double i, int count) {
return (((double) 1) / (double) count * (double) i);
}
public static void main(String[] args) {
launch(args);
}
}
If you run the supplied code you will see it draws a dog bone. The lineargradient colors of the dashed outline should continue through the cross hatch ImagePattern fill. I'm aware of why the hatched ImagePattern is colored like it is but this is the best compromise I have at present. As mentioned I'd like to be able to applied the LinearGradient fill to the whole shape once the ImagePattern fill has been applied so the LinearGradient affects the whole shape the same.
Thanks
There is no direct way to apply and combine two paints over one single node. We can overlay many different paints (like solid, linear gradients or even image patterns) using background color via css, but that won't combine.
So in order to combine two different paints, on one side a linear gradient, on the other a pattern fill, we need to apply them to two nodes, and use a blending effect between both paints.
According to the code posted, this is the SVGPath with the linear gradient:
#Override
public void start(Stage primaryStage) {
Node base = getNodeWithGradient();
BorderPane root = new BorderPane();
Group group = new Group(base);
root.setCenter(group);
Scene scene = new Scene(root, 400, 200);
primaryStage.setScene(scene);
primaryStage.show();
}
private SVGPath getNodeWithGradient() {
List<Color> colors = Arrays.asList(Color.RED, Color.YELLOW, Color.GREEN);
ArrayList<Stop> stops = new ArrayList<>(colors.size() * 2);
for (int i = 0; i < colors.size(); i++) {
stops.add(new Stop(getOffset(i, colors.size()), colors.get(i)));
stops.add(new Stop(getOffset(i + 1, colors.size()), colors.get(i)));
}
LinearGradient lg = new LinearGradient(0, 0, 20, 20, false, CycleMethod.REPEAT, stops);
SVGPath svgPath = getSVGPath();
svgPath.setFill(lg);
svgPath.setStroke(lg);
return svgPath;
}
private SVGPath getSVGPath() {
SVGPath svgPath = new SVGPath();
svgPath.setContent("M-84.1487,-15.8513 a22.4171,22.4171 0 1 0 0,31.7026 h168.2974 a22.4171,22.4171 0 1 0 0,-31.7026 Z");
return svgPath;
}
private double getOffset(double i, int count) {
return (((double) 1) / (double) count * (double) i);
}
While this is the SVGPath with the image pattern fill:
#Override
public void start(Stage primaryStage) {
Node overlay = getNodeWithPattern();
BorderPane root = new BorderPane();
Group group = new Group(overlay);
root.setCenter(group);
Scene scene = new Scene(root, 400, 200);
primaryStage.setScene(scene);
primaryStage.show();
}
private SVGPath getNodeWithPattern() {
Image hatch = createCrossHatch();
ImagePattern pattern = new ImagePattern(hatch, 0, 0, 10, 10, false);
SVGPath svgPath = getSVGPath();
svgPath.setFill(pattern);
return svgPath;
}
private SVGPath getSVGPath() {
SVGPath svgPath = new SVGPath();
svgPath.setContent("M-84.1487,-15.8513 a22.4171,22.4171 0 1 0 0,31.7026 h168.2974 a22.4171,22.4171 0 1 0 0,-31.7026 Z");
return svgPath;
}
private static Image createCrossHatch() {
Pane pane = new Pane();
pane.setPrefSize(20, 20);
Line fw = new Line(-5, -5, 25, 25);
Line bw = new Line(-5, 25, 25, -5);
fw.setStroke(Color.BLACK);
bw.setStroke(Color.BLACK);
fw.setStrokeWidth(3);
bw.setStrokeWidth(3);
pane.getChildren().addAll(fw, bw);
new Scene(pane);
SnapshotParameters sp = new SnapshotParameters();
return pane.snapshot(sp, null);
}
Now the trick is to combine both SVGPath nodes, adding a blending mode to the one on top.
According to JavaDoc for BlendMode.ADD:
The color and alpha components from the top input are added to those from the bottom input.
#Override
public void start(Stage primaryStage) {
Node base = getNodeWithGradient();
Node overlay = getNodeWithPattern();
overlay.setBlendMode(BlendMode.ADD);
BorderPane root = new BorderPane();
Group group = new Group(base, overlay);
root.setCenter(group);
Scene scene = new Scene(root, 400, 200);
primaryStage.setScene(scene);
primaryStage.show();
}
private SVGPath getNodeWithGradient() {
List<Color> colors = Arrays.asList(Color.RED, Color.YELLOW, Color.GREEN);
ArrayList<Stop> stops = new ArrayList<>(colors.size() * 2);
for (int i = 0; i < colors.size(); i++) {
stops.add(new Stop(getOffset(i, colors.size()), colors.get(i)));
stops.add(new Stop(getOffset(i + 1, colors.size()), colors.get(i)));
}
LinearGradient lg = new LinearGradient(0, 0, 20, 20, false, CycleMethod.REPEAT, stops);
SVGPath svgPath = getSVGPath();
svgPath.setFill(lg);
svgPath.setStroke(lg);
return svgPath;
}
private SVGPath getNodeWithPattern() {
Image hatch = createCrossHatch();
ImagePattern pattern = new ImagePattern(hatch, 0, 0, 10, 10, false);
SVGPath svgPath = getSVGPath();
svgPath.setFill(pattern);
return svgPath;
}
private SVGPath getSVGPath() {
SVGPath svgPath = new SVGPath();
svgPath.setContent("M-84.1487,-15.8513 a22.4171,22.4171 0 1 0 0,31.7026 h168.2974 a22.4171,22.4171 0 1 0 0,-31.7026 Z");
return svgPath;
}
private static Image createCrossHatch() {
Pane pane = new Pane();
pane.setPrefSize(20, 20);
Line fw = new Line(-5, -5, 25, 25);
Line bw = new Line(-5, 25, 25, -5);
fw.setStroke(Color.BLACK);
bw.setStroke(Color.BLACK);
fw.setStrokeWidth(3);
bw.setStrokeWidth(3);
pane.getChildren().addAll(fw, bw);
new Scene(pane);
SnapshotParameters sp = new SnapshotParameters();
return pane.snapshot(sp, null);
}
private double getOffset(double i, int count) {
return (((double) 1) / (double) count * (double) i);
}
And we get the desired result:
I can't seem to wrap my head around these shadow map matrices. It works great when light angles are about a 45 degree lookat position to 0,0,0. For example, if the sun angle is high(y) vs x,z. The shadows don't line up with the models. A light position of (-7,-10,7) or (-9,-10,0) works fine. (-1,-10,1) is skewed. Here is my code example. -x is left, -y is up and -z is far.
public Vector3f cameraPosition = new Vector3f(0f, -10f, 7f);
public float[] lightPosition = new float[]{-7f, -10.0f, 7f}
draw objects to shadow map texture
Matrix.setLookAtM(sunMatrix, 0, -lightPosition[0], lightPosition[1], lightPosition[2],
0, 0, 0,
0f, 1f, 0f);
//projection matrix plus sun
float width = (globals.glScreenSize/2);
float height = (globals.glScreenSize/2);
Matrix.orthoM(projectionMatrix, 0, -width, width,
-height, height,
-1f, 100f);
tempMatrix = new float[16];
Matrix.multiplyMM(tempMatrix, 0, projectionMatrix, 0, sunMatrix, 0);//add camera matrix to perspective
projectionMatrix = tempMatrix.clone();
//translate
Matrix.setIdentityM(modelMatrix, 0);//set to 0
//translate
Matrix.translateM(modelMatrix, 0, modelPos.location.x,
-(modelPos.location.y),
modelPos.location.z);//move
//rotate
Matrix.rotateM(modelMatrix, 0, modelPos.angles.x, 1f, 0f, 0f);
Matrix.rotateM(modelMatrix, 0, -modelPos.angles.y, 0f, 1f, 0f);
Matrix.rotateM(modelMatrix, 0, -modelPos.angles.z, 0f, 0f, 1f);
//scale
Matrix.scaleM(modelMatrix, 0, modelPos.scales.x,
modelPos.scales.y,
modelPos.scales.z);//scale
Matrix.multiplyMM(viewProjMatrix, 0, projectionMatrix, 0, modelMatrix, 0);
//Matrix.multiplyMM(projectionMatrix, 0, globals.viewProjMatrix, 0, modelMatrix, 0);//perspective/model/view projection matrix
finalMatrix = viewProjMatrix.clone();//final matrix created
rotate around y axis and draw shadow map to screen
Matrix.setLookAtM(modelMatrix, 0, 0f, 0f, 0f,
globals.lightPosition[0], 0f, -globals.lightPosition[2],
0f, 1f, 0f);
//scale
Matrix.scaleM(modelMatrix, 0, scale.x, scale.y, scale.y);//scale
Matrix.multiplyMM(projectionMatrix, 0, globals.viewProjMatrix, 0, modelMatrix, 0);//projection matrix
finalMatrix = projectionMatrix.clone();//final matrix created
enter image description here
Thanks for any help,
Norm
I changed,
float height = (globals.glScreenSize/2) * Math.abs(lightPos.y);
Shadows all line up with the models now. If you are looking for a simple shadow map and obj loader example, check out https://github.com/Normfly/myOpenglES20
Happy coding,
Norm
the inner shadow disappears during the transition it looks like the inner shadow is also scaled.
public void showTowerRange(int x0, int y0, double range) {
Circle circle = new Circle(
x0 * MAP_CELLS_WIDTH + MAP_CELLS_WIDTH / 2,
y0 * MAP_CELLS_HEIGHT + MAP_CELLS_HEIGHT / 2,
1,
Color.RED
);
circle.setOpacity(0.5);
gameRoot.getChildren().add(circle);
ScaleTransition scl = new ScaleTransition(Duration.millis(SHOW_RANGE_EFFECT_DURATION),circle);
scl.setByX(range * MAP_CELLS_WIDTH);
scl.setByY(range * MAP_CELLS_HEIGHT);
FadeTransition fd = new FadeTransition(Duration.millis(SHOW_RANGE_EFFECT_DURATION),circle);
fd.setByValue(-.3);
circle.setEffect(new InnerShadow(BlurType.GAUSSIAN, Color.GREEN, 4, 1, 0, 0));
circle.setEffect(new DropShadow(BlurType.GAUSSIAN, Color.WHITE, 2, 0, 0, 0));
ParallelTransition prl = new ParallelTransition(scl,fd);
prl.play();
}
Well, I have good news and bad news for you. The good one - scale transition does not ruin inner shadow. The bad one - it is you ruining inner shadow:)
The point is, setEffect() method does not add another effect, it just set effect property of the Node, replacing previous value. What you need to do is to chain you effects (see example here). So instead of this:
circle.setEffect(new InnerShadow(BlurType.GAUSSIAN, Color.GREEN, 4, 1, 0, 0));
circle.setEffect(new DropShadow(BlurType.GAUSSIAN, Color.WHITE, 2, 0, 0, 0));
You must do this:
InnerShadow is = new InnerShadow(BlurType.GAUSSIAN, Color.GREEN, 4, 1, 0, 0);
DropShadow ds = new DropShadow(BlurType.GAUSSIAN, Color.WHITE, 2, 0, 0, 0);
ds.setInput(is);
circle.setEffect(ds);
I have the following test code, where I try to clip a MeshView with a circle.
I also tried putting the meshView into a group then clipping that, but this result in a black circle.
Is there a way to clip a MeshView, preferably without putting it into a group?
import scalafx.application.JFXApp
import scalafx.application.JFXApp.PrimaryStage
import scalafx.scene.image.Image
import scalafx.scene.paint.{Color, PhongMaterial}
import scalafx.scene.shape.{TriangleMesh, Circle, MeshView}
import scalafx.scene.{Group, PerspectiveCamera, Scene, SceneAntialiasing}
object Test4 extends JFXApp {
stage = new PrimaryStage {
scene = new Scene(500, 500, true, SceneAntialiasing.Balanced) {
fill = Color.LightGray
val clipCircle = Circle(150.0)
val meshView = new MeshView(new RectangleMesh(500,500)) {
// takes a while to load
material = new PhongMaterial(Color.White, new Image("https://peach.blender.org/wp-content/uploads/bbb-splash.png"), null, null, null)
}
// val meshGroup = new Group(meshView)
meshView.setClip(clipCircle)
root = new Group {children = meshView; translateX = 250.0; translateY = 250.0; translateZ = 560.0}
camera = new PerspectiveCamera(false)
}
}
}
class RectangleMesh(Width: Float, Height: Float) extends TriangleMesh {
points = Array(
-Width / 2, Height / 2, 0,
-Width / 2, -Height / 2, 0,
Width / 2, Height / 2, 0,
Width / 2, -Height / 2, 0
)
texCoords = Array(
1, 1,
1, 0,
0, 1,
0, 0
)
faces = Array(
2, 2, 1, 1, 0, 0,
2, 2, 3, 3, 1, 1
)
The clippling actually works fine over the MeshView wrapped around a Group.
If you check JavaDoc for setClip():
There is a known limitation of mixing Clip with a 3D Transform. Clipping is essentially a 2D image operation. The result of a Clip set on a Group node with 3D transformed children will cause its children to be rendered in order without Z-buffering applied between those children.
As a result of this:
Group meshGroup = new Group(meshView);
meshGroup.setClip(clipCircle);
you will have a 2D image, and it seems Material is not applied. However you can check there's a mesh, by seting this:
meshView.setDrawMode(DrawMode.LINE);
So in your case, adjusting dimensions:
#Override
public void start(Stage primaryStage) {
Circle clipCircle = new Circle(220.0);
MeshView meshView = new MeshView(new RectangleMesh(400,400));
meshView.setDrawMode(DrawMode.LINE);
Group meshGroup = new Group(meshView);
meshGroup.setClip(clipCircle);
PerspectiveCamera camera = new PerspectiveCamera(false);
StackPane root = new StackPane();
final Circle circle = new Circle(220.0);
circle.setFill(Color.TRANSPARENT);
circle.setStroke(Color.RED);
root.getChildren().addAll(meshGroup, circle);
Scene scene = new Scene(root, 500, 500, true, SceneAntialiasing.BALANCED);
scene.setCamera(camera);
primaryStage.setTitle("Hello World!");
primaryStage.setScene(scene);
primaryStage.show();
}
will give this:
In the end, clipping doesn't make sense with 3D shapes. For that you can use just 2D shape to get the result you want.
If you want 3D clipping have a look at CSG operations. Check this question for a JavaFX based solution.