问题
so quicky, I am doing program which demonstrate methods used for computer graph drawing. I need to create timeline or history of actions like ( placeVertex(x,y), moveVertex(newX,newY) etc. ) and iterate through (forward and backwards, automatically or manual)
I already achieved that by using command design pattern but few of these commands are using transitions. First idea was to use Condition interface's lock, await and signal in setOnFinished between each commands but it led to gui freezing.
I tryed SequentialTransition but it's no use for my problem - can't change properties dynamically between transitions.
Is there a possibility to somehow inform generation that one transition ended and next can run without GUI freezing and drawing?
Thanks!
edit: I ll try to simplify it all Here is my Command interface and one of these commands:
public interface Command {
public void execute();
}
public class MoveVertex implements Command {
public MoveVertex(Data d, Vertex v, double changedX, double changedY){..};
@Override
public void execute() {
Path path = new Path();
path.getElements().add(new MoveTo(v.getCenterX(), v.getCenterY()));
path.getElements().add(new LineTo(changedX, changedY));
PathTransition pathTransition = new PathTransition();
pathTransition.setDuration(Duration.millis(velocity));
pathTransition.setPath(path);
pathTransition.setNode(v.getVertex());
pathTransition.play(); }
}
These Commands are stored in my history class which is basically
private List<Command> history;
And I do going through the list and executing Commands
public boolean executeNext() {
if (history.size() != position) {
history.get(position).execute();
position++;
return true;
}
return false;
}
And I am trying to achieve state when next Command is started only if previous finished. Tryed to put await/signal in between without success.
回答1:
The solution below uses Itachi's suggestion of providing an onFinished handler to move to a node to a new (random) location after we get to the next location.
It could probably be made more efficient (and simpler to understand) by re-using a single Transition rather than using recursion within the event handler. It is probably unnecessary to create a new Transition for each movement - but, as long as there aren't hundreds of thousands of movement iterations, it should be acceptable as is.
import javafx.animation.*;
import javafx.application.Application;
import javafx.event.*;
import javafx.geometry.Point2D;
import javafx.scene.*;
import javafx.scene.layout.Pane;
import javafx.scene.paint.Color;
import javafx.scene.shape.*;
import javafx.stage.Stage;
import javafx.util.Duration;
import java.util.Random;
// animates moving a node forever in a random walk pattern.
public class RandomWalk extends Application {
private static final Random random = new Random(42);
private static final double W = 200;
private static final double H = 200;
private static final double R = 10;
private static final Node node = new Circle(
R, Color.FORESTGREEN
);
@Override
public void start(Stage stage) {
// start from the center of the screen.
node.relocate(W / 2 - R, H / 2 - R);
stage.setScene(new Scene(new Pane(node), W, H));
stage.show();
walk();
}
// start walking from the current position to random points in sequence.
private void walk() {
final Point2D to = getRandomPoint();
final Transition transition = createMovementTransition(
node,
to
);
transition.setOnFinished(
walkFrom(to)
);
transition.play();
}
private EventHandler<ActionEvent> walkFrom(final Point2D from) {
return event -> {
// Finished handler might be called a frame before transition complete,
// leading to glitches if we relocate in the handler.
// The transition works by manipulating translation values,
// so zero the translation out before relocating the node.
node.setTranslateX(0);
node.setTranslateY(0);
// After the transition is complete, move the node to the new location.
// Relocation co-ordinates are adjusted by the circle's radius.
// For a standard node, the R adjustment would be unnecessary
// as most nodes are located at the top left corner of the node
// rather than at the center like a circle is.
node.relocate(
from.getX() - R,
from.getY() - R
);
// Generate the next random point and play a transition to walk to it.
// I'd rather not use recursion here as if you recurse long enough,
// then you will end up with a stack overflow, but I'm not quite sure
// how to do this without recursion.
final Point2D next = getRandomPoint();
final Transition transition = createMovementTransition(node, next);
transition.setOnFinished(walkFrom(next));
transition.play();
};
}
// We use a PathTransition to move from the current position to the next.
// For the simple straight-line movement we are doing,
// a straight TranslateTransition would have been fine.
// A PathTransition is just used to demonstrate that this
// can work for the generic path case, not just straight line movement.
private Transition createMovementTransition(Node node, Point2D to) {
Path path = new Path(
new MoveTo(
0,
0
),
new LineTo(
to.getX() - node.getLayoutX(),
to.getY() - node.getLayoutY()
)
);
return new PathTransition(
Duration.seconds(2),
path,
node
);
}
// @return a random location within a bounding rectangle (0, 0, W, H)
// with a margin of R kept between the point and the bounding rectangle edge.
private Point2D getRandomPoint() {
return new Point2D(
random.nextInt((int) (W - 2*R)) + R,
random.nextInt((int) (H - 2*R)) + R
);
}
public static void main(String[] args) {
launch(args);
}
}
来源：https://stackoverflow.com/questions/28093416/javafx-transition-animation-waiting