A
Rectangle specifies an area in a coordinate space that is enclosed by the
Rectangle object's upper-left point {@code (x,y)}in the coordinate space, its width, and its height.
A Rectangle object's width and height are public fields. The constructors that create a Rectangle, and the methods that can modify one, do not prevent setting a negative value for width or height.
A {@code Rectangle} whose width or height is exactly zero has locationalong those axes with zero dimension, but is otherwise considered empty. The {@link #isEmpty} method will return true for such a {@code Rectangle}. Methods which test if an empty {@code Rectangle} contains or intersectsa point or rectangle will always return false if either dimension is zero. Methods which combine such a {@code Rectangle} with a point or rectanglewill include the location of the {@code Rectangle} on that axis in theresult as if the {@link #add(Point)} method were being called.
A {@code Rectangle} whose width or height is negative has neitherlocation nor dimension along those axes with negative dimensions. Such a {@code Rectangle} is treated as non-existant along those axes.Such a {@code Rectangle} is also empty with respect to containmentcalculations and methods which test if it contains or intersects a point or rectangle will always return false. Methods which combine such a {@code Rectangle} with a point or rectanglewill ignore the {@code Rectangle} entirely in generating the result.If two {@code Rectangle} objects are combined and each has a negativedimension, the result will have at least one negative dimension.
Methods which affect only the location of a {@code Rectangle} willoperate on its location regardless of whether or not it has a negative or zero dimension along either axis.
Note that a {@code Rectangle} constructed with the default no-argumentconstructor will have dimensions of {@code 0x0} and therefore be empty.That {@code Rectangle} will still have a location of {@code (0,0)} andwill contribute that location to the union and add operations. Code attempting to accumulate the bounds of a set of points should therefore initially construct the {@code Rectangle} with a specificallynegative width and height or it should use the first point in the set to construct the {@code Rectangle}. For example:
Rectangle bounds = new Rectangle(0, 0, -1, -1); for (int i = 0; i < points.length; i++) { bounds.add(points[i]); } or if we know that the points array contains at least one point:
Rectangle bounds = new Rectangle(points[0]); for (int i = 1; i < points.length; i++) { bounds.add(points[i]); } This class uses 32-bit integers to store its location and dimensions. Frequently operations may produce a result that exceeds the range of a 32-bit integer. The methods will calculate their results in a way that avoids any 32-bit overflow for intermediate results and then choose the best representation to store the final results back into the 32-bit fields which hold the location and dimensions. The location of the result will be stored into the {@link #x} and{@link #y} fields by clipping the true result to the nearest 32-bit value.The values stored into the {@link #width} and {@link #height} dimensionfields will be chosen as the 32-bit values that encompass the largest part of the true result as possible. Generally this means that the dimension will be clipped independently to the range of 32-bit integers except that if the location had to be moved to store it into its pair of 32-bit fields then the dimensions will be adjusted relative to the "best representation" of the location. If the true result had a negative dimension and was therefore non-existant along one or both axes, the stored dimensions will be negative numbers in those axes. If the true result had a location that could be represented within the range of 32-bit integers, but zero dimension along one or both axes, then the stored dimensions will be zero in those axes.
@author Sami Shaio
@since 1.0