Point and Vector operations

Notes on compact table notation

  • Many arguments which might be strongly typed as Point3d, Vector3d, Point3d, Vector3d are weakly typed as XYandZ or XandY.
    • These allow any object that has x and y properties to be passed as inputs.
  • Many methods have optional result args.
    • The optional arg is NOT indicated here.
    • If the caller supplies the optional arg, that preexisting object will be reinitialized.
    • If a method is being called many times in a loop, reusing a result can give a significant performance benefit.

Typical names in the tables are:

name implied type remarks
x number x coordinate
y number y coordinate
z number z coordinate
p Point3d or Point2d point object
v Vector3d or Vector2d vector object
newVector Vector3d or Vector2d newly created vector object
basePoint Point3d or Point2d point object used as origin of the calculation
targetA, targetB Point3d, Vector3d, Point2d, or Vector3d point or vector used as target of a vector during the calculation

public members

category Point3d Vector3d Point2d Vector2d
components x: number; x: number; x: number; x: number;
y: number; y: number; y: number; y: number;
z: number; z: number;

static "create" methods

category Point3d Vector3d Point2d Vector2d
create by coordinates p = Point3d.create (x,y,z) v = Vector3d.create (x,y,z) p = Point2d.create (x,y) v = Vector2d.create (x,y)
vectors in principal axis directions, default length 1 v = Vector3d.unitX (length:number = 1) v = Vector3d.unitX (length:number = 1)
v = Vector3d.unitY (length:number = 1) v = Vector3d.unitY (length:number = 1)
v = Vector3d.unitZ (length:number = 1)
create all zeros p = Point3d.createZero () v = Vector3d.createZero () p = Point2d.createZero () v = Vector2d.createZero ()
create from variant sources Point3d.createFrom (other: Float64Array | XAndY | XAndYAndZ) Vector3d.createFrom (other: Float64Array | XAndY | XAndYAndZ) Point2d.createFrom (other: XAndY) Vector2d.createFrom (other: XAndY | Float64Array)
create from index in packed xyzxyz.. p = Point3d.createFromPacked (Float64Array, pointIndex)
unweight from indexed in packed xyzwxyzw.. p = Point3d.createFromPackedXYZW (Float64Array, pointIndex)
create scaled copy p = Point3d.createScale (pointA: XYAndZ, scalefactor)
create sum of 2 weighted p = Point3d.createAdd2Scaled (pointA: XYZAndZ, scaleA, pointB: XYAndZ: pointB, scaleB) v = Vector3d.createAdd2Scaled (pointA: XYZAndZ, scaleA, pointB: XYAndZ: pointB, scaleB)
create sum of 2 weighted v = Vector3d.createAdd2ScaledXYZ (ax, ay, az, scaleA, bx, by, bz, scaleB)
create sum of 3 weighted p = Point3d.createAdd3Scaled (pointA: XYZAndZ, scaleA, pointB: XYAndZ: pointB, scaleB, XYAndZ: pointC, scaleC) v = Vector3d.createAdd3Scaled (pointA: XYZAndZ, scaleA, pointB: XYAndZ: pointB, scaleB, XYAndZ: pointC, scaleC)
Create from polar radius, angle, and z p = Point3d.createPolar (radius, angle, z) p = Point2d.createPolar (radius, angle)
Create from spherical radius, xy angle, elevation angle p = Pointd.createSpherical (radius, xyAngle, elevationAngle)

"create" via instance methods on existing objects

category Point3d Vector3d Point2d Vector2d
vector from start to end newVector = point.vectorTo (otherXYAndZ) newVector = vector.vectorTo(otherXYandZ) newVector = point.vectorTo (otherXAndY) newVector = vector.vectorTo (otherXAndY)
vector = Vector3d.createStartEnd (pointA, pointB) vector = Vector2d.createStartEnd (pointA, pointB)
vector = Vector3d.createStartEndXYZXYZ (ax, ay, az, bx, by, bz)
vector from start to end newVector = vector.vectorTo(otherXYandZ) newVector = point.vectorTo (otherXYAndZ) newVector = vector.vectorTo (otherXAndY) newVector = point.vectorTo (otherXAndY)
clone as same type newPoint = p.clone () newVector = v.clone () newPoint = p.clone () newVector = v.clone ()
unit vector from start to end newVector = point.unitVectorTo (otherXYAndZ) newVector = point.unitVectorTo (otherXYAndZ) newVector = point.unitVectorTo (otherXYAndZ) newVector = point.unitVectorTo (otherXAndY)
vector of specified length from start to end newVector = point.scaledVectorTo (otherXYAndZ, length) newVector = vector.scaledVectorTo (otherXYAndZ, length)
clone as strongly type Point3d p.cloneAsPoint3d () v.cloneAsPoint3d ()
vector rotated around axis vector newVector = Vector3d.createRotateVectorAroundVector (oldVector, axisVector, angle)
divide x,y,z of existing vector by scalar newVector = oldVector.saveDivideOrNull (denominator) newVector = oldVector.safeDivideOrNull (denominator)
return normalized vector, packaged in an object with length property newVector = oldVector.normalizeWithLength (): {v: vector3d, mag: number}
return scaled copy of instance newVector = oldVector.scale (scaleFactor) newVector = oldVector.scale (scaleFactor)
return copy scaled to specific length newVector = oldVector.scaleToLength (newLength) newVector = oldVector.scaleToLength (newLength)
return normalized vector newVector = oldVector.normalize () : Vector3d | undefined newVector = oldVector.normalize () : Vector2d | undefined
attempt to normalize in place. If near zero length, leave unchanged and return false vector.normalizeInPlace () : boolean
clone negated newVector = oldVector.negate () newVector = oldVector.negate ()
vector rotated 90 degrees COUNERCLOCKWISE in XY plane, preserving z newVector = oldVector.rotate99CCWXY () newVector = oldVector.rotate99CCWXY ()
vector rotated 90 degrees CLOCKWISE in XY plane, preserving z newVector = oldVector.rotate99CWXY ()
vector rotated by angle in XY plane, preserving z newVector = oldVector.rotateXY () newVector = oldVector.rotateXY ()
unit vector rotated 90 degrees in xy plane newVector = oldVector.unitPerpendicularXY () newVector = oldVector.unitPerpendicularXY ()
vector rotated 90 degrees towards a target vector. Rotation is in the plane containing both inputs. newVector = oldVector.rotate90Towards ()
vector rotated 90 degrees around an axis vector. newVector = oldVector.rotate90Around ()
vector to the intersection of offsets vector = Vector2d.createOffsetBisector (unitPerpendicularA, unitPerpendicularB, offsetDistance) :Vector2d | undefined

create by Interpolation and addition

category Point3d Vector3d Point2d Vector2d
interpolate a point between instance and target newPoint = p.interpolate (fraction, target) newVector = v.interpolate (fraction, target) newPoint = basePoint.interpolate (fraction, target) newVector = v.interpolate (fraction, target)
ray with origin interpolated between points, vector from point to point newPoint = basePoint.interpolatePointAndTangent (fraction, target)
interpolate a point with distinct fractions for each x,y,z direction newPoint = basePoint.interpolateXYZ (fractionX, fractionY, fractionZ, target) newPoint = basePoint.interpolateXY (fractionX, fractionY, target)
fractional interpolate, then move perpendicular by a fraction of the XY rotation of the same vector newPoint = p.interpolatePerpendicularXY (fraction, target, perpendicularFraction)
from instance point, move a forwardFraction of vector, and perpendicularFraction of vector rotated 90 CCW newPoint = oldPoint.addForwardAndLeft (forwardFraction, leftFraction, vector)
from instance point, move a forwardFraction of towards target, and perpendicularFraction of vector rotated 90 CCW newPoint = oldPoint.addForwardAndLeft (forwardFraction, leftFraction, targetPoint)
new point/vector by adding a vector newPoint = basePoint.plus (vector: XYAndZ) newVector = v.plus (vector: XYAndZ) newPoint = basePoint.plus (vector: XAndY) newVector = v.plus (vector: XAndY)
new point by adding x,y,z args newPoint = basePoint.plusXYZ (x,y,z) newPoint = basePoint.plusXY (x,y)
new point/vector by subtracting a vector newPoint = basePoint.minus (vector: XYAndZ) newVector = v.minus (vector: XYAndZ) newPoint = basePoint.minus (vector: XAndY) newVector = v.minus (vector: XAndY)
new point/vector by adding scaled vector(s) newPoint = basePoint.plusScaled (vector: XYAndZ, scale) newVector = v.plusScaled (vector: XYAndZ, scale) newPoint = basePoint.plusScaled (vector: XAndY, scale) newVector = baseVector.plusScaled (vector: XAndY, scale)
newPoint = basePoint.plus2Scaled (vector: XYAndZ, scale1, vector2: XYAndZ, scale2) newVector = baseVector.plus2Scaled (vector: XYAndZ, scale1, vector2: XYAndZ, scale2) newPoint = basePoint.plus2Scaled (vector: XAndY, scale1, vector2: XAndY, scale2) newVector = baseVector.plus2Scaled (vector: XAndY, scale1, vector2: XAndY, scale2)
newPoint = basePoint.plus3Scaled (vector: XYAndZ, scale1, vector2: XYAndZ, scale2, vector3: XYAndZ, scale3) newVector = v.plus3Scaled (vector: XYAndZ, scale1, vector2: XYAndZ, scale2, vector3: XYAndZ, scale3) newPoint = basePoint.plus3Scaled (vector: XAndY, scale1, vector2: XAndY, scale2, vector3: XAndY, scale3) newVector = baseVector.plus3Scaled (vector: XAndY, scale1, vector2: XAndY, scale2, vector3: XAndY, scale3)

Dot, Cross and triple products

category Point3d Vector3d Point2d Vector2d
dot product of two vectors value = vectorA.dotProduct (vectorB) : number value = vectorA.dotProduct (vectorB) : number
dot product of XY parts of two vectors value = vectorA.dotProductXY (vectorB) : number
dot product with vector given as x,y,z value = vectorA.dotProductXYZ (x,y,z) : number
cross product of two vectors newVector = Vector3d.createCrossProduct (vectorA, vectorB) : Vector3d
value = vectorA.crossProduct (vectorB):number
cross product of XY parts of two vectors value = vectorA.crossProductXY (vectorB) : number
normalized cross product (or undefined if vectorA and vectorB are parallel) newVector = vectorA.unitCrossProduct (vectorB) : Vector3d | undefined
cross product scaled to given length (undefined if vectorA and vectorB are parallel) newVector = vectorA.sizedCrossProduct (vectorB, length) : Vector3d | undefined
normalized cross product (or use x,y,z for default if vectorA and vectorB are parallel) newVector = vectorA.unitCrossProductWithDefault (vectorB, x, y, z) : Vector3d
cross product of two vectors newVector = p.crossProductToPoints (targetA, targetB) : Vector3d newVector = Vector3d.createCrossProductToPoints (basePoint, targetA, targetB ) value = basePoint.crossProductToPoints (targetA, targetB) : number
cross product with vector between input points value = vectorA.crossProductStartEnd (startPoint, endPoint) : Vector3d
cross product with vector between input points, using only xy parts value = vectorA.crossProductStartEndXY (startPoint, endPoint) : Vector3d
cross product of vectors from instance point to 2 targets value = p.crossProductToPointsXY (targetA, targetB)
(scalar) triple product of three vectors value = p.tripleProductToPoints () : number
(scalar) triple product of three vectors value = vectorA.tripleProduct (vectorB, vectorC) : number
project instance onto a line segment, return fractional position fraction = spacePoint.fractionOfProjectionToLine (pointA, pointB) : number fraction = spaceVector.fractionOfProjectionToVector(targetVector) : number fraction = spacePoint.fractionOfProjectionToLine (pointA, pointB) : number fraction = spaceVector.fractionOfProjectionToVector(targetVector) : number
in the instance, accumulate cross product of vectors from (baseX. baseY, baseZ) to (ax, ay, az) and (bx, by, bz) vector.addCrossProductToTargetsInPlace (baseX, baseY, baseZ, ax, ay, az, bx, by, bz)
dot product of vectors from instance to 2 targets a = basePoint.dotVectorsToTargets (pointA, pointB) a = basePoint.dotVectorsToTargets (pointA, pointB)
dot product of instance vector with vector from startPoint to endPoint. a = vector.dotProductStartEnd (startPoint, endPoint) : number a = vector.dotProductStartEnd (startPoint, endPoint) : number
dot product of instance vector with vector from startPoint to endPoint. endPoint given as x,y,z,w to be unweighted. returns zero if weight is zero. a = vector.dotProductStartEndXYZW (startPoint, x,y,x,z) : number
squared magnitude of cross product value = vectorA.crossProductMagnitudeSquared (vectorB)
magnitude of cross product value = vectorA.crossProductMagnitude (vectorB)

Angles between vectors

  • Methods that return bare radians have radians in the method name.
  • Methods without specific radians indication return a (strongly typed) Angle object that can be queried for degrees or radians.
category Point3d Vector3d Point2d Vector2d
angle between vectors, 3D is "in their plane" and unsigned. 2D is in XY plane and signed angle = vectorA.angleTo (vectorB) : Angle angle = vectorA.angleTo (vectorB) : Angle
angle between vectors, in their plane but using outOfPlaneVector to define top and bottom. outOfPlane is not necessarily a perpendicular. angle = vectorA.signedAngleTo (vectorB, outOfPlaneVector) : Angle
radians between vectors, in their plane but using outOfPlaneVector to define top and bottom. outOfPlane is not necessarily a perpendicular. angle = vectorA.signedRadiansTo (vectorB, outOfPlaneVector) : number
angle between vectors, as viewed in xy plane angle = vectorA.angleToXY (vectorB) : Angle
angle between vectors, as viewed in plane perpendicular to planeNormal angle = vectorA.planarRadiansTo (vectorB, planeNormal) : number
angle = vectorA.planarAngleTo (vectorB, planeNormal) : Angle
parallel vector test vectorA.isParallelTo (vectorB, oppositeIsParallel : boolean = false, returnValueIfAnInputIsZero : boolean = false) vectorA.isParallelTo (vectorB, oppositeIsParallel : boolean = false)
perpendicular vector test vectorA.isPerpendicularTo (vectorB, returnValueIfAnInputIsZero : boolean = false) vectorA.isPerpendicularTo (vectorB)

Bisector of 2 unit vectors

>

Projection of a vector on a plane

>

In-place updates (instance methods)

category Point3d Vector3d Point2d Vector2d
set coordinates from number args p.set (x,y,z) v.set(x,y,z) p.set(x,y) v.set (x,y)
set coordinates to zero p.setZero () v.setZero () p.setZero v.setZero ()
set coordinates as vector between inputs v.setStartEnd (basePoint, targetPoint)
set coordinates from other objects p.setFrom (other: Float64Array | XAndY | XAndYAndZ) v.setFrom (other: Float64Array | XAndY | XAndYAndZ) p.setFrom (other?: XAndY) v.setFrom (other?: XAndY)
scale coordinates p.scaleInPlace (scaleFactor) v.scaleInPlace (scaleFactor)

Unary queries (instance methods)

category Point3d Vector3d Point2d Vector2d
metric zero test on all components p.isAlmostZero () v.isAlmostZero () p.isAlmostZero () v.isAlmostZero ()
largest absolute component (number) p.maxAbs () v.maxAbs () p.maxAbs () v.maxAbs ()
magnitude p.magnitude () v.magnitude () p.magnitude () v.magnitude ()
magnitude squared p.magnitudeSquared () v.magnitudeSquared () p.magnitudeSquared () v.magnitudeSquared ()
magnitude ignoring z p.magnitudeXY () v.magnitudeXY ()
magnitude squared ignoring z p.magnitudeSquaredXY () v.magnitudeSquaredXY ()

Property verification

These are static methods on the XYZ class. Hence they are inherited by Point3d and Vector3d. Their inputs are raw objects which may have x,y,z as properties but are not full-fledged Point3d, Vector3d, Point2d, Vector2d objects.

condition method
has x and y properties (z not tested) XYZ.isXAndY (anyObject)
has z propertiy XYZ.hasZ (anyObject)
has x, y and z properties (z not tested) XYZ.isXYAndZ (anyObject)

binary queries (instance methods)

(methods that take the "other" point or vector as direct numbers in the call list are considered "binary" for this table.)

category Point3d Vector3d Point2d Vector2d
near equality with metric tolerance p.isExactEqual (otherXYAndZ) v.isExactEqual (otherXYAndZ) p.isExactEqual (otherXAndY) v.isExactEqual (otherXAndY)
near equality with metric tolerance, unrolled xyz inputs p.isAlmostEqualXYZ (x,y,z) v.isAlmostEqual (x,y,z)
near equality with metric tolerance, unrolled xy inputs p.isAlmostEqualXYZ (x,y) v.isAlmostEqual (x,y)
exact equality p.isAlmostEqual (otherXYAndZ) v.isAlmostEqual (otherXYAndZ) p.isAlmostEqual (otherXAndY) v.isAlmostEqual (otherXAndY)
distance between p.distance (otherXYAndZ) v.distance (otherXYAndZ) p.distance (otherXAndY) v.distance (otherXAndY)
squared distance between p.distanceSquared (otherXYAndZ) v.distanceSquared (otherXYAndZ) p.distanceSquared (otherXAndY) v.distanceSquared (otherXAndY)
distance between, ignore z p.distanceXY (otherXYAndZ) v.distanceXY (otherXYAndZ)
squared distance between, ignore z p.distanceSquaredXY (otherXYAndZ) v.distanceSquaredXY (otherXYAndZ)
max component difference p.maxDiff (otherXYAndZ) v.maxDiff (otherXYAndZ) p.maxDiff (otherXAndY) v.maxDiff (otherXAndY)

Last Updated: 21 November, 2022