summaryrefslogtreecommitdiff
path: root/vendor/github.com/golang/geo/s2/edge_crossings.go
diff options
context:
space:
mode:
Diffstat (limited to 'vendor/github.com/golang/geo/s2/edge_crossings.go')
-rw-r--r--vendor/github.com/golang/geo/s2/edge_crossings.go396
1 files changed, 396 insertions, 0 deletions
diff --git a/vendor/github.com/golang/geo/s2/edge_crossings.go b/vendor/github.com/golang/geo/s2/edge_crossings.go
new file mode 100644
index 000000000..a98ec76ff
--- /dev/null
+++ b/vendor/github.com/golang/geo/s2/edge_crossings.go
@@ -0,0 +1,396 @@
+// Copyright 2017 Google Inc. All rights reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package s2
+
+import (
+ "fmt"
+ "math"
+
+ "github.com/golang/geo/r3"
+ "github.com/golang/geo/s1"
+)
+
+const (
+ // intersectionError can be set somewhat arbitrarily, because the algorithm
+ // uses more precision if necessary in order to achieve the specified error.
+ // The only strict requirement is that intersectionError >= dblEpsilon
+ // radians. However, using a larger error tolerance makes the algorithm more
+ // efficient because it reduces the number of cases where exact arithmetic is
+ // needed.
+ intersectionError = s1.Angle(8 * dblError)
+
+ // intersectionMergeRadius is used to ensure that intersection points that
+ // are supposed to be coincident are merged back together into a single
+ // vertex. This is required in order for various polygon operations (union,
+ // intersection, etc) to work correctly. It is twice the intersection error
+ // because two coincident intersection points might have errors in
+ // opposite directions.
+ intersectionMergeRadius = 2 * intersectionError
+)
+
+// A Crossing indicates how edges cross.
+type Crossing int
+
+const (
+ // Cross means the edges cross.
+ Cross Crossing = iota
+ // MaybeCross means two vertices from different edges are the same.
+ MaybeCross
+ // DoNotCross means the edges do not cross.
+ DoNotCross
+)
+
+func (c Crossing) String() string {
+ switch c {
+ case Cross:
+ return "Cross"
+ case MaybeCross:
+ return "MaybeCross"
+ case DoNotCross:
+ return "DoNotCross"
+ default:
+ return fmt.Sprintf("(BAD CROSSING %d)", c)
+ }
+}
+
+// CrossingSign reports whether the edge AB intersects the edge CD.
+// If AB crosses CD at a point that is interior to both edges, Cross is returned.
+// If any two vertices from different edges are the same it returns MaybeCross.
+// Otherwise it returns DoNotCross.
+// If either edge is degenerate (A == B or C == D), the return value is MaybeCross
+// if two vertices from different edges are the same and DoNotCross otherwise.
+//
+// Properties of CrossingSign:
+//
+// (1) CrossingSign(b,a,c,d) == CrossingSign(a,b,c,d)
+// (2) CrossingSign(c,d,a,b) == CrossingSign(a,b,c,d)
+// (3) CrossingSign(a,b,c,d) == MaybeCross if a==c, a==d, b==c, b==d
+// (3) CrossingSign(a,b,c,d) == DoNotCross or MaybeCross if a==b or c==d
+//
+// This method implements an exact, consistent perturbation model such
+// that no three points are ever considered to be collinear. This means
+// that even if you have 4 points A, B, C, D that lie exactly in a line
+// (say, around the equator), C and D will be treated as being slightly to
+// one side or the other of AB. This is done in a way such that the
+// results are always consistent (see RobustSign).
+func CrossingSign(a, b, c, d Point) Crossing {
+ crosser := NewChainEdgeCrosser(a, b, c)
+ return crosser.ChainCrossingSign(d)
+}
+
+// VertexCrossing reports whether two edges "cross" in such a way that point-in-polygon
+// containment tests can be implemented by counting the number of edge crossings.
+//
+// Given two edges AB and CD where at least two vertices are identical
+// (i.e. CrossingSign(a,b,c,d) == 0), the basic rule is that a "crossing"
+// occurs if AB is encountered after CD during a CCW sweep around the shared
+// vertex starting from a fixed reference point.
+//
+// Note that according to this rule, if AB crosses CD then in general CD
+// does not cross AB. However, this leads to the correct result when
+// counting polygon edge crossings. For example, suppose that A,B,C are
+// three consecutive vertices of a CCW polygon. If we now consider the edge
+// crossings of a segment BP as P sweeps around B, the crossing number
+// changes parity exactly when BP crosses BA or BC.
+//
+// Useful properties of VertexCrossing (VC):
+//
+// (1) VC(a,a,c,d) == VC(a,b,c,c) == false
+// (2) VC(a,b,a,b) == VC(a,b,b,a) == true
+// (3) VC(a,b,c,d) == VC(a,b,d,c) == VC(b,a,c,d) == VC(b,a,d,c)
+// (3) If exactly one of a,b equals one of c,d, then exactly one of
+// VC(a,b,c,d) and VC(c,d,a,b) is true
+//
+// It is an error to call this method with 4 distinct vertices.
+func VertexCrossing(a, b, c, d Point) bool {
+ // If A == B or C == D there is no intersection. We need to check this
+ // case first in case 3 or more input points are identical.
+ if a == b || c == d {
+ return false
+ }
+
+ // If any other pair of vertices is equal, there is a crossing if and only
+ // if OrderedCCW indicates that the edge AB is further CCW around the
+ // shared vertex O (either A or B) than the edge CD, starting from an
+ // arbitrary fixed reference point.
+
+ // Optimization: if AB=CD or AB=DC, we can avoid most of the calculations.
+ switch {
+ case a == c:
+ return (b == d) || OrderedCCW(Point{a.Ortho()}, d, b, a)
+ case b == d:
+ return OrderedCCW(Point{b.Ortho()}, c, a, b)
+ case a == d:
+ return (b == c) || OrderedCCW(Point{a.Ortho()}, c, b, a)
+ case b == c:
+ return OrderedCCW(Point{b.Ortho()}, d, a, b)
+ }
+
+ return false
+}
+
+// EdgeOrVertexCrossing is a convenience function that calls CrossingSign to
+// handle cases where all four vertices are distinct, and VertexCrossing to
+// handle cases where two or more vertices are the same. This defines a crossing
+// function such that point-in-polygon containment tests can be implemented
+// by simply counting edge crossings.
+func EdgeOrVertexCrossing(a, b, c, d Point) bool {
+ switch CrossingSign(a, b, c, d) {
+ case DoNotCross:
+ return false
+ case Cross:
+ return true
+ default:
+ return VertexCrossing(a, b, c, d)
+ }
+}
+
+// Intersection returns the intersection point of two edges AB and CD that cross
+// (CrossingSign(a,b,c,d) == Crossing).
+//
+// Useful properties of Intersection:
+//
+// (1) Intersection(b,a,c,d) == Intersection(a,b,d,c) == Intersection(a,b,c,d)
+// (2) Intersection(c,d,a,b) == Intersection(a,b,c,d)
+//
+// The returned intersection point X is guaranteed to be very close to the
+// true intersection point of AB and CD, even if the edges intersect at a
+// very small angle.
+func Intersection(a0, a1, b0, b1 Point) Point {
+ // It is difficult to compute the intersection point of two edges accurately
+ // when the angle between the edges is very small. Previously we handled
+ // this by only guaranteeing that the returned intersection point is within
+ // intersectionError of each edge. However, this means that when the edges
+ // cross at a very small angle, the computed result may be very far from the
+ // true intersection point.
+ //
+ // Instead this function now guarantees that the result is always within
+ // intersectionError of the true intersection. This requires using more
+ // sophisticated techniques and in some cases extended precision.
+ //
+ // - intersectionStable computes the intersection point using
+ // projection and interpolation, taking care to minimize cancellation
+ // error.
+ //
+ // - intersectionExact computes the intersection point using precision
+ // arithmetic and converts the final result back to an Point.
+ pt, ok := intersectionStable(a0, a1, b0, b1)
+ if !ok {
+ pt = intersectionExact(a0, a1, b0, b1)
+ }
+
+ // Make sure the intersection point is on the correct side of the sphere.
+ // Since all vertices are unit length, and edges are less than 180 degrees,
+ // (a0 + a1) and (b0 + b1) both have positive dot product with the
+ // intersection point. We use the sum of all vertices to make sure that the
+ // result is unchanged when the edges are swapped or reversed.
+ if pt.Dot((a0.Add(a1.Vector)).Add(b0.Add(b1.Vector))) < 0 {
+ pt = Point{pt.Mul(-1)}
+ }
+
+ return pt
+}
+
+// Computes the cross product of two vectors, normalized to be unit length.
+// Also returns the length of the cross
+// product before normalization, which is useful for estimating the amount of
+// error in the result. For numerical stability, the vectors should both be
+// approximately unit length.
+func robustNormalWithLength(x, y r3.Vector) (r3.Vector, float64) {
+ var pt r3.Vector
+ // This computes 2 * (x.Cross(y)), but has much better numerical
+ // stability when x and y are unit length.
+ tmp := x.Sub(y).Cross(x.Add(y))
+ length := tmp.Norm()
+ if length != 0 {
+ pt = tmp.Mul(1 / length)
+ }
+ return pt, 0.5 * length // Since tmp == 2 * (x.Cross(y))
+}
+
+/*
+// intersectionSimple is not used by the C++ so it is skipped here.
+*/
+
+// projection returns the projection of aNorm onto X (x.Dot(aNorm)), and a bound
+// on the error in the result. aNorm is not necessarily unit length.
+//
+// The remaining parameters (the length of aNorm (aNormLen) and the edge endpoints
+// a0 and a1) allow this dot product to be computed more accurately and efficiently.
+func projection(x, aNorm r3.Vector, aNormLen float64, a0, a1 Point) (proj, bound float64) {
+ // The error in the dot product is proportional to the lengths of the input
+ // vectors, so rather than using x itself (a unit-length vector) we use
+ // the vectors from x to the closer of the two edge endpoints. This
+ // typically reduces the error by a huge factor.
+ x0 := x.Sub(a0.Vector)
+ x1 := x.Sub(a1.Vector)
+ x0Dist2 := x0.Norm2()
+ x1Dist2 := x1.Norm2()
+
+ // If both distances are the same, we need to be careful to choose one
+ // endpoint deterministically so that the result does not change if the
+ // order of the endpoints is reversed.
+ var dist float64
+ if x0Dist2 < x1Dist2 || (x0Dist2 == x1Dist2 && x0.Cmp(x1) == -1) {
+ dist = math.Sqrt(x0Dist2)
+ proj = x0.Dot(aNorm)
+ } else {
+ dist = math.Sqrt(x1Dist2)
+ proj = x1.Dot(aNorm)
+ }
+
+ // This calculation bounds the error from all sources: the computation of
+ // the normal, the subtraction of one endpoint, and the dot product itself.
+ // dblError appears because the input points are assumed to be
+ // normalized in double precision.
+ //
+ // For reference, the bounds that went into this calculation are:
+ // ||N'-N|| <= ((1 + 2 * sqrt(3))||N|| + 32 * sqrt(3) * dblError) * epsilon
+ // |(A.B)'-(A.B)| <= (1.5 * (A.B) + 1.5 * ||A|| * ||B||) * epsilon
+ // ||(X-Y)'-(X-Y)|| <= ||X-Y|| * epsilon
+ bound = (((3.5+2*math.Sqrt(3))*aNormLen+32*math.Sqrt(3)*dblError)*dist + 1.5*math.Abs(proj)) * epsilon
+ return proj, bound
+}
+
+// compareEdges reports whether (a0,a1) is less than (b0,b1) with respect to a total
+// ordering on edges that is invariant under edge reversals.
+func compareEdges(a0, a1, b0, b1 Point) bool {
+ if a0.Cmp(a1.Vector) != -1 {
+ a0, a1 = a1, a0
+ }
+ if b0.Cmp(b1.Vector) != -1 {
+ b0, b1 = b1, b0
+ }
+ return a0.Cmp(b0.Vector) == -1 || (a0 == b0 && b0.Cmp(b1.Vector) == -1)
+}
+
+// intersectionStable returns the intersection point of the edges (a0,a1) and
+// (b0,b1) if it can be computed to within an error of at most intersectionError
+// by this function.
+//
+// The intersection point is not guaranteed to have the correct sign because we
+// choose to use the longest of the two edges first. The sign is corrected by
+// Intersection.
+func intersectionStable(a0, a1, b0, b1 Point) (Point, bool) {
+ // Sort the two edges so that (a0,a1) is longer, breaking ties in a
+ // deterministic way that does not depend on the ordering of the endpoints.
+ // This is desirable for two reasons:
+ // - So that the result doesn't change when edges are swapped or reversed.
+ // - It reduces error, since the first edge is used to compute the edge
+ // normal (where a longer edge means less error), and the second edge
+ // is used for interpolation (where a shorter edge means less error).
+ aLen2 := a1.Sub(a0.Vector).Norm2()
+ bLen2 := b1.Sub(b0.Vector).Norm2()
+ if aLen2 < bLen2 || (aLen2 == bLen2 && compareEdges(a0, a1, b0, b1)) {
+ return intersectionStableSorted(b0, b1, a0, a1)
+ }
+ return intersectionStableSorted(a0, a1, b0, b1)
+}
+
+// intersectionStableSorted is a helper function for intersectionStable.
+// It expects that the edges (a0,a1) and (b0,b1) have been sorted so that
+// the first edge passed in is longer.
+func intersectionStableSorted(a0, a1, b0, b1 Point) (Point, bool) {
+ var pt Point
+
+ // Compute the normal of the plane through (a0, a1) in a stable way.
+ aNorm := a0.Sub(a1.Vector).Cross(a0.Add(a1.Vector))
+ aNormLen := aNorm.Norm()
+ bLen := b1.Sub(b0.Vector).Norm()
+
+ // Compute the projection (i.e., signed distance) of b0 and b1 onto the
+ // plane through (a0, a1). Distances are scaled by the length of aNorm.
+ b0Dist, b0Error := projection(b0.Vector, aNorm, aNormLen, a0, a1)
+ b1Dist, b1Error := projection(b1.Vector, aNorm, aNormLen, a0, a1)
+
+ // The total distance from b0 to b1 measured perpendicularly to (a0,a1) is
+ // |b0Dist - b1Dist|. Note that b0Dist and b1Dist generally have
+ // opposite signs because b0 and b1 are on opposite sides of (a0, a1). The
+ // code below finds the intersection point by interpolating along the edge
+ // (b0, b1) to a fractional distance of b0Dist / (b0Dist - b1Dist).
+ //
+ // It can be shown that the maximum error in the interpolation fraction is
+ //
+ // (b0Dist * b1Error - b1Dist * b0Error) / (distSum * (distSum - errorSum))
+ //
+ // We save ourselves some work by scaling the result and the error bound by
+ // "distSum", since the result is normalized to be unit length anyway.
+ distSum := math.Abs(b0Dist - b1Dist)
+ errorSum := b0Error + b1Error
+ if distSum <= errorSum {
+ return pt, false // Error is unbounded in this case.
+ }
+
+ x := b1.Mul(b0Dist).Sub(b0.Mul(b1Dist))
+ err := bLen*math.Abs(b0Dist*b1Error-b1Dist*b0Error)/
+ (distSum-errorSum) + 2*distSum*epsilon
+
+ // Finally we normalize the result, compute the corresponding error, and
+ // check whether the total error is acceptable.
+ xLen := x.Norm()
+ maxError := intersectionError
+ if err > (float64(maxError)-epsilon)*xLen {
+ return pt, false
+ }
+
+ return Point{x.Mul(1 / xLen)}, true
+}
+
+// intersectionExact returns the intersection point of (a0, a1) and (b0, b1)
+// using precise arithmetic. Note that the result is not exact because it is
+// rounded down to double precision at the end. Also, the intersection point
+// is not guaranteed to have the correct sign (i.e., the return value may need
+// to be negated).
+func intersectionExact(a0, a1, b0, b1 Point) Point {
+ // Since we are using presice arithmetic, we don't need to worry about
+ // numerical stability.
+ a0P := r3.PreciseVectorFromVector(a0.Vector)
+ a1P := r3.PreciseVectorFromVector(a1.Vector)
+ b0P := r3.PreciseVectorFromVector(b0.Vector)
+ b1P := r3.PreciseVectorFromVector(b1.Vector)
+ aNormP := a0P.Cross(a1P)
+ bNormP := b0P.Cross(b1P)
+ xP := aNormP.Cross(bNormP)
+
+ // The final Normalize() call is done in double precision, which creates a
+ // directional error of up to 2*dblError. (Precise conversion and Normalize()
+ // each contribute up to dblError of directional error.)
+ x := xP.Vector()
+
+ if x == (r3.Vector{}) {
+ // The two edges are exactly collinear, but we still consider them to be
+ // "crossing" because of simulation of simplicity. Out of the four
+ // endpoints, exactly two lie in the interior of the other edge. Of
+ // those two we return the one that is lexicographically smallest.
+ x = r3.Vector{10, 10, 10} // Greater than any valid S2Point
+
+ aNorm := Point{aNormP.Vector()}
+ bNorm := Point{bNormP.Vector()}
+ if OrderedCCW(b0, a0, b1, bNorm) && a0.Cmp(x) == -1 {
+ return a0
+ }
+ if OrderedCCW(b0, a1, b1, bNorm) && a1.Cmp(x) == -1 {
+ return a1
+ }
+ if OrderedCCW(a0, b0, a1, aNorm) && b0.Cmp(x) == -1 {
+ return b0
+ }
+ if OrderedCCW(a0, b1, a1, aNorm) && b1.Cmp(x) == -1 {
+ return b1
+ }
+ }
+
+ return Point{x}
+}