API change and bump go version

.github/workflows/go.yml

@@ -12,7 +12,7 @@ jobs:
   test:
     strategy:
       matrix:
-        go-version: ['1.20']
+        go-version: ['1.21']
         os: [ubuntu-latest, macos-latest, windows-latest]
     runs-on: ${{ matrix.os }}
     steps:
@@ -27,7 +27,7 @@ jobs:
     steps:
       - uses: golang/govulncheck-action@v1
         with:
-          go-version-input: 1.20
+          go-version-input: 1.21
           check-latest: true
 
   coverage:
@@ -36,10 +36,10 @@ jobs:
       - uses: actions/checkout@v3         
       - uses: actions/setup-go@v4
         with:
-          go-version: 1.20
+          go-version: 1.21
 
       - name: Test Coverage
-        run: go test -v -coverprofile=profile.cov ./...
+        run: go test -coverprofile=profile.cov ./...
 
       - uses: shogo82148/actions-goveralls@v1
         with:
@@ -51,7 +51,7 @@ jobs:
       - uses: actions/checkout@v3
       - uses: actions/setup-go@v4
         with:
-          go-version: 1.20
+          go-version: 1.21
 
       - name: golangci-lint
         uses: golangci/golangci-lint-action@v3

README.md

@@ -8,7 +8,7 @@
 
 ## API CHANGE !!!
 
-The API has changed from v0.1.0 to v0.2.0
+The API has changed from v0.9.1 to v0.10.0
 
 ## Overview
 
@@ -104,15 +104,6 @@ The benchmark for `Insert()` shows the values for inserting an item into trees w
 
 The trees are randomly generated, as is the item to be inserted.
 
-The trees are immutable, insertions and deletions generate new nodes on the path. The expected depth
-of the trees is **O(log(n))** and the **allocs/op** represent this well.
-
-The data structure is a randomized BST, the expected depth is determined with very
-high probability (for large n) but not deterministic.
-
-If the original tree is allowed to mutate during insert and delete because the old state is no longer needed,
-then the values are correspondingly better.
-
 ```
 $ go test -benchmem -bench='Insert'
 goos: linux
@@ -131,8 +122,6 @@ BenchmarkInsert/Into1_000_000-8         645523          1863 ns/op        64 B/o
 
 ### Delete
 
-The benchmark for `Delete()` shows the same asymptotic behavior:
-
 ```
 $ go test -benchmem -bench='Delete'
 goos: linux

example_period_test.go

@@ -1,6 +1,7 @@
 package interval_test
 
 import (
+	"cmp"
 	"fmt"
 	"os"
 
@@ -12,18 +13,10 @@ type uintInterval [2]uint
 
 // cmp function for uintInterval
 func cmpUintInterval(p, q uintInterval) (ll, rr, lr, rl int) {
-	return cmpUint(p[0], q[0]), cmpUint(p[1], q[1]), cmpUint(p[0], q[1]), cmpUint(p[1], q[0])
-}
-
-// little helper
-func cmpUint(a, b uint) int {
-	switch {
-	case a == b:
-		return 0
-	case a < b:
-		return -1
-	}
-	return 1
+	return cmp.Compare(p[0], q[0]),
+		cmp.Compare(p[1], q[1]),
+		cmp.Compare(p[0], q[1]),
+		cmp.Compare(p[1], q[0])
 }
 
 // example data

helpers.go

@@ -327,7 +327,7 @@ func (t Tree[T]) Statistics() (size int, maxDepth int, average, deviation float6
 	variance = variance / float64(sum)
 	deviation = math.Sqrt(variance)
 
-	return size, maxDepth, average, deviation
+	return size, maxDepth, math.Round(average*10000) / 10000, math.Round(deviation*10000) / 10000
 }
 
 // Min returns the min item in tree.
@@ -392,9 +392,9 @@ func (t Tree[T]) Visit(start, stop T, visitFn func(item T) bool) {
 
 // Clone, deep cloning of the tree structure.
 func (t Tree[T]) Clone() *Tree[T] {
-	c := NewTree[T](t.cmp)
+	c := t
 	c.root = t.clone(t.root)
-	return c
+	return &c
 }
 
 // clone rec-descent

treap_test.go

@@ -282,7 +282,7 @@ func TestMutable(t *testing.T) {
 	clone := tree1.Clone()
 
 	if !equalStatistics(tree1, clone) {
-		t.Fatalf("Clone, something wrong, statistics differs")
+		t.Error("Clone, something wrong, statistics differs")
 	}
 
 	min := tree1.Min()