fix: public facing key/value fields in entry

## Description

The `cuckoo.MinimumLoad()` option was not a very useful option, and prone to error. By removing the ability to modify it, and setting it to something reasonable (like 5%), we can remove a whole set of errors that the user may stumble into.

## Changes

- Remove `MinimumLoad()` option.
- Privated `DefaultMinimumLoad`.

### Design Decisions

- `DefaultMinimumLoad` should be privated because it is no longer an option. The user should not need to interact with it.

## Checklist

- [x] Tests pass
- [x] Docs updated

Reviewed-on: #17

.claude/settings.json

@@ -1,7 +0,0 @@
-{
-  "permissions": {
-    "allow": [
-      "Bash(curl -s -w \"\\\\n---HTTP_STATUS:%{http_code}---\" https://raw.githubusercontent.com/kubernetes/kubernetes/master/.markdownlint.yaml)"
-    ]
-  }
-}

.gitea/ISSUE_TEMPLATE/BUG_REPORT.yml

@@ -0,0 +1,26 @@
+# yaml-language-server: $schema=https://www.schemastore.org/gitea-issue-forms.json
+name: 🐛 Bug Report
+about: Report a bug in this project
+title: "[BUG]: "
+body:
+  - type: textarea
+    id: context
+    attributes:
+      label: Context
+      placeholder: What circumstances led to the bug?
+    validations:
+      required: true
+  - type: textarea
+    id: expected-behavior
+    attributes:
+      label: Expected Behavior
+      placeholder: What did you expect would happen?
+    validations:
+      required: true
+  - type: textarea
+    id: actual-behavior
+    attributes:
+      label: Actual Behavior
+      placeholder: What happened, and why was it unexpected?
+    validations:
+      required: true

.gitea/ISSUE_TEMPLATE/FEATURE_REQUEST.yml

@@ -0,0 +1,37 @@
+# 'Feature Request Template' By @cheehwatang
+# https://github.com/cheehwatang/.github/blob/master/.github/ISSUE_TEMPLATE/feature_request.yml
+#
+# yaml-language-server: $schema=https://www.schemastore.org/gitea-issue-forms.json
+name: ✨ Feature Request
+about: Suggest an idea for this project
+title: "[FEATURE]: "
+body:
+  - type: dropdown
+    attributes:
+      multiple: false
+      label: Feature Type
+      options:
+        - "✨ New Feature"
+        - "📝 Documentation"
+        - "🎨 Style and UI"
+        - "🔨 Code Refactor"
+        - "⚡ Performance Improvements"
+        - "✅ New Test"
+    validations:
+      required: true
+  - type: textarea
+    id: description
+    attributes:
+      label: Description
+      placeholder: |
+        Give us a brief description of the feature or enhancement you would
+        like!
+    validations:
+      required: true
+  - type: textarea
+    id: additional-information
+    attributes:
+      label: Additional Information
+      placeholder: |
+        Give us some additional information on the feature request like proposed
+        solutions, links, screenshots, etc.

.gitea/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,2 @@
+# yaml-language-server: $schema=https://www.schemastore.org/gitea-issue-config.json
+blank_issues_enabled: false

.gitea/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,17 @@
+---
+name: "New Pull Request"
+about: "Standard PR template"
+title: ""
+ref: "main"
+---
+
+## Description
+
+## Changes
+
+### Design Decisions
+
+## Checklist
+
+- [ ] Tests pass
+- [ ] Docs updated

.gitea/workflows/ci.yml

@@ -6,6 +6,19 @@ on:
   pull_request:
 
 jobs:
+  check-pr-title:
+    name: Check PR Title
+    runs-on: ubuntu-latest
+    if: github.event_name == 'pull_request'
+    env:
+      TITLE: ${{ gitea.event.pull_request.title }}
+    steps:
+      - run: |
+          if ! echo "$TITLE" | grep -qE '^(WIP: )?(feat|fix|docs|chore|ci|test|refactor|perf|build|style|revert)(\(.+\))?(!)?: .+'; then
+            echo "::error::Pull Request title must follow conventional commits"
+            exit 1
+          fi
+
   lint-go:
     name: Go Lint
     runs-on: ubuntu-latest

.gitignore

@@ -24,4 +24,3 @@ go.work.sum
 
 # env file
 .env
-

bucket.go

@@ -1,12 +1,13 @@
 package cuckoo
 
-type entry[K, V any] struct {
+// An Entry is a key-value pair.
-	key   K
+type Entry[K, V any] struct {
-	value V
+	Key   K
+	Value V
 }
 
 type slot[K, V any] struct {
-	entry[K, V]
+	Entry[K, V]
 	occupied bool
 }
 
@@ -29,7 +30,7 @@ func (b bucket[K, V]) get(key K) (value V, found bool) {
 	}
 
 	slot := b.slots[b.location(key)]
-	return slot.value, slot.occupied && b.compare(slot.key, key)
+	return slot.Value, slot.occupied && b.compare(slot.Key, key)
 }
 
 func (b *bucket[K, V]) drop(key K) (occupied bool) {
@@ -39,7 +40,7 @@ func (b *bucket[K, V]) drop(key K) (occupied bool) {
 
 	slot := &b.slots[b.location(key)]
 
-	if slot.occupied && b.compare(slot.key, key) {
+	if slot.occupied && b.compare(slot.Key, key) {
 		slot.occupied = false
 		b.size--
 		return true
@@ -48,10 +49,13 @@ func (b *bucket[K, V]) drop(key K) (occupied bool) {
 	return false
 }
 
-func (b *bucket[K, V]) resize(capacity uint64) {
+func (b *bucket[K, V]) resized(capacity uint64) bucket[K, V] {
-	b.slots = make([]slot[K, V], capacity)
+	return bucket[K, V]{
-	b.capacity = capacity
+		slots:    make([]slot[K, V], capacity),
-	b.size = 0
+		capacity: capacity,
+		hash:     b.hash,
+		compare:  b.compare,
+	}
 }
 
 func (b bucket[K, V]) update(key K, value V) (updated bool) {
@@ -61,34 +65,34 @@ func (b bucket[K, V]) update(key K, value V) (updated bool) {
 
 	slot := &b.slots[b.location(key)]
 
-	if slot.occupied && b.compare(slot.key, key) {
+	if slot.occupied && b.compare(slot.Key, key) {
-		slot.value = value
+		slot.Value = value
 		return true
 	}
 
 	return false
 }
 
-func (b *bucket[K, V]) evict(insertion entry[K, V]) (evicted entry[K, V], eviction bool) {
+func (b *bucket[K, V]) insert(insertion Entry[K, V]) (evicted Entry[K, V], eviction bool) {
 	if b.capacity == 0 {
 		return insertion, true
 	}
 
-	slot := &b.slots[b.location(insertion.key)]
+	slot := &b.slots[b.location(insertion.Key)]
 
 	if !slot.occupied {
-		slot.entry = insertion
+		slot.Entry = insertion
 		slot.occupied = true
 		b.size++
 		return
 	}
 
-	if b.compare(slot.key, insertion.key) {
+	if b.compare(slot.Key, insertion.Key) {
-		slot.value = insertion.value
+		slot.Value = insertion.Value
 		return
 	}
 
-	insertion, slot.entry = slot.entry, insertion
+	insertion, slot.Entry = slot.Entry, insertion
 	return insertion, true
 }
 

cuckoo_fuzz_test.go

@@ -1,7 +1,9 @@
 package cuckoo_test
 
 import (
+	"fmt"
 	"maps"
+	"os"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
@@ -26,6 +28,7 @@ type fuzzStep struct {
 
 type fuzzScenario struct {
 	seedA, seedB           uint32
+	capacity, growthFactor uint8
 	steps                  []fuzzStep
 }
 
@@ -40,29 +43,41 @@ func FuzzInsertLookup(f *testing.F) {
 			return
 		}
 
-		if scenario.seedA == scenario.seedB {
+		seedA, seedB := scenario.seedA, scenario.seedB
-			return
+		growthFactor := max(2, int(scenario.growthFactor))
+		capacity := int(scenario.capacity)
+
+		// If they are the same number, the hashes will clash, always causing an
+		// error.
+		if seedA == seedB {
+			t.Skip()
 		}
 
+		fmt.Fprintf(os.Stderr, "seedA=%d seedB=%d capacity=%d growthFactor=%d\n",
+			seedA, seedB, capacity, growthFactor)
+
 		actual := cuckoo.NewCustomTable[uint32, uint32](
-			offsetHash(scenario.seedA),
+			offsetHash(seedA),
-			offsetHash(scenario.seedB),
+			offsetHash(seedB),
 			func(a, b uint32) bool { return a == b },
+			cuckoo.Capacity(capacity),
+			cuckoo.GrowthFactor(growthFactor),
 		)
 
 		expected := map[uint32]uint32{}
 
 		for _, step := range scenario.steps {
 			if step.drop {
-				err := actual.Drop(step.key)
+				ok := actual.Drop(step.key)
-				assert.NoError(err)
+				_, has := expected[step.key]
+				assert.Equal(ok, has)
 
 				delete(expected, step.key)
 
-				_, err = actual.Get(step.key)
+				_, err := actual.Get(step.key)
 				assert.Error(err)
 			} else {
-				err := actual.Put(step.key, step.value)
+				_, err := actual.Put(step.key, step.value)
 				assert.NoError(err)
 
 				expected[step.key] = step.value

cuckoo_internal_test.go

@@ -23,7 +23,7 @@ func TestLoad(t *testing.T) {
 	table := NewTable[int, bool](Capacity(8))
 
 	for i := range 16 {
-		err := table.Put(i, true)
+		_, err := table.Put(i, true)
 		assert.NoError(err)
 		assert.Equal(float64(table.Size())/float64(table.TotalCapacity()), table.load())
 	}

cuckoo_test.go

@@ -25,7 +25,7 @@ func TestAddItem(t *testing.T) {
 	key, value := 0, true
 	table := cuckoo.NewTable[int, bool]()
 
-	err := table.Put(key, value)
+	_, err := table.Put(key, value)
 
 	assert.NoError(err)
 	assert.Equal(1, table.Size())
@@ -38,7 +38,7 @@ func TestPutOverwrite(t *testing.T) {
 	table := cuckoo.NewTable[int, int]()
 	(table.Put(key, value))
 
-	err := table.Put(key, newValue)
+	_, err := table.Put(key, newValue)
 
 	assert.NoError(err)
 	assert.Equal(1, table.Size())
@@ -52,9 +52,9 @@ func TestSameHash(t *testing.T) {
 	hash := func(int) uint64 { return 0 }
 	table := cuckoo.NewCustomTable[int, bool](hash, hash, cuckoo.DefaultEqualFunc[int])
 
-	errA := table.Put(0, true)
+	_, errA := table.Put(0, true)
-	errB := table.Put(1, true)
+	_, errB := table.Put(1, true)
-	errC := table.Put(2, true)
+	_, errC := table.Put(2, true)
 
 	assert.NoError(errA)
 	assert.NoError(errB)
@@ -76,7 +76,7 @@ func TestResizeCapacity(t *testing.T) {
 	)
 
 	for table.TotalCapacity() == 16 {
-		err := table.Put(rand.Int(), true)
+		_, err := table.Put(rand.Int(), true)
 		assert.NoError(err)
 	}
 
@@ -89,7 +89,7 @@ func TestPutMany(t *testing.T) {
 
 	for i := range 1_000 {
 		expected[i] = true
-		err := actual.Put(i, true)
+		_, err := actual.Put(i, true)
 
 		assert.NoError(err)
 	}
@@ -103,7 +103,7 @@ func TestGetMany(t *testing.T) {
 	table := cuckoo.NewTable[int, bool]()
 
 	for i := range 1_000 {
-		err := table.Put(i, true)
+		_, err := table.Put(i, true)
 		assert.NoError(err)
 	}
 
@@ -124,9 +124,9 @@ func TestDropExistingItem(t *testing.T) {
 	table := cuckoo.NewTable[int, bool]()
 	(table.Put(key, value))
 
-	err := table.Drop(key)
+	had := table.Drop(key)
 
-	assert.NoError(err)
+	assert.True(had)
 	assert.Equal(0, table.Size())
 	assert.False(table.Has(key))
 }
@@ -136,9 +136,9 @@ func TestDropNoItem(t *testing.T) {
 	key := 0
 	table := cuckoo.NewTable[int, bool]()
 
-	err := table.Drop(key)
+	had := table.Drop(key)
 
-	assert.NoError(err)
+	assert.False(had)
 	assert.Equal(0, table.Size())
 	assert.False(table.Has(key))
 }
@@ -152,10 +152,9 @@ func TestDropItemCapacity(t *testing.T) {
 	)
 
 	startingCapacity := table.TotalCapacity()
-	err := table.Drop(key)
+	table.Drop(key)
 	endingCapacity := table.TotalCapacity()
 
-	assert.NoError(err)
 	assert.Equal(0, table.Size())
 	assert.Equal(uint64(128), startingCapacity)
 	assert.Equal(uint64(64), endingCapacity)
@@ -168,7 +167,7 @@ func TestPutNoCapacity(t *testing.T) {
 		cuckoo.Capacity(0),
 	)
 
-	err := table.Put(key, value)
+	_, err := table.Put(key, value)
 
 	assert.NoError(err)
 	assert.Equal(1, table.Size())
@@ -184,9 +183,9 @@ func TestBadHashCapacity(t *testing.T) {
 		cuckoo.Capacity(20),
 	)
 
-	err1 := table.Put(0, true)
+	_, err1 := table.Put(0, true)
-	err2 := table.Put(1, true)
+	_, err2 := table.Put(1, true)
-	err3 := table.Put(2, true)
+	_, err3 := table.Put(2, true)
 
 	assert.NoError(err1)
 	assert.NoError(err2)
@@ -201,11 +200,11 @@ func TestDropResizeCapacity(t *testing.T) {
 		cuckoo.Capacity(10),
 	)
 
-	err1 := table.Put(0, true)
+	_, err1 := table.Put(0, true)
-	err2 := table.Put(1, true)
+	_, err2 := table.Put(1, true)
-	err3 := table.Drop(1)
+	table.Drop(1)
 
-	assert.NoError(errors.Join(err1, err2, err3))
+	assert.NoError(errors.Join(err1, err2))
 	assert.Equal(uint64(20), table.TotalCapacity())
 }
 
@@ -221,7 +220,7 @@ func TestNewTableBy(t *testing.T) {
 		func(u User) string { return u.id },
 	)
 
-	err := table.Put(User{nil, "1", "Robert"}, true)
+	_, err := table.Put(User{nil, "1", "Robert"}, true)
 
 	assert.NoError(err)
 	assert.Equal(1, table.Size())

doc_example_test.go

@@ -10,7 +10,7 @@ import (
 func Example_basic() {
 	table := cuckoo.NewTable[int, string]()
 
-	if err := table.Put(1, "Hello, World!"); err != nil {
+	if _, err := table.Put(1, "Hello, World!"); err != nil {
 		fmt.Println("Put error:", err)
 	}
 

settings.go

@@ -1,5 +1,7 @@
 package cuckoo
 
+import "fmt"
+
 // DefaultCapacity is the initial capacity of a [Table]. It is inspired from
 // Java's [HashMap] implementation, which also uses 16.
 //
@@ -10,11 +12,17 @@ const DefaultCapacity uint64 = 16
 // hash table implementations use 2.
 const DefaultGrowthFactor uint64 = 2
 
-// DefaultMinimumLoad is the default lowest acceptable occupancy of a [Table].
+// defaultMinimumLoad is the default lowest acceptable occupancy of a [Table].
-// The value of 5% is taken from [libcuckoo].
+// The higher the minimum load, the more likely that a [Table.Put] will not
+// succeed. The value of 5% is taken from [libcuckoo].
 //
 // [libcuckoo]: https://github.com/efficient/libcuckoo/blob/656714705a055df2b7a605eb3c71586d9da1e119/libcuckoo/cuckoohash_config.hh#L21
-const DefaultMinimumLoad float64 = 0.05
+const defaultMinimumLoad float64 = 0.05
+
+// defaultGrowthLimit is the maximum number of times a [Table] can grow in a
+// single [Table.Put], before the library infers it will lead to a stack
+// overflow. The value of '64' was chosen arbirarily.
+const defaultGrowthLimit uint64 = 64
 
 type settings struct {
 	growthFactor  uint64
@@ -27,19 +35,21 @@ type settings struct {
 type Option func(*settings)
 
 // Capacity modifies the starting capacity of each bucket of the [Table]. The
-// value must be greater than 0.
+// value must be non-negative.
 func Capacity(value int) Option {
-	return func(s *settings) { s.bucketSize = uint64(value) }
+	if value < 0 {
-}
+		panic(fmt.Sprintf("go-cuckoo: Capacity must be non-negative, got %d", value))
+	}
 
-// MinimumLoad modifies the [DefaultMinimumLoad] of the [Table]. The value must
+	return func(s *settings) { s.bucketSize = uint64(value) }
-// be between 0.00 and 1.00.
-func MinimumLoad(value float64) Option {
-	return func(s *settings) { s.minLoadFactor = value }
 }
 
 // GrowthFactor controls how much the capacity of the [Table] multiplies when
 // it must resize. The value must be greater than 1.
 func GrowthFactor(value int) Option {
+	if value < 2 {
+		panic(fmt.Sprintf("go-cuckoo: GrowthFactor must be greater than 1, got %d", value))
+	}
+
 	return func(s *settings) { s.growthFactor = uint64(value) }
 }

table.go

@@ -45,30 +45,61 @@ func (t Table[K, V]) load() float64 {
 	return float64(t.Size()) / float64(t.TotalCapacity())
 }
 
-// resize clears all buckets, changes the sizes of them to a specific capacity,
+// insert attempts to put/update an entry in the table, without modifying the
-// and fills them back up again. It is a helper function for [Table.grow] and
+// size of the table. Returns a displaced entry and 'homeless = true' if an
-// [Table.shrink]; use them instead.
+// entry could not be placed after exhausting evictions.
-func (t *Table[K, V]) resize(capacity uint64) error {
+func (t *Table[K, V]) insert(entry Entry[K, V]) (displaced Entry[K, V], homeless bool) {
-	entries := make([]entry[K, V], 0, t.Size())
+	if t.bucketA.update(entry.Key, entry.Value) {
+		return
+	}
+
+	if t.bucketB.update(entry.Key, entry.Value) {
+		return
+	}
+
+	for range t.maxEvictions() {
+		if entry, homeless = t.bucketA.insert(entry); !homeless {
+			return
+		}
+
+		if entry, homeless = t.bucketB.insert(entry); !homeless {
+			return
+		}
+	}
+
+	return entry, true
+}
+
+// resized creates an empty copy of the table, with a new capacity for each
+// bucket.
+func (t Table[K, V]) resized(capacity uint64) Table[K, V] {
+	return Table[K, V]{
+		growthFactor:  t.growthFactor,
+		minLoadFactor: t.minLoadFactor,
+		bucketA:       t.bucketA.resized(capacity),
+		bucketB:       t.bucketB.resized(capacity),
+	}
+}
+
+// resize creates a new [Table.resized] with 'capacity', inserts all items into
+// the array, and replaces the current table. It is a helper function for
+// [Table.grow] and [Table.shrink]; use them instead.
+func (t *Table[K, V]) resize(capacity uint64) bool {
+	updated := t.resized(capacity)
+
 	for k, v := range t.Entries() {
-		entries = append(entries, entry[K, V]{k, v})
+		if _, failed := updated.insert(Entry[K, V]{k, v}); failed {
-	}
+			return false
-
-	t.bucketA.resize(capacity)
-	t.bucketB.resize(capacity)
-
-	for _, entry := range entries {
-		if err := t.Put(entry.key, entry.value); err != nil {
-			return err
 		}
 	}
 
-	return nil
+	*t = updated
+	return true
 }
 
 // grow increases the table's capacity by the [Table.growthFactor]. If the
 // capacity is 0, it increases it to 1.
-func (t *Table[K, V]) grow() error {
+func (t *Table[K, V]) grow() bool {
 	var newCapacity uint64
 
 	if t.TotalCapacity() == 0 {
@@ -82,7 +113,7 @@ func (t *Table[K, V]) grow() error {
 
 // shrink reduces the table's capacity by the [Table.growthFactor]. It may
 // reduce it down to 0.
-func (t *Table[K, V]) shrink() error {
+func (t *Table[K, V]) shrink() bool {
 	return t.resize(t.bucketA.capacity / t.growthFactor)
 }
 
@@ -106,49 +137,52 @@ func (t Table[K, V]) Has(key K) (exists bool) {
 	return err == nil
 }
 
-// Put sets the value for a key. Returns error if its value cannot be set.
+// Put sets the value for a key. If it cannot be set, an error is returned,
-func (t *Table[K, V]) Put(key K, value V) (err error) {
+// along with the last displaced entry.
-	if t.bucketA.update(key, value) {
+//
-		return nil
+// On failure, the returned entry and the current table contents together
-	}
+// preserve all previously inserted entries and the attempted entry.
-
+func (t *Table[K, V]) Put(key K, value V) (displaced Entry[K, V], err error) {
-	if t.bucketB.update(key, value) {
+	var (
-		return nil
+		entry    = Entry[K, V]{key, value}
-	}
+		homeless bool
-
+	)
-	entry, eviction := entry[K, V]{key, value}, false
+
-	for range t.maxEvictions() {
+	for range defaultGrowthLimit {
-		if entry, eviction = t.bucketA.evict(entry); !eviction {
+		if entry, homeless = t.insert(entry); !homeless {
-			return nil
+			return
-		}
-
-		if entry, eviction = t.bucketB.evict(entry); !eviction {
-			return nil
-		}
 		}
 
+		// Both this and the growth limit are necessary: this catches bad hashes
+		// early when the table is sparse, while the latter catches cases where
+		// growing never helps.
 		if t.load() < t.minLoadFactor {
-		return fmt.Errorf("bad hash: resize on load %d/%d = %f", t.Size(), t.TotalCapacity(), t.load())
+			return entry, fmt.Errorf("bad hash: resize on load %d/%d", t.Size(), t.TotalCapacity())
 		}
 
-	if err := t.grow(); err != nil {
+		// It is theoretically possible to have a table with a larger capacity
-		return err
+		// that is valid. But this chance is astronomically small, so we ignore
+		// it in this implementation.
+		if grew := t.grow(); !grew {
+			return entry, fmt.Errorf("bad hash: could not redistribute entries into larger table")
+		}
 	}
 
-	return t.Put(entry.key, entry.value)
+	return entry, fmt.Errorf("bad hash: could not place entry after %d resizes", defaultGrowthLimit)
 }
 
-// Drop removes a value for a key in the table. Returns an error if its value
+// Drop removes a value for a key in the table. Returns whether the key had
-// cannot be removed.
+// existed.
-func (t *Table[K, V]) Drop(key K) (err error) {
+func (t *Table[K, V]) Drop(key K) bool {
-	t.bucketA.drop(key)
+	occupied := t.bucketA.drop(key) || t.bucketB.drop(key)
-	t.bucketB.drop(key)
 
 	if t.load() < t.minLoadFactor {
-		return t.shrink()
+		// The error is not handled here, because table-shrinking is an internal
+		// optimization.
+		t.shrink()
 	}
 
-	return nil
+	return occupied
 }
 
 // Entries returns an unordered sequence of all key-value pairs in the table.
@@ -156,7 +190,7 @@ func (t Table[K, V]) Entries() iter.Seq2[K, V] {
 	return func(yield func(K, V) bool) {
 		for _, slot := range t.bucketA.slots {
 			if slot.occupied {
-				if !yield(slot.key, slot.value) {
+				if !yield(slot.Key, slot.Value) {
 					return
 				}
 			}
@@ -164,7 +198,7 @@ func (t Table[K, V]) Entries() iter.Seq2[K, V] {
 
 		for _, slot := range t.bucketB.slots {
 			if slot.occupied {
-				if !yield(slot.key, slot.value) {
+				if !yield(slot.Key, slot.Value) {
 					return
 				}
 			}
@@ -198,7 +232,7 @@ func NewCustomTable[K, V any](hashA, hashB Hash[K], compare EqualFunc[K], option
 	settings := &settings{
 		growthFactor:  DefaultGrowthFactor,
 		bucketSize:    DefaultCapacity,
-		minLoadFactor: DefaultMinimumLoad,
+		minLoadFactor: defaultMinimumLoad,
 	}
 
 	for _, option := range options {