Merge remote-tracking branch 'origin' into feat/safe-put

cuckoo_fuzz_test.go

@@ -68,15 +68,16 @@ func FuzzInsertLookup(f *testing.F) {
 
 		for _, step := range scenario.steps {
 			if step.drop {
-				err := actual.Drop(step.key)
-				assert.NoError(err)
+				ok := actual.Drop(step.key)
+				_, has := expected[step.key]
+				assert.Equal(ok, has)
 
 				delete(expected, step.key)
 
-				_, ok := actual.Get(step.key)
-				assert.False(ok)
+				_, ok = actual.Get(step.key)
+				assert.True(ok)
 			} 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 := New[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.New[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.New[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.NewCustom[int, bool](hash, hash, cuckoo.DefaultEqualFunc[int])
 
-	errA := table.Put(0, true)
-	errB := table.Put(1, true)
-	errC := table.Put(2, true)
+	_, errA := table.Put(0, true)
+	_, errB := table.Put(1, 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.New[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.New[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.New[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)
-	err2 := table.Put(1, true)
-	err3 := table.Put(2, true)
+	_, err1 := table.Put(0, true)
+	_, err2 := table.Put(1, 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)
-	err2 := table.Put(1, true)
-	err3 := table.Drop(1)
+	_, err1 := table.Put(0, true)
+	_, err2 := table.Put(1, true)
+	table.Drop(1)
 
-	assert.NoError(errors.Join(err1, err2, err3))
+	assert.NoError(errors.Join(err1, err2))
 	assert.Equal(uint64(20), table.TotalCapacity())
 }
 
@@ -219,7 +218,7 @@ func TestNewTableBy(t *testing.T) {
 	assert := assert.New(t)
 	table := cuckoo.NewBy[User, bool](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.New[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

@@ -19,6 +19,11 @@ const DefaultGrowthFactor uint64 = 2
 // [libcuckoo]: https://github.com/efficient/libcuckoo/blob/656714705a055df2b7a605eb3c71586d9da1e119/libcuckoo/cuckoohash_config.hh#L21
 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
 	minLoadFactor float64

subtable.go

@@ -1,12 +1,13 @@
 package cuckoo
 
-type entry[K, V any] struct {
-	key   K
-	value V
+// An Entry is a key-value pair.
+type Entry[K, V any] struct {
+	Key   K
+	Value V
 }
 
 type slot[K, V any] struct {
-	entry[K, V]
+	Entry[K, V]
 	occupied bool
 }
 
@@ -17,83 +18,86 @@ type subtable[K, V any] struct {
 	compare        EqualFunc[K]
 }
 
-// location determines where in the subtable a certain key would be placed. If
-// the capacity is 0, this will panic.
-func (t *subtable[K, V]) location(key K) uint64 {
-	return t.hash(key) % t.capacity
+// location determines where in the bucket a certain key would be placed. If the
+// capacity is 0, this will panic.
+func (b *subtable[K, V]) location(key K) uint64 {
+	return b.hash(key) % b.capacity
 }
 
-func (t *subtable[K, V]) get(key K) (value V, found bool) {
-	if t.capacity == 0 {
+func (b *subtable[K, V]) get(key K) (value V, found bool) {
+	if b.capacity == 0 {
 		return
 	}
 
-	slot := t.slots[t.location(key)]
-	return slot.value, slot.occupied && t.compare(slot.key, key)
+	slot := b.slots[b.location(key)]
+	return slot.Value, slot.occupied && b.compare(slot.Key, key)
 }
 
-func (t *subtable[K, V]) drop(key K) (occupied bool) {
-	if t.capacity == 0 {
+func (b *subtable[K, V]) drop(key K) (occupied bool) {
+	if b.capacity == 0 {
 		return
 	}
 
-	slot := &t.slots[t.location(key)]
+	slot := &b.slots[b.location(key)]
 
-	if slot.occupied && t.compare(slot.key, key) {
+	if slot.occupied && b.compare(slot.Key, key) {
 		slot.occupied = false
-		t.size--
+		b.size--
 		return true
 	}
 
 	return false
 }
 
-func (t *subtable[K, V]) resize(capacity uint64) {
-	t.slots = make([]slot[K, V], capacity)
-	t.capacity = capacity
-	t.size = 0
+func (b *subtable[K, V]) resized(capacity uint64) *subtable[K, V] {
+	return &subtable[K, V]{
+		slots:    make([]slot[K, V], capacity),
+		capacity: capacity,
+		hash:     b.hash,
+		compare:  b.compare,
+	}
 }
 
-func (t *subtable[K, V]) update(key K, value V) (updated bool) {
-	if t.capacity == 0 {
+func (b *subtable[K, V]) update(key K, value V) (updated bool) {
+	if b.capacity == 0 {
 		return
 	}
 
-	slot := &t.slots[t.location(key)]
+	slot := &b.slots[b.location(key)]
 
-	if slot.occupied && t.compare(slot.key, key) {
-		slot.value = value
+	if slot.occupied && b.compare(slot.Key, key) {
+		slot.Value = value
 		return true
 	}
 
 	return false
 }
 
-func (t *subtable[K, V]) evict(insertion entry[K, V]) (evicted entry[K, V], eviction bool) {
-	if t.capacity == 0 {
+func (b *subtable[K, V]) insert(insertion Entry[K, V]) (evicted Entry[K, V], eviction bool) {
+	if b.capacity == 0 {
 		return insertion, true
 	}
 
-	slot := &t.slots[t.location(insertion.key)]
+	slot := &b.slots[b.location(insertion.Key)]
 
 	if !slot.occupied {
-		slot.entry = insertion
+		slot.Entry = insertion
 		slot.occupied = true
-		t.size++
+		b.size++
 		return
 	}
 
-	if t.compare(slot.key, insertion.key) {
-		slot.value = insertion.value
+	if b.compare(slot.Key, insertion.Key) {
+		slot.Value = insertion.Value
 		return
 	}
 
-	insertion, slot.entry = slot.entry, insertion
+	insertion, slot.Entry = slot.Entry, insertion
 	return insertion, true
 }
 
-func newSubtable[K, V any](capacity uint64, hash Hash[K], compare EqualFunc[K]) subtable[K, V] {
-	return subtable[K, V]{
+func newSubtable[K, V any](capacity uint64, hash Hash[K], compare EqualFunc[K]) *subtable[K, V] {
+	return &subtable[K, V]{
 		hash:     hash,
 		capacity: capacity,
 		compare:  compare,

table.go

@@ -20,7 +20,7 @@ var ErrBadHash = errors.New("bad hash")
 // one with [New]. Or if you want more granularity, use [NewBy] or
 // [NewCustom].
 type Table[K, V any] struct {
-	tableA, tableB subtable[K, V]
+	tableA, tableB *subtable[K, V]
 	growthFactor   uint64
 	minLoadFactor  float64
 }
@@ -54,30 +54,61 @@ func (t *Table[K, V]) load() float64 {
 	return float64(t.Size()) / float64(t.TotalCapacity())
 }
 
-// resize clears all tables, changes the sizes of them to a specific capacity,
-// and fills them back up again. It is a helper function for [Table.grow] and
-// [Table.shrink]; use them instead.
-func (t *Table[K, V]) resize(capacity uint64) error {
-	entries := make([]entry[K, V], 0, t.Size())
+// insert attempts to put/update an entry in the table, without modifying the
+// size of the table. Returns a displaced entry and 'homeless = true' if an
+// entry could not be placed after exhausting evictions.
+func (t *Table[K, V]) insert(entry Entry[K, V]) (displaced Entry[K, V], homeless bool) {
+	if t.tableA.update(entry.Key, entry.Value) {
+		return
+	}
+
+	if t.tableB.update(entry.Key, entry.Value) {
+		return
+	}
+
+	for range t.maxEvictions() {
+		if entry, homeless = t.tableA.insert(entry); !homeless {
+			return
+		}
+
+		if entry, homeless = t.tableB.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,
+		tableA:        t.tableA.resized(capacity),
+		tableB:        t.tableB.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})
-	}
-
-	t.tableA.resize(capacity)
-	t.tableB.resize(capacity)
-
-	for _, entry := range entries {
-		if err := t.Put(entry.key, entry.value); err != nil {
-			return err
+		if _, failed := updated.insert(Entry[K, V]{k, v}); failed {
+			return false
 		}
 	}
 
-	return nil
+	*t = *updated
+	return true
 }
 
 // grow increases the table's capacity by the growth factor. 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 {
@@ -91,7 +122,7 @@ func (t *Table[K, V]) grow() error {
 
 // shrink reduces the table's capacity by the growth factor. 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.tableA.capacity / t.growthFactor)
 }
 
@@ -122,49 +153,52 @@ func (t *Table[K, V]) Has(key K) (exists bool) {
 	return
 }
 
-// Put sets the value for a key. Returns error if its value cannot be set.
-func (t *Table[K, V]) Put(key K, value V) (err error) {
-	if t.tableA.update(key, value) {
-		return nil
-	}
-
-	if t.tableB.update(key, value) {
-		return nil
-	}
-
-	entry, eviction := entry[K, V]{key, value}, false
-	for range t.maxEvictions() {
-		if entry, eviction = t.tableA.evict(entry); !eviction {
-			return nil
-		}
-
-		if entry, eviction = t.tableB.evict(entry); !eviction {
-			return nil
-		}
+// Put sets the value for a key. If it cannot be set, an error is returned,
+// along with the last displaced entry.
+//
+// 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) {
+	var (
+		entry    = Entry[K, V]{key, value}
+		homeless bool
+	)
+
+	for range defaultGrowthLimit {
+		if entry, homeless = t.insert(entry); !homeless {
+			return
 		}
 
+		// 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("hash functions produced a cycle at load %d/%d: %w", t.Size(), t.TotalCapacity(), ErrBadHash)
+			return entry, fmt.Errorf("bad hash: resize on load %d/%d", t.Size(), t.TotalCapacity())
 		}
 
-	if err := t.grow(); err != nil {
-		return err
+		// It is theoretically possible to have a table with a larger capacity
+		// 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
-// cannot be removed.
-func (t *Table[K, V]) Drop(key K) (err error) {
-	t.tableA.drop(key)
-	t.tableB.drop(key)
+// Drop removes a value for a key in the table. Returns whether the key had
+// existed.
+func (t *Table[K, V]) Drop(key K) bool {
+	occupied := t.tableA.drop(key) || t.tableB.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.
@@ -172,7 +206,7 @@ func (t *Table[K, V]) Entries() iter.Seq2[K, V] {
 	return func(yield func(K, V) bool) {
 		for _, slot := range t.tableA.slots {
 			if slot.occupied {
-				if !yield(slot.key, slot.value) {
+				if !yield(slot.Key, slot.Value) {
 					return
 				}
 			}
@@ -180,7 +214,7 @@ func (t *Table[K, V]) Entries() iter.Seq2[K, V] {
 
 		for _, slot := range t.tableB.slots {
 			if slot.occupied {
-				if !yield(slot.key, slot.value) {
+				if !yield(slot.Key, slot.Value) {
 					return
 				}
 			}