diff --git a/go.mod b/go.mod
index 7c148f8..a2cf39d 100644
--- a/go.mod
+++ b/go.mod
@@ -3,6 +3,10 @@ module github.com/prologic/bitcask
 go 1.13
 
 require (
+	github.com/boltdb/bolt v1.3.1
+	github.com/facebookgo/ensure v0.0.0-20200202191622-63f1cf65ac4c
+	github.com/facebookgo/stack v0.0.0-20160209184415-751773369052 // indirect
+	github.com/facebookgo/subset v0.0.0-20200203212716-c811ad88dec4 // indirect
 	github.com/gofrs/flock v0.7.1
 	github.com/pelletier/go-toml v1.6.0 // indirect
 	github.com/pkg/errors v0.9.1
diff --git a/go.sum b/go.sum
index 707aa7b..90a7921 100644
--- a/go.sum
+++ b/go.sum
@@ -25,6 +25,8 @@ github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24
 github.com/beorn7/perks v1.0.0/go.mod h1:KWe93zE9D1o94FZ5RNwFwVgaQK1VOXiVxmqh+CedLV8=
 github.com/bgentry/speakeasy v0.1.0/go.mod h1:+zsyZBPWlz7T6j88CTgSN5bM796AkVf0kBD4zp0CCIs=
 github.com/bketelsen/crypt v0.0.3-0.20200106085610-5cbc8cc4026c/go.mod h1:MKsuJmJgSg28kpZDP6UIiPt0e0Oz0kqKNGyRaWEPv84=
+github.com/boltdb/bolt v1.3.1 h1:JQmyP4ZBrce+ZQu0dY660FMfatumYDLun9hBCUVIkF4=
+github.com/boltdb/bolt v1.3.1/go.mod h1:clJnj/oiGkjum5o1McbSZDSLxVThjynRyGBgiAx27Ps=
 github.com/cespare/xxhash v1.1.0/go.mod h1:XrSqR1VqqWfGrhpAt58auRo0WTKS1nRRg3ghfAqPWnc=
 github.com/client9/misspell v0.3.4/go.mod h1:qj6jICC3Q7zFZvVWo7KLAzC3yx5G7kyvSDkc90ppPyw=
 github.com/coreos/bbolt v1.3.2/go.mod h1:iRUV2dpdMOn7Bo10OQBFzIJO9kkE559Wcmn+qkEiiKk=
@@ -40,6 +42,12 @@ github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/dgrijalva/jwt-go v3.2.0+incompatible/go.mod h1:E3ru+11k8xSBh+hMPgOLZmtrrCbhqsmaPHjLKYnJCaQ=
 github.com/dgryski/go-sip13 v0.0.0-20181026042036-e10d5fee7954/go.mod h1:vAd38F8PWV+bWy6jNmig1y/TA+kYO4g3RSRF0IAv0no=
+github.com/facebookgo/ensure v0.0.0-20200202191622-63f1cf65ac4c h1:8ISkoahWXwZR41ois5lSJBSVw4D0OV19Ht/JSTzvSv0=
+github.com/facebookgo/ensure v0.0.0-20200202191622-63f1cf65ac4c/go.mod h1:Yg+htXGokKKdzcwhuNDwVvN+uBxDGXJ7G/VN1d8fa64=
+github.com/facebookgo/stack v0.0.0-20160209184415-751773369052 h1:JWuenKqqX8nojtoVVWjGfOF9635RETekkoH6Cc9SX0A=
+github.com/facebookgo/stack v0.0.0-20160209184415-751773369052/go.mod h1:UbMTZqLaRiH3MsBH8va0n7s1pQYcu3uTb8G4tygF4Zg=
+github.com/facebookgo/subset v0.0.0-20200203212716-c811ad88dec4 h1:7HZCaLC5+BZpmbhCOZJ293Lz68O7PYrF2EzeiFMwCLk=
+github.com/facebookgo/subset v0.0.0-20200203212716-c811ad88dec4/go.mod h1:5tD+neXqOorC30/tWg0LCSkrqj/AR6gu8yY8/fpw1q0=
 github.com/fatih/color v1.7.0/go.mod h1:Zm6kSWBoL9eyXnKyktHP6abPY2pDugNf5KwzbycvMj4=
 github.com/fsnotify/fsnotify v1.4.7 h1:IXs+QLmnXW2CcXuY+8Mzv/fWEsPGWxqefPtCP5CnV9I=
 github.com/fsnotify/fsnotify v1.4.7/go.mod h1:jwhsz4b93w/PPRr/qN1Yymfu8t87LnFCMoQvtojpjFo=
diff --git a/hash.go b/hash.go
new file mode 100644
index 0000000..4b4841c
--- /dev/null
+++ b/hash.go
@@ -0,0 +1,114 @@
+package bitcask
+
+import (
+	"bytes"
+	"errors"
+)
+
+func (b *Bitcask) Hash(key []byte) *Hash {
+	return &Hash{db: b, key: key}
+}
+
+// Hash ...
+// 	+key,h = ""
+// 	h[key]name = "latermoon"
+// 	h[key]age = "27"
+// 	h[key]sex = "Male"
+type Hash struct {
+	db  *Bitcask
+	key []byte
+}
+
+func (h *Hash) Get(field []byte) ([]byte, error) {
+	return h.db.Get(h.fieldKey(field))
+}
+
+func (h *Hash) MGet(fields ...[]byte) ([][]byte, error) {
+	vals := make([][]byte, 0, len(fields))
+	for _, field := range fields {
+		val, err := h.db.Get(h.fieldKey(field))
+		if err != nil {
+			return nil, err
+		}
+		vals = append(vals, val)
+	}
+	return vals, nil
+}
+
+// GetAll ...
+func (h *Hash) GetAll() (map[string][]byte, error) {
+	keyVals := map[string][]byte{}
+	prefix := h.fieldPrefix()
+	err := h.db.Scan(prefix, func(key []byte) error {
+		val, err := h.db.Get(key)
+		if err != nil {
+			return err
+		}
+		keyVals[string(h.fieldInKey(key))] = val
+		return nil
+	})
+	return keyVals, err
+}
+
+func (h *Hash) Set(field, value []byte) error {
+	return h.MSet(field, value)
+}
+
+func (h *Hash) MSet(fieldVals ...[]byte) error {
+	if len(fieldVals) == 0 || len(fieldVals)%2 != 0 {
+		return errors.New("invalid field value pairs")
+	}
+
+	for i := 0; i < len(fieldVals); i += 2 {
+		field, val := fieldVals[i], fieldVals[i+1]
+		if err := h.db.Put(h.fieldKey(field), val); err != nil {
+			return err
+		}
+	}
+	return h.db.Put(h.rawKey(), nil)
+}
+
+func (h *Hash) Remove(fields ...[]byte) error {
+	for _, field := range fields {
+		if err := h.db.Delete(h.fieldKey(field)); err != nil {
+			return err
+		}
+	}
+	// clean up
+	prefix := h.fieldPrefix()
+	return h.db.Scan(prefix, func(key []byte) error {
+		return h.db.Delete(key)
+	})
+}
+
+func (h *Hash) Drop() error {
+	prefix := h.fieldPrefix()
+	err := h.db.Scan(prefix, func(key []byte) error {
+		return h.db.Delete(key)
+	})
+	if err != nil {
+		return err
+	}
+	return h.db.Delete(h.rawKey())
+}
+
+// +key,h
+func (h *Hash) rawKey() []byte {
+	return rawKey(h.key, HASH)
+}
+
+// h[key]field
+func (h *Hash) fieldKey(field []byte) []byte {
+	return bytes.Join([][]byte{h.fieldPrefix(), field}, nil)
+}
+
+// h[key]
+func (h *Hash) fieldPrefix() []byte {
+	return bytes.Join([][]byte{[]byte{byte(HASH)}, SOK, h.key, EOK}, nil)
+}
+
+// split h[key]field into field
+func (h *Hash) fieldInKey(fieldKey []byte) []byte {
+	right := bytes.Index(fieldKey, EOK)
+	return fieldKey[right+1:]
+}
diff --git a/list.go b/list.go
new file mode 100644
index 0000000..f6b0ad1
--- /dev/null
+++ b/list.go
@@ -0,0 +1,200 @@
+package bitcask
+
+import (
+	"bytes"
+	"errors"
+	"log"
+)
+
+func (b *Bitcask) List(key []byte) *List {
+	return &List{db: b, key: key}
+}
+
+// List ...
+// +key,l = ""
+// l[key]0 = "a"
+// l[key]1 = "b"
+// l[key]2 = "c"
+type List struct {
+	db  *Bitcask
+	key []byte
+}
+
+func (l *List) Index(i int64) ([]byte, error) {
+	x, err := l.leftIndex()
+	if err != nil {
+		return nil, err
+	}
+	return l.db.Get(l.indexKey(x + i))
+}
+
+// Range enumerate value by index
+// <start> must >= 0
+// <stop> should equal to -1 or lager than <start>
+func (l *List) Range(start, stop int64, fn func(i int64, value []byte, quit *bool)) error {
+	if start < 0 || (stop != -1 && start > stop) {
+		return errors.New("bad start/stop index")
+	}
+	x, y, err := l.rangeIndex()
+	if err != nil {
+		return err
+	}
+	if stop == -1 {
+		stop = (y - x + 1) - 1 // (size) - 1
+	}
+	min := l.indexKey(x + int64(start))
+	max := l.indexKey(x + int64(stop))
+	var i int64 // 0
+	ErrStopIteration := errors.New("err: stop iteration")
+	err = l.db.Scan(min, func(key []byte) error {
+		if key != nil && bytes.Compare(key, max) <= 0 {
+			val, err := l.db.Get(key)
+			if err != nil {
+				return err
+			}
+			quit := false
+			if fn(start+i, val, &quit); quit {
+				return ErrStopIteration
+			}
+			i++
+			return nil
+		}
+		return ErrStopIteration
+	})
+	if err == ErrStopIteration {
+		return nil
+	}
+	return err
+}
+
+// Append ...
+func (l *List) Append(vals ...[]byte) error {
+	x, y, err := l.rangeIndex()
+	if err != nil {
+		return err
+	}
+	if x == 0 && y == -1 {
+		if err := l.db.Put(l.rawKey(), nil); err != nil {
+			return err
+		}
+	}
+	for i, val := range vals {
+		if err := l.db.Put(l.indexKey(y+int64(i)+1), val); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+// Pop ...
+func (l *List) Pop() ([]byte, error) {
+	x, y, err := l.rangeIndex()
+	if err != nil {
+		return nil, err
+	}
+
+	size := y - x + 1
+	if size == 0 {
+		return nil, nil
+	} else if size < 0 { // double check
+		return nil, errors.New("bad list struct")
+	}
+
+	idxkey := l.indexKey(y)
+
+	val, err := l.db.Get(idxkey)
+	if err != nil {
+		return nil, err
+	}
+	if err := l.db.Delete(idxkey); err != nil {
+		return nil, err
+	}
+	if size == 1 { // clean up
+		return nil, l.db.Delete(l.rawKey())
+	}
+
+	return val, nil
+}
+
+// Len ...
+func (l *List) Len() (int64, error) {
+	x, y, err := l.rangeIndex()
+	return y - x + 1, err
+}
+
+func (l *List) rangeIndex() (int64, int64, error) {
+	left, err := l.leftIndex()
+	if err != nil {
+		return 0, -1, err
+	}
+	right, err := l.rightIndex()
+	if err != nil {
+		return 0, -1, err
+	}
+	log.Printf("left: %d\n", left)
+	log.Printf("right: %d\n", right)
+	return left, right, nil
+}
+
+func (l *List) leftIndex() (int64, error) {
+	log.Println("leftIndex:")
+	idx := int64(0) // default 0
+	prefix := l.keyPrefix()
+	log.Printf(" prefix: %s\n", prefix)
+	ErrStopIteration := errors.New("err: stop iteration")
+	err := l.db.Scan(prefix, func(key []byte) error {
+		log.Printf(" key: %v\n", key)
+		if bytes.HasPrefix(key, prefix) {
+			idx = l.indexInKey(key)
+			log.Printf("  idx: %d\n", idx)
+		}
+		return ErrStopIteration
+	})
+	if err == ErrStopIteration {
+		return idx, nil
+	}
+	return idx, err
+}
+
+func (l *List) rightIndex() (int64, error) {
+	log.Println("rightIndex:")
+	idx := int64(-1) // default -1
+	prefix := l.keyPrefix()
+	log.Printf(" prefix: %s\n", prefix)
+	err := l.db.Scan(prefix, func(key []byte) error {
+		log.Printf(" key: %v\n", key)
+		if bytes.HasPrefix(key, prefix) {
+			idx = l.indexInKey(key)
+			log.Printf(" idx: %d\n", idx)
+		}
+		return nil
+	})
+	return idx, err
+}
+
+// +key,l = ""
+func (l *List) rawKey() []byte {
+	return rawKey(l.key, ElemType(LIST))
+}
+
+// l[key]
+func (l *List) keyPrefix() []byte {
+	return bytes.Join([][]byte{[]byte{byte(LIST)}, SOK, l.key, EOK}, nil)
+}
+
+// l[key]0 = "a"
+func (l *List) indexKey(i int64) []byte {
+	sign := []byte{0}
+	if i >= 0 {
+		sign = []byte{1}
+	}
+	b := bytes.Join([][]byte{l.keyPrefix(), sign, itob(i)}, nil)
+	log.Printf("indexKeu: %x\n", b)
+	return b
+}
+
+// split l[key]index into index
+func (l *List) indexInKey(key []byte) int64 {
+	idxbuf := bytes.TrimPrefix(key, l.keyPrefix())
+	return btoi(idxbuf[1:]) // skip sign "0/1"
+}
diff --git a/score.go b/score.go
new file mode 100644
index 0000000..31a1b5e
--- /dev/null
+++ b/score.go
@@ -0,0 +1,63 @@
+package bitcask
+
+import (
+	"encoding/binary"
+	"math"
+)
+
+// Score indicated that a number can be encoded to sorted []byte
+// Score is use in SortedSet
+// you can implement your own decode & encode function just like below
+type Score []byte
+
+// Int64ToScore ...
+func Int64ToScore(i int64) Score {
+	b := make([]byte, 9)
+	// store sign in the first byte to keep the score order
+	if i < 0 {
+		b[0] = byte(0)
+	} else {
+		b[0] = byte(1)
+	}
+	binary.LittleEndian.PutUint64(b[1:], uint64(i))
+	return b
+}
+
+// ScoreToInt64 ...
+func ScoreToInt64(b Score) int64 {
+	return int64(binary.LittleEndian.Uint64(b[1:]))
+}
+
+// Float64ToScore ...
+func Float64ToScore(f float64) Score {
+	b := make([]byte, 8)
+	binary.LittleEndian.PutUint64(b, float64ToUint64(f))
+	return b
+}
+
+// ScoreToFloat64 ...
+func ScoreToFloat64(b Score) float64 {
+	return uint64ToFloat64(binary.LittleEndian.Uint64(b))
+}
+
+// Copy from https://github.com/reborndb/qdb/blob/master/pkg/store/util.go
+// We can not use lexicographically bytes comparison for negative and positive float directly.
+// so here we will do a trick below.
+func float64ToUint64(f float64) uint64 {
+	u := math.Float64bits(f)
+	if f >= 0 {
+		u |= 0x8000000000000000
+	} else {
+		u = ^u
+	}
+	return u
+}
+
+func uint64ToFloat64(u uint64) float64 {
+	if u&0x8000000000000000 > 0 {
+		u &= ^uint64(0x8000000000000000)
+	} else {
+		u = ^u
+	}
+	return math.Float64frombits(u)
+}
diff --git a/sortedset.go b/sortedset.go
new file mode 100644
index 0000000..c3cf86f
--- /dev/null
+++ b/sortedset.go
@@ -0,0 +1,163 @@
+package bitcask
+
+import (
+	"bytes"
+	"errors"
+)
+
+func (b *Bitcask) SortedSet(key []byte) *SortedSet {
+	return &SortedSet{db: b, key: key}
+}
+
+// SortedSet ...
+// +key,z = ""
+// z[key]m member = score
+// z[key]s score member = ""
+type SortedSet struct {
+	db  *Bitcask
+	key []byte
+}
+
+// Add add score & member pairs
+// SortedSet.Add(Score, []byte, Score, []byte ...)
+func (s *SortedSet) Add(scoreMembers ...[]byte) (int, error) {
+	count := len(scoreMembers)
+	if count < 2 || count%2 != 0 {
+		return 0, errors.New("invalid score/member pairs")
+	}
+	added := 0
+	for i := 0; i < count; i += 2 {
+		score, member := scoreMembers[i], scoreMembers[i+1]
+		skey, mkey := s.scoreKey(score, member), s.memberKey(member)
+		oldscore, err := s.db.Get(mkey)
+		if err != nil && err != ErrKeyNotFound {
+			return added, err
+		}
+		// remove old score key
+		if oldscore != nil {
+			oldskey := s.scoreKey(oldscore, member)
+			if err := s.db.Delete(oldskey); err != nil {
+				return added, err
+			}
+		} else {
+			added++
+		}
+		if err := s.db.Put(mkey, score); err != nil {
+			return added, err
+		}
+		if err := s.db.Put(skey, nil); err != nil {
+			return added, err
+		}
+	}
+	if err := s.db.Put(s.rawKey(), nil); err != nil {
+		return added, err
+	}
+	return added, nil
+}
+
+func (s SortedSet) Score(member []byte) (Score, error) {
+	return s.db.Get(s.memberKey(member))
+}
+
+func (s *SortedSet) Remove(members ...[]byte) (int, error) {
+	removed := 0 // not including non existing members
+	for _, member := range members {
+		score, err := s.db.Get(s.memberKey(member))
+		if err != nil {
+			return removed, err
+		}
+		if score == nil {
+			continue
+		}
+		if err := s.db.Delete(s.scoreKey(score, member)); err != nil {
+			return removed, err
+		}
+		if err := s.db.Delete(s.memberKey(member)); err != nil {
+			return removed, err
+		}
+		removed++
+	}
+	// clean up
+	prefix := s.keyPrefix()
+	ErrStopIteration := errors.New("err: stop iteration")
+	err := s.db.Scan(prefix, func(key []byte) error {
+		if !bytes.HasPrefix(key, prefix) {
+			if err := s.db.Delete(s.rawKey()); err != nil {
+				return err
+			}
+		}
+		return ErrStopIteration
+	})
+	if err != ErrStopIteration {
+		return removed, err
+	}
+	return removed, nil
+}
+
+// Range ...
+// <fr> is less than <to>
+func (s *SortedSet) Range(fr, to Score, fn func(i int64, score Score, member []byte, quit *bool)) error {
+	min := s.scorePrefix(fr)
+	max := append(s.scorePrefix(to), MAXBYTE)
+	var i int64 // 0
+	ErrStopIteration := errors.New("err: stop iteration")
+	err := s.db.Scan(min, func(key []byte) error {
+		if bytes.Compare(key, max) <= 0 {
+			quit := false
+			score, member, err := s.splitScoreKey(key)
+			if err != nil {
+				return err
+			}
+			if fn(i, score, member, &quit); quit {
+				return ErrStopIteration
+			}
+			i++
+		}
+		return nil
+	})
+	if err != ErrStopIteration {
+		return err
+	}
+	return nil
+}
+
+// +key,z = ""
+func (s *SortedSet) rawKey() []byte {
+	return rawKey(s.key, ElemType(SORTEDSET))
+}
+
+// z[key]
+func (s *SortedSet) keyPrefix() []byte {
+	return bytes.Join([][]byte{[]byte{byte(SORTEDSET)}, SOK, s.key, EOK}, nil)
+}
+
+// z[key]m
+func (s *SortedSet) memberKey(member []byte) []byte {
+	return bytes.Join([][]byte{s.keyPrefix(), []byte{'m'}, member}, nil)
+}
+
+// z[key]s score
+func (s *SortedSet) scorePrefix(score []byte) []byte {
+	return bytes.Join([][]byte{s.keyPrefix(), []byte{'s'}, score, []byte{' '}}, nil)
+}
+
+// z[key]s score member
+func (s *SortedSet) scoreKey(score, member []byte) []byte {
+	return bytes.Join([][]byte{s.keyPrefix(), []byte{'s'}, score, []byte{' '}, member}, nil)
+}
+
+// split (z[key]s score member) into (score, member)
+func (s *SortedSet) splitScoreKey(skey []byte) ([]byte, []byte, error) {
+	buf := bytes.TrimPrefix(skey, s.keyPrefix())
+	pairs := bytes.Split(buf[1:], []byte{' '}) // skip score mark 's'
+	if len(pairs) != 2 {
+		return nil, nil, errors.New("invalid score/member key: " + string(skey))
+	}
+	return pairs[0], pairs[1], nil
+}
+
+// split (z[key]m member) into (member)
+func (s *SortedSet) splitMemberKey(mkey []byte) ([]byte, error) {
+	buf := bytes.TrimPrefix(mkey, s.keyPrefix())
+	return buf[1:], nil // skip member mark 'm'
+}
diff --git a/tests/test_hash.go b/tests/test_hash.go
new file mode 100644
index 0000000..8be6b8d
--- /dev/null
+++ b/tests/test_hash.go
@@ -0,0 +1,27 @@
+package main
+
+import (
+	"log"
+
+	"github.com/prologic/bitcask"
+)
+
+func main() {
+	db, err := bitcask.Open("test.db")
+	if err != nil {
+		log.Fatal(err)
+	}
+	l := db.Hash([]byte("foo"))
+	err = l.Set([]byte("1"), []byte("one"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	err = l.Set([]byte("2"), []byte("two"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	err = l.Set([]byte("3"), []byte("three"))
+	if err != nil {
+		log.Fatal(err)
+	}
+}
diff --git a/tests/test_list.go b/tests/test_list.go
new file mode 100644
index 0000000..cc18229
--- /dev/null
+++ b/tests/test_list.go
@@ -0,0 +1,33 @@
+package main
+
+import (
+	"fmt"
+	"log"
+
+	"github.com/prologic/bitcask"
+)
+
+func main() {
+	db, err := bitcask.Open("test.db")
+	if err != nil {
+		log.Fatal(err)
+	}
+	l := db.List([]byte("foo"))
+	err = l.Append([]byte("one"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	err = l.Append([]byte("two"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	err = l.Append([]byte("three"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	len, err := l.Len()
+	if err != nil {
+		log.Fatal(err)
+	}
+	fmt.Printf("len: %d\n", len)
+}
diff --git a/tests/test_sortedset.go b/tests/test_sortedset.go
new file mode 100644
index 0000000..012eb20
--- /dev/null
+++ b/tests/test_sortedset.go
@@ -0,0 +1,31 @@
+package main
+
+import (
+	"fmt"
+	"log"
+
+	"github.com/prologic/bitcask"
+)
+
+func main() {
+	db, err := bitcask.Open("test.db")
+	if err != nil {
+		log.Fatal(err)
+	}
+	z := db.SortedSet([]byte("foo"))
+	added, err := z.Add(
+		bitcask.Int64ToScore(1), []byte("a"),
+		bitcask.Int64ToScore(2), []byte("b"),
+		bitcask.Int64ToScore(3), []byte("c"),
+	)
+	if err != nil {
+		log.Fatal(err)
+	}
+	fmt.Printf("added %d\n", added)
+
+	score, err := z.Score([]byte("b"))
+	if err != nil {
+		log.Fatal(err)
+	}
+	fmt.Printf("score: %d\n", bitcask.ScoreToInt64(score))
+}
diff --git a/tests/tests b/tests/tests
new file mode 100755
index 0000000..6547d3a
Binary files /dev/null and b/tests/tests differ
diff --git a/util.go b/util.go
new file mode 100644
index 0000000..38ccd30
--- /dev/null
+++ b/util.go
@@ -0,0 +1,80 @@
+package bitcask
+
+import (
+	"bytes"
+	"encoding/binary"
+	"errors"
+	"math"
+)
+
+var ErrInvalidKeyFormat = errors.New("invalid key format includes +[],")
+
+// Raw key:
+// +key,type = value
+// +name,s = "latermoon"
+
+var (
+	SEP = []byte{','}
+	KEY = []byte{'+'} // Key Prefix
+	SOK = []byte{'['} // Start of Key
+	EOK = []byte{']'} // End of Key
+)
+
+const (
+	MINBYTE byte = 0
+	MAXBYTE byte = math.MaxUint8
+)
+
+type ElemType byte
+
+const (
+	STRING    ElemType = 's'
+	HASH      ElemType = 'h'
+	LIST      ElemType = 'l'
+	SORTEDSET ElemType = 'z'
+	NONE      ElemType = '0'
+)
+
+func (e ElemType) String() string {
+	switch byte(e) {
+	case 's':
+		return "string"
+	case 'h':
+		return "hash"
+	case 'l':
+		return "list"
+	case 'z':
+		return "sortedset"
+	default:
+		return "none"
+	}
+}
+
+func rawKey(key []byte, t ElemType) []byte {
+	return bytes.Join([][]byte{KEY, key, SEP, []byte{byte(t)}}, nil)
+}
+
+func verifyKey(key []byte) error {
+	err := ErrInvalidKeyFormat
+	if bytes.Contains(key, SEP) {
+		return err
+	} else if bytes.Contains(key, KEY) {
+		return err
+	} else if bytes.Contains(key, SOK) {
+		return err
+	} else if bytes.Contains(key, EOK) {
+		return err
+	}
+	return nil
+}
+
+// itob returns an 8-byte big endian representation of v.
+func itob(i int64) []byte {
+	b := make([]byte, 8)
+	binary.LittleEndian.PutUint64(b, uint64(i))
+	return b
+}
+
+func btoi(b []byte) int64 {
+	return int64(binary.LittleEndian.Uint64(b))
+}