bolt/bucket.go

package bolt

import (
	"bytes"
	"errors"
	"fmt"
	"unsafe"
)

const (
	// MaxKeySize is the maximum length of a key, in bytes.
	MaxKeySize = 32768

	// MaxValueSize is the maximum length of a value, in bytes.
	MaxValueSize = 4294967295
)

var (
	// ErrBucketNotFound is returned when trying to access a bucket that has
	// not been created yet.
	ErrBucketNotFound = errors.New("bucket not found")

	// ErrBucketExists is returned when creating a bucket that already exists.
	ErrBucketExists = errors.New("bucket already exists")

	// ErrBucketNameRequired is returned when creating a bucket with a blank name.
	ErrBucketNameRequired = errors.New("bucket name required")

	// ErrKeyRequired is returned when inserting a zero-length key.
	ErrKeyRequired = errors.New("key required")

	// ErrKeyTooLarge is returned when inserting a key that is larger than MaxKeySize.
	ErrKeyTooLarge = errors.New("key too large")

	// ErrValueTooLarge is returned when inserting a value that is larger than MaxValueSize.
	ErrValueTooLarge = errors.New("value too large")

	// ErrIncompatibleValue is returned when trying create or delete a bucket
	// on an existing non-bucket key or when trying to create or delete a
	// non-bucket key on an existing bucket key.
	ErrIncompatibleValue = errors.New("incompatible value")

	// ErrSequenceOverflow is returned when the next sequence number will be
	// larger than the maximum integer size.
	ErrSequenceOverflow = errors.New("sequence overflow")
)

const (
	maxUint = ^uint(0)
	minUint = 0
	maxInt  = int(^uint(0) >> 1)
	minInt  = -maxInt - 1
)

const bucketHeaderSize = int(unsafe.Sizeof(bucket{}))

// Bucket represents a collection of key/value pairs inside the database.
type Bucket struct {
	*bucket
	tx       *Tx                // the associated transaction
	buckets  map[string]*Bucket // subbucket cache
	page     *page              // inline page reference
	rootNode *node              // materialized node for the root page.
	nodes    map[pgid]*node     // node cache
}

// bucket represents the on-file representation of a bucket.
// This is stored as the "value" of a bucket key. If the bucket is small enough,
// then its root page can be stored inline in the "value", after the bucket
// header. In the case of inline buckets, the "root" will be 0.
type bucket struct {
	root     pgid   // page id of the bucket's root-level page
	sequence uint64 // monotonically incrementing, used by NextSequence()
}

// newBucket returns a new bucket associated with a transaction.
func newBucket(tx *Tx) Bucket {
	var b = Bucket{tx: tx}
	b.buckets = make(map[string]*Bucket)
	if tx.writable {
		b.nodes = make(map[pgid]*node)
	}
	return b
}

// Tx returns the tx of the bucket.
func (b *Bucket) Tx() *Tx {
	return b.tx
}

// Root returns the root of the bucket.
func (b *Bucket) Root() pgid {
	return b.root
}

// Writable returns whether the bucket is writable.
func (b *Bucket) Writable() bool {
	return b.tx.writable
}

// Cursor creates a cursor associated with the bucket.
// The cursor is only valid as long as the transaction is open.
// Do not use a cursor after the transaction is closed.
func (b *Bucket) Cursor() *Cursor {
	// Update transaction statistics.
	b.tx.stats.CursorCount++

	// Allocate and return a cursor.
	return &Cursor{
		bucket: b,
		stack:  make([]elemRef, 0),
	}
}

// Bucket retrieves a nested bucket by name.
// Returns nil if the bucket does not exist.
func (b *Bucket) Bucket(name []byte) *Bucket {
	if child := b.buckets[string(name)]; child != nil {
		return child
	}

	// Move cursor to key.
	c := b.Cursor()
	k, v, flags := c.seek(name)

	// Return nil if the key doesn't exist or it is not a bucket.
	if !bytes.Equal(name, k) || (flags&bucketLeafFlag) == 0 {
		return nil
	}

	// Otherwise create a bucket and cache it.
	var child = b.openBucket(v)
	b.buckets[string(name)] = child

	return child
}

// Helper method that re-interprets a sub-bucket value
// from a parent into a Bucket
func (b *Bucket) openBucket(value []byte) *Bucket {
	var child = newBucket(b.tx)
	child.bucket = &bucket{}
	*child.bucket = *(*bucket)(unsafe.Pointer(&value[0]))

	// Save a reference to the inline page if the bucket is inline.
	if child.root == 0 {
		child.page = (*page)(unsafe.Pointer(&value[bucketHeaderSize]))
	}

	return &child
}

// CreateBucket creates a new bucket at the given key and returns the new bucket.
// Returns an error if the key already exists, if the bucket name is blank, or if the bucket name is too long.
func (b *Bucket) CreateBucket(key []byte) (*Bucket, error) {
	if b.tx.db == nil {
		return nil, ErrTxClosed
	} else if !b.tx.writable {
		return nil, ErrTxNotWritable
	} else if len(key) == 0 {
		return nil, ErrBucketNameRequired
	}

	// Move cursor to correct position.
	c := b.Cursor()
	k, _, flags := c.seek(key)

	// Return an error if there is an existing key.
	if bytes.Equal(key, k) {
		if (flags & bucketLeafFlag) != 0 {
			return nil, ErrBucketExists
		} else {
			return nil, ErrIncompatibleValue
		}
	}

	// Create empty, inline bucket.
	var bucket = Bucket{bucket: &bucket{}, rootNode: &node{isLeaf: true}}
	var value = bucket.write()

	// Insert into node.
	key = cloneBytes(key)
	c.node().put(key, key, value, 0, bucketLeafFlag)

	// Since subbuckets are not allowed on inline buckets, we need to
	// dereference the inline page, if it exists. This will cause the bucket
	// to be treated as a regular, non-inline bucket for the rest of the tx.
	b.page = nil

	return b.Bucket(key), nil
}

// CreateBucketIfNotExists creates a new bucket if it doesn't already exist and returns a reference to it.
// Returns an error if the bucket name is blank, or if the bucket name is too long.
func (b *Bucket) CreateBucketIfNotExists(key []byte) (*Bucket, error) {
	child, err := b.CreateBucket(key)
	if err == ErrBucketExists {
		return b.Bucket(key), nil
	} else if err != nil {
		return nil, err
	}
	return child, nil
}

// DeleteBucket deletes a bucket at the given key.
// Returns an error if the bucket does not exists, or if the key represents a non-bucket value.
func (b *Bucket) DeleteBucket(key []byte) error {
	if b.tx.db == nil {
		return ErrTxClosed
	} else if !b.Writable() {
		return ErrTxNotWritable
	}

	// Move cursor to correct position.
	c := b.Cursor()
	k, _, flags := c.seek(key)

	// Return an error if bucket doesn't exist or is not a bucket.
	if !bytes.Equal(key, k) {
		return ErrBucketNotFound
	} else if (flags & bucketLeafFlag) == 0 {
		return ErrIncompatibleValue
	}

	// Recursively delete all child buckets.
	child := b.Bucket(key)
	err := child.ForEach(func(k, v []byte) error {
		if v == nil {
			if err := child.DeleteBucket(k); err != nil {
				return fmt.Errorf("delete bucket: %s", err)
			}
		}
		return nil
	})
	if err != nil {
		return err
	}

	// Remove cached copy.
	delete(b.buckets, string(key))

	// Release all bucket pages to freelist.
	child.nodes = nil
	child.rootNode = nil
	child.free()

	// Delete the node if we have a matching key.
	c.node().del(key)

	return nil
}

// Get retrieves the value for a key in the bucket.
// Returns a nil value if the key does not exist or if the key is a nested bucket.
func (b *Bucket) Get(key []byte) []byte {
	k, v, flags := b.Cursor().seek(key)

	// Return nil if this is a bucket.
	if (flags & bucketLeafFlag) != 0 {
		return nil
	}

	// If our target node isn't the same key as what's passed in then return nil.
	if !bytes.Equal(key, k) {
		return nil
	}
	return v
}

// Put sets the value for a key in the bucket.
// If the key exist then its previous value will be overwritten.
// Returns an error if the bucket was created from a read-only transaction, if the key is blank, if the key is too large, or if the value is too large.
func (b *Bucket) Put(key []byte, value []byte) error {
	if b.tx.db == nil {
		return ErrTxClosed
	} else if !b.Writable() {
		return ErrTxNotWritable
	} else if len(key) == 0 {
		return ErrKeyRequired
	} else if len(key) > MaxKeySize {
		return ErrKeyTooLarge
	} else if int64(len(value)) > MaxValueSize {
		return ErrValueTooLarge
	}

	// Move cursor to correct position.
	c := b.Cursor()
	k, _, flags := c.seek(key)

	// Return an error if there is an existing key with a bucket value.
	if bytes.Equal(key, k) && (flags&bucketLeafFlag) != 0 {
		return ErrIncompatibleValue
	}

	// Insert into node.
	key = cloneBytes(key)
	c.node().put(key, key, value, 0, 0)

	return nil
}

// Delete removes a key from the bucket.
// If the key does not exist then nothing is done and a nil error is returned.
// Returns an error if the bucket was created from a read-only transaction.
func (b *Bucket) Delete(key []byte) error {
	if b.tx.db == nil {
		return ErrTxClosed
	} else if !b.Writable() {
		return ErrTxNotWritable
	}

	// Move cursor to correct position.
	c := b.Cursor()
	_, _, flags := c.seek(key)

	// Return an error if there is already existing bucket value.
	if (flags & bucketLeafFlag) != 0 {
		return ErrIncompatibleValue
	}

	// Delete the node if we have a matching key.
	c.node().del(key)

	return nil
}

// NextSequence returns an autoincrementing integer for the bucket.
func (b *Bucket) NextSequence() (int, error) {
	if b.tx.db == nil {
		return 0, ErrTxClosed
	} else if !b.Writable() {
		return 0, ErrTxNotWritable
	}

	// Make sure next sequence number will not be larger than the maximum
	// integer size of the system.
	if b.bucket.sequence == uint64(maxInt) {
		return 0, ErrSequenceOverflow
	}

	// Increment and return the sequence.
	b.bucket.sequence++

	return int(b.bucket.sequence), nil
}

// ForEach executes a function for each key/value pair in a bucket.
// If the provided function returns an error then the iteration is stopped and
// the error is returned to the caller.
func (b *Bucket) ForEach(fn func(k, v []byte) error) error {
	if b.tx.db == nil {
		return ErrTxClosed
	}
	c := b.Cursor()
	for k, v := c.First(); k != nil; k, v = c.Next() {
		if err := fn(k, v); err != nil {
			return err
		}
	}
	return nil
}

// Stat returns stats on a bucket.
func (b *Bucket) Stats() BucketStats {
	var s, subStats BucketStats
	pageSize := b.tx.db.pageSize
	s.BucketN += 1
	if b.root == 0 {
		s.InlineBucketN += 1
	}
	b.forEachPage(func(p *page, depth int) {
		if (p.flags & leafPageFlag) != 0 {
			s.KeyN += int(p.count)
			used := pageHeaderSize
			if p.count != 0 {
				used += leafPageElementSize * int(p.count-1)
				lastElement := p.leafPageElement(p.count - 1)
				used += int(lastElement.pos + lastElement.ksize + lastElement.vsize)
			}
			if b.root == 0 {
				s.InlineBucketInuse += used
			} else {
				s.LeafPageN++
				s.LeafInuse += used
				s.LeafOverflowN += int(p.overflow)

				// Collect stats from sub-buckets
				for i := uint16(0); i < p.count; i++ {
					e := p.leafPageElement(i)
					if (e.flags & bucketLeafFlag) != 0 {
						subStats.Add(b.openBucket(e.value()).Stats())
					}
				}
			}
		} else if (p.flags & branchPageFlag) != 0 {
			s.BranchPageN++
			lastElement := p.branchPageElement(p.count - 1)
			used := pageHeaderSize + (branchPageElementSize * int(p.count-1))
			used += int(lastElement.pos + lastElement.ksize)
			s.BranchInuse += used
			s.BranchOverflowN += int(p.overflow)
		}

		if depth+1 > s.Depth {
			s.Depth = (depth + 1)
		}
	})
	s.BranchAlloc = (s.BranchPageN + s.BranchOverflowN) * pageSize
	s.LeafAlloc = (s.LeafPageN + s.LeafOverflowN) * pageSize

	// add the max depth of sub-buckets to get total nested depth
	s.Depth += subStats.Depth
	s.Add(subStats)
	return s
}

// forEachPage iterates over every page in a bucket, including inline pages.
func (b *Bucket) forEachPage(fn func(*page, int)) {
	// If we have an inline page then just use that.
	if b.page != nil {
		fn(b.page, 0)
		return
	}

	// Otherwise traverse the page hierarchy.
	b.tx.forEachPage(b.root, 0, fn)
}

// forEachPageNode iterates over every page (or node) in a bucket.
// This also includes inline pages.
func (b *Bucket) forEachPageNode(fn func(*page, *node, int)) {
	// If we have an inline page or root node then just use that.
	if b.page != nil {
		fn(b.page, nil, 0)
		return
	}
	b._forEachPageNode(b.root, 0, fn)
}

func (b *Bucket) _forEachPageNode(pgid pgid, depth int, fn func(*page, *node, int)) {
	var p, n = b.pageNode(pgid)

	// Execute function.
	fn(p, n, depth)

	// Recursively loop over children.
	if p != nil {
		if (p.flags & branchPageFlag) != 0 {
			for i := 0; i < int(p.count); i++ {
				elem := p.branchPageElement(uint16(i))
				b._forEachPageNode(elem.pgid, depth+1, fn)
			}
		}
	} else {
		if !n.isLeaf {
			for _, inode := range n.inodes {
				b._forEachPageNode(inode.pgid, depth+1, fn)
			}
		}
	}
}

// spill writes all the nodes for this bucket to dirty pages.
func (b *Bucket) spill() error {
	// Spill all child buckets first.
	for name, child := range b.buckets {
		// If the child bucket is small enough and it has no child buckets then
		// write it inline into the parent bucket's page. Otherwise spill it
		// like a normal bucket and make the parent value a pointer to the page.
		var value []byte
		if child.inlineable() {
			child.free()
			value = child.write()
		} else {
			if err := child.spill(); err != nil {
				return err
			}

			// Update the child bucket header in this bucket.
			value = make([]byte, unsafe.Sizeof(bucket{}))
			var bucket = (*bucket)(unsafe.Pointer(&value[0]))
			*bucket = *child.bucket
		}

		// Skip writing the bucket if there are no materialized nodes.
		if child.rootNode == nil {
			continue
		}

		// Update parent node.
		var c = b.Cursor()
		k, _, flags := c.seek([]byte(name))
		_assert(bytes.Equal([]byte(name), k), "misplaced bucket header: %x -> %x", []byte(name), k)
		_assert(flags&bucketLeafFlag != 0, "unexpected bucket header flag: %x", flags)
		c.node().put([]byte(name), []byte(name), value, 0, bucketLeafFlag)
	}

	// Ignore if there's not a materialized root node.
	if b.rootNode == nil {
		return nil
	}

	// Spill nodes.
	if err := b.rootNode.spill(); err != nil {
		return err
	}
	b.rootNode = b.rootNode.root()

	// Update the root node for this bucket.
	_assert(b.rootNode.pgid < b.tx.meta.pgid, "pgid (%d) above high water mark (%d)", b.rootNode.pgid, b.tx.meta.pgid)
	b.root = b.rootNode.pgid

	return nil
}

// inlineable returns true if a bucket is small enough to be written inline
// and if it contains no subbuckets. Otherwise returns false.
func (b *Bucket) inlineable() bool {
	var n = b.rootNode

	// Bucket must only contain a single leaf node.
	if n == nil || !n.isLeaf {
		return false
	}

	// Bucket is not inlineable if it contains subbuckets or if it goes beyond
	// our threshold for inline bucket size.
	var size = pageHeaderSize
	for _, inode := range n.inodes {
		size += leafPageElementSize + len(inode.key) + len(inode.value)

		if inode.flags&bucketLeafFlag != 0 {
			return false
		} else if size > b.maxInlineBucketSize() {
			return false
		}
	}

	return true
}

// Returns the maximum total size of a bucket to make it a candidate for inlining.
func (b *Bucket) maxInlineBucketSize() int {
	return b.tx.db.pageSize / 4
}

// write allocates and writes a bucket to a byte slice.
func (b *Bucket) write() []byte {
	// Allocate the appropriate size.
	var n = b.rootNode
	var value = make([]byte, bucketHeaderSize+pageHeaderSize+n.size())

	// Write a bucket header.
	var bucket = (*bucket)(unsafe.Pointer(&value[0]))
	*bucket = *b.bucket

	// Convert byte slice to a fake page and write the root node.
	var p = (*page)(unsafe.Pointer(&value[bucketHeaderSize]))
	n.write(p)

	return value
}

// rebalance attempts to balance all nodes.
func (b *Bucket) rebalance() {
	for _, n := range b.nodes {
		n.rebalance()
	}
	for _, child := range b.buckets {
		child.rebalance()
	}
}

// node creates a node from a page and associates it with a given parent.
func (b *Bucket) node(pgid pgid, parent *node) *node {
	_assert(b.nodes != nil, "nodes map expected")
	// Retrieve node if it's already been created.
	if n := b.nodes[pgid]; n != nil {
		return n
	}

	// Otherwise create a node and cache it.
	n := &node{bucket: b, parent: parent}
	if parent == nil {
		b.rootNode = n
	} else {
		parent.children = append(parent.children, n)
	}

	// Use the inline page if this is an inline bucket.
	var p = b.page
	if p == nil {
		p = b.tx.page(pgid)
	}

	// Read the page into the node and cache it.
	n.read(p)
	b.nodes[pgid] = n

	// Update statistics.
	b.tx.stats.NodeCount++

	return n
}

// free recursively frees all pages in the bucket.
func (b *Bucket) free() {
	if b.root == 0 {
		return
	}

	var tx = b.tx
	b.forEachPageNode(func(p *page, n *node, _ int) {
		if p != nil {
			tx.db.freelist.free(tx.id(), p)
		} else {
			n.free()
		}
	})
	b.root = 0
}

// dereference removes all references to the old mmap.
func (b *Bucket) dereference() {
	if b.rootNode != nil {
		b.rootNode.dereference()
	}

	for _, child := range b.buckets {
		child.dereference()
	}
}

// pageNode returns the in-memory node, if it exists.
// Otherwise returns the underlying page.
func (b *Bucket) pageNode(id pgid) (*page, *node) {
	// Inline buckets have a fake page embedded in their value so treat them
	// differently. We'll return the rootNode (if available) or the fake page.
	if b.root == 0 {
		_assert(id == 0, "inline bucket non-zero page access(2): %d != 0", id)
		if b.rootNode != nil {
			return nil, b.rootNode
		}
		return b.page, nil
	}

	// Check the node cache for non-inline buckets.
	if b.nodes != nil {
		if n := b.nodes[id]; n != nil {
			return nil, n
		}
	}

	// Finally lookup the page from the transaction if no node is materialized.
	return b.tx.page(id), nil
}

// BucketStats records statistics about resources used by a bucket.
type BucketStats struct {
	// Page count statistics.
	BranchPageN     int // number of logical branch pages
	BranchOverflowN int // number of physical branch overflow pages
	LeafPageN       int // number of logical leaf pages
	LeafOverflowN   int // number of physical leaf overflow pages

	// Tree statistics.
	KeyN  int // number of keys/value pairs
	Depth int // number of levels in B+tree

	// Page size utilization.
	BranchAlloc int // bytes allocated for physical branch pages
	BranchInuse int // bytes actually used for branch data
	LeafAlloc   int // bytes allocated for physical leaf pages
	LeafInuse   int // bytes actually used for leaf data

	// Bucket statistics
	BucketN           int // total number of buckets including the top bucket
	InlineBucketN     int // total number on inlined buckets
	InlineBucketInuse int // bytes used for inlined buckets (also accounted for in LeafInuse)
}

func (s *BucketStats) Add(other BucketStats) {
	s.BranchPageN += other.BranchPageN
	s.BranchOverflowN += other.BranchOverflowN
	s.LeafPageN += other.LeafPageN
	s.LeafOverflowN += other.LeafOverflowN
	s.KeyN += other.KeyN
	if s.Depth < other.Depth {
		s.Depth = other.Depth
	}
	s.BranchAlloc += other.BranchAlloc
	s.BranchInuse += other.BranchInuse
	s.LeafAlloc += other.LeafAlloc
	s.LeafInuse += other.LeafInuse

	s.BucketN += other.BucketN
	s.InlineBucketN += other.InlineBucketN
	s.InlineBucketInuse += other.InlineBucketInuse
}

// cloneBytes returns a copy of a given slice.
func cloneBytes(v []byte) []byte {
	var clone = make([]byte, len(v))
	copy(clone, v)
	return clone
}