Merge pull request #3 from benbjohnson/spill

Spill to dirty pages, write to disk

branch.go

@@ -7,6 +7,8 @@ import (
 
 // branch represents a temporary in-memory branch page.
 type branch struct {
+	pgid   pgid
+	depth  int
 	parent *branch
 	items  branchItems
 }
@@ -42,11 +44,11 @@ func (b *branch) put(id pgid, newid pgid, key []byte, replace bool) {
 }
 
 // read initializes the item data from an on-disk page.
-func (b *branch) read(page *page) {
-	ncount := int(page.count)
-	b.items = make(branchItems, ncount)
-	bnodes := (*[maxNodesPerPage]bnode)(unsafe.Pointer(&page.ptr))
-	for i := 0; i < ncount; i++ {
+func (b *branch) read(p *page) {
+	b.pgid = p.id
+	b.items = make(branchItems, int(p.count))
+	bnodes := (*[maxNodesPerPage]bnode)(unsafe.Pointer(&p.ptr))
+	for i := 0; i < int(p.count); i++ {
 		bnode := &bnodes[i]
 		item := &b.items[i]
 		item.pgid = bnode.pgid
@@ -109,6 +111,12 @@ func (b *branch) split(pageSize int) []*branch {
 	return branches
 }
 
+type branches []*branch
+
+func (s branches) Len() int           { return len(s) }
+func (s branches) Swap(i, j int)      { s[i], s[j] = s[j], s[i] }
+func (s branches) Less(i, j int) bool { return s[i].depth < s[j].depth }
+
 type branchItems []branchItem
 
 type branchItem struct {

bucket.go

@@ -1,5 +1,7 @@
 package bolt
 
+const MaxBucketNameSize = 255
+
 type Bucket struct {
 	*bucket
 	name        string

db.go

@@ -118,7 +118,9 @@ func (db *DB) mmap() error {
 	} else if int(info.Size()) < db.pageSize*2 {
 		return &Error{"file size too small", err}
 	}
-	size := int(info.Size())
+
+	// TEMP(benbjohnson): Set max size to 1MB.
+	size := 2 << 20
 
 	// Memory-map the data file as a byte slice.
 	if db.data, err = db.syscall.Mmap(int(db.file.Fd()), 0, size, syscall.PROT_READ, syscall.MAP_SHARED); err != nil {
@@ -159,7 +161,9 @@ func (db *DB) init() error {
 		m.pageSize = uint32(db.pageSize)
 		m.version = Version
 		m.free = 2
-		m.sys.root = 3
+		m.sys = 3
+		m.pgid = 4
+		m.txnid = txnid(i)
 	}
 
 	// Write an empty freelist at page 3.
@@ -171,7 +175,7 @@ func (db *DB) init() error {
 	// Write an empty leaf page at page 4.
 	p = db.pageInBuffer(buf[:], pgid(3))
 	p.id = pgid(3)
-	p.flags = p_leaf
+	p.flags = p_sys
 	p.count = 0
 
 	// Write the buffer to our data file.
@@ -206,7 +210,7 @@ func (db *DB) Transaction() (*Transaction, error) {
 
 	// Create a transaction associated with the database.
 	t := &Transaction{}
-	t.init(db, db.meta())
+	t.init(db)
 
 	return t, nil
 }
@@ -230,7 +234,7 @@ func (db *DB) RWTransaction() (*RWTransaction, error) {
 		branches: make(map[pgid]*branch),
 		leafs:    make(map[pgid]*leaf),
 	}
-	t.init(db, db.meta())
+	t.init(db)
 
 	return t, nil
 }

leaf.go

@@ -8,6 +8,7 @@ import (
 
 // leaf represents an in-memory, deserialized leaf page.
 type leaf struct {
+	pgid   pgid
 	parent *branch
 	items  leafItems
 }
@@ -39,10 +40,10 @@ func (l *leaf) put(key []byte, value []byte) {
 
 // read initializes the item data from an on-disk page.
 func (l *leaf) read(p *page) {
-	ncount := int(p.count)
-	l.items = make(leafItems, ncount)
+	l.pgid = p.id
+	l.items = make(leafItems, int(p.count))
 	lnodes := (*[maxNodesPerPage]lnode)(unsafe.Pointer(&p.ptr))
-	for i := 0; i < ncount; i++ {
+	for i := 0; i < int(p.count); i++ {
 		lnode := &lnodes[i]
 		item := &l.items[i]
 		item.key = lnode.key()

meta.go

@@ -14,8 +14,8 @@ type meta struct {
 	pageSize uint32
 	pgid     pgid
 	free     pgid
+	sys      pgid
 	txnid    txnid
-	sys      bucket
 }
 
 // validate checks the marker bytes and version of the meta page to ensure it matches this binary.
@@ -30,8 +30,20 @@ func (m *meta) validate() error {
 
 // copy copies one meta object to another.
 func (m *meta) copy(dest *meta) {
+	dest.magic = m.magic
+	dest.version = m.version
 	dest.pageSize = m.pageSize
 	dest.pgid = m.pgid
+	dest.free = m.free
 	dest.txnid = m.txnid
 	dest.sys = m.sys
 }
+
+// write writes the meta onto a page.
+func (m *meta) write(p *page) {
+	// Page id is either going to be 0 or 1 which we can determine by the Txn ID.
+	p.id = pgid(m.txnid % 2)
+	p.flags |= p_meta
+
+	m.copy(p.meta())
+}

page.go

@@ -14,7 +14,8 @@ const (
 	p_branch   = 0x01
 	p_leaf     = 0x02
 	p_meta     = 0x04
-	p_freelist = 0x08
+	p_sys      = 0x08
+	p_freelist = 0x10
 )
 
 type pgid uint64
@@ -56,3 +57,9 @@ func (p *page) bnodes() []bnode {
 func (p *page) freelist() []pgid {
 	return ((*[maxNodesPerPage]pgid)(unsafe.Pointer(&p.ptr)))[0:p.count]
 }
+
+type pages []*page
+
+func (s pages) Len() int           { return len(s) }
+func (s pages) Swap(i, j int)      { s[i], s[j] = s[j], s[i] }
+func (s pages) Less(i, j int) bool { return s[i].id < s[j].id }

rwtransaction.go

@@ -1,6 +1,7 @@
 package bolt
 
 import (
+	"sort"
 	"unsafe"
 )
 
@@ -12,31 +13,43 @@ type RWTransaction struct {
 	leafs    map[pgid]*leaf
 }
 
+// init initializes the transaction.
+func (t *RWTransaction) init(db *DB) {
+	t.Transaction.init(db)
+	t.pages = make(map[pgid]*page)
+
+	// Copy the meta and increase the transaction id. 
+	t.meta = &meta{}
+	db.meta().copy(t.meta)
+	t.meta.txnid += txnid(2)
+}
+
 // CreateBucket creates a new bucket.
 func (t *RWTransaction) CreateBucket(name string) error {
 	// Check if bucket already exists.
 	if b := t.Bucket(name); b != nil {
 		return &Error{"bucket already exists", nil}
+	} else if len(name) == 0 {
+		return &Error{"bucket name cannot be blank", nil}
+	} else if len(name) > MaxBucketNameSize {
+		return &Error{"bucket name too large", nil}
 	}
 
-	// Create a new bucket entry.
-	var buf [unsafe.Sizeof(bucket{})]byte
-	var raw = (*bucket)(unsafe.Pointer(&buf[0]))
-	raw.root = 0
+	// Create a blank root leaf page.
+	p := t.allocate(1)
+	p.flags = p_leaf
 
-	// Move cursor to insertion location.
-	c := t.sys.Cursor()
-	c.Goto([]byte(name))
-
-	// Load the leaf node from the cursor and insert the key/value.
-	t.leaf(c).put([]byte(name), buf[:])
+	// Add bucket to system page.
+	t.sys.put(name, &bucket{root: p.id})
 
 	return nil
 }
 
 // DropBucket deletes a bucket.
-func (t *RWTransaction) DeleteBucket(b *Bucket) error {
-	// TODO: Remove from main DB.
+func (t *RWTransaction) DeleteBucket(name string) error {
+	// Remove from system page.
+	t.sys.del(name)
+
 	// TODO: Delete entry from system bucket.
 	// TODO: Free all pages.
 	// TODO: Remove cursor.
@@ -74,13 +87,31 @@ func (t *RWTransaction) Delete(key []byte) error {
 	return nil
 }
 
+// Commit writes all changes to disk.
 func (t *RWTransaction) Commit() error {
 	// TODO(benbjohnson): Use vectorized I/O to write out dirty pages.
 
-	// TODO: Flush data.
+	// TODO: Rebalance.
+
+	// Spill data onto dirty pages.
+	t.spill()
 
-	// TODO: Update meta.
-	// TODO: Write meta.
+	// Spill system page.
+	p := t.allocate((t.sys.size() / t.db.pageSize) + 1)
+	t.sys.write(p)
+
+	// Write dirty pages to disk.
+	if err := t.write(); err != nil {
+		return err
+	}
+
+	// Update the meta.
+	t.meta.sys = p.id
+
+	// Write meta to disk.
+	if err := t.writeMeta(); err != nil {
+		return err
+	}
 
 	return nil
 }
@@ -99,10 +130,107 @@ func (t *RWTransaction) close() error {
 }
 
 // allocate returns a contiguous block of memory starting at a given page.
-func (t *RWTransaction) allocate(size int) (*page, error) {
-	// TODO: Find a continuous block of free pages.
-	// TODO: If no free pages are available, resize the mmap to allocate more.
-	return nil, nil
+func (t *RWTransaction) allocate(count int) *page {
+	// TODO(benbjohnson): Use pages from the freelist.
+
+	// Allocate a set of contiguous pages from the end of the file.
+	buf := make([]byte, count*t.db.pageSize)
+	p := (*page)(unsafe.Pointer(&buf[0]))
+	p.id = t.meta.pgid
+	p.overflow = uint32(count - 1)
+
+	// Increment the last page id.
+	t.meta.pgid += pgid(count)
+
+	// Save it in our page cache.
+	t.pages[p.id] = p
+
+	return p
+}
+
+// spill writes all the leafs and branches to dirty pages.
+func (t *RWTransaction) spill() {
+	// Spill leafs first.
+	for _, l := range t.leafs {
+		t.spillLeaf(l)
+	}
+
+	// Sort branches by highest depth first.
+	branches := make(branches, 0, len(t.branches))
+	for _, b := range t.branches {
+		branches = append(branches, b)
+	}
+	sort.Sort(branches)
+
+	// Spill branches by deepest first.
+	for _, b := range branches {
+		t.spillBranch(b)
+	}
+}
+
+// spillLeaf writes a leaf to one or more dirty pages.
+func (t *RWTransaction) spillLeaf(l *leaf) {
+	parent := l.parent
+
+	// Split leaf, if necessary.
+	leafs := l.split(t.db.pageSize)
+
+	// TODO: If this is a root leaf and we split then add a parent branch.
+
+	// Process each resulting leaf.
+	previd := leafs[0].pgid
+	for index, l := range leafs {
+		// Allocate contiguous space for the leaf.
+		p := t.allocate((l.size() / t.db.pageSize) + 1)
+
+		// Write the leaf to the page.
+		l.write(p)
+
+		// Insert or replace the node in the parent branch with the new pgid.
+		if parent != nil {
+			parent.put(previd, p.id, l.items[0].key, (index == 0))
+			previd = l.pgid
+		}
+	}
+}
+
+// spillBranch writes a branch to one or more dirty pages.
+func (t *RWTransaction) spillBranch(l *branch) {
+	warn("[pending] RWTransaction.spillBranch()") // TODO
+}
+
+// write writes any dirty pages to disk.
+func (t *RWTransaction) write() error {
+	// TODO(benbjohnson): If our last page id is greater than the mmap size then lock the DB and resize.
+
+	// Sort pages by id.
+	pages := make(pages, 0, len(t.pages))
+	for _, p := range t.pages {
+		pages = append(pages, p)
+	}
+	sort.Sort(pages)
+
+	// Write pages to disk in order.
+	for _, p := range pages {
+		size := (int(p.overflow) + 1) * t.db.pageSize
+		buf := (*[maxAllocSize]byte)(unsafe.Pointer(p))[:size]
+		t.db.file.WriteAt(buf, int64(p.id)*int64(t.db.pageSize))
+	}
+
+	return nil
+}
+
+// writeMeta writes the meta to the disk.
+func (t *RWTransaction) writeMeta() error {
+	// Create a temporary buffer for the meta page.
+	buf := make([]byte, t.db.pageSize)
+	p := t.db.pageInBuffer(buf, 0)
+	t.meta.write(p)
+
+	// Write the meta page to file.
+	t.db.metafile.WriteAt(buf, int64(p.id)*int64(t.db.pageSize))
+
+	return nil
 }
 
 // leaf retrieves a leaf object based on the current position of a cursor.
@@ -141,6 +269,7 @@ func (t *RWTransaction) branch(stack []elem) *branch {
 	// Otherwise create a branch and cache it.
 	b := &branch{}
 	b.read(t.page(id))
+	b.depth = len(stack) - 1
 	b.parent = t.branch(stack[:len(stack)-1])
 	t.branches[id] = b