plumbing/cache: change FIFO to LRU cache

4 files changed, 94 insertions, 91 deletions

diff --git a/plumbing/cache/object.go b/plumbing/cache/object.go
deleted file mode 100644
index 44e0d32..0000000
--- a/plumbing/cache/object.go
+++ /dev/null
@@ -1,73 +0,0 @@
-package cache
-
-import (
-	"gopkg.in/src-d/go-git.v4/plumbing"
-)
-
-const (
-	initialQueueSize = 20
-	MaxSize          = 10 * MiByte
-)
-
-type ObjectFIFO struct {
-	objects map[plumbing.Hash]plumbing.EncodedObject
-	order   *queue
-
-	maxSize    FileSize
-	actualSize FileSize
-}
-
-// NewObjectFIFO returns an Object cache that keeps the newest objects that fit
-// into the specific memory size
-func NewObjectFIFO(size FileSize) *ObjectFIFO {
-	return &ObjectFIFO{
-		objects: make(map[plumbing.Hash]plumbing.EncodedObject),
-		order:   newQueue(initialQueueSize),
-		maxSize: size,
-	}
-}
-
-// Put puts a new object to the cache. If the object size is greater than the
-// cache size, the object is not added.
-func (c *ObjectFIFO) Put(o plumbing.EncodedObject) {
-	objSize := FileSize(o.Size())
-
-	// if the size of the object is bigger or equal than the cache size,
-	// skip it
-	if objSize >= c.maxSize {
-		return
-	}
-
-	// if the object is into the cache, do not add it again
-	if _, ok := c.objects[o.Hash()]; ok {
-		return
-	}
-
-	// delete the oldest object if cache is full
-	if c.actualSize >= c.maxSize {
-		h := c.order.Pop()
-		o := c.objects[h]
-		if o != nil {
-			c.actualSize -= FileSize(o.Size())
-			delete(c.objects, h)
-		}
-	}
-
-	c.objects[o.Hash()] = o
-	c.order.Push(o.Hash())
-	c.actualSize += FileSize(o.Size())
-}
-
-// Get returns an object by his hash. If the object is not found in the cache, it
-// returns nil
-func (c *ObjectFIFO) Get(k plumbing.Hash) (plumbing.EncodedObject, bool) {
-	obj, ok := c.objects[k]
-	return obj, ok
-}
-
-// Clear the content of this object cache
-func (c *ObjectFIFO) Clear() {
-	c.objects = make(map[plumbing.Hash]plumbing.EncodedObject)
-	c.order = newQueue(initialQueueSize)
-	c.actualSize = 0
-}
diff --git a/plumbing/cache/object_lru.go b/plumbing/cache/object_lru.go
new file mode 100644
index 0000000..e4c3160
--- /dev/null
+++ b/plumbing/cache/object_lru.go
@@ -0,0 +1,84 @@
+package cache
+
+import (
+	"container/list"
+
+	"gopkg.in/src-d/go-git.v4/plumbing"
+)
+
+// ObjectLRU implements an object cache with an LRU eviction policy and a
+// maximum size (measured in object size).
+type ObjectLRU struct {
+	MaxSize FileSize
+
+	actualSize FileSize
+	ll         *list.List
+	cache      map[interface{}]*list.Element
+}
+
+// NewObjectLRU creates a new ObjectLRU with the given maximum size. The maximum
+// size will never be exceeded.
+func NewObjectLRU(maxSize FileSize) *ObjectLRU {
+	return &ObjectLRU{MaxSize: maxSize}
+}
+
+// Put puts an object into the cache. If the object is already in the cache, it
+// will be marked as used. Otherwise, it will be inserted. A single object might
+// be evicted to make room for the new object.
+func (c *ObjectLRU) Put(obj plumbing.EncodedObject) {
+	if c.cache == nil {
+		c.actualSize = 0
+		c.cache = make(map[interface{}]*list.Element, 1000)
+		c.ll = list.New()
+	}
+
+	key := obj.Hash()
+	if ee, ok := c.cache[key]; ok {
+		c.ll.MoveToFront(ee)
+		ee.Value = obj
+		return
+	}
+
+	objSize := FileSize(obj.Size())
+
+	if objSize >= c.MaxSize {
+		return
+	}
+
+	if c.actualSize+objSize > c.MaxSize {
+		last := c.ll.Back()
+		lastObj := last.Value.(plumbing.EncodedObject)
+		lastSize := FileSize(lastObj.Size())
+
+		c.ll.Remove(last)
+		delete(c.cache, lastObj.Hash())
+		c.actualSize -= lastSize
+
+		if c.actualSize+objSize > c.MaxSize {
+			return
+		}
+	}
+
+	ee := c.ll.PushFront(obj)
+	c.cache[key] = ee
+	c.actualSize += objSize
+}
+
+// Get returns an object by its hash. It marks the object as used. If the object
+// is not in the cache, (nil, false) will be returned.
+func (c *ObjectLRU) Get(k plumbing.Hash) (plumbing.EncodedObject, bool) {
+	ee, ok := c.cache[k]
+	if !ok {
+		return nil, false
+	}
+
+	c.ll.MoveToFront(ee)
+	return ee.Value.(plumbing.EncodedObject), true
+}
+
+// Clear the content of this object cache.
+func (c *ObjectLRU) Clear() {
+	c.ll = nil
+	c.cache = nil
+	c.actualSize = 0
+}
diff --git a/plumbing/cache/object_test.go b/plumbing/cache/object_test.go
index 2a55acf..9359455 100644
--- a/plumbing/cache/object_test.go
+++ b/plumbing/cache/object_test.go
@@ -12,7 +12,7 @@ import (
 func Test(t *testing.T) { TestingT(t) }
 
 type ObjectSuite struct {
-	c       *ObjectFIFO
+	c       Object
 	aObject plumbing.EncodedObject
 	bObject plumbing.EncodedObject
 	cObject plumbing.EncodedObject
@@ -27,32 +27,26 @@ func (s *ObjectSuite) SetUpTest(c *C) {
 	s.cObject = newObject("cccccccccccccccccccccccccccccccccccccccc", 1*Byte)
 	s.dObject = newObject("dddddddddddddddddddddddddddddddddddddddd", 1*Byte)
 
-	s.c = NewObjectFIFO(2 * Byte)
+	s.c = NewObjectLRU(2 * Byte)
 }
 
-func (s *ObjectSuite) TestAdd_SameObject(c *C) {
+func (s *ObjectSuite) TestPutSameObject(c *C) {
 	s.c.Put(s.aObject)
-	c.Assert(s.c.actualSize, Equals, 1*Byte)
 	s.c.Put(s.aObject)
-	c.Assert(s.c.actualSize, Equals, 1*Byte)
+	_, ok := s.c.Get(s.aObject.Hash())
+	c.Assert(ok, Equals, true)
 }
 
-func (s *ObjectSuite) TestAdd_BigObject(c *C) {
+func (s *ObjectSuite) TestPutBigObject(c *C) {
 	s.c.Put(s.bObject)
-	c.Assert(s.c.actualSize, Equals, 0*Byte)
-	c.Assert(s.c.actualSize, Equals, 0*KiByte)
-	c.Assert(s.c.actualSize, Equals, 0*MiByte)
-	c.Assert(s.c.actualSize, Equals, 0*GiByte)
-	c.Assert(len(s.c.objects), Equals, 0)
+	_, ok := s.c.Get(s.aObject.Hash())
+	c.Assert(ok, Equals, false)
 }
 
-func (s *ObjectSuite) TestAdd_CacheOverflow(c *C) {
+func (s *ObjectSuite) TestPutCacheOverflow(c *C) {
 	s.c.Put(s.aObject)
-	c.Assert(s.c.actualSize, Equals, 1*Byte)
 	s.c.Put(s.cObject)
-	c.Assert(len(s.c.objects), Equals, 2)
 	s.c.Put(s.dObject)
-	c.Assert(len(s.c.objects), Equals, 2)
 
 	obj, ok := s.c.Get(s.aObject.Hash())
 	c.Assert(ok, Equals, false)
@@ -67,9 +61,7 @@ func (s *ObjectSuite) TestAdd_CacheOverflow(c *C) {
 
 func (s *ObjectSuite) TestClear(c *C) {
 	s.c.Put(s.aObject)
-	c.Assert(s.c.actualSize, Equals, 1*Byte)
 	s.c.Clear()
-	c.Assert(s.c.actualSize, Equals, 0*Byte)
 	obj, ok := s.c.Get(s.aObject.Hash())
 	c.Assert(ok, Equals, false)
 	c.Assert(obj, IsNil)
diff --git a/storage/filesystem/object.go b/storage/filesystem/object.go
index 4a67a00..6dd910b 100644
--- a/storage/filesystem/object.go
+++ b/storage/filesystem/object.go
@@ -29,7 +29,7 @@ type ObjectStorage struct {
 
 func newObjectStorage(dir *dotgit.DotGit) (ObjectStorage, error) {
 	s := ObjectStorage{
-		DeltaBaseCache: cache.NewObjectFIFO(DefaultMaxDeltaBaseCacheSize),
+		DeltaBaseCache: cache.NewObjectLRU(DefaultMaxDeltaBaseCacheSize),
 		dir:            dir,
 	}