storage/repository: add new functions for garbage collection

object_walker.go

@@ -0,0 +1,105 @@
+package git
+
+import (
+	"fmt"
+
+	"gopkg.in/src-d/go-git.v4/plumbing"
+	"gopkg.in/src-d/go-git.v4/plumbing/filemode"
+	"gopkg.in/src-d/go-git.v4/plumbing/object"
+	"gopkg.in/src-d/go-git.v4/storage"
+)
+
+type objectWalker struct {
+	Storer storage.Storer
+	// seen is the set of objects seen in the repo.
+	// seen map can become huge if walking over large
+	// repos. Thus using struct{} as the value type.
+	seen map[plumbing.Hash]struct{}
+}
+
+func newObjectWalker(s storage.Storer) *objectWalker {
+	return &objectWalker{s, map[plumbing.Hash]struct{}{}}
+}
+
+// walkAllRefs walks all (hash) refererences from the repo.
+func (p *objectWalker) walkAllRefs() error {
+	// Walk over all the references in the repo.
+	it, err := p.Storer.IterReferences()
+	if err != nil {
+		return err
+	}
+	defer it.Close()
+	err = it.ForEach(func(ref *plumbing.Reference) error {
+		// Exit this iteration early for non-hash references.
+		if ref.Type() != plumbing.HashReference {
+			return nil
+		}
+		return p.walkObjectTree(ref.Hash())
+	})
+	if err != nil {
+		return err
+	}
+	return nil
+}
+
+func (p *objectWalker) isSeen(hash plumbing.Hash) bool {
+	_, seen := p.seen[hash]
+	return seen
+}
+
+func (p *objectWalker) add(hash plumbing.Hash) {
+	p.seen[hash] = struct{}{}
+}
+
+// walkObjectTree walks over all objects and remembers references
+// to them in the objectWalker. This is used instead of the revlist
+// walks because memory usage is tight with huge repos.
+func (p *objectWalker) walkObjectTree(hash plumbing.Hash) error {
+	// Check if we have already seen, and mark this object
+	if p.isSeen(hash) {
+		return nil
+	}
+	p.add(hash)
+	// Fetch the object.
+	obj, err := object.GetObject(p.Storer, hash)
+	if err != nil {
+		return fmt.Errorf("Getting object %s failed: %v", hash, err)
+	}
+	// Walk all children depending on object type.
+	switch obj := obj.(type) {
+	case *object.Commit:
+		err = p.walkObjectTree(obj.TreeHash)
+		if err != nil {
+			return err
+		}
+		for _, h := range obj.ParentHashes {
+			err = p.walkObjectTree(h)
+			if err != nil {
+				return err
+			}
+		}
+	case *object.Tree:
+		for i := range obj.Entries {
+			// Shortcut for blob objects:
+			// 'or' the lower bits of a mode and check that it
+			// it matches a filemode.Executable. The type information
+			// is in the higher bits, but this is the cleanest way
+			// to handle plain files with different modes.
+			// Other non-tree objects are somewhat rare, so they
+			// are not special-cased.
+			if obj.Entries[i].Mode|0755 == filemode.Executable {
+				p.add(obj.Entries[i].Hash)
+				continue
+			}
+			// Normal walk for sub-trees (and symlinks etc).
+			err = p.walkObjectTree(obj.Entries[i].Hash)
+			if err != nil {
+				return err
+			}
+		}
+	default:
+		// Error out on unhandled object types.
+		return fmt.Errorf("Unknown object %X %s %T\n", obj.ID(), obj.Type(), obj)
+	}
+	return nil
+}

plumbing/storer/object.go

@@ -3,6 +3,7 @@ package storer
 import (
 	"errors"
 	"io"
+	"time"
 
 	"gopkg.in/src-d/go-git.v4/plumbing"
 )
@@ -36,6 +37,9 @@ type EncodedObjectStorer interface {
 	//
 	// Valid plumbing.ObjectType values are CommitObject, BlobObject, TagObject,
 	IterEncodedObjects(plumbing.ObjectType) (EncodedObjectIter, error)
+	// HasEncodedObject returns ErrObjNotFound if the object doesn't
+	// exist.  If the object does exist, it returns nil.
+	HasEncodedObject(plumbing.Hash) error
 }
 
 // DeltaObjectStorer is an EncodedObjectStorer that can return delta
@@ -53,6 +57,34 @@ type Transactioner interface {
 	Begin() Transaction
 }
 
+// LooseObjectStorer is an optional interface for managing "loose"
+// objects, i.e. those not in packfiles.
+type LooseObjectStorer interface {
+	// ForEachObjectHash iterates over all the (loose) object hashes
+	// in the repository without necessarily having to read those objects.
+	// Objects only inside pack files may be omitted.
+	// If ErrStop is sent the iteration is stop but no error is returned.
+	ForEachObjectHash(func(plumbing.Hash) error) error
+	// LooseObjectTime looks up the (m)time associated with the
+	// loose object (that is not in a pack file). Some
+	// implementations (e.g. without loose objects)
+	// always return an error.
+	LooseObjectTime(plumbing.Hash) (time.Time, error)
+	// DeleteLooseObject deletes a loose object if it exists.
+	DeleteLooseObject(plumbing.Hash) error
+}
+
+// PackedObjectStorer is an optional interface for managing objects in
+// packfiles.
+type PackedObjectStorer interface {
+	// ObjectPacks returns hashes of object packs if the underlying
+	// implementation has pack files.
+	ObjectPacks() ([]plumbing.Hash, error)
+	// DeleteOldObjectPackAndIndex deletes an object pack and the corresponding index file if they exist.
+	// Deletion is only performed if the pack is older than the supplied time (or the time is zero).
+	DeleteOldObjectPackAndIndex(plumbing.Hash, time.Time) error
+}
+
 // PackfileWriter is a optional method for ObjectStorer, it enable direct write
 // of packfile to the storage
 type PackfileWriter interface {

plumbing/storer/object_test.go

@@ -132,6 +132,15 @@ func (o *MockObjectStorage) SetEncodedObject(obj plumbing.EncodedObject) (plumbi
 	return plumbing.ZeroHash, nil
 }
 
+func (o *MockObjectStorage) HasEncodedObject(h plumbing.Hash) error {
+	for _, o := range o.db {
+		if o.Hash() == h {
+			return nil
+		}
+	}
+	return plumbing.ErrObjectNotFound
+}
+
 func (o *MockObjectStorage) EncodedObject(t plumbing.ObjectType, h plumbing.Hash) (plumbing.EncodedObject, error) {
 	for _, o := range o.db {
 		if o.Hash() == h {

plumbing/storer/reference.go

@@ -24,6 +24,8 @@ type ReferenceStorer interface {
 	Reference(plumbing.ReferenceName) (*plumbing.Reference, error)
 	IterReferences() (ReferenceIter, error)
 	RemoveReference(plumbing.ReferenceName) error
+	CountLooseRefs() (int, error)
+	PackRefs() error
 }
 
 // ReferenceIter is a generic closable interface for iterating over references.

prune.go

@@ -0,0 +1,66 @@
+package git
+
+import (
+	"errors"
+	"time"
+
+	"gopkg.in/src-d/go-git.v4/plumbing"
+	"gopkg.in/src-d/go-git.v4/plumbing/storer"
+)
+
+type PruneHandler func(unreferencedObjectHash plumbing.Hash) error
+type PruneOptions struct {
+	// OnlyObjectsOlderThan if set to non-zero value
+	// selects only objects older than the time provided.
+	OnlyObjectsOlderThan time.Time
+	// Handler is called on matching objects
+	Handler PruneHandler
+}
+
+var ErrLooseObjectsNotSupported = errors.New("Loose objects not supported")
+
+// DeleteObject deletes an object from a repository.
+// The type conveniently matches PruneHandler.
+func (r *Repository) DeleteObject(hash plumbing.Hash) error {
+	los, ok := r.Storer.(storer.LooseObjectStorer)
+	if !ok {
+		return ErrLooseObjectsNotSupported
+	}
+
+	return los.DeleteLooseObject(hash)
+}
+
+func (r *Repository) Prune(opt PruneOptions) error {
+	los, ok := r.Storer.(storer.LooseObjectStorer)
+	if !ok {
+		return ErrLooseObjectsNotSupported
+	}
+
+	pw := newObjectWalker(r.Storer)
+	err := pw.walkAllRefs()
+	if err != nil {
+		return err
+	}
+	// Now walk all (loose) objects in storage.
+	return los.ForEachObjectHash(func(hash plumbing.Hash) error {
+		// Get out if we have seen this object.
+		if pw.isSeen(hash) {
+			return nil
+		}
+		// Otherwise it is a candidate for pruning.
+		// Check out for too new objects next.
+		if opt.OnlyObjectsOlderThan != (time.Time{}) {
+			// Errors here are non-fatal. The object may be e.g. packed.
+			// Or concurrently deleted. Skip such objects.
+			t, err := los.LooseObjectTime(hash)
+			if err != nil {
+				return nil
+			}
+			// Skip too new objects.
+			if !t.Before(opt.OnlyObjectsOlderThan) {
+				return nil
+			}
+		}
+		return opt.Handler(hash)
+	})
+}

prune_test.go

@@ -0,0 +1,73 @@
+package git
+
+import (
+	"time"
+
+	"gopkg.in/src-d/go-git.v4/plumbing"
+	"gopkg.in/src-d/go-git.v4/plumbing/storer"
+	"gopkg.in/src-d/go-git.v4/storage"
+	"gopkg.in/src-d/go-git.v4/storage/filesystem"
+
+	. "gopkg.in/check.v1"
+	"gopkg.in/src-d/go-git-fixtures.v3"
+)
+
+type PruneSuite struct {
+	BaseSuite
+}
+
+var _ = Suite(&PruneSuite{})
+
+func (s *PruneSuite) testPrune(c *C, deleteTime time.Time) {
+	srcFs := fixtures.ByTag("unpacked").One().DotGit()
+	var sto storage.Storer
+	var err error
+	sto, err = filesystem.NewStorage(srcFs)
+	c.Assert(err, IsNil)
+
+	los := sto.(storer.LooseObjectStorer)
+	c.Assert(los, NotNil)
+
+	count := 0
+	err = los.ForEachObjectHash(func(_ plumbing.Hash) error {
+		count++
+		return nil
+	})
+	c.Assert(err, IsNil)
+
+	r, err := Open(sto, srcFs)
+	c.Assert(err, IsNil)
+	c.Assert(r, NotNil)
+
+	// Remove a branch so we can prune some objects.
+	err = sto.RemoveReference(plumbing.ReferenceName("refs/heads/v4"))
+	c.Assert(err, IsNil)
+	err = sto.RemoveReference(plumbing.ReferenceName("refs/remotes/origin/v4"))
+	c.Assert(err, IsNil)
+
+	err = r.Prune(PruneOptions{
+		OnlyObjectsOlderThan: deleteTime,
+		Handler:              r.DeleteObject,
+	})
+	c.Assert(err, IsNil)
+
+	newCount := 0
+	err = los.ForEachObjectHash(func(_ plumbing.Hash) error {
+		newCount++
+		return nil
+	})
+	if deleteTime.IsZero() {
+		c.Assert(newCount < count, Equals, true)
+	} else {
+		// Assume a delete time older than any of the objects was passed in.
+		c.Assert(newCount, Equals, count)
+	}
+}
+
+func (s *PruneSuite) TestPrune(c *C) {
+	s.testPrune(c, time.Time{})
+}
+
+func (s *PruneSuite) TestPruneWithNoDelete(c *C) {
+	s.testPrune(c, time.Unix(0, 1))
+}