aboutsummaryrefslogtreecommitdiffstats
path: root/storage
diff options
context:
space:
mode:
authorMáximo Cuadros <mcuadros@gmail.com>2016-09-09 20:18:08 +0200
committerMáximo Cuadros <mcuadros@gmail.com>2016-09-09 20:18:08 +0200
commit6fe1b93e96d4384f34f0562f81116ae565c6954d (patch)
tree408e677b936c4af75d236b98d07ff20b80bdb14f /storage
parent7c0ca21db6741af7735e3be332987e362393fb07 (diff)
downloadgo-git-6fe1b93e96d4384f34f0562f81116ae565c6954d.tar.gz
storage: filesystem iter implementation
Diffstat (limited to 'storage')
-rw-r--r--storage/filesystem/object.go145
-rw-r--r--storage/filesystem/object_test.go38
2 files changed, 169 insertions, 14 deletions
diff --git a/storage/filesystem/object.go b/storage/filesystem/object.go
index 1da24e5..379c359 100644
--- a/storage/filesystem/object.go
+++ b/storage/filesystem/object.go
@@ -11,6 +11,7 @@ import (
"gopkg.in/src-d/go-git.v4/formats/packfile"
"gopkg.in/src-d/go-git.v4/storage/filesystem/internal/dotgit"
"gopkg.in/src-d/go-git.v4/storage/memory"
+ "gopkg.in/src-d/go-git.v4/utils/fs"
)
// ObjectStorage is an implementation of core.ObjectStorage that stores
@@ -165,34 +166,50 @@ func (s *ObjectStorage) findObjectInPackfile(h core.Hash) (core.Hash, int64) {
// Iter returns an iterator for all the objects in the packfile with the
// given type.
func (s *ObjectStorage) Iter(t core.ObjectType) (core.ObjectIter, error) {
- var objects []core.Object
+ objects, err := s.dir.Objects()
+ if err != nil {
+ return nil, err
+ }
+
+ seen := make(map[core.Hash]bool, 0)
+ var iters []core.ObjectIter
+ if len(objects) != 0 {
+ iters = append(iters, &objectsIter{s: s, t: t, h: objects})
+ seen = hashListAsMap(objects)
+ }
- hashes, err := s.dir.Objects()
+ packi, err := s.buildPackfileIters(t, seen)
if err != nil {
return nil, err
}
- for _, hash := range hashes {
- object, err := s.getFromUnpacked(hash)
+ iters = append(iters, packi...)
+ return core.NewMultiObjectIter(iters), nil
+}
+
+func (s *ObjectStorage) buildPackfileIters(
+ t core.ObjectType, seen map[core.Hash]bool) ([]core.ObjectIter, error) {
+ packs, err := s.dir.ObjectPacks()
+ if err != nil {
+ return nil, err
+ }
+
+ var iters []core.ObjectIter
+ for _, h := range packs {
+ pack, err := s.dir.ObjectPack(h)
if err != nil {
return nil, err
}
- if object.Type() == t {
- objects = append(objects, object)
- }
- }
- for hash := range s.index {
- object, err := s.getFromPackfile(hash)
+ iter, err := newPackfileIter(pack, t, seen)
if err != nil {
return nil, err
}
- if t == core.AnyObject || object.Type() == t {
- objects = append(objects, object)
- }
+
+ iters = append(iters, iter)
}
- return core.NewObjectSliceIter(objects), nil
+ return iters, nil
}
func (o *ObjectStorage) Begin() core.TxObjectStorage {
@@ -233,3 +250,103 @@ func (i index) Decode(r io.Reader) error {
return nil
}
+
+type packfileIter struct {
+ f fs.File
+ d *packfile.Decoder
+ t core.ObjectType
+
+ seen map[core.Hash]bool
+ position uint32
+ total uint32
+}
+
+func newPackfileIter(
+ f fs.File,
+ t core.ObjectType,
+ seen map[core.Hash]bool,
+) (core.ObjectIter, error) {
+ s := packfile.NewScanner(f)
+ _, total, err := s.Header()
+ if err != nil {
+ return nil, err
+ }
+
+ d := packfile.NewDecoder(s, memory.NewStorage().ObjectStorage())
+ return &packfileIter{f: f, d: d, t: t, total: total, seen: seen}, nil
+}
+
+func (iter *packfileIter) Next() (core.Object, error) {
+ if iter.position >= iter.total {
+ return nil, io.EOF
+ }
+
+ obj, err := iter.d.ReadObject()
+ if err != nil {
+ return nil, err
+ }
+
+ iter.position++
+ if iter.seen[obj.Hash()] {
+ return iter.Next()
+ }
+
+ if iter.t != core.AnyObject && iter.t != obj.Type() {
+ return iter.Next()
+ }
+
+ return obj, nil
+}
+
+// ForEach is never called since is used inside of a MultiObjectIterator
+func (iter *packfileIter) ForEach(cb func(core.Object) error) error {
+ return nil
+}
+
+func (iter *packfileIter) Close() {
+ iter.f.Close()
+ iter.d.Close()
+}
+
+type objectsIter struct {
+ s *ObjectStorage
+ t core.ObjectType
+ h []core.Hash
+}
+
+func (iter *objectsIter) Next() (core.Object, error) {
+ if len(iter.h) == 0 {
+ return nil, io.EOF
+ }
+
+ obj, err := iter.s.getFromUnpacked(iter.h[0])
+ iter.h = iter.h[1:]
+
+ if err != nil {
+ return nil, err
+ }
+
+ if iter.t != core.AnyObject && iter.t != obj.Type() {
+ return iter.Next()
+ }
+
+ return obj, err
+}
+
+// ForEach is never called since is used inside of a MultiObjectIterator
+func (iter *objectsIter) ForEach(cb func(core.Object) error) error {
+ return nil
+}
+
+func (iter *objectsIter) Close() {
+ iter.h = []core.Hash{}
+}
+
+func hashListAsMap(l []core.Hash) map[core.Hash]bool {
+ m := make(map[core.Hash]bool, len(l))
+ for _, h := range l {
+ m[h] = true
+ }
+
+ return m
+}
diff --git a/storage/filesystem/object_test.go b/storage/filesystem/object_test.go
index 99266ab..14c77e4 100644
--- a/storage/filesystem/object_test.go
+++ b/storage/filesystem/object_test.go
@@ -53,3 +53,41 @@ func (s *FsSuite) TestGetFromPackfileMultiplePackfiles(c *C) {
c.Assert(err, IsNil)
c.Assert(obj.Hash(), Equals, expected)
}
+
+func (s *FsSuite) TestIter(c *C) {
+ fixtures.ByTag(".git").Test(c, func(f *fixtures.Fixture) {
+ fs := f.DotGit()
+ o, err := newObjectStorage(dotgit.New(fs))
+ c.Assert(err, IsNil)
+
+ iter, err := o.Iter(core.AnyObject)
+ c.Assert(err, IsNil)
+
+ var count int32
+ err = iter.ForEach(func(o core.Object) error {
+ count++
+ return nil
+ })
+
+ c.Assert(err, IsNil)
+ c.Assert(count, Equals, f.ObjectsCount)
+ })
+}
+
+func (s *FsSuite) TestIterWithType(c *C) {
+ fixtures.ByTag(".git").Test(c, func(f *fixtures.Fixture) {
+ fs := f.DotGit()
+ o, err := newObjectStorage(dotgit.New(fs))
+ c.Assert(err, IsNil)
+
+ iter, err := o.Iter(core.CommitObject)
+ c.Assert(err, IsNil)
+
+ err = iter.ForEach(func(o core.Object) error {
+ c.Assert(o.Type(), Equals, core.CommitObject)
+ return nil
+ })
+
+ c.Assert(err, IsNil)
+ })
+}