Add LFS Migration and Mirror (#14726)

* Implemented LFS client.

* Implemented scanning for pointer files.

* Implemented downloading of lfs files.

* Moved model-dependent code into services.

* Removed models dependency. Added TryReadPointerFromBuffer.

* Migrated code from service to module.

* Centralised storage creation.

* Removed dependency from models.

* Moved ContentStore into modules.

* Share structs between server and client.

* Moved method to services.

* Implemented lfs download on clone.

* Implemented LFS sync on clone and mirror update.

* Added form fields.

* Updated templates.

* Fixed condition.

* Use alternate endpoint.

* Added missing methods.

* Fixed typo and make linter happy.

* Detached pointer parser from gogit dependency.

* Fixed TestGetLFSRange test.

* Added context to support cancellation.

* Use ReadFull to probably read more data.

* Removed duplicated code from models.

* Moved scan implementation into pointer_scanner_nogogit.

* Changed method name.

* Added comments.

* Added more/specific log/error messages.

* Embedded lfs.Pointer into models.LFSMetaObject.

* Moved code from models to module.

* Moved code from models to module.

* Moved code from models to module.

* Reduced pointer usage.

* Embedded type.

* Use promoted fields.

* Fixed unexpected eof.

* Added unit tests.

* Implemented migration of local file paths.

* Show an error on invalid LFS endpoints.

* Hide settings if not used.

* Added LFS info to mirror struct.

* Fixed comment.

* Check LFS endpoint.

* Manage LFS settings from mirror page.

* Fixed selector.

* Adjusted selector.

* Added more tests.

* Added local filesystem migration test.

* Fixed typo.

* Reset settings.

* Added special windows path handling.

* Added unit test for HTTPClient.

* Added unit test for BasicTransferAdapter.

* Moved into util package.

* Test if LFS endpoint is allowed.

* Added support for git://

* Just use a static placeholder as the displayed url may be invalid.

* Reverted to original code.

* Added "Advanced Settings".

* Updated wording.

* Added discovery info link.

* Implemented suggestion.

* Fixed missing format parameter.

* Added Pointer.IsValid().

* Always remove model on error.

* Added suggestions.

* Use channel instead of array.

* Update routers/repo/migrate.go

* fmt

Signed-off-by: Andrew Thornton <art27@cantab.net>

Co-authored-by: zeripath <art27@cantab.net>

2 files changed, 29 insertions, 36 deletions

diff --git a/modules/repofiles/update.go b/modules/repofiles/update.go
index d25e109b29..ad984c465a 100644
--- a/modules/repofiles/update.go
+++ b/modules/repofiles/update.go
@@ -18,7 +18,6 @@ import (
 	"code.gitea.io/gitea/modules/log"
 	repo_module "code.gitea.io/gitea/modules/repository"
 	"code.gitea.io/gitea/modules/setting"
-	"code.gitea.io/gitea/modules/storage"
 	"code.gitea.io/gitea/modules/structs"
 
 	stdcharset "golang.org/x/net/html/charset"
@@ -70,30 +69,29 @@ func detectEncodingAndBOM(entry *git.TreeEntry, repo *models.Repository) (string
 	buf = buf[:n]
 
 	if setting.LFS.StartServer {
-		meta := lfs.IsPointerFile(&buf)
-		if meta != nil {
-			meta, err = repo.GetLFSMetaObjectByOid(meta.Oid)
+		pointer, _ := lfs.ReadPointerFromBuffer(buf)
+		if pointer.IsValid() {
+			meta, err := repo.GetLFSMetaObjectByOid(pointer.Oid)
 			if err != nil && err != models.ErrLFSObjectNotExist {
 				// return default
 				return "UTF-8", false
 			}
-		}
-		if meta != nil {
-			dataRc, err := lfs.ReadMetaObject(meta)
-			if err != nil {
-				// return default
-				return "UTF-8", false
-			}
-			defer dataRc.Close()
-			buf = make([]byte, 1024)
-			n, err = dataRc.Read(buf)
-			if err != nil {
-				// return default
-				return "UTF-8", false
+			if meta != nil {
+				dataRc, err := lfs.ReadMetaObject(pointer)
+				if err != nil {
+					// return default
+					return "UTF-8", false
+				}
+				defer dataRc.Close()
+				buf = make([]byte, 1024)
+				n, err = dataRc.Read(buf)
+				if err != nil {
+					// return default
+					return "UTF-8", false
+				}
+				buf = buf[:n]
 			}
-			buf = buf[:n]
 		}
-
 	}
 
 	encoding, err := charset.DetectEncoding(buf)
@@ -387,12 +385,12 @@ func CreateOrUpdateRepoFile(repo *models.Repository, doer *models.User, opts *Up
 
 		if filename2attribute2info[treePath] != nil && filename2attribute2info[treePath]["filter"] == "lfs" {
 			// OK so we are supposed to LFS this data!
-			oid, err := models.GenerateLFSOid(strings.NewReader(opts.Content))
+			pointer, err := lfs.GeneratePointer(strings.NewReader(opts.Content))
 			if err != nil {
 				return nil, err
 			}
-			lfsMetaObject = &models.LFSMetaObject{Oid: oid, Size: int64(len(opts.Content)), RepositoryID: repo.ID}
-			content = lfsMetaObject.Pointer()
+			lfsMetaObject = &models.LFSMetaObject{Pointer: pointer, RepositoryID: repo.ID}
+			content = pointer.StringContent()
 		}
 	}
 	// Add the object to the database
@@ -435,13 +433,13 @@ func CreateOrUpdateRepoFile(repo *models.Repository, doer *models.User, opts *Up
 		if err != nil {
 			return nil, err
 		}
-		contentStore := &lfs.ContentStore{ObjectStorage: storage.LFS}
-		exist, err := contentStore.Exists(lfsMetaObject)
+		contentStore := lfs.NewContentStore()
+		exist, err := contentStore.Exists(lfsMetaObject.Pointer)
 		if err != nil {
 			return nil, err
 		}
 		if !exist {
-			if err := contentStore.Put(lfsMetaObject, strings.NewReader(opts.Content)); err != nil {
+			if err := contentStore.Put(lfsMetaObject.Pointer, strings.NewReader(opts.Content)); err != nil {
 				if _, err2 := repo.RemoveLFSMetaObjectByOid(lfsMetaObject.Oid); err2 != nil {
 					return nil, fmt.Errorf("Error whilst removing failed inserted LFS object %s: %v (Prev Error: %v)", lfsMetaObject.Oid, err2, err)
 				}
diff --git a/modules/repofiles/upload.go b/modules/repofiles/upload.go
index 8716e1c8f1..e97f55a656 100644
--- a/modules/repofiles/upload.go
+++ b/modules/repofiles/upload.go
@@ -14,7 +14,6 @@ import (
 	"code.gitea.io/gitea/modules/git"
 	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/setting"
-	"code.gitea.io/gitea/modules/storage"
 )
 
 // UploadRepoFileOptions contains the uploaded repository file options
@@ -137,7 +136,7 @@ func UploadRepoFiles(repo *models.Repository, doer *models.User, opts *UploadRep
 
 	// OK now we can insert the data into the store - there's no way to clean up the store
 	// once it's in there, it's in there.
-	contentStore := &lfs.ContentStore{ObjectStorage: storage.LFS}
+	contentStore := lfs.NewContentStore()
 	for _, info := range infos {
 		if err := uploadToLFSContentStore(info, contentStore); err != nil {
 			return cleanUpAfterFailure(&infos, t, err)
@@ -163,18 +162,14 @@ func copyUploadedLFSFileIntoRepository(info *uploadInfo, filename2attribute2info
 	if setting.LFS.StartServer && filename2attribute2info[info.upload.Name] != nil && filename2attribute2info[info.upload.Name]["filter"] == "lfs" {
 		// Handle LFS
 		// FIXME: Inefficient! this should probably happen in models.Upload
-		oid, err := models.GenerateLFSOid(file)
-		if err != nil {
-			return err
-		}
-		fileInfo, err := file.Stat()
+		pointer, err := lfs.GeneratePointer(file)
 		if err != nil {
 			return err
 		}
 
-		info.lfsMetaObject = &models.LFSMetaObject{Oid: oid, Size: fileInfo.Size(), RepositoryID: t.repo.ID}
+		info.lfsMetaObject = &models.LFSMetaObject{Pointer: pointer, RepositoryID: t.repo.ID}
 
-		if objectHash, err = t.HashObject(strings.NewReader(info.lfsMetaObject.Pointer())); err != nil {
+		if objectHash, err = t.HashObject(strings.NewReader(pointer.StringContent())); err != nil {
 			return err
 		}
 	} else if objectHash, err = t.HashObject(file); err != nil {
@@ -189,7 +184,7 @@ func uploadToLFSContentStore(info uploadInfo, contentStore *lfs.ContentStore) er
 	if info.lfsMetaObject == nil {
 		return nil
 	}
-	exist, err := contentStore.Exists(info.lfsMetaObject)
+	exist, err := contentStore.Exists(info.lfsMetaObject.Pointer)
 	if err != nil {
 		return err
 	}
@@ -202,7 +197,7 @@ func uploadToLFSContentStore(info uploadInfo, contentStore *lfs.ContentStore) er
 		defer file.Close()
 		// FIXME: Put regenerates the hash and copies the file over.
 		// I guess this strictly ensures the soundness of the store but this is inefficient.
-		if err := contentStore.Put(info.lfsMetaObject, file); err != nil {
+		if err := contentStore.Put(info.lfsMetaObject.Pointer, file); err != nil {
 			// OK Now we need to cleanup
 			// Can't clean up the store, once uploaded there they're there.
 			return err