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>

cmd/serv.go

@@ -17,11 +17,11 @@ import (
 	"time"
 
 	"code.gitea.io/gitea/models"
-	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/pprof"
 	"code.gitea.io/gitea/modules/private"
 	"code.gitea.io/gitea/modules/setting"
+	"code.gitea.io/gitea/services/lfs"
 
 	"github.com/dgrijalva/jwt-go"
 	jsoniter "github.com/json-iterator/go"

integrations/api_repo_lfs_migrate_test.go

@@ -0,0 +1,49 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package integrations
+
+import (
+	"net/http"
+	"path"
+	"testing"
+
+	"code.gitea.io/gitea/models"
+	"code.gitea.io/gitea/modules/lfs"
+	"code.gitea.io/gitea/modules/setting"
+	api "code.gitea.io/gitea/modules/structs"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestAPIRepoLFSMigrateLocal(t *testing.T) {
+	defer prepareTestEnv(t)()
+
+	oldImportLocalPaths := setting.ImportLocalPaths
+	oldAllowLocalNetworks := setting.Migrations.AllowLocalNetworks
+	setting.ImportLocalPaths = true
+	setting.Migrations.AllowLocalNetworks = true
+
+	user := models.AssertExistsAndLoadBean(t, &models.User{ID: 1}).(*models.User)
+	session := loginUser(t, user.Name)
+	token := getTokenForLoggedInUser(t, session)
+
+	req := NewRequestWithJSON(t, "POST", "/api/v1/repos/migrate?token="+token, &api.MigrateRepoOptions{
+		CloneAddr:   path.Join(setting.RepoRootPath, "migration/lfs-test.git"),
+		RepoOwnerID: user.ID,
+		RepoName:    "lfs-test-local",
+		LFS:         true,
+	})
+	resp := MakeRequest(t, req, NoExpectedStatus)
+	assert.EqualValues(t, http.StatusCreated, resp.Code)
+
+	store := lfs.NewContentStore()
+	ok, _ := store.Verify(lfs.Pointer{Oid: "fb8f7d8435968c4f82a726a92395be4d16f2f63116caf36c8ad35c60831ab041", Size: 6})
+	assert.True(t, ok)
+	ok, _ = store.Verify(lfs.Pointer{Oid: "d6f175817f886ec6fbbc1515326465fa96c3bfd54a4ea06cfd6dbbd8340e0152", Size: 6})
+	assert.True(t, ok)
+
+	setting.ImportLocalPaths = oldImportLocalPaths
+	setting.Migrations.AllowLocalNetworks = oldAllowLocalNetworks
+}

integrations/git_test.go

@@ -18,6 +18,7 @@ import (
 
 	"code.gitea.io/gitea/models"
 	"code.gitea.io/gitea/modules/git"
+	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/setting"
 	api "code.gitea.io/gitea/modules/structs"
 	"code.gitea.io/gitea/modules/util"
@@ -218,7 +219,7 @@ func rawTest(t *testing.T, ctx *APITestContext, little, big, littleLFS, bigLFS s
 			assert.NotEqual(t, littleSize, resp.Body.Len())
 			assert.LessOrEqual(t, resp.Body.Len(), 1024)
 			if resp.Body.Len() != littleSize && resp.Body.Len() <= 1024 {
-				assert.Contains(t, resp.Body.String(), models.LFSMetaFileIdentifier)
+				assert.Contains(t, resp.Body.String(), lfs.MetaFileIdentifier)
 			}
 		}
 
@@ -232,7 +233,7 @@ func rawTest(t *testing.T, ctx *APITestContext, little, big, littleLFS, bigLFS s
 				resp := session.MakeRequest(t, req, http.StatusOK)
 				assert.NotEqual(t, bigSize, resp.Body.Len())
 				if resp.Body.Len() != bigSize && resp.Body.Len() <= 1024 {
-					assert.Contains(t, resp.Body.String(), models.LFSMetaFileIdentifier)
+					assert.Contains(t, resp.Body.String(), lfs.MetaFileIdentifier)
 				}
 			}
 		}

integrations/gitea-repositories-meta/migration/lfs-test.git/HEAD

@@ -0,0 +1 @@
+ref: refs/heads/master

integrations/gitea-repositories-meta/migration/lfs-test.git/config

@@ -0,0 +1,7 @@
+[core]
+	bare = false
+	repositoryformatversion = 0
+	filemode = false
+	symlinks = false
+	ignorecase = true
+	logallrefupdates = true

integrations/gitea-repositories-meta/migration/lfs-test.git/description

@@ -0,0 +1 @@
+Unnamed repository; edit this file 'description' to name the repository.

integrations/gitea-repositories-meta/migration/lfs-test.git/hooks/post-checkout

@@ -0,0 +1,3 @@
+#!/bin/sh
+command -v git-lfs >/dev/null 2>&1 || { echo >&2 "\nThis repository is configured for Git LFS but 'git-lfs' was not found on your path. If you no longer wish to use Git LFS, remove this hook by deleting .git/hooks/post-checkout.\n"; exit 2; }
+git lfs post-checkout "$@"

integrations/gitea-repositories-meta/migration/lfs-test.git/hooks/post-commit

@@ -0,0 +1,3 @@
+#!/bin/sh
+command -v git-lfs >/dev/null 2>&1 || { echo >&2 "\nThis repository is configured for Git LFS but 'git-lfs' was not found on your path. If you no longer wish to use Git LFS, remove this hook by deleting .git/hooks/post-commit.\n"; exit 2; }
+git lfs post-commit "$@"

integrations/gitea-repositories-meta/migration/lfs-test.git/hooks/post-merge

@@ -0,0 +1,3 @@
+#!/bin/sh
+command -v git-lfs >/dev/null 2>&1 || { echo >&2 "\nThis repository is configured for Git LFS but 'git-lfs' was not found on your path. If you no longer wish to use Git LFS, remove this hook by deleting .git/hooks/post-merge.\n"; exit 2; }
+git lfs post-merge "$@"

integrations/gitea-repositories-meta/migration/lfs-test.git/hooks/pre-push

@@ -0,0 +1,3 @@
+#!/bin/sh
+command -v git-lfs >/dev/null 2>&1 || { echo >&2 "\nThis repository is configured for Git LFS but 'git-lfs' was not found on your path. If you no longer wish to use Git LFS, remove this hook by deleting .git/hooks/pre-push.\n"; exit 2; }
+git lfs pre-push "$@"

integrations/gitea-repositories-meta/migration/lfs-test.git/lfs/objects/d6/f1/d6f175817f886ec6fbbc1515326465fa96c3bfd54a4ea06cfd6dbbd8340e0152

@@ -0,0 +1 @@
+dummy2

integrations/gitea-repositories-meta/migration/lfs-test.git/lfs/objects/fb/8f/fb8f7d8435968c4f82a726a92395be4d16f2f63116caf36c8ad35c60831ab041

@@ -0,0 +1 @@
+dummy1

integrations/gitea-repositories-meta/migration/lfs-test.git/refs/heads/master

@@ -0,0 +1 @@
+546244003622c64b2fc3c2cd544d7a29882c8383

integrations/lfs_getobject_test.go

@@ -7,9 +7,6 @@ package integrations
 import (
 	"archive/zip"
 	"bytes"
-	"crypto/sha256"
-	"encoding/hex"
-	"io"
 	"io/ioutil"
 	"net/http"
 	"net/http/httptest"
@@ -18,46 +15,36 @@ import (
 	"code.gitea.io/gitea/models"
 	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/setting"
-	"code.gitea.io/gitea/modules/storage"
 	"code.gitea.io/gitea/routers/routes"
 
 	gzipp "github.com/klauspost/compress/gzip"
 	"github.com/stretchr/testify/assert"
 )
 
-func GenerateLFSOid(content io.Reader) (string, error) {
-	h := sha256.New()
-	if _, err := io.Copy(h, content); err != nil {
-		return "", err
-	}
-	sum := h.Sum(nil)
-	return hex.EncodeToString(sum), nil
-}
-
 var lfsID = int64(20000)
 
 func storeObjectInRepo(t *testing.T, repositoryID int64, content *[]byte) string {
-	oid, err := GenerateLFSOid(bytes.NewReader(*content))
+	pointer, err := lfs.GeneratePointer(bytes.NewReader(*content))
 	assert.NoError(t, err)
 	var lfsMetaObject *models.LFSMetaObject
 
 	if setting.Database.UsePostgreSQL {
-		lfsMetaObject = &models.LFSMetaObject{ID: lfsID, Oid: oid, Size: int64(len(*content)), RepositoryID: repositoryID}
+		lfsMetaObject = &models.LFSMetaObject{ID: lfsID, Pointer: pointer, RepositoryID: repositoryID}
 	} else {
-		lfsMetaObject = &models.LFSMetaObject{Oid: oid, Size: int64(len(*content)), RepositoryID: repositoryID}
+		lfsMetaObject = &models.LFSMetaObject{Pointer: pointer, RepositoryID: repositoryID}
 	}
 
 	lfsID++
 	lfsMetaObject, err = models.NewLFSMetaObject(lfsMetaObject)
 	assert.NoError(t, err)
-	contentStore := &lfs.ContentStore{ObjectStorage: storage.LFS}
-	exist, err := contentStore.Exists(lfsMetaObject)
+	contentStore := lfs.NewContentStore()
+	exist, err := contentStore.Exists(pointer)
 	assert.NoError(t, err)
 	if !exist {
-		err := contentStore.Put(lfsMetaObject, bytes.NewReader(*content))
+		err := contentStore.Put(pointer, bytes.NewReader(*content))
 		assert.NoError(t, err)
 	}
-	return oid
+	return pointer.Oid
 }
 
 func storeAndGetLfs(t *testing.T, content *[]byte, extraHeader *http.Header, expectedStatus int) *httptest.ResponseRecorder {

integrations/lfs_local_endpoint_test.go

@@ -0,0 +1,117 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package integrations
+
+import (
+	"fmt"
+	"io/ioutil"
+	"net/url"
+	"os"
+	"path/filepath"
+	"testing"
+
+	"code.gitea.io/gitea/modules/lfs"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func str2url(raw string) *url.URL {
+	u, _ := url.Parse(raw)
+	return u
+}
+
+func TestDetermineLocalEndpoint(t *testing.T) {
+	defer prepareTestEnv(t)()
+
+	root, _ := ioutil.TempDir("", "lfs_test")
+	defer os.RemoveAll(root)
+
+	rootdotgit, _ := ioutil.TempDir("", "lfs_test")
+	defer os.RemoveAll(rootdotgit)
+	os.Mkdir(filepath.Join(rootdotgit, ".git"), 0700)
+
+	lfsroot, _ := ioutil.TempDir("", "lfs_test")
+	defer os.RemoveAll(lfsroot)
+
+	// Test cases
+	var cases = []struct {
+		cloneurl string
+		lfsurl   string
+		expected *url.URL
+	}{
+		// case 0
+		{
+			cloneurl: root,
+			lfsurl:   "",
+			expected: str2url(fmt.Sprintf("file://%s", root)),
+		},
+		// case 1
+		{
+			cloneurl: root,
+			lfsurl:   lfsroot,
+			expected: str2url(fmt.Sprintf("file://%s", lfsroot)),
+		},
+		// case 2
+		{
+			cloneurl: "https://git.com/repo.git",
+			lfsurl:   lfsroot,
+			expected: str2url(fmt.Sprintf("file://%s", lfsroot)),
+		},
+		// case 3
+		{
+			cloneurl: rootdotgit,
+			lfsurl:   "",
+			expected: str2url(fmt.Sprintf("file://%s", filepath.Join(rootdotgit, ".git"))),
+		},
+		// case 4
+		{
+			cloneurl: "",
+			lfsurl:   rootdotgit,
+			expected: str2url(fmt.Sprintf("file://%s", filepath.Join(rootdotgit, ".git"))),
+		},
+		// case 5
+		{
+			cloneurl: rootdotgit,
+			lfsurl:   rootdotgit,
+			expected: str2url(fmt.Sprintf("file://%s", filepath.Join(rootdotgit, ".git"))),
+		},
+		// case 6
+		{
+			cloneurl: fmt.Sprintf("file://%s", root),
+			lfsurl:   "",
+			expected: str2url(fmt.Sprintf("file://%s", root)),
+		},
+		// case 7
+		{
+			cloneurl: fmt.Sprintf("file://%s", root),
+			lfsurl:   fmt.Sprintf("file://%s", lfsroot),
+			expected: str2url(fmt.Sprintf("file://%s", lfsroot)),
+		},
+		// case 8
+		{
+			cloneurl: root,
+			lfsurl:   fmt.Sprintf("file://%s", lfsroot),
+			expected: str2url(fmt.Sprintf("file://%s", lfsroot)),
+		},
+		// case 9
+		{
+			cloneurl: "",
+			lfsurl:   "/does/not/exist",
+			expected: nil,
+		},
+		// case 10
+		{
+			cloneurl: "",
+			lfsurl:   "file:///does/not/exist",
+			expected: str2url("file:///does/not/exist"),
+		},
+	}
+
+	for n, c := range cases {
+		ep := lfs.DetermineEndpoint(c.cloneurl, c.lfsurl)
+
+		assert.Equal(t, c.expected, ep, "case %d: error should match", n)
+	}
+}

models/lfs.go

@@ -5,13 +5,9 @@
 package models
 
 import (
-	"crypto/sha256"
-	"encoding/hex"
 	"errors"
-	"fmt"
-	"io"
-	"path"
 
+	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/timeutil"
 
 	"xorm.io/builder"
@@ -20,27 +16,12 @@ import (
 // LFSMetaObject stores metadata for LFS tracked files.
 type LFSMetaObject struct {
 	ID           int64 `xorm:"pk autoincr"`
-	Oid          string             `xorm:"UNIQUE(s) INDEX NOT NULL"`
-	Size         int64              `xorm:"NOT NULL"`
+	lfs.Pointer  `xorm:"extends"`
 	RepositoryID int64              `xorm:"UNIQUE(s) INDEX NOT NULL"`
 	Existing     bool               `xorm:"-"`
 	CreatedUnix  timeutil.TimeStamp `xorm:"created"`
 }
 
-// RelativePath returns the relative path of the lfs object
-func (m *LFSMetaObject) RelativePath() string {
-	if len(m.Oid) < 5 {
-		return m.Oid
-	}
-
-	return path.Join(m.Oid[0:2], m.Oid[2:4], m.Oid[4:])
-}
-
-// Pointer returns the string representation of an LFS pointer file
-func (m *LFSMetaObject) Pointer() string {
-	return fmt.Sprintf("%s\n%s%s\nsize %d\n", LFSMetaFileIdentifier, LFSMetaFileOidPrefix, m.Oid, m.Size)
-}
-
 // LFSTokenResponse defines the JSON structure in which the JWT token is stored.
 // This structure is fetched via SSH and passed by the Git LFS client to the server
 // endpoint for authorization.
@@ -53,15 +34,6 @@ type LFSTokenResponse struct {
 // to differentiate between database and missing object errors.
 var ErrLFSObjectNotExist = errors.New("LFS Meta object does not exist")
 
-const (
-	// LFSMetaFileIdentifier is the string appearing at the first line of LFS pointer files.
-	// https://github.com/git-lfs/git-lfs/blob/master/docs/spec.md
-	LFSMetaFileIdentifier = "version https://git-lfs.github.com/spec/v1"
-
-	// LFSMetaFileOidPrefix appears in LFS pointer files on a line before the sha256 hash.
-	LFSMetaFileOidPrefix = "oid sha256:"
-)
-
 // NewLFSMetaObject stores a given populated LFSMetaObject structure in the database
 // if it is not already present.
 func NewLFSMetaObject(m *LFSMetaObject) (*LFSMetaObject, error) {
@@ -90,16 +62,6 @@ func NewLFSMetaObject(m *LFSMetaObject) (*LFSMetaObject, error) {
 	return m, sess.Commit()
 }
 
-// GenerateLFSOid generates a Sha256Sum to represent an oid for arbitrary content
-func GenerateLFSOid(content io.Reader) (string, error) {
-	h := sha256.New()
-	if _, err := io.Copy(h, content); err != nil {
-		return "", err
-	}
-	sum := h.Sum(nil)
-	return hex.EncodeToString(sum), nil
-}
-
 // GetLFSMetaObjectByOid selects a LFSMetaObject entry from database by its OID.
 // It may return ErrLFSObjectNotExist or a database error. If the error is nil,
 // the returned pointer is a valid LFSMetaObject.
@@ -108,7 +70,7 @@ func (repo *Repository) GetLFSMetaObjectByOid(oid string) (*LFSMetaObject, error
 		return nil, ErrLFSObjectNotExist
 	}
 
-	m := &LFSMetaObject{Oid: oid, RepositoryID: repo.ID}
+	m := &LFSMetaObject{Pointer: lfs.Pointer{Oid: oid}, RepositoryID: repo.ID}
 	has, err := x.Get(m)
 	if err != nil {
 		return nil, err
@@ -131,12 +93,12 @@ func (repo *Repository) RemoveLFSMetaObjectByOid(oid string) (int64, error) {
 		return -1, err
 	}
 
-	m := &LFSMetaObject{Oid: oid, RepositoryID: repo.ID}
+	m := &LFSMetaObject{Pointer: lfs.Pointer{Oid: oid}, RepositoryID: repo.ID}
 	if _, err := sess.Delete(m); err != nil {
 		return -1, err
 	}
 
-	count, err := sess.Count(&LFSMetaObject{Oid: oid})
+	count, err := sess.Count(&LFSMetaObject{Pointer: lfs.Pointer{Oid: oid}})
 	if err != nil {
 		return count, err
 	}
@@ -168,11 +130,11 @@ func (repo *Repository) CountLFSMetaObjects() (int64, error) {
 // LFSObjectAccessible checks if a provided Oid is accessible to the user
 func LFSObjectAccessible(user *User, oid string) (bool, error) {
 	if user.IsAdmin {
-		count, err := x.Count(&LFSMetaObject{Oid: oid})
+		count, err := x.Count(&LFSMetaObject{Pointer: lfs.Pointer{Oid: oid}})
 		return (count > 0), err
 	}
 	cond := accessibleRepositoryCondition(user)
-	count, err := x.Where(cond).Join("INNER", "repository", "`lfs_meta_object`.repository_id = `repository`.id").Count(&LFSMetaObject{Oid: oid})
+	count, err := x.Where(cond).Join("INNER", "repository", "`lfs_meta_object`.repository_id = `repository`.id").Count(&LFSMetaObject{Pointer: lfs.Pointer{Oid: oid}})
 	return (count > 0), err
 }
 

models/migrations/migrations.go

@@ -302,6 +302,8 @@ var migrations = []Migration{
 	NewMigration("Remove invalid labels from comments", removeInvalidLabels),
 	// v177 -> v178
 	NewMigration("Delete orphaned IssueLabels", deleteOrphanedIssueLabels),
+	// v178 -> v179
+	NewMigration("Add LFS columns to Mirror", addLFSMirrorColumns),
 }
 
 // GetCurrentDBVersion returns the current db version

models/migrations/v178.go

@@ -0,0 +1,18 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package migrations
+
+import (
+	"xorm.io/xorm"
+)
+
+func addLFSMirrorColumns(x *xorm.Engine) error {
+	type Mirror struct {
+		LFS         bool   `xorm:"lfs_enabled NOT NULL DEFAULT false"`
+		LFSEndpoint string `xorm:"lfs_endpoint TEXT"`
+	}
+
+	return x.Sync2(new(Mirror))
+}

models/repo.go

@@ -25,6 +25,7 @@ import (
 	"strings"
 	"time"
 
+	"code.gitea.io/gitea/modules/lfs"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/markup"
 	"code.gitea.io/gitea/modules/options"
@@ -1531,7 +1532,7 @@ func DeleteRepository(doer *User, uid, repoID int64) error {
 	}
 
 	for _, v := range lfsObjects {
-		count, err := sess.Count(&LFSMetaObject{Oid: v.Oid})
+		count, err := sess.Count(&LFSMetaObject{Pointer: lfs.Pointer{Oid: v.Oid}})
 		if err != nil {
 			return err
 		}

models/repo_mirror.go

@@ -25,6 +25,9 @@ type Mirror struct {
 	UpdatedUnix    timeutil.TimeStamp `xorm:"INDEX"`
 	NextUpdateUnix timeutil.TimeStamp `xorm:"INDEX"`
 
+	LFS         bool   `xorm:"lfs_enabled NOT NULL DEFAULT false"`
+	LFSEndpoint string `xorm:"lfs_endpoint TEXT"`
+
 	Address string `xorm:"-"`
 }
 

modules/lfs/client.go

@@ -0,0 +1,24 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package lfs
+
+import (
+	"context"
+	"io"
+	"net/url"
+)
+
+// Client is used to communicate with a LFS source
+type Client interface {
+	Download(ctx context.Context, oid string, size int64) (io.ReadCloser, error)
+}
+
+// NewClient creates a LFS client
+func NewClient(endpoint *url.URL) Client {
+	if endpoint.Scheme == "file" {
+		return newFilesystemClient(endpoint)
+	}
+	return newHTTPClient(endpoint)
+}

modules/lfs/client_test.go

@@ -0,0 +1,23 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package lfs
+
+import (
+	"net/url"
+
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestNewClient(t *testing.T) {
+	u, _ := url.Parse("file:///test")
+	c := NewClient(u)
+	assert.IsType(t, &FilesystemClient{}, c)
+
+	u, _ = url.Parse("https://test.com/lfs")
+	c = NewClient(u)
+	assert.IsType(t, &HTTPClient{}, c)
+}

modules/lfs/content_store.go

@@ -13,14 +13,15 @@ import (
 	"io"
 	"os"
 
-	"code.gitea.io/gitea/models"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/storage"
 )
 
 var (
-	errHashMismatch = errors.New("Content hash does not match OID")
-	errSizeMismatch = errors.New("Content size does not match")
+	// ErrHashMismatch occurs if the content has does not match OID
+	ErrHashMismatch = errors.New("Content hash does not match OID")
+	// ErrSizeMismatch occurs if the content size does not match
+	ErrSizeMismatch = errors.New("Content size does not match")
 )
 
 // ErrRangeNotSatisfiable represents an error which request range is not satisfiable.
@@ -28,61 +29,67 @@ type ErrRangeNotSatisfiable struct {
 	FromByte int64
 }
 
-func (err ErrRangeNotSatisfiable) Error() string {
-	return fmt.Sprintf("Requested range %d is not satisfiable", err.FromByte)
-}
-
 // IsErrRangeNotSatisfiable returns true if the error is an ErrRangeNotSatisfiable
 func IsErrRangeNotSatisfiable(err error) bool {
 	_, ok := err.(ErrRangeNotSatisfiable)
 	return ok
 }
 
+func (err ErrRangeNotSatisfiable) Error() string {
+	return fmt.Sprintf("Requested range %d is not satisfiable", err.FromByte)
+}
+
 // ContentStore provides a simple file system based storage.
 type ContentStore struct {
 	storage.ObjectStorage
 }
 
+// NewContentStore creates the default ContentStore
+func NewContentStore() *ContentStore {
+	contentStore := &ContentStore{ObjectStorage: storage.LFS}
+	return contentStore
+}
+
 // Get takes a Meta object and retrieves the content from the store, returning
 // it as an io.ReadSeekCloser.
-func (s *ContentStore) Get(meta *models.LFSMetaObject) (storage.Object, error) {
-	f, err := s.Open(meta.RelativePath())
+func (s *ContentStore) Get(pointer Pointer) (storage.Object, error) {
+	f, err := s.Open(pointer.RelativePath())
 	if err != nil {
-		log.Error("Whilst trying to read LFS OID[%s]: Unable to open Error: %v", meta.Oid, err)
+		log.Error("Whilst trying to read LFS OID[%s]: Unable to open Error: %v", pointer.Oid, err)
 		return nil, err
 	}
 	return f, err
 }
 
 // Put takes a Meta object and an io.Reader and writes the content to the store.
-func (s *ContentStore) Put(meta *models.LFSMetaObject, r io.Reader) error {
-	p := meta.RelativePath()
+func (s *ContentStore) Put(pointer Pointer, r io.Reader) error {
+	p := pointer.RelativePath()
 
 	// Wrap the provided reader with an inline hashing and size checker
-	wrappedRd := newHashingReader(meta.Size, meta.Oid, r)
+	wrappedRd := newHashingReader(pointer.Size, pointer.Oid, r)
 
 	// now pass the wrapped reader to Save - if there is a size mismatch or hash mismatch then
 	// the errors returned by the newHashingReader should percolate up to here
-	written, err := s.Save(p, wrappedRd, meta.Size)
+	written, err := s.Save(p, wrappedRd, pointer.Size)
 	if err != nil {
-		log.Error("Whilst putting LFS OID[%s]: Failed to copy to tmpPath: %s Error: %v", meta.Oid, p, err)
+		log.Error("Whilst putting LFS OID[%s]: Failed to copy to tmpPath: %s Error: %v", pointer.Oid, p, err)
 		return err
 	}
 
 	// This shouldn't happen but it is sensible to test
-	if written != meta.Size {
+	if written != pointer.Size {
 		if err := s.Delete(p); err != nil {
-			log.Error("Cleaning the LFS OID[%s] failed: %v", meta.Oid, err)
+			log.Error("Cleaning the LFS OID[%s] failed: %v", pointer.Oid, err)
 		}
-		return errSizeMismatch
+		return ErrSizeMismatch
 	}
 
 	return nil
 }
 
 // Exists returns true if the object exists in the content store.
-func (s *ContentStore) Exists(meta *models.LFSMetaObject) (bool, error) {
-	_, err := s.ObjectStorage.Stat(meta.RelativePath())
+func (s *ContentStore) Exists(pointer Pointer) (bool, error) {
+	_, err := s.ObjectStorage.Stat(pointer.RelativePath())
 	if err != nil {
 		if os.IsNotExist(err) {
 			return false, nil
@@ -93,19 +100,25 @@ func (s *ContentStore) Exists(meta *models.LFSMetaObject) (bool, error) {
 }
 
 // Verify returns true if the object exists in the content store and size is correct.
-func (s *ContentStore) Verify(meta *models.LFSMetaObject) (bool, error) {
-	p := meta.RelativePath()
+func (s *ContentStore) Verify(pointer Pointer) (bool, error) {
+	p := pointer.RelativePath()
 	fi, err := s.ObjectStorage.Stat(p)
-	if os.IsNotExist(err) || (err == nil && fi.Size() != meta.Size) {
+	if os.IsNotExist(err) || (err == nil && fi.Size() != pointer.Size) {
 		return false, nil
 	} else if err != nil {
-		log.Error("Unable stat file: %s for LFS OID[%s] Error: %v", p, meta.Oid, err)
+		log.Error("Unable stat file: %s for LFS OID[%s] Error: %v", p, pointer.Oid, err)
 		return false, err
 	}
 
 	return true, nil
 }
 
+// ReadMetaObject will read a models.LFSMetaObject and return a reader
+func ReadMetaObject(pointer Pointer) (io.ReadCloser, error) {
+	contentStore := NewContentStore()
+	return contentStore.Get(pointer)
+}
+
 type hashingReader struct {
 	internal     io.Reader
 	currentSize  int64
@@ -127,12 +140,12 @@ func (r *hashingReader) Read(b []byte) (int, error) {
 
 	if err != nil && err == io.EOF {
 		if r.currentSize != r.expectedSize {
-			return n, errSizeMismatch
+			return n, ErrSizeMismatch
 		}
 
 		shaStr := hex.EncodeToString(r.hash.Sum(nil))
 		if shaStr != r.expectedHash {
-			return n, errHashMismatch
+			return n, ErrHashMismatch
 		}
 	}
 

modules/lfs/endpoint.go

@@ -0,0 +1,106 @@
+// Copyright 2021 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package lfs
+
+import (
+	"fmt"
+	"net/url"
+	"os"
+	"path"
+	"path/filepath"
+	"strings"
+
+	"code.gitea.io/gitea/modules/log"
+)
+
+// DetermineEndpoint determines an endpoint from the clone url or uses the specified LFS url.
+func DetermineEndpoint(cloneurl, lfsurl string) *url.URL {
+	if len(lfsurl) > 0 {
+		return endpointFromURL(lfsurl)
+	}
+	return endpointFromCloneURL(cloneurl)
+}
+
+func endpointFromCloneURL(rawurl string) *url.URL {
+	ep := endpointFromURL(rawurl)
+	if ep == nil {
+		return ep
+	}
+
+	if strings.HasSuffix(ep.Path, "/") {
+		ep.Path = ep.Path[:len(ep.Path)-1]
+	}
+
+	if ep.Scheme == "file" {
+		return ep
+	}
+
+	if path.Ext(ep.Path) == ".git" {
+		ep.Path += "/info/lfs"
+	} else {
+		ep.Path += ".git/info/lfs"
+	}
+
+	return ep
+}
+
+func endpointFromURL(rawurl string) *url.URL {
+	if strings.HasPrefix(rawurl, "/") {
+		return endpointFromLocalPath(rawurl)
+	}
+
+	u, err := url.Parse(rawurl)
+	if err != nil {
+		log.Error("lfs.endpointFromUrl: %v", err)
+		return nil
+	}
+
+	switch u.Scheme {
+	case "http", "https":
+		return u
+	case "git":
+		u.Scheme = "https"
+		return u
+	case "file":
+		return u
+	default:
+		if _, err := os.Stat(rawurl); err == nil {
+			return endpointFromLocalPath(rawurl)
+		}
+
+		log.Error("lfs.endpointFromUrl: unknown url")
+		return nil
+	}
+}
+
+func endpointFromLocalPath(path string) *url.URL {
+	var slash string
+	if abs, err := filepath.Abs(path); err == nil {
+		if !strings.HasPrefix(abs, "/") {
+			slash = "/"
+		}
+		path = abs
+	}
+
+	var gitpath string
+	if filepath.Base(path) == ".git" {
+		gitpath = path
+		path = filepath.Dir(path)
+	} else {
+		gitpath = filepath.Join(path, ".git")
+	}
+
+	if _, err := os.Stat(gitpath); err == nil {
+		path = gitpath
+	} else if _, err := os.Stat(path); err != nil {
+		return nil
+	}
+
+	path = fmt.Sprintf("file://%s%s", slash, filepath.ToSlash(path))
+
+	u, _ := url.Parse(path)
+
+	return u
+}