Add sites metadata in Groups backup (#4184)

Add sites metadata (previous paths) under `groupID/libraries/sites/previouspath`.

<!-- PR description-->

---

#### Does this PR need a docs update or release note?

- [ ] ✅ Yes, it's included
- [ ] 🕐 Yes, but in a later PR
- [x] ⛔ No

#### Type of change

<!--- Please check the type of change your PR introduces: --->
- [x] 🌻 Feature
- [ ] 🐛 Bugfix
- [ ] 🗺️ Documentation
- [ ] 🤖 Supportability/Tests
- [ ] 💻 CI/Deployment
- [ ] 🧹 Tech Debt/Cleanup

#### Issue(s)

<!-- Can reference multiple issues. Use one of the following "magic words" - "closes, fixes" to auto-close the Github issue. -->
* #<issue>

#### Test Plan

<!-- How will this be tested prior to merging.-->
- [ ] 💪 Manual
- [x] ⚡ Unit test
- [ ] 💚 E2E

src/internal/m365/backup_test.go

@@ -13,6 +13,7 @@ import (
 	inMock "github.com/alcionai/corso/src/internal/common/idname/mock"
 	"github.com/alcionai/corso/src/internal/m365/resource"
 	"github.com/alcionai/corso/src/internal/m365/service/exchange"
+	odConsts "github.com/alcionai/corso/src/internal/m365/service/onedrive/consts"
 	"github.com/alcionai/corso/src/internal/m365/service/sharepoint"
 	"github.com/alcionai/corso/src/internal/operations/inject"
 	"github.com/alcionai/corso/src/internal/tester"
@@ -473,6 +474,7 @@ func (suite *SPCollectionIntgSuite) TestCreateSharePointCollection_Lists() {
 type GroupsCollectionIntgSuite struct {
 	tester.Suite
 	connector *Controller
+	tenantID  string
 	user      string
 }
 
@@ -485,13 +487,21 @@ func TestGroupsCollectionIntgSuite(t *testing.T) {
 }
 
 func (suite *GroupsCollectionIntgSuite) SetupSuite() {
-	ctx, flush := tester.NewContext(suite.T())
+	t := suite.T()
+
+	ctx, flush := tester.NewContext(t)
 	defer flush()
 
-	suite.connector = newController(ctx, suite.T(), resource.Sites, path.GroupsService)
+	suite.connector = newController(ctx, t, resource.Sites, path.GroupsService)
-	suite.user = tconfig.M365UserID(suite.T())
+	suite.user = tconfig.M365UserID(t)
 
-	tester.LogTimeOfTest(suite.T())
+	acct := tconfig.NewM365Account(t)
+	creds, err := acct.M365Config()
+	require.NoError(t, err, clues.ToCore(err))
+
+	suite.tenantID = creds.AzureTenantID
+
+	tester.LogTimeOfTest(t)
 }
 
 func (suite *GroupsCollectionIntgSuite) TestCreateGroupsCollection_SharePoint() {
@@ -535,14 +545,37 @@ func (suite *GroupsCollectionIntgSuite) TestCreateGroupsCollection_SharePoint()
 	// but it should be more than one.
 	assert.Greater(t, len(collections), 1)
 
+	// TODO(meain): Switch to using BuildMetadata
+	// https://github.com/alcionai/corso/pull/4184#discussion_r1316139701
+	p, err := path.Builder{}.ToServiceCategoryMetadataPath(
+		suite.tenantID,
+		groupID,
+		path.GroupsService,
+		path.LibrariesCategory,
+		false)
+	require.NoError(t, err, clues.ToCore(err))
+
+	p, err = p.Append(false, odConsts.SitesPathDir)
+	require.NoError(t, err, clues.ToCore(err))
+
+	foundSitesMetadata := false
+
 	for _, coll := range collections {
+		sitesMetadataCollection := coll.FullPath().String() == p.String()
+
 		for object := range coll.Items(ctx, fault.New(true)) {
+			if object.ID() == "previouspath" && sitesMetadataCollection {
+				foundSitesMetadata = true
+			}
+
 			buf := &bytes.Buffer{}
 			_, err := buf.ReadFrom(object.ToReader())
 			assert.NoError(t, err, "reading item", clues.ToCore(err))
 		}
 	}
 
+	assert.True(t, foundSitesMetadata, "missing sites metadata")
+
 	status := ctrl.Wait()
 	assert.NotZero(t, status.Successes)
 	t.Log(status.String())

src/internal/m365/collection/drive/group_handler.go

@@ -65,7 +65,7 @@ func (h groupBackupHandler) MetadataPathPrefix(tenantID string) (path.Path, erro
 
 	p, err = p.Append(false, odConsts.SitesPathDir, h.siteID)
 	if err != nil {
-		return nil, clues.Wrap(err, "appending sites to metadata path")
+		return nil, clues.Wrap(err, "appending site id to metadata path")
 	}
 
 	return p, nil
@@ -86,6 +86,17 @@ func (h groupBackupHandler) CanonicalPath(
 	)
 }
 
+func (h groupBackupHandler) SitePathPrefix(tenantID string) (path.Path, error) {
+	return path.Build(
+		tenantID,
+		h.groupID,
+		h.service,
+		path.LibrariesCategory,
+		false,
+		odConsts.SitesPathDir,
+		h.siteID)
+}
+
 func (h groupBackupHandler) IsAllPass() bool {
 	return h.scope.IsAny(selectors.GroupsLibraryFolder)
 }

src/internal/m365/collection/drive/group_handler_test.go

@@ -49,6 +49,35 @@ func (suite *GroupBackupHandlerUnitSuite) TestPathPrefix() {
 	}
 }
 
+func (suite *GroupBackupHandlerUnitSuite) TestSitePathPrefix() {
+	tenantID, resourceOwner := "tenant", "resourceOwner"
+
+	table := []struct {
+		name      string
+		expect    string
+		expectErr assert.ErrorAssertionFunc
+	}{
+		{
+			name:      "group",
+			expect:    "tenant/groups/resourceOwner/libraries/sites/site-id",
+			expectErr: assert.NoError,
+		},
+	}
+	for _, test := range table {
+		suite.Run(test.name, func() {
+			t := suite.T()
+			h := NewGroupBackupHandler(resourceOwner, "site-id", api.Drives{}, nil)
+
+			result, err := h.SitePathPrefix(tenantID)
+			test.expectErr(t, err, clues.ToCore(err))
+
+			if result != nil {
+				assert.Equal(t, test.expect, result.String())
+			}
+		})
+	}
+}
+
 func (suite *GroupBackupHandlerUnitSuite) TestMetadataPathPrefix() {
 	tenantID, resourceOwner := "tenant", "resourceOwner"
 

src/internal/m365/service/groups/backup.go

@@ -13,6 +13,7 @@ import (
 	"github.com/alcionai/corso/src/internal/m365/collection/groups"
 	"github.com/alcionai/corso/src/internal/m365/collection/site"
 	"github.com/alcionai/corso/src/internal/m365/graph"
+	odConsts "github.com/alcionai/corso/src/internal/m365/service/onedrive/consts"
 	"github.com/alcionai/corso/src/internal/m365/support"
 	"github.com/alcionai/corso/src/internal/observe"
 	"github.com/alcionai/corso/src/internal/operations/inject"
@@ -41,6 +42,7 @@ func ProduceBackupCollections(
 		categories           = map[path.CategoryType]struct{}{}
 		ssmb                 = prefixmatcher.NewStringSetBuilder()
 		canUsePreviousBackup bool
+		sitesPreviousPaths   = map[string]string{}
 	)
 
 	ctx = clues.Add(
@@ -78,15 +80,23 @@ func ProduceBackupCollections(
 				Selector:          bpc.Selector,
 			}
 
+			bh := drive.NewGroupBackupHandler(
+				bpc.ProtectedResource.ID(),
+				ptr.Val(resp.GetId()),
+				ac.Drives(),
+				scope)
+
+			cp, err := bh.SitePathPrefix(creds.AzureTenantID)
+			if err != nil {
+				return nil, nil, false, clues.Wrap(err, "getting canonical path")
+			}
+
+			sitesPreviousPaths[ptr.Val(resp.GetId())] = cp.String()
+
 			dbcs, canUsePreviousBackup, err = site.CollectLibraries(
 				ctx,
 				sbpc,
-				drive.NewGroupBackupHandler(
+				bh,
-					bpc.ProtectedResource.ID(),
-					ptr.Val(resp.GetId()),
-					ac.Drives(),
-					scope,
-				),
 				creds.AzureTenantID,
 				ssmb,
 				su,
@@ -133,5 +143,49 @@ func ProduceBackupCollections(
 		collections = append(collections, baseCols...)
 	}
 
+	// Add metadata about sites
+	md, err := getSitesMetadataCollection(
+		creds.AzureTenantID,
+		bpc.ProtectedResource.ID(),
+		sitesPreviousPaths,
+		su)
+	if err != nil {
+		return nil, nil, false, err
+	}
+
+	collections = append(collections, md)
+
 	return collections, ssmb.ToReader(), canUsePreviousBackup, el.Failure()
 }
+
+func getSitesMetadataCollection(
+	tenantID, groupID string,
+	sites map[string]string,
+	su support.StatusUpdater,
+) (data.BackupCollection, error) {
+	// TODO(meain): Switch to using BuildMetadata
+	// https://github.com/alcionai/corso/pull/4184#discussion_r1316139701
+	p, err := path.Builder{}.ToServiceCategoryMetadataPath(
+		tenantID,
+		groupID,
+		path.GroupsService,
+		path.LibrariesCategory,
+		false)
+	if err != nil {
+		return nil, clues.Wrap(err, "making metadata path")
+	}
+
+	p, err = p.Append(false, odConsts.SitesPathDir)
+	if err != nil {
+		return nil, clues.Wrap(err, "appending sites to metadata path")
+	}
+
+	md, err := graph.MakeMetadataCollection(
+		p,
+		[]graph.MetadataCollectionEntry{
+			graph.NewMetadataEntry(graph.PreviousPathFileName, sites),
+		},
+		su)
+
+	return md, err
+}