Prevent hang in git cat-file if the repository is not a valid repository (Partial go-gitea#17991)

Unfortunately it appears that if git cat-file is run in an invalid
repository it will hang until stdin is closed. This will result in
deadlocked /pulls pages and dangling git cat-file calls if a broken
repository is tried to be reviewed or pulls exists for a broken
repository.

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

Author: zeripath

modules/git/batch_reader.go

@@ -27,6 +27,17 @@ type WriteCloserError interface {
 	CloseWithError(err error) error
 }
 
+func EnsureValidGitRepository(ctx context.Context, repoPath string) error {
+	stderr := strings.Builder{}
+	err := NewCommandContext(ctx, "rev-parse").
+		SetDescription(fmt.Sprintf("%s rev-parse [repo_path: %s]", GitExecutable, repoPath)).
+		RunInDirFullPipeline(repoPath, nil, &stderr, nil)
+	if err != nil {
+		return ConcatenateError(err, (&stderr).String())
+	}
+	return nil
+}
+
 // CatFileBatchCheck opens git cat-file --batch-check in the provided repo and returns a stdin pipe, a stdout reader and cancel function
 func CatFileBatchCheck(repoPath string) (WriteCloserError, *bufio.Reader, func()) {
 	batchStdinReader, batchStdinWriter := io.Pipe()

modules/git/repo_base_nogogit.go

@@ -43,6 +43,11 @@ func OpenRepository(repoPath string) (*Repository, error) {
 		return nil, errors.New("no such file or directory")
 	}
 
+	// Now because of some insanity with git cat-file not immediately failing if not run in a valid git directory we need to run git rev-parse first!
+	if err := EnsureValidGitRepository(ctx, repoPath); err != nil {
+		return nil, err
+	}
+
 	repo := &Repository{
 		Path:     repoPath,
 		tagCache: newObjectCache(),

modules/git/repo_commit_nogogit.go

@@ -38,7 +38,10 @@ func (repo *Repository) ResolveReference(name string) (string, error) {
 func (repo *Repository) GetRefCommitID(name string) (string, error) {
 	wr, rd, cancel := repo.CatFileBatchCheck()
 	defer cancel()
-	_, _ = wr.Write([]byte(name + "\n"))
+	_, err := wr.Write([]byte(name + "\n"))
+	if err != nil {
+		return "", err
+	}
 	shaBs, _, _, err := ReadBatchLine(rd)
 	if IsErrNotExist(err) {
 		return "", ErrNotExist{name, ""}

modules/indexer/code/bleve.go

@@ -276,6 +276,12 @@ func (b *BleveIndexer) Index(repo *models.Repository, sha string, changes *repoC
 	batch := gitea_bleve.NewFlushingBatch(b.indexer, maxBatchSize)
 	if len(changes.Updates) > 0 {
 
+		// Now because of some insanity with git cat-file not immediately failing if not run in a valid git directory we need to run git rev-parse first!
+		if err := git.EnsureValidGitRepository(git.DefaultContext, repo.RepoPath()); err != nil {
+			log.Error("Unable to open git repo: %s for %-v: %v", repo.RepoPath(), repo, err)
+			return err
+		}
+
 		batchWriter, batchReader, cancel := git.CatFileBatch(repo.RepoPath())
 		defer cancel()
 

modules/indexer/code/elastic_search.go

@@ -248,6 +248,11 @@ func (b *ElasticSearchIndexer) addDelete(filename string, repo *models.Repositor
 func (b *ElasticSearchIndexer) Index(repo *models.Repository, sha string, changes *repoChanges) error {
 	reqs := make([]elastic.BulkableRequest, 0)
 	if len(changes.Updates) > 0 {
+		// Now because of some insanity with git cat-file not immediately failing if not run in a valid git directory we need to run git rev-parse first!
+		if err := git.EnsureValidGitRepository(git.DefaultContext, repo.RepoPath()); err != nil {
+			log.Error("Unable to open git repo: %s for %-v: %v", repo.RepoPath(), repo, err)
+			return err
+		}
 
 		batchWriter, batchReader, cancel := git.CatFileBatch(repo.RepoPath())
 		defer cancel()

services/pull/pull.go

@@ -713,7 +713,8 @@ func GetIssuesLastCommitStatus(issues models.IssueList) (map[int64]*models.Commi
 		if !ok {
 			gitRepo, err = git.OpenRepository(issue.Repo.RepoPath())
 			if err != nil {
-				return nil, err
+				log.Error("Cannot open git repository %-v for issue #%d[%d]. Error: %v", issue.Repo, issue.Index, issue.ID, err)
+				continue
 			}
 			gitRepos[issue.RepoID] = gitRepo
 		}