controllers: control tar ignores w/ exclude files

This commit changes the file excludes for tarballs generated for
Git repository artifacts from a fixed set of strings to an exclude
file based system. It currently takes `.sourceignore` and
`.tarignore` in the root of the given directory into account.

In addition to this, it adds the `--exclude-vcs` option to the tar
command so that all VCS related files are always excluded.

Author: hiddeco

controllers/gitrepository_controller.go

@@ -345,7 +345,7 @@ func (r *GitRepositoryReconciler) sync(ctx context.Context, repository sourcev1.
 	defer unlock()
 
 	// archive artifact and check integrity
-	err = r.Storage.Archive(artifact, tmpGit, "", true)
+	err = r.Storage.Archive(artifact, tmpGit, true)
 	if err != nil {
 		err = fmt.Errorf("storage archive error: %w", err)
 		return sourcev1.GitRepositoryNotReady(repository, sourcev1.StorageOperationFailedReason, err.Error()), err

controllers/storage.go

@@ -33,6 +33,11 @@ import (
 	"github.com/fluxcd/source-controller/internal/lockedfile"
 )
 
+var excludeFiles = []string{
+	".sourceignore",
+	".tarignore",
+}
+
 // Storage manages artifacts
 type Storage struct {
 	// BasePath is the local directory path where the source artifacts are stored.
@@ -112,17 +117,17 @@ func (s *Storage) ArtifactExist(artifact sourcev1.Artifact) bool {
 	return true
 }
 
-// Archive creates a tar.gz to the artifact path from the given dir excluding the provided file extensions
-func (s *Storage) Archive(artifact sourcev1.Artifact, dir string, excludes string, integrityCheck bool) error {
-	if excludes == "" {
-		excludes = "jpg,jpeg,gif,png,wmv,flv,tar.gz,zip"
-	}
-
+// Archive creates a tar.gz to the artifact path from the given dir excluding any VCS specific
+// files and directories, or any of the excludes defined in the excludeFiles.
+func (s *Storage) Archive(artifact sourcev1.Artifact, dir string, integrityCheck bool) error {
 	ctx, cancel := context.WithTimeout(context.Background(), s.Timeout)
 	defer cancel()
 
-	tarExcludes := fmt.Sprintf("--exclude=\\*.{%s} --exclude .git", excludes)
-	cmd := fmt.Sprintf("cd %s && tar -c %s -f - . | gzip > %s", dir, tarExcludes, artifact.Path)
+	tarExcludes := []string{"--exclude-vcs"}
+	for _, f := range excludeFiles {
+		tarExcludes = append(tarExcludes, "--exclude-file="+f)
+	}
+	cmd := fmt.Sprintf("cd %s && tar -c %s -f - . | gzip > %s", dir, strings.Join(tarExcludes, " "), artifact.Path)
 	command := exec.CommandContext(ctx, "/bin/sh", "-c", cmd)
 
 	err := command.Run()