Let shallow clone work with any ref

Currently the shallow clone only works if ref is either not set, or set
to the default branch.  This is becuase go-getter does not set the ref
to checkout during a shallow clone so Git assumes the default branch,
and then when go-getter tries to checkout a ref other than the default
branch it fails because the shallow clone did not fetch that ref.

To fix this, I have used `--branch` to explicitly set the ref to fetch
during a clone so that when a shallow clone is done go-getter will clone
the ref given and not the default branch.  This will also make clones of
non-default branches just a tad bit more optimized since the clone will
not waste time checking out a branch it ultimately will not use.

Author: zachwhaley

get_git.go

@@ -27,6 +27,7 @@ type GitGetter struct {
 }
 
 var defaultBranchRegexp = regexp.MustCompile(`\s->\sorigin/(.*)`)
+var lsRemoteSymRefRegexp = regexp.MustCompile(`ref: refs/heads/([^\s]+).*`)
 
 func (g *GitGetter) ClientMode(_ *url.URL) (ClientMode, error) {
 	return ClientModeDir, nil
@@ -114,7 +115,7 @@ func (g *GitGetter) Get(dst string, u *url.URL) error {
 	if err == nil {
 		err = g.update(ctx, dst, sshKeyFile, ref, depth)
 	} else {
-		err = g.clone(ctx, dst, sshKeyFile, u, depth)
+		err = g.clone(ctx, dst, sshKeyFile, u, ref, depth)
 	}
 	if err != nil {
 		return err
@@ -166,14 +167,17 @@ func (g *GitGetter) checkout(dst string, ref string) error {
 	return getRunCommand(cmd)
 }
 
-func (g *GitGetter) clone(ctx context.Context, dst, sshKeyFile string, u *url.URL, depth int) error {
+func (g *GitGetter) clone(ctx context.Context, dst, sshKeyFile string, u *url.URL, ref string, depth int) error {
 	args := []string{"clone"}
 
+	if ref == "" {
+		ref = findRemoteDefaultBranch(u)
+	}
 	if depth > 0 {
 		args = append(args, "--depth", strconv.Itoa(depth))
 	}
 
-	args = append(args, u.String(), dst)
+	args = append(args, "--branch", ref, u.String(), dst)
 	cmd := exec.CommandContext(ctx, "git", args...)
 	setupGitEnv(cmd, sshKeyFile)
 	return getRunCommand(cmd)
@@ -236,6 +240,20 @@ func findDefaultBranch(dst string) string {
 	return matches[len(matches)-1]
 }
 
+// findRemoteDefaultBranch checks the remote repo's HEAD symref to return the remote repo's
+// default branch. "master" is returned if no HEAD symref exists.
+func findRemoteDefaultBranch(u *url.URL) string {
+	var stdoutbuf bytes.Buffer
+	cmd := exec.Command("git", "ls-remote", "--symref", u.String(), "HEAD")
+	cmd.Stdout = &stdoutbuf
+	err := cmd.Run()
+	matches := lsRemoteSymRefRegexp.FindStringSubmatch(stdoutbuf.String())
+	if err != nil || matches == nil {
+		return "master"
+	}
+	return matches[len(matches)-1]
+}
+
 // setupGitEnv sets up the environment for the given command. This is used to
 // pass configuration data to git and ssh and enables advanced cloning methods.
 func setupGitEnv(cmd *exec.Cmd, sshKeyFile string) {

get_git_test.go

@@ -162,6 +162,44 @@ func TestGitGetter_shallowClone(t *testing.T) {
 	}
 }
 
+func TestGitGetter_shallowCloneWithTag(t *testing.T) {
+	if !testHasGit {
+		t.Log("git not found, skipping")
+		t.Skip()
+	}
+
+	g := new(GitGetter)
+	dst := tempDir(t)
+
+	repo := testGitRepo(t, "upstream")
+	repo.commitFile("upstream.txt", "0")
+	repo.commitFile("upstream.txt", "1")
+	repo.git("tag", "v1.0")
+
+	// Specifiy a clone depth of 1 with a tag
+	q := repo.url.Query()
+	q.Add("ref", "v1.0")
+	q.Add("depth", "1")
+	repo.url.RawQuery = q.Encode()
+
+	if err := g.Get(dst, repo.url); err != nil {
+		t.Fatalf("err: %s", err)
+	}
+
+	// Assert rev-list count is '1'
+	cmd := exec.Command("git", "rev-list", "HEAD", "--count")
+	cmd.Dir = dst
+	b, err := cmd.Output()
+	if err != nil {
+		t.Fatalf("err: %s", err)
+	}
+
+	out := strings.TrimSpace(string(b))
+	if out != "1" {
+		t.Fatalf("expected rev-list count to be '1' but got %v", out)
+	}
+}
+
 func TestGitGetter_branchUpdate(t *testing.T) {
 	if !testHasGit {
 		t.Skip("git not found, skipping")