From 0bcf6f9d46fe30b6301ffb85169998b20e1bf801 Mon Sep 17 00:00:00 2001
From: Xing Xin <xingxin.xx@bytedance.com>
Date: Thu, 4 May 2023 19:52:22 +0800
Subject: [PATCH] localrepo: Avoid getting all branches for default branch
 detection

To detect the default branch of a repo, we are currently getting all its
branches and compare one by one. This approach is unnecessarily
expensive, especially for repos having too many branches, giving HEAD is
actually available for the vast majority of repos.

This commit adjusts the logic of GetDefaultBranch, we now prefer
detection using git-rev-parse to exhaustion using git-for-each-ref.

For example, gitlab-org/gitlab has around 14,000 branches, getting all
its branches took around 40ms for all refs packed in packed-refs.

  $ time git for-each-ref --format="%(refname)%00%(objectname)%00%(symref)" refs/heads/ >/dev/null

  real	0m0.042s
  user	0m0.032s
  sys	0m0.010s

And it's getting terribly bad for all loose-refs.

  $ time git for-each-ref --format="%(refname)%00%(objectname)%00%(symref)" refs/heads/ >/dev/null

  real	0m0.177s
  user	0m0.062s
  sys	0m0.111s

But it's much quicker to detect the existence of a branch.

  $ time git rev-parse --verify refs/heads/master >/dev/null

  real	0m0.003s
  user	0m0.002s
  sys	0m0.001s

In addition to the comparison of git commands mentioned above, we can
also save more time and memory for not iterating and keeping the
branches list.

Changelog: performance
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
---
 internal/git/localrepo/refs.go | 41 ++++++++++------------------------
 1 file changed, 12 insertions(+), 29 deletions(-)

diff --git a/internal/git/localrepo/refs.go b/internal/git/localrepo/refs.go
index d88f95911a..dd3248f026 100644
--- a/internal/git/localrepo/refs.go
+++ b/internal/git/localrepo/refs.go
@@ -326,17 +326,6 @@ func (repo *Repo) GetRemoteReferences(ctx context.Context, remote string, opts .
 
 // GetDefaultBranch determines the default branch name
 func (repo *Repo) GetDefaultBranch(ctx context.Context) (git.ReferenceName, error) {
-	branches, err := repo.GetBranches(ctx)
-	if err != nil {
-		return "", err
-	}
-	switch len(branches) {
-	case 0:
-		return "", nil
-	case 1:
-		return branches[0].Name, nil
-	}
-
 	headReference, err := repo.HeadReference(ctx)
 	if err != nil {
 		return "", err
@@ -344,30 +333,24 @@ func (repo *Repo) GetDefaultBranch(ctx context.Context) (git.ReferenceName, erro
 
 	// Ideally we would only use HEAD to determine the default branch, but
 	// gitlab-rails depends on the branch being determined like this.
-	var defaultRef, legacyDefaultRef git.ReferenceName
-	for _, branch := range branches {
-		if len(headReference) != 0 && headReference == branch.Name {
-			return branch.Name, nil
+	for _, headCandidate := range []git.ReferenceName{
+		headReference, git.DefaultRef, git.LegacyDefaultRef,
+	} {
+		has, err := repo.HasRevision(ctx, headCandidate.Revision())
+		if err != nil {
+			return "", err
 		}
-
-		if git.DefaultRef == branch.Name {
-			defaultRef = branch.Name
+		if has {
+			return headCandidate, nil
 		}
-
-		if git.LegacyDefaultRef == branch.Name {
-			legacyDefaultRef = branch.Name
-		}
-	}
-
-	if len(defaultRef) != 0 {
-		return defaultRef, nil
 	}
 
-	if len(legacyDefaultRef) != 0 {
-		return legacyDefaultRef, nil
+	// If all else fails, return the first branch name
+	branches, err := repo.getReferences(ctx, 1, "refs/heads/")
+	if err != nil || len(branches) == 0 {
+		return "", err
 	}
 
-	// If all else fails, return the first branch name
 	return branches[0].Name, nil
 }
 
-- 
GitLab