perf(git): speed up activity top authors and subdirectory listings (#39526)

Speeds up two slow git paths. Results match `main` apart
from the `log.follow` fix.

- Activity top authors skip `--numstat` and an unused `rev-list
--count`, they only need names and emails
- Subdirectory listings pass only the directory as pathspec, which
already matches all its entries
- Directory listings pass `--no-follow` to `git log`, a configured
`log.follow` disabled parent rewriting and gave wrong last commits

| Benchmark | main | PR | Change |
|---|--:|--:|--:|
| Top authors, gitea, 1 month | 632 ms | 21 ms | -97% |
| Top authors, gitea, 1 year | 3216 ms | 68 ms | -98% |
| Top authors, tea | 62 ms | 12 ms | -80% |
| Listing, git `Documentation/technical` | 189 ms | 91 ms | -52% |
| Listing, gitea `options/license` | 329 ms | 208 ms | -37% |
| Listing, gitea `templates/repo` | 294 ms | 223 ms | -24% |

Tested with unit and sqlite integration tests in default and `gogit`
builds on git 2.25 and 2.56, and by comparing listing results and cache
writes with `main` on randomized histories and the gitea, tea and git
repos. Benchmarks are medians of 8 interleaved macOS runs.
This commit is contained in:
silverwind authored and GitHub committed 2026-10-01 15:51:07 +00:00
1 parent 2f5cdbd5c1
commit 075ff8f516
5 files changed
+38 -31

No files matched your search

+23 -6
View File
@@ -61,10 +61,28 @@ func (repo *Repository) GetCodeActivityStats(ctx context.Context, fromTime time.
} else {
gitCmd.AddArguments("--first-parent").AddDynamicArguments(branch)
}
if err := parseCodeActivityLog(ctx, repo, gitCmd, stats); err != nil {
return nil, err
}
return stats, nil
}
// GetCodeActivityAuthors returns the authors of non-merge commits in all branches since fromTime, sorted by commit count
func GetCodeActivityAuthors(ctx context.Context, repo RepositoryFacade, fromTime time.Time) ([]*CodeActivityAuthor, error) {
stats := &CodeActivityStats{}
gitCmd := gitcmd.NewCommand("log", "--no-merges", "--pretty=format:---%n%H%n%aN%n%aE%n", "--branches=*").
AddOptionFormat("--since=%s", fromTime.Format(time.RFC3339))
if err := parseCodeActivityLog(ctx, repo, gitCmd, stats); err != nil {
return nil, err
}
return stats.Authors, nil
}
func parseCodeActivityLog(ctx context.Context, repo RepositoryFacade, gitCmd *gitcmd.Command, stats *CodeActivityStats) error {
stdoutReader, stdoutReaderClose := gitCmd.MakeStdoutPipe()
defer stdoutReaderClose()
err = gitCmd.
err := gitCmd.
WithRepo(repo).
WithPipelineFunc(func(ctx gitcmd.Context) error {
scanner := bufio.NewScanner(stdoutReader)
@@ -116,8 +134,8 @@ func (repo *Repository) GetCodeActivityStats(ctx context.Context, fromTime time.
}
}
}
if err = scanner.Err(); err != nil {
return fmt.Errorf("GetCodeActivityStats scan: %w", err)
if err := scanner.Err(); err != nil {
return fmt.Errorf("parseCodeActivityLog scan: %w", err)
}
a := make([]*CodeActivityAuthor, 0, len(authors))
for _, v := range authors {
@@ -134,8 +152,7 @@ func (repo *Repository) GetCodeActivityStats(ctx context.Context, fromTime time.
}).
RunWithStderr(ctx)
if err != nil {
return nil, fmt.Errorf("GetCodeActivityStats: %w", err)
return fmt.Errorf("parseCodeActivityLog: %w", err)
}
return stats, nil
return nil
}