cmd: Add stand-alone command gitaly-license-detecttoon-bench-go-license

This change extracts all the code related to go-license-detect so we can
use it without spawning a gitaly process.

3 files changed, 122 insertions, 19 deletions

diff --git a/cmd/gitaly-license-detect/main.go b/cmd/gitaly-license-detect/main.go
new file mode 100644
index 000000000..c02fff419
--- /dev/null
+++ b/cmd/gitaly-license-detect/main.go
@@ -0,0 +1,89 @@
+package main
+
+import (
+	"context"
+	"flag"
+	"fmt"
+	"log"
+	_ "net/http/pprof"
+	"os"
+	"runtime"
+	"runtime/pprof"
+
+	"github.com/go-enry/go-license-detector/v4/licensedb"
+	"gitlab.com/gitlab-org/gitaly/v15/internal/git"
+	"gitlab.com/gitlab-org/gitaly/v15/internal/git/catfile"
+	"gitlab.com/gitlab-org/gitaly/v15/internal/git/localrepo"
+	"gitlab.com/gitlab-org/gitaly/v15/internal/gitaly/config"
+	"gitlab.com/gitlab-org/gitaly/v15/internal/gitaly/service/repository"
+	"gitlab.com/gitlab-org/gitaly/v15/proto/go/gitalypb"
+)
+
+func main() {
+	cpuprofile := flag.String("cpuprofile", "", "write cpu profile to `file`")
+	memprofile := flag.String("memprofile", "", "write memory profile to `file`")
+
+	flag.Parse()
+
+	licensedb.Preload()
+
+	if *cpuprofile != "" {
+		f, err := os.Create(*cpuprofile)
+		if err != nil {
+			log.Fatal("could not create CPU profile: ", err)
+		}
+		defer f.Close() // error handling omitted for example
+		if err := pprof.StartCPUProfile(f); err != nil {
+			log.Fatal("could not start CPU profile: ", err)
+		}
+		defer pprof.StopCPUProfile()
+	}
+
+	// ... rest of the program ...
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	cfg := config.Cfg{
+		BinDir: "/usr/bin",
+		Storages: []config.Storage{
+			{
+				Name: "default",
+				Path: "/home",
+			},
+		},
+	}
+	cmdFactory, cleanup, err := git.NewExecCommandFactory(cfg)
+	if err != nil {
+		panic(err)
+	}
+	defer cleanup()
+
+	catfileCache := catfile.NewCache(cfg)
+	locator := config.NewLocator(cfg)
+	repoProto := &gitalypb.Repository{
+		StorageName:  "default",
+		RelativePath: flag.Args()[0],
+	}
+	repo := localrepo.New(locator, cmdFactory, catfileCache, repoProto)
+	headOID, err := repo.ResolveRevision(ctx, "HEAD")
+	repoFiler := &repository.GitFiler{Ctx: ctx, Repo: repo, TreeishID: headOID}
+	detectedLicenses, err := licensedb.Detect(repoFiler)
+	if err != nil {
+		panic(err)
+	}
+
+	fmt.Println(detectedLicenses)
+
+	if *memprofile != "" {
+		f, err := os.Create(*memprofile)
+		if err != nil {
+			log.Fatal("could not create memory profile: ", err)
+		}
+		defer f.Close() // error handling omitted for example
+		runtime.GC()    // get up-to-date statistics
+		if err := pprof.WriteHeapProfile(f); err != nil {
+			log.Fatal("could not write memory profile: ", err)
+		}
+	}
+}
diff --git a/internal/gitaly/service/repository/license.go b/internal/gitaly/service/repository/license.go
index c0b12833a..a63a08bfd 100644
--- a/internal/gitaly/service/repository/license.go
+++ b/internal/gitaly/service/repository/license.go
@@ -85,7 +85,7 @@ func (s *server) FindLicense(ctx context.Context, req *gitalypb.FindLicenseReque
 }
 
 func findLicense(ctx context.Context, repo *localrepo.Repo, commitID git.ObjectID) (*gitalypb.FindLicenseResponse, error) {
-	repoFiler := &gitFiler{ctx: ctx, repo: repo, treeishID: commitID}
+	repoFiler := &GitFiler{Ctx: ctx, Repo: repo, TreeishID: commitID}
 	detectedLicenses, err := licensedb.Detect(repoFiler)
 	if err != nil {
 		if errors.Is(err, licensedb.ErrNoLicenseFound) {
@@ -175,16 +175,16 @@ func trimDeprecatedPrefix(name string) string {
 
 var readmeRegexp = regexp.MustCompile(`(readme|guidelines)(\.md|\.rst|\.html|\.txt)?$`)
 
-type gitFiler struct {
-	ctx          context.Context
-	repo         *localrepo.Repo
+type GitFiler struct {
+	Ctx          context.Context
+	Repo         *localrepo.Repo
 	foundLicense bool
 	path         string
-	treeishID    git.ObjectID
+	TreeishID    git.ObjectID
 }
 
-func (f *gitFiler) ReadFile(path string) ([]byte, error) {
-	data, err := f.repo.ReadObject(f.ctx, git.ObjectID(fmt.Sprintf("%s:%s", f.treeishID, path)))
+func (f *GitFiler) ReadFile(path string) ([]byte, error) {
+	data, err := f.Repo.ReadObject(f.Ctx, git.ObjectID(fmt.Sprintf("%s:%s", f.TreeishID, path)))
 	if err != nil {
 		return nil, fmt.Errorf("read file: %w", err)
 	}
@@ -207,14 +207,14 @@ func (f *gitFiler) ReadFile(path string) ([]byte, error) {
 	return data, nil
 }
 
-func (f *gitFiler) ReadDir(string) ([]filer.File, error) {
+func (f *GitFiler) ReadDir(string) ([]filer.File, error) {
 	var stderr bytes.Buffer
-	cmd, err := f.repo.Exec(f.ctx, git.Command{
+	cmd, err := f.Repo.Exec(f.Ctx, git.Command{
 		Name: "ls-tree",
 		Flags: []git.Option{
 			git.Flag{Name: "-z"},
 		},
-		Args: []string{f.treeishID.String()},
+		Args: []string{f.TreeishID.String()},
 	}, git.WithStderr(&stderr))
 	if err != nil {
 		return nil, err
@@ -249,9 +249,9 @@ func (f *gitFiler) ReadDir(string) ([]filer.File, error) {
 	return files, nil
 }
 
-func (f *gitFiler) Close() {}
+func (f *GitFiler) Close() {}
 
-func (f *gitFiler) PathsAreAlwaysSlash() bool {
+func (f *GitFiler) PathsAreAlwaysSlash() bool {
 	// git ls-files uses unix slash `/`
 	return true
 }
diff --git a/internal/gitaly/service/repository/license_test.go b/internal/gitaly/service/repository/license_test.go
index 3384d0cad..d518ec7a3 100644
--- a/internal/gitaly/service/repository/license_test.go
+++ b/internal/gitaly/service/repository/license_test.go
@@ -4,6 +4,7 @@ package repository
 
 import (
 	"context"
+	_ "net/http/pprof"
 	"os"
 	"testing"
 
@@ -339,6 +340,11 @@ func BenchmarkFindLicense(b *testing.B) {
 		SkipCreationViaService: true,
 	})
 
+	repoKicad, _ := gittest.CreateRepository(b, ctx, cfg, gittest.CreateRepositoryConfig{
+		SkipCreationViaService: true,
+		Seed:                   "kicad.git",
+	})
+
 	// Based on https://github.com/go-enry/go-license-detector/blob/18a439e5437cd46905b074ac24c27cbb6cac4347/licensedb/internal/investigation.go#L28-L38
 	fileNames := []string{
 		"licence",
@@ -397,16 +403,24 @@ func BenchmarkFindLicense(b *testing.B) {
 		)
 
 		for _, tc := range []struct {
-			desc string
-			repo *gitalypb.Repository
+			desc      string
+			repo      *gitalypb.Repository
+			shortname string
 		}{
 			{
-				desc: "gitlab-org/gitlab.git",
-				repo: repoGitLab,
+				desc:      "gitlab-org/gitlab.git",
+				repo:      repoGitLab,
+				shortname: "mit",
+			},
+			{
+				desc:      "stress.git",
+				repo:      repoStress,
+				shortname: "mit",
 			},
 			{
-				desc: "stress.git",
-				repo: repoStress,
+				desc:      "kicad.git",
+				repo:      repoKicad,
+				shortname: "bsl-1.0",
 			},
 		} {
 			b.Run(tc.desc, func(b *testing.B) {
@@ -415,7 +429,7 @@ func BenchmarkFindLicense(b *testing.B) {
 				for i := 0; i < b.N; i++ {
 					resp, err := client.FindLicense(ctx, &gitalypb.FindLicenseRequest{Repository: tc.repo})
 					require.NoError(b, err)
-					require.Equal(b, "mit", resp.GetLicenseShortName())
+					require.Equal(b, tc.shortname, resp.GetLicenseShortName())
 				}
 
 				if featureflag.LocalrepoReadObjectCached.IsEnabled(ctx) {