Support repo license

go.mod

@@ -56,6 +56,7 @@ require (
 	github.com/gogs/go-gogs-client v0.0.0-20210131175652-1d7215cd8d85
 	github.com/golang-jwt/jwt/v4 v4.5.0
 	github.com/google/go-github/v51 v51.0.0
+	github.com/google/licensecheck v0.3.1
 	github.com/google/pprof v0.0.0-20230502171905-255e3b9b56de
 	github.com/google/uuid v1.3.0
 	github.com/gorilla/feeds v1.1.1
@@ -203,7 +204,6 @@ require (
 	github.com/golang/snappy v0.0.4 // indirect
 	github.com/google/go-querystring v1.1.0 // indirect
 	github.com/google/go-tpm v0.3.3 // indirect
-	github.com/google/licensecheck v0.3.1 // indirect
 	github.com/gorilla/css v1.0.0 // indirect
 	github.com/gorilla/handlers v1.5.1 // indirect
 	github.com/gorilla/mux v1.8.0 // indirect

models/repo/license.go

@@ -0,0 +1,38 @@
+// Copyright 2023 The Gitea Authors. All rights reserved.
+// SPDX-License-Identifier: MIT
+
+package repo
+
+import (
+	"context"
+
+	"code.gitea.io/gitea/models/db"
+)
+
+// UpdateLanguageStats updates the license statistics for repository
+func UpdateLicenseStats(repo *Repository, commitID string, licenses []string) error {
+	ctx, committer, err := db.TxContext(db.DefaultContext)
+	if err != nil {
+		return err
+	}
+	defer committer.Close()
+
+	if err = UpdateLicense(ctx, repo.ID, licenses); err != nil {
+		return err
+	}
+
+	// Update indexer status
+	if err = UpdateIndexerStatus(ctx, repo, RepoIndexerTypeStats, commitID); err != nil {
+		return err
+	}
+
+	return committer.Commit()
+}
+
+// UpdateRepoSize updates the repository size, calculating it using getDirectorySize
+func UpdateLicense(ctx context.Context, repoID int64, licenses []string) error {
+	_, err := db.GetEngine(ctx).ID(repoID).Cols("licenses").NoAutoTime().Update(&Repository{
+		Licenses: licenses,
+	})
+	return err
+}

models/repo/repo.go

@@ -157,6 +157,8 @@ type Repository struct {
 	Units                  []*RepoUnit       `xorm:"-"`
 	PrimaryLanguage        *LanguageStat     `xorm:"-"`
 
+	Licenses []string `xorm:"TEXT JSON"`
+
 	IsFork                          bool               `xorm:"INDEX NOT NULL DEFAULT false"`
 	ForkID                          int64              `xorm:"INDEX"`
 	BaseRepo                        *Repository        `xorm:"-"`

modules/context/repo.go

@@ -399,6 +399,9 @@ func repoAssignment(ctx *Context, repo *repo_model.Repository) {
 	ctx.Data["PushMirrors"] = pushMirrors
 	ctx.Data["RepoName"] = ctx.Repo.Repository.Name
 	ctx.Data["IsEmptyRepo"] = ctx.Repo.Repository.IsEmpty
+	ctx.Data["Licenses"] = ctx.Repo.Repository.Licenses
+	// TODO: support ext
+	ctx.Data["LicenseFileName"] = "LICENSE"
 }
 
 // RepoIDAssignment returns a handler which assigns the repo to the context.

modules/git/repo_license_gogit.go

@@ -0,0 +1,68 @@
+// Copyright 2023 The Gitea Authors. All rights reserved.
+// SPDX-License-Identifier: MIT
+
+//go:build gogit
+
+package git
+
+import (
+	"io"
+
+	"github.com/go-git/go-git/v5"
+	"github.com/go-git/go-git/v5/plumbing"
+	"github.com/go-git/go-git/v5/plumbing/object"
+	"github.com/google/licensecheck"
+)
+
+// GetLicenseStats calculates license stats for git repository at specified commit
+func (repo *Repository) GetLicenseStats(commitID string) ([]string, error) {
+	r, err := git.PlainOpen(repo.Path)
+	if err != nil {
+		return nil, err
+	}
+
+	rev, err := r.ResolveRevision(plumbing.Revision(commitID))
+	if err != nil {
+		return nil, err
+	}
+
+	commit, err := r.CommitObject(*rev)
+	if err != nil {
+		return nil, err
+	}
+
+	tree, err := commit.Tree()
+	if err != nil {
+		return nil, err
+	}
+
+	var licenses []string
+	// TODO: fix unnecessary check
+	err = tree.Files().ForEach(func(f *object.File) error {
+		if f.Size == 0 {
+			return nil
+		}
+
+		// TODO: support ext
+		if f.Name == "LICENSE" {
+			r, err := f.Reader()
+			if err != nil {
+				return err
+			}
+			defer r.Close()
+
+			content, err := io.ReadAll(r)
+			if err != nil {
+				return err
+			}
+
+			cov := licensecheck.Scan(content)
+			for _, m := range cov.Match {
+				licenses = append(licenses, m.ID)
+			}
+		}
+		return nil
+	})
+
+	return licenses, nil
+}

modules/git/repo_license_nogogit.go

@@ -0,0 +1,90 @@
+// Copyright 2023 The Gitea Authors. All rights reserved.
+// SPDX-License-Identifier: MIT
+
+//go:build !gogit
+
+package git
+
+import (
+	"bytes"
+	"io"
+
+	"code.gitea.io/gitea/modules/log"
+	"github.com/google/licensecheck"
+)
+
+// GetLicenseStats calculates language stats for git repository at specified commit
+func (repo *Repository) GetLicenseStats(commitID string) ([]string, error) {
+	// We will feed the commit IDs in order into cat-file --batch, followed by blobs as necessary.
+	// so let's create a batch stdin and stdout
+	batchStdinWriter, batchReader, cancel := repo.CatFileBatch(repo.Ctx)
+	defer cancel()
+
+	writeID := func(id string) error {
+		_, err := batchStdinWriter.Write([]byte(id + "\n"))
+		return err
+	}
+
+	if err := writeID(commitID); err != nil {
+		return nil, err
+	}
+	shaBytes, typ, size, err := ReadBatchLine(batchReader)
+	if typ != "commit" {
+		log.Debug("Unable to get commit for: %s. Err: %v", commitID, err)
+		return nil, ErrNotExist{commitID, ""}
+	}
+
+	sha, err := NewIDFromString(string(shaBytes))
+	if err != nil {
+		log.Debug("Unable to get commit for: %s. Err: %v", commitID, err)
+		return nil, ErrNotExist{commitID, ""}
+	}
+
+	commit, err := CommitFromReader(repo, sha, io.LimitReader(batchReader, size))
+	if err != nil {
+		log.Debug("Unable to get commit for: %s. Err: %v", commitID, err)
+		return nil, err
+	}
+	if _, err = batchReader.Discard(1); err != nil {
+		return nil, err
+	}
+
+	tree := commit.Tree
+
+	entries, err := tree.ListEntriesRecursiveWithSize()
+	if err != nil {
+		return nil, err
+	}
+
+	var licenses []string
+	for _, f := range entries {
+		if f.Size() == 0 {
+			continue
+		}
+
+		// TODO: support ext
+		if f.Name() == "LICENSE" {
+			// Need to read all of them or can not get the right result from licensecheck.Scan
+			if err := writeID(f.ID.String()); err != nil {
+				return nil, err
+			}
+			_, _, size, err := ReadBatchLine(batchReader)
+			if err != nil {
+				log.Debug("Error reading blob: %s Err: %v", f.ID.String(), err)
+				return nil, err
+			}
+			contentBuf := bytes.Buffer{}
+			_, err = contentBuf.ReadFrom(io.LimitReader(batchReader, size))
+			if err != nil {
+				return nil, err
+			}
+			cov := licensecheck.Scan(contentBuf.Bytes())
+			for _, m := range cov.Match {
+				licenses = append(licenses, m.ID)
+			}
+			break
+		}
+	}
+
+	return licenses, nil
+}