Add Gitlab V3 Detector

pkg/detectors/gitlab/v1/gitlab_integration_test.go

@@ -297,6 +297,100 @@ func TestGitlab_FromChunk_WithV2Secrets(t *testing.T) {
 	}
 }
 
+// This test ensures gitlab v1 detector does not work on gitlab v3 secrets
+func TestGitlab_FromChunk_WithV3Secrets(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors6")
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+	secret := testSecrets.MustGetField("GITLABV3")
+	secretInactive := testSecrets.MustGetField("GITLABV3_INACTIVE")
+	type args struct {
+		ctx    context.Context
+		data   []byte
+		verify bool
+	}
+	tests := []struct {
+		name                string
+		s                   Scanner
+		args                args
+		want                []detectors.Result
+		wantErr             bool
+		wantVerificationErr bool
+	}{
+		{
+			name: "verified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a gitlab super secret %s within", secret)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "verified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("gitlab %s", secret)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "unverified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a gitlab secret %s within", secretInactive)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte("You cannot find the secret within"),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.s.SetCloudEndpoint("https://gitlab.com")
+			tt.s.UseCloudEndpoint(true)
+			got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("Gitlab.FromData() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			for i := range got {
+				if len(got[i].Raw) == 0 {
+					t.Fatal("no raw secret present")
+				}
+				if (got[i].VerificationError() != nil) != tt.wantVerificationErr {
+					t.Fatalf(" wantVerificationError = %v, verification error = %v,", tt.wantVerificationErr, got[i].VerificationError())
+				}
+				got[i].AnalysisInfo = nil
+			}
+			opts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "RawV2", "verificationError", "primarySecret")
+			if diff := cmp.Diff(got, tt.want, opts); diff != "" {
+				t.Errorf("Gitlab.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			}
+		})
+	}
+}
+
 func BenchmarkV2FromData(benchmark *testing.B) {
 	ctx := context.Background()
 	s := Scanner{}

pkg/detectors/gitlab/v2/gitlab_integration_test.go

@@ -111,6 +111,100 @@ func TestGitlabV2_FromChunk_WithV1Secrets(t *testing.T) {
 	}
 }
 
+// This test ensures gitlab v2 detector does not work on gitlab v3 secrets
+func TestGitlabV2_FromChunk_WithV3Secrets(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors6")
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+	secret := testSecrets.MustGetField("GITLABV3")
+	secretInactive := testSecrets.MustGetField("GITLABV3_INACTIVE")
+	type args struct {
+		ctx    context.Context
+		data   []byte
+		verify bool
+	}
+	tests := []struct {
+		name                string
+		s                   Scanner
+		args                args
+		want                []detectors.Result
+		wantErr             bool
+		wantVerificationErr bool
+	}{
+		{
+			name: "verified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a gitlab super secret %s within", secret)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "verified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("gitlab %s", secret)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "unverified v3 secret, not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a gitlab secret %s within", secretInactive)),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+		{
+			name: "not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte("You cannot find the secret within"),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.s.SetCloudEndpoint("https://gitlab.com")
+			tt.s.UseCloudEndpoint(true)
+			got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("Gitlab.FromData() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			for i := range got {
+				if len(got[i].Raw) == 0 {
+					t.Fatal("no raw secret present")
+				}
+				if (got[i].VerificationError() != nil) != tt.wantVerificationErr {
+					t.Fatalf(" wantVerificationError = %v, verification error = %v,", tt.wantVerificationErr, got[i].VerificationError())
+				}
+				got[i].AnalysisInfo = nil
+			}
+			opts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "RawV2", "verificationError", "primarySecret")
+			if diff := cmp.Diff(got, tt.want, opts); diff != "" {
+				t.Errorf("Gitlab.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			}
+		})
+	}
+}
+
 func BenchmarkFromData(benchmark *testing.B) {
 	ctx := context.Background()
 	s := Scanner{}

pkg/detectors/gitlab/v3/gitlab_v3.go

@@ -0,0 +1,102 @@
+package gitlab
+
+import (
+	"context"
+	"fmt"
+	"maps"
+	"net/http"
+	"strings"
+
+	regexp "github.com/wasilibs/go-re2"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	v1 "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/gitlab/v1"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+type Scanner struct {
+	client *http.Client
+	detectors.EndpointSetter
+}
+
+// Ensure the Scanner satisfies the interfaces at compile time.
+var _ detectors.Detector = (*Scanner)(nil)
+var _ detectors.EndpointCustomizer = (*Scanner)(nil)
+var _ detectors.Versioner = (*Scanner)(nil)
+
+func (Scanner) Version() int          { return 3 }
+func (Scanner) CloudEndpoint() string { return "https://gitlab.com" }
+
+var (
+	defaultClient = common.SaneHttpClient()
+	// pattern taken from gitlab's PR for the format change: https://gitlab.com/gitlab-org/gitlab/-/merge_requests/169322
+	keyPat = regexp.MustCompile(`\b(glpat-[a-zA-Z0-9\-=_]{27,300}.[0-9a-z]{2}.[a-z0-9]{9})\b`)
+)
+
+func (s Scanner) getClient() *http.Client {
+	if s.client != nil {
+		return s.client
+	}
+
+	return defaultClient
+}
+
+// Keywords are used for efficiently pre-filtering chunks.
+// Use identifiers in the secret preferably, or the provider name.
+func (s Scanner) Keywords() []string { return []string{"glpat-"} }
+
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_Gitlab
+}
+
+func (s Scanner) Description() string {
+	return "GitLab is a web-based DevOps lifecycle tool that provides a Git repository manager providing wiki, issue-tracking, and CI/CD pipeline features. GitLab Personal Access Tokens (PATs) can be used to authenticate and access GitLab resources."
+}
+
+// FromData will find and optionally verify Gitlab secrets in a given set of bytes.
+func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (results []detectors.Result, err error) {
+	dataStr := string(data)
+
+	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	for _, match := range matches {
+		resMatch := strings.TrimSpace(match[1])
+
+		for _, endpoint := range s.Endpoints() {
+			s1 := detectors.Result{
+				DetectorType: detectorspb.DetectorType_Gitlab,
+				Raw:          []byte(resMatch),
+				RawV2:        []byte(resMatch + endpoint),
+				ExtraData: map[string]string{
+					"rotation_guide": "https://howtorotate.com/docs/tutorials/gitlab/",
+					"version":        fmt.Sprintf("%d", s.Version()),
+				},
+			}
+
+			if verify {
+
+				isVerified, extraData, verificationErr := v1.VerifyGitlab(ctx, s.getClient(), endpoint, resMatch)
+				s1.Verified = isVerified
+				maps.Copy(s1.ExtraData, extraData)
+
+				s1.SetVerificationError(verificationErr)
+
+				// for verified keys set the analysis info
+				if s1.Verified {
+					s1.AnalysisInfo = map[string]string{
+						"key":  resMatch,
+						"host": endpoint,
+					}
+
+					// if secret is verified with one endpoint, break the loop to continue to next secret
+					results = append(results, s1)
+					break
+				}
+			}
+
+			results = append(results, s1)
+		}
+	}
+
+	return results, nil
+}