Adding detector for Nvidia NGC (#797)

* template

* minimum viable regex

* valid api 401

* passing tests

* snake to camelcase

pkg/detectors/ngc/ngc.go

@@ -0,0 +1,83 @@
+package ngc
+
+import (
+	"context"
+	"encoding/base64"
+	"net/http"
+	"regexp"
+	"strings"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+type Scanner struct{}
+
+// Ensure the Scanner satisfies the interface at compile time.
+var _ detectors.Detector = (*Scanner)(nil)
+
+var (
+	client = common.SaneHttpClient()
+
+	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
+	// keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"ngc"}) + `\b([[:alnum:]]{26}:[[:alnum:]]{8}-[[:alnum:]]{4}-[[:alnum:]]{4}-[[:alnum:]]{4}-[[:alnum:]]{12})\b`)
+	keyPat1 = regexp.MustCompile(`\b([[:alnum:]]{84})\b`)
+	keyPat2 = regexp.MustCompile(`\b([[:alnum:]]{26}:[[:alnum:]]{8}-[[:alnum:]]{4}-[[:alnum:]]{4}-[[:alnum:]]{4}-[[:alnum:]]{12})\b`)
+)
+
+// 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{"ngc"}
+}
+
+// FromData will find and optionally verify NGC 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 := keyPat1.FindAllStringSubmatch(dataStr, -1)
+
+	for _, match := range matches {
+		if len(match) != 2 {
+			continue
+		}
+		resMatch := strings.TrimSpace(match[1])
+		decode, _ := base64.StdEncoding.DecodeString(resMatch)
+
+		containsKey := keyPat2.MatchString(string(decode))
+		if containsKey {
+			s1 := detectors.Result{
+				DetectorType: detectorspb.DetectorType_NGC,
+				Raw:          []byte(resMatch),
+			}
+
+			if verify {
+				key := "Basic " + string(base64.StdEncoding.EncodeToString([]byte("$oauthtoken:"+resMatch)))
+				req, err := http.NewRequestWithContext(ctx, "GET", "https://authn.nvidia.com/token?service=ngc", nil)
+				if err != nil {
+					continue
+				}
+				req.Header = http.Header{
+					"accept":        {"*/*"},
+					"Authorization": {key},
+				}
+				res, err := client.Do(req)
+				if err == nil {
+					defer res.Body.Close()
+					if res.StatusCode >= 200 && res.StatusCode < 300 {
+						s1.Verified = true
+					} else {
+						// This function will check false positives for common test words, but also it will make sure the key appears 'random' enough to be a real key.
+						if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+							continue
+						}
+					}
+				}
+			}
+
+			results = append(results, s1)
+		}
+
+	}
+	return detectors.CleanResults(results), nil
+}

pkg/detectors/ngc/ngc_test.go

@@ -0,0 +1,119 @@
+//go:build detectors
+// +build detectors
+
+package ngc
+
+import (
+	"context"
+	"fmt"
+	"testing"
+	"time"
+
+	"github.com/kylelemons/godebug/pretty"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+func TestNGC_FromChunk(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetSecret(ctx, "hackthisai", "th-ngc-valid")
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+	secret := testSecrets.MustGetField("NGC")
+	inactiveSecret := testSecrets.MustGetField("NGC_INACTIVE")
+
+	type args struct {
+		ctx    context.Context
+		data   []byte
+		verify bool
+	}
+	tests := []struct {
+		name    string
+		s       Scanner
+		args    args
+		want    []detectors.Result
+		wantErr bool
+	}{
+		{
+			name: "found, verified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a ngc secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_NGC,
+					Verified:     true,
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a ngc secret %s within but not valid", inactiveSecret)), // the secret would satisfy the regex but not pass validation
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_NGC,
+					Verified:     false,
+				},
+			},
+			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) {
+			s := Scanner{}
+			got, err := s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("NGC.FromData() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			for i := range got {
+				if len(got[i].Raw) == 0 {
+					t.Fatalf("no raw secret present: \n %+v", got[i])
+				}
+				got[i].Raw = nil
+			}
+			if diff := pretty.Compare(got, tt.want); diff != "" {
+				t.Errorf("NGC.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			}
+		})
+	}
+}
+
+func BenchmarkFromData(benchmark *testing.B) {
+	ctx := context.Background()
+	s := Scanner{}
+	for name, data := range detectors.MustGetBenchmarkData() {
+		benchmark.Run(name, func(b *testing.B) {
+			for n := 0; n < b.N; n++ {
+				_, err := s.FromData(ctx, false, data)
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
+	}
+}

pkg/pb/detectorspb/detectors.pb.go

@@ -915,6 +915,7 @@ const (
 	DetectorType_Teamup                        DetectorType = 893
 	DetectorType_Workday                       DetectorType = 894
 	DetectorType_MongoDB                       DetectorType = 895
+	DetectorType_NGC                           DetectorType = 896
 )
 
 // Enum value maps for DetectorType.
@@ -1812,6 +1813,7 @@ var (
 		893: "Teamup",
 		894: "Workday",
 		895: "MongoDB",
+		896: "NGC",
 	}
 	DetectorType_value = map[string]int32{
 		"Alibaba":                       0,
@@ -2706,6 +2708,7 @@ var (
 		"Teamup":                        893,
 		"Workday":                       894,
 		"MongoDB":                       895,
+		"NGC":                           896,
 	}
 )
 
@@ -3055,7 +3058,7 @@ var file_detectors_proto_rawDesc = []byte{
 	0x16, 0x70, 0x75, 0x62, 0x6c, 0x69, 0x63, 0x5f, 0x6b, 0x65, 0x79, 0x5f, 0x66, 0x69, 0x6e, 0x67,
 	0x65, 0x72, 0x70, 0x72, 0x69, 0x6e, 0x74, 0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x14, 0x70,
 	0x75, 0x62, 0x6c, 0x69, 0x63, 0x4b, 0x65, 0x79, 0x46, 0x69, 0x6e, 0x67, 0x65, 0x72, 0x70, 0x72,
-	0x69, 0x6e, 0x74, 0x2a, 0x92, 0x70, 0x0a, 0x0c, 0x44, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72,
+	0x69, 0x6e, 0x74, 0x2a, 0x9c, 0x70, 0x0a, 0x0c, 0x44, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72,
 	0x54, 0x79, 0x70, 0x65, 0x12, 0x0b, 0x0a, 0x07, 0x41, 0x6c, 0x69, 0x62, 0x61, 0x62, 0x61, 0x10,
 	0x00, 0x12, 0x08, 0x0a, 0x04, 0x41, 0x4d, 0x51, 0x50, 0x10, 0x01, 0x12, 0x07, 0x0a, 0x03, 0x41,
 	0x57, 0x53, 0x10, 0x02, 0x12, 0x09, 0x0a, 0x05, 0x41, 0x7a, 0x75, 0x72, 0x65, 0x10, 0x03, 0x12,
@@ -3952,11 +3955,12 @@ var file_detectors_proto_rawDesc = []byte{
 	0x06, 0x12, 0x0c, 0x0a, 0x07, 0x45, 0x64, 0x75, 0x73, 0x69, 0x67, 0x6e, 0x10, 0xfc, 0x06, 0x12,
 	0x0b, 0x0a, 0x06, 0x54, 0x65, 0x61, 0x6d, 0x75, 0x70, 0x10, 0xfd, 0x06, 0x12, 0x0c, 0x0a, 0x07,
 	0x57, 0x6f, 0x72, 0x6b, 0x64, 0x61, 0x79, 0x10, 0xfe, 0x06, 0x12, 0x0c, 0x0a, 0x07, 0x4d, 0x6f,
-	0x6e, 0x67, 0x6f, 0x44, 0x42, 0x10, 0xff, 0x06, 0x42, 0x3d, 0x5a, 0x3b, 0x67, 0x69, 0x74, 0x68,
+	0x6e, 0x67, 0x6f, 0x44, 0x42, 0x10, 0xff, 0x06, 0x12, 0x08, 0x0a, 0x03, 0x4e, 0x47, 0x43, 0x10,
-	0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x73, 0x65,
+	0x80, 0x07, 0x42, 0x3d, 0x5a, 0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d,
-	0x63, 0x75, 0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x68, 0x6f,
+	0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75, 0x72, 0x69, 0x74, 0x79,
-	0x67, 0x2f, 0x76, 0x33, 0x2f, 0x70, 0x6b, 0x67, 0x2f, 0x70, 0x62, 0x2f, 0x64, 0x65, 0x74, 0x65,
+	0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f, 0x76, 0x33, 0x2f, 0x70,
-	0x63, 0x74, 0x6f, 0x72, 0x73, 0x70, 0x62, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
+	0x6b, 0x67, 0x2f, 0x70, 0x62, 0x2f, 0x64, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72, 0x73, 0x70,
+	0x62, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
 }
 
 var (

proto/detectors.proto

@@ -897,6 +897,7 @@ enum DetectorType {
   Teamup = 893;
   Workday = 894;
   MongoDB = 895;
+  NGC = 896;
 }
 
 message Result {