added roaring detector

pkg/detectors/roaring/roaring.go

@@ -0,0 +1,87 @@
+package roaring
+
+import (
+	"context"
+	b64 "encoding/base64"
+	"fmt"
+
+	// "log"
+	"regexp"
+	"strings"
+
+	"net/http"
+
+	"github.com/trufflesecurity/trufflehog/pkg/common"
+	"github.com/trufflesecurity/trufflehog/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/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 boundry characters such as below to reduce false positives
+	clientPat = regexp.MustCompile(detectors.PrefixRegex([]string{"roaring"}) + `\b([0-9A-Za-z_-]{28})\b`)
+	secretPat = regexp.MustCompile(detectors.PrefixRegex([]string{"roaring"}) + `\b([0-9A-Za-z_-]{28})\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{"roaring"}
+}
+
+// FromData will find and optionally verify Roaring 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)
+
+	clientMatches := clientPat.FindAllStringSubmatch(dataStr, -1)
+	secretMatches := secretPat.FindAllStringSubmatch(dataStr, -1)
+
+	for _, clientMatch := range clientMatches {
+		if len(clientMatch) != 2 {
+			continue
+		}
+		resClient := strings.TrimSpace(clientMatch[1])
+
+		for _, secretMatch := range secretMatches {
+			if len(secretMatch) != 2 {
+				continue
+			}
+			resSecret := strings.TrimSpace(secretMatch[1])
+
+			s1 := detectors.Result{
+				DetectorType: detectorspb.DetectorType_Roaring,
+				Raw:          []byte(resClient),
+			}
+
+			if verify {
+				data := fmt.Sprintf("%s:%s", resClient, resSecret)
+				sEnc := b64.StdEncoding.EncodeToString([]byte(data))
+				payload := strings.NewReader("grant_type=client_credentials")
+				req, _ := http.NewRequest("POST", "https://api.roaring.io/token", payload)
+				req.Header.Add("Authorization", fmt.Sprintf("Basic %s", sEnc))
+				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(resSecret, detectors.DefaultFalsePositives, true) {
+							continue
+						}
+					}
+				}
+			}
+
+			results = append(results, s1)
+		}
+	}
+
+	return detectors.CleanResults(results), nil
+}

pkg/detectors/roaring/roaring_test.go

@@ -0,0 +1,114 @@
+package roaring
+
+import (
+	"context"
+	"fmt"
+	"testing"
+	"time"
+
+	"github.com/kylelemons/godebug/pretty"
+	"github.com/trufflesecurity/trufflehog/pkg/detectors"
+
+	"github.com/trufflesecurity/trufflehog/pkg/common"
+	"github.com/trufflesecurity/trufflehog/pkg/pb/detectorspb"
+)
+
+func TestRoaring_FromChunk(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetScannersTestSecretsVersion(ctx, 2)
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+	client := testSecrets.MustGetField("ROARING_CLIENT")
+	secret := testSecrets.MustGetField("ROARING_SECRET")
+	inactiveClient := testSecrets.MustGetField("ROARING_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 roaring secret %s within roaring client %s", secret, client)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_Roaring,
+					Verified:     true,
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a roaring client %s within roaring secret %s but not valid", inactiveClient, secret)), // the secret would satisfy the regex but not pass validation
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_Roaring,
+					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("Roaring.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("Roaring.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++ {
+				s.FromData(ctx, false, data)
+			}
+		})
+	}
+}