Merge branch 'feature-bytification' into detectors-36

* fix sentry detector to check response.

* use err.

* address comments.

pkg/detectors/sentrytoken/sentrytoken.go

@@ -2,7 +2,10 @@ package sentrytoken
 
 import (
 	"context"
+	"encoding/json"
+	"errors"
 	"fmt"
+	"io"
 	"net/http"
 	"regexp"
 	"strings"
@@ -12,16 +15,20 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
 
-type Scanner struct{}
+type Scanner struct {
+	client *http.Client
+}
 
 // Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
-	client = common.SaneHttpClient()
+	defaultClient = 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{"sentry"}) + `\b([a-f0-9]{64})\b`)
+
+	errUnauthorized = fmt.Errorf("token unauthorized")
 )
 
 // Keywords are used for efficiently pre-filtering chunks.
@@ -47,32 +54,78 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", "https://sentry.io/api/0/projects/", nil)
-			if err != nil {
-				continue
+			client := s.client
+			if client == nil {
+				client = defaultClient
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
-			// req.Header.Add("Cookie", "sc=g2ceDPcsBwPLGXugHUv0coGBi09TQYKqmtp0u4JYdIvOTFd28yVvn1yiuXgmJJa8; session=.eJwNyE0OwiAQQGGsPwtP4kqoGQa4gUdgZ2CGiaZJCUqjGxOPbnfve7_h29QpHpVSvbw61To9StvEwzre9TkVbkPcr7jN5dPbNl7X1jUt_a4vLpEHgUycQSgFZ0YBRkGxnthxwIyBAgdngcF4YxBssiMjekGv2245_wHyySMG:1m0J5S:FjpRlBVuFx577UQp6GcSGWTq0uk")
-			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
-					}
-				}
+			isVerified, verificationErr := verifyToken(ctx, client, resMatch)
+
+			switch {
+			case errors.Is(verificationErr, errUnauthorized):
+				s1.Verified = false
+			case isVerified:
+				s1.Verified = true
+			default:
+				s1.VerificationError = verificationErr
 			}
 		}
 
+		if !s1.Verified && detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+			continue
+		}
 		results = append(results, s1)
 	}
 
 	return results, nil
 }
 
+type Response []Project
+
+type Project struct {
+	Organization Organization `json:"organization"`
+}
+
+type Organization struct {
+	ID   string `json:"id"`
+	Name string `json:"name"`
+}
+
+func verifyToken(ctx context.Context, client *http.Client, token string) (bool, error) {
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://sentry.io/api/0/projects/", nil)
+	if err != nil {
+		return false, err
+	}
+	req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", token))
+
+	res, err := client.Do(req)
+	if err != nil {
+		return false, err
+	}
+	defer res.Body.Close()
+
+	var isVerified bool
+	switch res.StatusCode {
+	case http.StatusOK, http.StatusForbidden:
+		isVerified = true
+	case http.StatusUnauthorized:
+		return false, errUnauthorized
+	default:
+		return false, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
+	}
+
+	bytes, readErr := io.ReadAll(res.Body)
+	if readErr != nil {
+		return false, readErr
+	}
+
+	var resp Response
+	if err = json.Unmarshal(bytes, &resp); err != nil {
+		return false, err
+	}
+
+	return isVerified, err
+}
+
 func (s Scanner) Type() detectorspb.DetectorType {
 	return detectorspb.DetectorType_SentryToken
 }

pkg/detectors/sentrytoken/sentrytoken_test.go

@@ -9,7 +9,9 @@ import (
 	"testing"
 	"time"
 
-	"github.com/kylelemons/godebug/pretty"
+	"github.com/google/go-cmp/cmp"
+	"github.com/google/go-cmp/cmp/cmpopts"
+
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
@@ -32,11 +34,12 @@ func TestSentryToken_FromChunk(t *testing.T) {
 		verify bool
 	}
 	tests := []struct {
-		name    string
-		s       Scanner
-		args    args
-		want    []detectors.Result
-		wantErr bool
+		name                string
+		s                   Scanner
+		args                args
+		want                []detectors.Result
+		wantErr             bool
+		wantVerificationErr bool
 	}{
 		{
 			name: "found, verified",
@@ -70,6 +73,56 @@ func TestSentryToken_FromChunk(t *testing.T) {
 			},
 			wantErr: false,
 		},
+		{
+			name: "found, would be verified but for timeout",
+			s:    Scanner{client: common.SaneHttpClientTimeOut(1 * time.Microsecond)},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+		{
+			name: "found and valid but unexpected api response",
+			s:    Scanner{client: common.ConstantResponseHttpClient(500, "")},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+		{
+			name: "found, good key but wrong scope",
+			s:    Scanner{client: common.ConstantResponseHttpClient(403, responseBody403)},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     true,
+				},
+			},
+			wantErr: false,
+		},
 		{
 			name: "not found",
 			s:    Scanner{},
@@ -82,27 +135,46 @@ func TestSentryToken_FromChunk(t *testing.T) {
 			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)
+			got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
 			if (err != nil) != tt.wantErr {
-				t.Errorf("SentryToken.FromData() error = %v, wantErr %v", err, 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.Fatalf("no raw secret present: \n %+v", got[i])
+					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].Raw = nil
 			}
-			if diff := pretty.Compare(got, tt.want); diff != "" {
-				t.Errorf("SentryToken.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			opts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "VerificationError")
+			if diff := cmp.Diff(got, tt.want, opts); diff != "" {
+				t.Errorf("Gitlab.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
 			}
 		})
 	}
 }
 
+const responseBody403 = `
+[
+  {
+    "organization": {
+      "id": "911964",
+      "slug": "wigslap",
+      "status": {
+        "id": "active",
+        "name": "active"
+      },
+      "name": "wigslap"
+    }
+  }
+]
+`
+
 func BenchmarkFromData(benchmark *testing.B) {
 	ctx := context.Background()
 	s := Scanner{}

pkg/detectors/sourcegraph/sourcegraph.go

@@ -22,13 +22,13 @@ var _ detectors.Detector = (*Scanner)(nil)
 var (
 	defaultClient = 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{"sourcegraph", "sg_token"}) + `\b(sgp_[a-z0-9]{40})\b`)
+	keyPat = regexp.MustCompile(`\b(sgp_[a-f0-9]{40})\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{"sourcegraph", "sg_token"}
+	return []string{"sgp_"}
 }
 
 // FromData will find and optionally verify Sourcegraph secrets in a given set of bytes.

pkg/detectors/twilio/twilio.go

@@ -28,30 +28,28 @@ var (
 
 // 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{"sid"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("sid")}
 }
 
 // FromData will find and optionally verify Twilio 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)
-
-	identifierMatches := identifierPat.FindAllString(dataStr, -1)
+	identifierMatches := identifierPat.FindAll(data, -1)
 
 	if len(identifierMatches) == 0 {
 		return
 	}
 
-	keyMatches := keyPat.FindAllString(dataStr, -1)
-	sidMatches := sidPat.FindAllString(dataStr, -1)
+	keyMatches := keyPat.FindAll(data, -1)
+	sidMatches := sidPat.FindAll(data, -1)
 
 	for _, sid := range sidMatches {
 		for _, key := range keyMatches {
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Twilio,
-				Raw:          []byte(sid),
-				RawV2:        []byte(sid + key),
-				Redacted:     sid,
+				Raw:          sid,
+				RawV2:        append(append([]byte(nil), sid...), key...),
+				Redacted:     string(sid),
 			}
 
 			if verify {
@@ -67,10 +65,10 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				}
 				req.Header.Add("Content-Type", "application/x-www-form-urlencoded")
 				req.Header.Add("Accept", "*/*")
-				req.SetBasicAuth(sid, key)
+				req.SetBasicAuth(string(sid), string(key))
 				res, err := client.Do(req)
 				if err == nil {
-					res.Body.Close() // The request body is unused.
+					res.Body.Close()
 
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						s1.Verified = true
@@ -84,7 +82,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				}
 			}
 
-			if !s1.Verified && detectors.IsKnownFalsePositive(string(s1.Raw), detectors.DefaultFalsePositives, true) {
+			if !s1.Verified && detectors.IsKnownFalsePositive(s1.Raw, detectors.DefaultFalsePositives, true) {
 				continue
 			}
 

pkg/detectors/twist/twist.go

@@ -1,11 +1,11 @@
 package twist
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -26,30 +26,28 @@ var (
 
 // 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{"twist"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("twist")}
 }
 
 // FromData will find and optionally verify Twist 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 		setAuth := resMatch
 
-		if strings.Contains(match[0], "oauth") {
-			setAuth = fmt.Sprintf("oauth2:%s", strings.TrimSpace(match[1]))
+		if bytes.Contains(match[0], []byte("oauth")) {
+			setAuth = append([]byte("oauth2:"), bytes.TrimSpace(match[1])...)
 		}
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Twist,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -57,7 +55,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", setAuth))
+			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(setAuth)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()

pkg/detectors/twitch/twitch.go

@@ -1,12 +1,10 @@
 package twitch
 
 import (
+	"bytes"
 	"context"
 	"net/http"
-	"net/url"
 	"regexp"
-	"strconv"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -15,68 +13,57 @@ import (
 
 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{"twitch"}) + `\b([0-9a-z]{30})\b`)
 	idPat  = regexp.MustCompile(detectors.PrefixRegex([]string{"twitch"}) + `\b([0-9a-z]{30})\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{"twitch"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("twitch")}
 }
 
-// FromData will find and optionally verify Twitch 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)
-	idMatches := idPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	idMatches := idPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		for _, idMatch := range idMatches {
 			if len(idMatch) != 2 {
 				continue
 			}
 
-			resIdMatch := strings.TrimSpace(idMatch[1])
+			resIdMatch := bytes.TrimSpace(idMatch[1])
 
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Twitch,
-				Raw:          []byte(resMatch),
+				Raw:          resMatch,
 			}
 
 			if verify {
-				data := url.Values{}
-				data.Set("client_id", resIdMatch)
-				data.Set("client_secret", resMatch)
-				data.Set("grant_type", "client_credentials")
-				encodedData := data.Encode()
-				req, err := http.NewRequestWithContext(ctx, "POST", "https://id.twitch.tv/oauth2/token", strings.NewReader(encodedData))
+				data := "client_id=" + string(resIdMatch) + "&client_secret=" + string(resMatch) + "&grant_type=client_credentials"
+				req, err := http.NewRequestWithContext(ctx, "POST", "https://id.twitch.tv/oauth2/token", bytes.NewBufferString(data))
 				if err != nil {
 					continue
 				}
 				req.Header.Add("Content-Type", "application/x-www-form-urlencoded")
-				req.Header.Add("Content-Length", strconv.Itoa(len(data.Encode())))
+				req.Header.Add("Content-Length", string(len(data)))
 				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) {
+						if detectors.IsKnownFalsePositive([]byte(resMatch), detectors.DefaultFalsePositives, true) {
 							continue
 						}
 					}

pkg/detectors/twitter/twitter.go

@@ -1,11 +1,11 @@
 package twitter
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,30 @@ import (
 
 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{"twitter"}) + `\b([A-Z]{22}%[a-zA-Z-0-9]{23}%[a-zA-Z-0-9]{6}%[a-zA-Z-0-9]{3}%[a-zA-Z-0-9]{9}%[a-zA-Z-0-9]{52})\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{"twitter"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("twitter")}
 }
 
-// FromData will find and optionally verify Twitter 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
-
+		resMatch := bytes.TrimSpace(match[1])
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Twitter,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -61,7 +54,11 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				}
 			}
 		}
-
+		if !s1.Verified {
+			if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+				continue
+			}
+		}
 		results = append(results, s1)
 	}
 

pkg/detectors/tyntec/tyntec.go

@@ -1,10 +1,10 @@
 package tyntec
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -25,25 +25,23 @@ var (
 
 // 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{"tyntec"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("tyntec")}
 }
 
 // FromData will find and optionally verify Tyntec 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Tyntec,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -51,7 +49,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("apiKey", resMatch)
+			req.Header.Add("apiKey", string(resMatch))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()

pkg/detectors/typeform/typeform.go

@@ -1,11 +1,11 @@
 package typeform
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -26,25 +26,24 @@ var (
 
 // 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{"typeform"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("typeform")}
 }
 
 // FromData will find and optionally verify Typeform 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Typeform,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,7 +51,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(resMatch)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()

pkg/detectors/typetalk/typetalk.go

@@ -1,11 +1,11 @@
 package typetalk
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -27,36 +27,34 @@ var (
 
 // 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{"typetalk"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("typetalk")}
 }
 
 // FromData will find and optionally verify Typetalk 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)
-	idMatches := idPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	idMatches := idPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		for _, idMatch := range idMatches {
 			if len(idMatch) != 2 {
 				continue
 			}
+			resIdMatch := bytes.TrimSpace(idMatch[1])
 
-			resIdMatch := strings.TrimSpace(idMatch[1])
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Typetalk,
-				Raw:          []byte(resMatch),
+				Raw:          resMatch,
 			}
 
 			if verify {
-				req, err := http.NewRequestWithContext(ctx, "POST", fmt.Sprintf("https://typetalk.com/oauth2/access_token?client_id=%s&client_secret=%s&grant_type=client_credentials", resIdMatch, resMatch), nil)
+				req, err := http.NewRequestWithContext(ctx, "POST", fmt.Sprintf("https://typetalk.com/oauth2/access_token?client_id=%s&client_secret=%s&grant_type=client_credentials", string(resIdMatch), string(resMatch)), nil)
 				if err != nil {
 					continue
 				}
@@ -66,7 +64,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 					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
 						}

pkg/detectors/ubidots/ubidots.go

@@ -1,10 +1,10 @@
 package ubidots
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -25,25 +25,23 @@ var (
 
 // 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{"BBFF-"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("BBFF-")}
 }
 
 // FromData will find and optionally verify Ubidots 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Ubidots,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,14 +50,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				continue
 			}
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("X-Auth-Token", resMatch)
+			req.Header.Add("X-Auth-Token", string(resMatch))
 			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
 					}

pkg/detectors/uclassify/uclassify.go

@@ -1,11 +1,11 @@
 package uclassify
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,54 +14,46 @@ import (
 
 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{"uclassify"}) + `\b([a-z0-9A-Z]{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{"uclassify"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("uclassify")}
 }
 
-// FromData will find and optionally verify Uclassify 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Uclassify,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			payload := strings.NewReader(`{"texts":["I am so happy today"]}`)
-			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.uclassify.com/v1/uClassify/Sentiment/classify", payload)
+			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.uclassify.com/v1/uClassify/Sentiment/classify", bytes.NewReader([]byte(`{"texts":["I am so happy today"]}`)))
 			if err != nil {
 				continue
 			}
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("Authorization", fmt.Sprintf("Token %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Token %s", string(resMatch)))
 			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
 					}
@@ -71,7 +63,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 
 		results = append(results, s1)
 	}
-
 	return results, nil
 }
 

pkg/detectors/unifyid/unifyid.go

@@ -1,11 +1,11 @@
 package unifyid
 
 import (
+	"bytes"
 	"context"
 	"io"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,59 +14,55 @@ import (
 
 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{"unify"}) + `\b([0-9A-Za-z_=-]{44})`)
 )
 
-// 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{"unify"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("unify")}
 }
 
-// FromData will find and optionally verify Unifyid 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_UnifyID,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			payload := strings.NewReader(`{"token": "sample"}`)
+			payload := bytes.NewReader([]byte(`{"token": "sample"}`))
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.unify.id/v1/humandetect/verify", payload)
+
 			if err != nil {
 				continue
 			}
+
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("X-API-Key", resMatch)
+			req.Header.Add("X-API-Key", string(resMatch))
 			res, err := client.Do(req)
+
 			if err == nil {
 				defer res.Body.Close()
 				bodyBytes, err := io.ReadAll(res.Body)
+
 				if err != nil {
 					continue
 				}
-				body := string(bodyBytes)
-				if (res.StatusCode >= 200 && res.StatusCode < 300) || (res.StatusCode == 400 && strings.Contains(body, "invalid token")) {
+
+				if (res.StatusCode >= 200 && res.StatusCode < 300) || (res.StatusCode == 400 && bytes.Contains(bodyBytes, []byte("invalid token"))) {
 					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
 					}

pkg/detectors/unplugg/unplugg.go

@@ -1,10 +1,10 @@
 package unplugg
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -13,54 +13,46 @@ import (
 
 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{"unplu"}) + `\b([a-z0-9]{64})\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{"unplu"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("unplu")}
 }
 
-// FromData will find and optionally verify Unplugg 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Unplugg,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			payload := strings.NewReader(`{"data":[{"timestamp":1466640000,"value":0.27589941523037853},{"timestamp":1466640900,"value":0.4059699097648263}],"forecast_to":1458136800,"callback":"https://yourdomain.com/samplecallback"}`)
+			payload := bytes.NewReader([]byte(`{"data":[{"timestamp":1466640000,"value":0.27589941523037853},{"timestamp":1466640900,"value":0.4059699097648263}],"forecast_to":1458136800,"callback":"https://yourdomain.com/samplecallback"}`))
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.unplu.gg/forecast", payload)
 			if err != nil {
 				continue
 			}
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("x-access-token", resMatch)
+			req.Header.Add("x-access-token", string(resMatch))
 			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
 					}

pkg/detectors/unsplash/unsplash.go

@@ -1,10 +1,11 @@
 package unsplash
 
 import (
+	"bytes"
 	"context"
+	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -19,45 +20,40 @@ 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{"unsplash"}) + `\b([0-9A-Za-z_]{43})\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{"unsplash"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("unsplash")}
 }
 
-// FromData will find and optionally verify Unsplash 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Unsplash,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", "https://api.unsplash.com/photos/?client_id="+resMatch, nil)
+			req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://api.unsplash.com/photos/?client_id=%s", resMatch), nil)
 			if err != nil {
 				continue
 			}
+
 			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
 					}

pkg/detectors/upcdatabase/upcdatabase.go

@@ -1,12 +1,12 @@
 package upcdatabase
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"io"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -26,26 +26,23 @@ var (
 )
 
 // 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{"upcdatabase"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("upcdatabase")}
 }
 
 // FromData will find and optionally verify UPCDatabase 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_UPCDatabase,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -57,8 +54,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err == nil {
 				bodyBytes, err := io.ReadAll(res.Body)
 				if err == nil {
-					bodyString := string(bodyBytes)
-					validResponse := strings.Contains(bodyString, `added_time`)
+					validResponse := bytes.Contains(bodyBytes, []byte(`added_time`))
 					defer res.Body.Close()
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						if validResponse {
@@ -67,16 +63,15 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 							s1.Verified = false
 						}
 					} 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)
+			results = append(results, s1)
+		}
 	}
 
 	return results, nil

pkg/detectors/uplead/uplead.go

@@ -1,10 +1,10 @@
 package uplead
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -13,37 +13,30 @@ import (
 
 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{"uplead"}) + `\b([a-z0-9-]{32})\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{"uplead"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("uplead")}
 }
 
-// FromData will find and optionally verify Uplead 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Uplead,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -51,14 +44,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", resMatch)
+			req.Header.Add("Authorization", string(resMatch))
 			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
 					}

pkg/detectors/uploadcare/uploadcare.go

@@ -1,11 +1,11 @@
 package uploadcare
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,44 +14,38 @@ import (
 
 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{"uploadcare"}) + `\b([a-z0-9]{20})\b`)
 	publicKeyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"uploadcare"}) + `\b([a-z0-9]{20})\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{"uploadcare"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("uploadcare")}
 }
 
-// FromData will find and optionally verify UploadCare 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)
-	publicMatches := publicKeyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	publicMatches := publicKeyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		for _, publicMatch := range publicMatches {
 			if len(publicMatch) != 2 {
 				continue
 			}
-			publicKeyMatch := strings.TrimSpace(publicMatch[1])
+			publicKeyMatch := bytes.TrimSpace(publicMatch[1])
+
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_UploadCare,
-				Raw:          []byte(resMatch),
+				Raw:          resMatch,
 			}
 
 			if verify {
@@ -60,17 +54,14 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 					continue
 				}
 				req.Header.Add("Accept", "application/vnd.uploadcare-v0.5+json")
-				req.Header.Add("Authorization", fmt.Sprintf("Uploadcare.Simple %s:%s", publicKeyMatch, resMatch))
+				req.Header.Add("Authorization", fmt.Sprintf("Uploadcare.Simple %s:%s", string(publicKeyMatch), string(resMatch)))
 				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
-						}
+					} else if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+						continue
 					}
 				}
 			}

pkg/detectors/uptimerobot/uptimerobot.go

@@ -1,11 +1,12 @@
 package uptimerobot
 
 import (
+	"bytes"
 	"context"
+	"fmt"
 	"io"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -26,29 +27,28 @@ var (
 
 // 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{"uptimerobot"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("uptimerobot")}
 }
 
 // FromData will find and optionally verify UptimeRobot 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_UptimeRobot,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.uptimerobot.com/v2/getMonitors?api_key="+resMatch, nil)
+			req, err := http.NewRequestWithContext(ctx, "POST", fmt.Sprintf("https://api.uptimerobot.com/v2/getMonitors?api_key=%s", string(resMatch)), nil)
 			if err != nil {
 				continue
 			}
@@ -58,17 +58,11 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				if err != nil {
 					continue
 				}
-				bodyString := string(bodyBytes)
-				validResponse := strings.Contains(bodyString, `"ok"`)
+				validResponse := bytes.Contains(bodyBytes, []byte(`"ok"`))
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
-					if validResponse {
-						s1.Verified = true
-					} else {
-						s1.Verified = false
-					}
+					s1.Verified = validResponse
 				} 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
 					}

pkg/detectors/upwave/upwave.go

@@ -1,11 +1,11 @@
 package upwave
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,30 @@ import (
 
 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{"upwave"}) + `\b([0-9a-z]{32})\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{"upwave"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("upwave")}
 }
 
-// FromData will find and optionally verify Upwave 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Upwave,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,15 +45,15 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
+
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("Authorization", fmt.Sprintf("Token %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Token %s", string(resMatch)))
 			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
 					}

pkg/detectors/uri/uri.go

@@ -1,6 +1,7 @@
 package uri
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"net/url"
@@ -26,39 +27,32 @@ var (
 	client = common.SaneHttpClient()
 )
 
-// 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{"http"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("http")}
 }
 
-// FromData will find and optionally verify URI 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
+		// do not convert the match[0] into a string.
 
 		if !s.allowKnownTestSites {
-			if strings.Contains(match[0], "httpbin.org") {
+			if bytes.Contains(match[0], []byte("httpbin.org")) {
 				continue
 			}
-			if strings.Contains(match[0], "httpwatch.com") {
+
+			if bytes.Contains(match[0], []byte("httpwatch.com")) {
 				continue
 			}
 		}
 
-		urlMatch := match[0]
 		password := match[1]
-
-		// Skip findings where the password only has "*" characters, this is a redacted password
-		// Also include the url encoded "*" characters: "%2A"
-		if strings.Trim(password, "*") == "" || strings.Trim(password, "%2A") == "" {
+		if bytes.Trim(password, "*") == nil || bytes.Trim(password, "%2A") == nil {
 			continue
 		}
 
-		parsedURL, err := url.Parse(urlMatch)
+		parsedURL, err := url.Parse(string(match[0]))
 		if err != nil {
 			continue
 		}
@@ -66,15 +60,14 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			continue
 		}
 
-		rawURL, _ := url.Parse(urlMatch)
-		rawURLStr := rawURL.String()
-		// Removing the path causes possible deduplication issues if some paths have basic auth and some do not.
+		rawURL, _ := url.Parse(string(match[0]))
+
 		rawURL.Path = ""
 
 		s := detectors.Result{
 			DetectorType: detectorspb.DetectorType_URI,
 			Raw:          []byte(rawURL.String()),
-			RawV2:        []byte(rawURLStr),
+			RawV2:        match[0],
 			Redacted:     detectors.RedactURL(*rawURL),
 		}
 
@@ -82,14 +75,11 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			s.Verified = verifyURL(ctx, parsedURL)
 		}
 
-		if !s.Verified {
-			// Skip unverified findings where the password starts with a `$` - it's almost certainly a variable.
-			if strings.HasPrefix(password, "$") {
-				continue
-			}
+		if bytes.HasPrefix(password, []byte("$")) {
+			continue
 		}
 
-		if !s.Verified && detectors.IsKnownFalsePositive(string(s.Raw), detectors.DefaultFalsePositives, false) {
+		if !s.Verified && detectors.IsKnownFalsePositive(s.Raw, detectors.DefaultFalsePositives, false) {
 			continue
 		}
 

pkg/detectors/urlscan/urlscan.go

@@ -1,10 +1,10 @@
 package urlscan
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -18,32 +18,24 @@ 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{"urlscan"}) + `\b([a-z0-9-]{36})\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{"urlscan"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("urlscan")}
 }
 
-// FromData will find and optionally verify Urlscan 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)
-
+	matches := keyPat.FindAllSubmatch(data, -1)
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Urlscan,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -51,18 +43,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("API-Key", resMatch)
+			req.Header.Add("API-Key", string(resMatch))
 			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
-					}
-				}
+
+			if err == nil && res.StatusCode >= 200 && res.StatusCode < 300 {
+				s1.Verified = true
+			} else if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+				continue
 			}
 		}
 

pkg/detectors/user/user.go

@@ -1,11 +1,11 @@
 package user
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,29 @@ import (
 
 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{"user"}) + `\b([a-zA-Z0-9-._+=]{64})\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{"user"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("user")}
 }
 
-// FromData will find and optionally verify User 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_User,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,14 +44,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Token %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Token %s", string(resMatch)))
 			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
 					}

pkg/giturl/giturl.go

@@ -132,9 +132,9 @@ func GenerateLink(repo, commit, file string, line int64) string {
 		return baseLink
 
 	case providerAzure:
-		baseLink := repo + "?path=" + file + "&version=GB" + commit
+		baseLink := repo + "/commit/" + commit + "/" + file
 		if line > 0 {
-			baseLink += "&line=" + strconv.FormatInt(line, 10)
+			baseLink += "?line=" + strconv.FormatInt(line, 10)
 		}
 		return baseLink
 

pkg/giturl/giturl_test.go

@@ -148,7 +148,7 @@ func TestGenerateLink(t *testing.T) {
 				commit: "abcdef",
 				file:   "main.go",
 			},
-			want: "https://dev.azure.com/org/project/_git/repo?path=main.go&version=GBabcdef",
+			want: "https://dev.azure.com/org/project/_git/repo/commit/abcdef/main.go",
 		},
 		{
 			name: "Azure link gen with line",
@@ -158,7 +158,7 @@ func TestGenerateLink(t *testing.T) {
 				file:   "main.go",
 				line:   int64(20),
 			},
-			want: "https://dev.azure.com/org/project/_git/repo?path=main.go&version=GBabcdef&line=20",
+			want: "https://dev.azure.com/org/project/_git/repo/commit/abcdef/main.go?line=20",
 		},
 	}
 	for _, tt := range tests {