add extra data to github detector

* [chore] Add SourceUnitEnumChunker filesystem tests

* Ensure reported units are exactly what is expected

main.go

@@ -152,6 +152,9 @@ func init() {
 
 	cli.Version("trufflehog " + version.BuildVersion)
 
+	//Support -h for help
+	cli.HelpFlag.Short('h')
+
 	if len(os.Args) <= 1 && isatty.IsTerminal(os.Stdout.Fd()) {
 		args := tui.Run()
 		if len(args) == 0 {

pkg/decoders/base64.go

@@ -4,7 +4,6 @@ import (
 	"bytes"
 	"encoding/base64"
 
-	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
 )
 
@@ -26,8 +25,7 @@ func init() {
 	}
 }
 
-func (d *Base64) FromChunk(chunk *sources.Chunk) *DecodableChunk {
-	decodableChunk := &DecodableChunk{Chunk: chunk, DecoderType: detectorspb.DecoderType_BASE64}
+func (d *Base64) FromChunk(chunk *sources.Chunk) *sources.Chunk {
 	encodedSubstrings := getSubstringsOfCharacterSet(chunk.Data, 20, b64CharsetMapping, b64EndChars)
 	decodedSubstrings := make(map[string][]byte)
 
@@ -63,7 +61,7 @@ func (d *Base64) FromChunk(chunk *sources.Chunk) *DecodableChunk {
 		}
 		result.Write(chunk.Data[start:])
 		chunk.Data = result.Bytes()
-		return decodableChunk
+		return chunk
 	}
 
 	return nil

pkg/decoders/decoders.go

@@ -1,7 +1,6 @@
 package decoders
 
 import (
-	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
 )
 
@@ -14,15 +13,8 @@ func DefaultDecoders() []Decoder {
 	}
 }
 
-// DecodableChunk is a chunk that includes the type of decoder used.
-// This allows us to avoid a type assertion on each decoder.
-type DecodableChunk struct {
-	*sources.Chunk
-	DecoderType detectorspb.DecoderType
-}
-
 type Decoder interface {
-	FromChunk(chunk *sources.Chunk) *DecodableChunk
+	FromChunk(chunk *sources.Chunk) *sources.Chunk
 }
 
 // Fuzz is an entrypoint for go-fuzz, which is an AFL-style fuzzing tool.

pkg/decoders/utf16.go

@@ -5,24 +5,22 @@ import (
 	"encoding/binary"
 	"unicode/utf8"
 
-	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
 )
 
 type UTF16 struct{}
 
-func (d *UTF16) FromChunk(chunk *sources.Chunk) *DecodableChunk {
+func (d *UTF16) FromChunk(chunk *sources.Chunk) *sources.Chunk {
 	if chunk == nil || len(chunk.Data) == 0 {
 		return nil
 	}
 
-	decodableChunk := &DecodableChunk{Chunk: chunk, DecoderType: detectorspb.DecoderType_UTF16}
 	if utf16Data, err := utf16ToUTF8(chunk.Data); err == nil {
 		if len(utf16Data) == 0 {
 			return nil
 		}
 		chunk.Data = utf16Data
-		return decodableChunk
+		return chunk
 	}
 
 	return nil

pkg/decoders/utf8.go

@@ -4,25 +4,22 @@ import (
 	"bytes"
 	"unicode/utf8"
 
-	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
 )
 
 type UTF8 struct{}
 
-func (d *UTF8) FromChunk(chunk *sources.Chunk) *DecodableChunk {
+func (d *UTF8) FromChunk(chunk *sources.Chunk) *sources.Chunk {
 	if chunk == nil || len(chunk.Data) == 0 {
 		return nil
 	}
 
-	decodableChunk := &DecodableChunk{Chunk: chunk, DecoderType: detectorspb.DecoderType_PLAIN}
-
 	if !utf8.Valid(chunk.Data) {
 		chunk.Data = extractSubstrings(chunk.Data)
-		return decodableChunk
+		return chunk
 	}
 
-	return decodableChunk
+	return chunk
 }
 
 // extractSubstrings performs similarly to the strings binutil,

pkg/detectors/airtableapikey/airtableapikey.go

@@ -2,6 +2,7 @@ package airtableapikey
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
 	"net/http"
 	"regexp"
@@ -20,8 +21,9 @@ var _ detectors.Detector = (*Scanner)(nil)
 var (
 	client = common.SaneHttpClient()
 
-	appPat = regexp.MustCompile(`(app[a-zA-Z0-9_-]{14})`) // could be part of url
-	keyPat = regexp.MustCompile(`\b(key[a-zA-Z0-9_-]{14})\b`)
+	appPat      = regexp.MustCompile(`(app[a-zA-Z0-9_-]{14})`) // could be part of url
+	keyPat      = regexp.MustCompile(`\b(key[a-zA-Z0-9_-]{14})\b`)
+	personalPat = regexp.MustCompile(`(\bpat[[:alnum:]]{14}\.[[:alnum:]]{64}\b)`)
 )
 
 // Keywords are used for efficiently pre-filtering chunks.
@@ -30,17 +32,31 @@ func (s Scanner) Keywords() []string {
 	return []string{"airtable"}
 }
 
+type response struct {
+	Error struct {
+		Type    string `json:"type"`
+		Message string `json:"message"`
+	} `json:"error"`
+}
+
 // FromData will find and optionally verify AirtableApiKey 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)
 
 	appMatches := appPat.FindAllStringSubmatch(dataStr, -1)
 	keyMatches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	personalKeyMatches := personalPat.FindAllStringSubmatch(dataStr, -1)
+
+	if len(keyMatches) == 0 {
+		keyMatches = personalKeyMatches
+
+	}
 
 	for _, keyMatch := range keyMatches {
 		if len(keyMatch) != 2 {
 			continue
 		}
+
 		keyRes := strings.TrimSpace(keyMatch[1])
 
 		for _, appMatch := range appMatches {
@@ -67,6 +83,15 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 					defer res.Body.Close()
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						s1.Verified = true
+					} else if res.StatusCode == 403 {
+						var resp response
+						if err = json.NewDecoder(res.Body).Decode(&resp); err == nil {
+							// check if the error is due to invalid permissions or model not found
+							if resp.Error.Type == "INVALID_PERMISSIONS_OR_MODEL_NOT_FOUND" {
+								// The key is verified as it works, but the user must enumerate the tables or permissions for the key.
+								s1.Verified = true
+							}
+						}
 					} else {
 						if detectors.IsKnownFalsePositive(keyRes, detectors.DefaultFalsePositives, true) {
 							continue

pkg/detectors/airtableapikey/airtableapikey_test.go

@@ -9,9 +9,10 @@ import (
 	"testing"
 	"time"
 
-	"github.com/kylelemons/godebug/pretty"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
 
+	"github.com/google/go-cmp/cmp"
+	"github.com/google/go-cmp/cmp/cmpopts"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
@@ -99,8 +100,9 @@ func TestAirtableApiKey_FromChunk(t *testing.T) {
 				}
 				got[i].Raw = nil
 			}
-			if diff := pretty.Compare(got, tt.want); diff != "" {
-				t.Errorf("AirtableApiKey.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			ignoreOpts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "RawV2", "VerificationError")
+			if diff := cmp.Diff(got, tt.want, ignoreOpts); diff != "" {
+				t.Errorf("Airtable.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
 			}
 		})
 	}

pkg/detectors/falsepositives_test.go

@@ -74,6 +74,13 @@ func TestStringShannonEntropy(t *testing.T) {
 			},
 			want: 0.22228483068568816,
 		},
+		{
+			name: "empty",
+			args: args{
+				input: "",
+			},
+			want: 0.0,
+		},
 	}
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {

pkg/detectors/github/github.go

@@ -97,6 +97,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 							s1.ExtraData["site_admin"] = fmt.Sprintf("%t", userResponse.SiteAdmin)
 							s1.ExtraData["name"] = userResponse.Name
 							s1.ExtraData["company"] = userResponse.Company
+							s1.ExtraData["scopes"] = res.Header.Get("X-OAuth-Scopes")
 						}
 					}
 				}

pkg/detectors/github/github_test.go

@@ -28,6 +28,7 @@ func TestGitHub_FromChunk(t *testing.T) {
 	unverifiedGhu := testSecrets.MustGetField("GITHUB_UNVERIFIED_GHU")
 	unverifiedGhs := testSecrets.MustGetField("GITHUB_UNVERIFIED_GHS")
 	unverifiedGhr := testSecrets.MustGetField("GITHUB_UNVERIFIED_GHR")
+	verifiedGhp := testSecrets.MustGetField("GITHUB_VERIFIED_GHP")
 
 	type args struct {
 		ctx    context.Context
@@ -41,6 +42,32 @@ func TestGitHub_FromChunk(t *testing.T) {
 		want    []detectors.Result
 		wantErr bool
 	}{
+		{
+			name: "found, verified ghp",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a github secret %s within", verifiedGhp)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_Github,
+					Verified:     true,
+					ExtraData: map[string]string{
+						"account_type":   "User",
+						"company":        "",
+						"name":           "",
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+						"scopes":         "notifications",
+						"site_admin":     "false",
+						"url":            "https://github.com/truffle-sandbox",
+						"username":       "truffle-sandbox",
+					},
+				},
+			},
+			wantErr: false,
+		},
 		{
 			name: "found, unverified ghp",
 			s:    Scanner{},
@@ -53,6 +80,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,
@@ -69,6 +99,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,
@@ -85,6 +118,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,
@@ -101,6 +137,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,
@@ -117,6 +156,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,
@@ -133,6 +175,9 @@ func TestGitHub_FromChunk(t *testing.T) {
 				{
 					DetectorType: detectorspb.DetectorType_Github,
 					Verified:     false,
+					ExtraData: map[string]string{
+						"rotation_guide": "https://howtorotate.com/docs/tutorials/github/",
+					},
 				},
 			},
 			wantErr: false,

pkg/detectors/github_old/github_old.go

@@ -38,6 +38,7 @@ type userRes struct {
 	SiteAdmin bool   `json:"site_admin"`
 	Name      string `json:"name"`
 	Company   string `json:"company"`
+	UserURL   string `json:"html_url"`
 }
 
 // Keywords are used for efficiently pre-filtering chunks.
@@ -91,6 +92,17 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 						res.Body.Close()
 						if err == nil {
 							s1.Verified = true
+
+							if err == nil {
+								s1.Verified = true
+								s1.ExtraData["username"] = userResponse.Login
+								s1.ExtraData["url"] = userResponse.UserURL
+								s1.ExtraData["account_type"] = userResponse.Type
+								s1.ExtraData["site_admin"] = fmt.Sprintf("%t", userResponse.SiteAdmin)
+								s1.ExtraData["name"] = userResponse.Name
+								s1.ExtraData["company"] = userResponse.Company
+								s1.ExtraData["scopes"] = res.Header.Get("X-OAuth-Scopes")
+							}
 						}
 					}
 				}

pkg/detectors/sourcegraphcody/sourcegraphcody.go

@@ -0,0 +1,91 @@
+package sourcegraphcody
+
+import (
+	"context"
+	"fmt"
+	"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 {
+	client *http.Client
+}
+
+// Ensure the Scanner satisfies the interface at compile time.
+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(`\b(slk_[a-f0-9]{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{"slk_"}
+}
+
+// FromData will find and optionally verify Sourcegraphcody 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 {
+		if len(match) != 2 {
+			continue
+		}
+		resMatch := strings.TrimSpace(match[1])
+
+		s1 := detectors.Result{
+			DetectorType: detectorspb.DetectorType_SourcegraphCody,
+			Raw:          []byte(resMatch),
+		}
+
+		if verify {
+			client := s.client
+			if client == nil {
+				client = defaultClient
+			}
+			req, err := http.NewRequestWithContext(ctx, "GET", "https://cody-gateway.sourcegraph.com/v1/limits", nil)
+			req.Header.Add("Authorization", "Bearer "+resMatch)
+			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 if res.StatusCode == 403 || res.StatusCode == 401 {
+					// 403 when the gateway is not configured to accept the token,
+					// 401 when the token is invalid.
+					// The secret is determinately not verified (nothing to do)
+				} else {
+					s1.VerificationError = fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
+				}
+			} else {
+				s1.VerificationError = err
+			}
+		}
+
+		// 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 !s1.Verified && detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+			continue
+		}
+
+		results = append(results, s1)
+	}
+
+	return results, nil
+}
+
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_SourcegraphCody
+}

pkg/detectors/sourcegraphcody/sourcegraphcody_test.go

@@ -0,0 +1,162 @@
+//go:build detectors
+// +build detectors
+
+package sourcegraphcody
+
+import (
+	"context"
+	"fmt"
+	"testing"
+	"time"
+
+	"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"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+func TestSourcegraphcody_FromChunk(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors5")
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+	secret := testSecrets.MustGetField("SOURCEGRAPHCODY")
+	inactiveSecret := testSecrets.MustGetField("SOURCEGRAPHCODY_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: "found, verified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sourcegraphcody secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SourcegraphCody,
+					Verified:     true,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sourcegraphcody 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_SourcegraphCody,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: 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,
+			wantVerificationErr: false,
+		},
+		{
+			name: "found, would be verified if not for timeout",
+			s:    Scanner{client: common.SaneHttpClientTimeOut(1 * time.Microsecond)},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sourcegraphcody secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SourcegraphCody,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+		{
+			name: "found, verified but unexpected api surface",
+			s:    Scanner{client: common.ConstantResponseHttpClient(404, "")},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sourcegraphcody secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SourcegraphCody,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("Sourcegraphcody.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])
+				}
+				if (got[i].VerificationError != nil) != tt.wantVerificationErr {
+					t.Fatalf("wantVerificationError = %v, verification error = %v", tt.wantVerificationErr, got[i].VerificationError)
+				}
+			}
+			ignoreOpts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "VerificationError")
+			if diff := cmp.Diff(got, tt.want, ignoreOpts); diff != "" {
+				t.Errorf("Sourcegraphcody.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) {
+			b.ResetTimer()
+			for n := 0; n < b.N; n++ {
+				_, err := s.FromData(ctx, false, data)
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
+	}
+}

pkg/detectors/twilio/twilio.go

@@ -23,7 +23,6 @@ var (
 	identifierPat = regexp.MustCompile(`(?i)sid.{0,20}AC[0-9a-f]{32}`) // Should we have this? Seems restrictive.
 	sidPat        = regexp.MustCompile(`\bAC[0-9a-f]{32}\b`)
 	keyPat        = regexp.MustCompile(`\b[0-9a-f]{32}\b`)
-	client        = common.SaneHttpClient()
 )
 
 // Keywords are used for efficiently pre-filtering chunks.
@@ -59,7 +58,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			}
 
 			if verify {
-				client = s.client
+				client := s.client
 				if client == nil {
 					client = defaultClient
 				}

pkg/engine/ahocorasickcore.go

@@ -1,112 +0,0 @@
-package engine
-
-import (
-	"strings"
-
-	ahocorasick "github.com/BobuSumisu/aho-corasick"
-
-	"github.com/trufflesecurity/trufflehog/v3/pkg/context"
-	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
-	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
-)
-
-// detectorKey is used to identify a detector in the keywordsToDetectors map.
-// Multiple detectors can have the same detector type but different versions.
-// This allows us to identify a detector by its type and version.
-type detectorKey struct {
-	detectorType detectorspb.DetectorType
-	version      int
-}
-
-// detectorInfo is used to store a detector and whether it should be verified.
-type detectorInfo struct {
-	detectors.Detector
-	shouldVerify bool
-}
-
-// AhoCorasickCore encapsulates the operations and data structures used for keyword matching via the
-// Aho-Corasick algorithm. It is responsible for constructing and managing the trie for efficient
-// substring searches, as well as mapping keywords to their associated detectors for rapid lookups.
-type AhoCorasickCore struct {
-	// prefilter is a ahocorasick struct used for doing efficient string
-	// matching given a set of words. (keywords from the rules in the config)
-	prefilter ahocorasick.Trie
-	// Maps for efficient lookups during detection.
-	detectorTypeToDetectorInfo map[detectorKey]detectorInfo
-	detectors                  map[bool][]detectors.Detector
-	keywordsToDetectors        map[string][]detectorKey
-}
-
-// NewAhoCorasickCore allocates and initializes a new instance of AhoCorasickCore.
-// It creates an empty keyword-to-detectors map for future string matching operations.
-// The map detectorTypeToDetectorInfo is pre-allocated based on the size of detectors
-// provided, for efficient storage and lookup of detector information.
-func NewAhoCorasickCore(detectors map[bool][]detectors.Detector) *AhoCorasickCore {
-	return &AhoCorasickCore{
-		keywordsToDetectors:        make(map[string][]detectorKey),
-		detectors:                  detectors,
-		detectorTypeToDetectorInfo: make(map[detectorKey]detectorInfo, len(detectors[true])+len(detectors[false])),
-	}
-}
-
-// Setup initializes the internal state of AhoCorasickCore to prepare it for keyword matching.
-// This involves pre-filtering setup and lookup optimization, critical for the engine's performance.
-func (ac *AhoCorasickCore) Setup(ctx context.Context) {
-	// Prepare maps for fast detector lookups, instead of scanning through an array of detectors for every chunk.
-	var keywords []string
-	for verify, detectorsSet := range ac.detectors {
-		for _, d := range detectorsSet {
-			key := createDetectorKey(d)
-			ac.detectorTypeToDetectorInfo[key] = detectorInfo{Detector: d, shouldVerify: verify}
-			keywords = ac.extractAndMapKeywords(d, key, keywords)
-		}
-	}
-
-	// Use the Ahocorasick algorithm to create a trie structure for efficient keyword matching.
-	// This ensures that we can rapidly match against a vast set of keywords without individually comparing each one.
-	ac.prefilter = *ahocorasick.NewTrieBuilder().AddStrings(keywords).Build()
-	ctx.Logger().V(4).Info("AhoCorasickCore lookups built")
-	ctx.Logger().V(4).Info("AhoCorasickCore Setup complete")
-}
-
-// createDetectorKey creates a unique key for each detector. This key based on type and version,
-// it ensures faster lookups and reduces redundancy in our main detector store.
-func createDetectorKey(d detectors.Detector) detectorKey {
-	detectorType := d.Type()
-	var version int
-	if v, ok := d.(detectors.Versioner); ok {
-		version = v.Version()
-	}
-	return detectorKey{detectorType: detectorType, version: version}
-}
-
-// extractAndMapKeywords captures keywords associated with each detector and maps them.
-// This allows us to quickly determine which detectors are relevant based on the presence of certain keywords.
-func (ac *AhoCorasickCore) extractAndMapKeywords(d detectors.Detector, key detectorKey, keywords []string) []string {
-	for _, kw := range d.Keywords() {
-		kwLower := strings.ToLower(kw)
-		keywords = append(keywords, kwLower)
-		ac.keywordsToDetectors[kwLower] = append(ac.keywordsToDetectors[kwLower], key)
-	}
-	return keywords
-}
-
-// MatchString performs a string match using the Aho-Corasick algorithm, returning an array of matches.
-// Designed for internal use within the AhoCorasickCore component.
-func (ac *AhoCorasickCore) MatchString(input string) []*ahocorasick.Match {
-	return ac.prefilter.MatchString(strings.ToLower(input))
-}
-
-// PopulateDetectorsByMatch populates the given detectorMap based on the Aho-Corasick match results.
-// This method is designed to reuse the same map for performance optimization,
-// reducing the need for repeated allocations within each detector worker in the engine.
-func (ac *AhoCorasickCore) PopulateDetectorsByMatch(match *ahocorasick.Match, detectors map[detectorspb.DetectorType]detectorInfo) bool {
-	matchedKeys, ok := ac.keywordsToDetectors[match.MatchString()]
-	if !ok {
-		return false
-	}
-	for _, key := range matchedKeys {
-		detectors[key.detectorType] = ac.detectorTypeToDetectorInfo[key]
-	}
-	return true
-}

pkg/engine/defaults.go

@@ -8,6 +8,7 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/salesforce"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/snowflake"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/sourcegraph"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/sourcegraphcody"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/tailscale"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/trufflehogenterprise"
 
@@ -1552,6 +1553,7 @@ func DefaultDetectors() []detectors.Detector {
 		web3storage.Scanner{},
 		&ramp.Scanner{},
 		&anthropic.Scanner{},
+		&sourcegraphcody.Scanner{},
 	}
 
 }

pkg/engine/engine.go

@@ -3,11 +3,14 @@ package engine
 import (
 	"bytes"
 	"fmt"
+	"reflect"
 	"runtime"
+	"strings"
 	"sync"
 	"sync/atomic"
 	"time"
 
+	ahocorasick "github.com/BobuSumisu/aho-corasick"
 	lru "github.com/hashicorp/golang-lru"
 	"google.golang.org/protobuf/proto"
 
@@ -63,8 +66,9 @@ type Engine struct {
 	onlyVerified         bool
 	printAvgDetectorTime bool
 
-	// ahoCorasickHandler manages the Aho-Corasick trie and related keyword lookups.
-	ahoCorasickCore *AhoCorasickCore
+	// prefilter is a ahocorasick struct used for doing efficient string
+	// matching given a set of words (keywords from the rules in the config)
+	prefilter ahocorasick.Trie
 
 	// Engine synchronization primitives.
 	sourceManager        *sources.SourceManager
@@ -89,10 +93,9 @@ type Engine struct {
 	dedupeCache *lru.Cache
 }
 
-// Option is used to configure the engine during initialization using functional options.
-type Option func(*Engine)
+type EngineOption func(*Engine)
 
-func WithConcurrency(concurrency uint8) Option {
+func WithConcurrency(concurrency uint8) EngineOption {
 	return func(e *Engine) {
 		e.concurrency = concurrency
 	}
@@ -100,7 +103,7 @@ func WithConcurrency(concurrency uint8) Option {
 
 const ignoreTag = "trufflehog:ignore"
 
-func WithDetectors(verify bool, d ...detectors.Detector) Option {
+func WithDetectors(verify bool, d ...detectors.Detector) EngineOption {
 	return func(e *Engine) {
 		if e.detectors == nil {
 			e.detectors = make(map[bool][]detectors.Detector)
@@ -113,7 +116,7 @@ func WithDetectors(verify bool, d ...detectors.Detector) Option {
 	}
 }
 
-func WithDecoders(decoders ...decoders.Decoder) Option {
+func WithDecoders(decoders ...decoders.Decoder) EngineOption {
 	return func(e *Engine) {
 		e.decoders = decoders
 	}
@@ -121,14 +124,14 @@ func WithDecoders(decoders ...decoders.Decoder) Option {
 
 // WithFilterUnverified sets the filterUnverified flag on the engine. If set to
 // true, the engine will only return the first unverified result for a chunk for a detector.
-func WithFilterUnverified(filter bool) Option {
+func WithFilterUnverified(filter bool) EngineOption {
 	return func(e *Engine) {
 		e.filterUnverified = filter
 	}
 }
 
 // WithFilterEntropy filters out unverified results using Shannon entropy.
-func WithFilterEntropy(entropy float64) Option {
+func WithFilterEntropy(entropy float64) EngineOption {
 	return func(e *Engine) {
 		if entropy > 0 {
 			e.filterEntropy = &entropy
@@ -138,7 +141,7 @@ func WithFilterEntropy(entropy float64) Option {
 
 // WithOnlyVerified sets the onlyVerified flag on the engine. If set to true,
 // the engine will only print verified results.
-func WithOnlyVerified(onlyVerified bool) Option {
+func WithOnlyVerified(onlyVerified bool) EngineOption {
 	return func(e *Engine) {
 		e.onlyVerified = onlyVerified
 	}
@@ -150,7 +153,7 @@ func WithOnlyVerified(onlyVerified bool) Option {
 // the engine is configured to print the results.
 // Calculating the average time taken by each detector is an expensive operation
 // and should be avoided unless specified by the user.
-func WithPrintAvgDetectorTime(printAvgDetectorTime bool) Option {
+func WithPrintAvgDetectorTime(printAvgDetectorTime bool) EngineOption {
 	return func(e *Engine) {
 		e.printAvgDetectorTime = printAvgDetectorTime
 	}
@@ -160,7 +163,7 @@ func WithPrintAvgDetectorTime(printAvgDetectorTime bool) Option {
 // the filterFunc returns true, the detector will be included for scanning.
 // This option applies to the existing list of detectors configured, so the
 // order this option appears matters. All filtering happens before scanning.
-func WithFilterDetectors(filterFunc func(detectors.Detector) bool) Option {
+func WithFilterDetectors(filterFunc func(detectors.Detector) bool) EngineOption {
 	return func(e *Engine) {
 		// If no detectors are configured, do nothing.
 		if e.detectors == nil {
@@ -172,7 +175,7 @@ func WithFilterDetectors(filterFunc func(detectors.Detector) bool) Option {
 }
 
 // WithPrinter sets the Printer on the engine.
-func WithPrinter(printer Printer) Option {
+func WithPrinter(printer Printer) EngineOption {
 	return func(e *Engine) {
 		e.printer = printer
 	}
@@ -266,60 +269,32 @@ func (e *Engine) DetectorAvgTime() map[string][]time.Duration {
 	return avgTime
 }
 
-// Start initializes and activates the engine's processing pipeline.
-// It sets up various default configurations, prepares lookup structures for
-// detectors, conducts basic sanity checks, and kickstarts all necessary workers.
-// Once started, the engine begins processing input data to identify secrets.
-func Start(ctx context.Context, options ...Option) (*Engine, error) {
-	e := &Engine{}
-
-	if err := e.initialize(ctx, options...); err != nil {
-		return nil, err
-	}
-	e.setDefaults(ctx)
-	ctx.Logger().V(4).Info("setting up aho-corasick core")
-	e.ahoCorasickCore.Setup(ctx)
-	e.sanityChecks(ctx)
-	e.startWorkers(ctx)
-
-	return e, nil
-}
-
-const defaultChannelBuffer = 1
-
-// initialize prepares the engine's internal structures. The LRU cache optimizes
-// deduplication efforts, allowing the engine to quickly check if a chunk has
-// been processed before, thereby saving computational overhead.
-func (e *Engine) initialize(ctx context.Context, options ...Option) error {
-	// TODO (ahrav): Determine the optimal cache size.
-	const cacheSize = 512 // number of entries in the LRU cache
+// Start the engine with options.
+func Start(ctx context.Context, options ...EngineOption) (*Engine, error) {
+	const (
+		defaultChannelBuffer = 1
+		// TODO (ahrav): Determine the optimal cache size.
+		cacheSize = 512 // number of entries in the LRU cache
+	)
 
 	cache, err := lru.New(cacheSize)
 	if err != nil {
-		return fmt.Errorf("failed to initialize LRU cache: %w", err)
+		return nil, fmt.Errorf("failed to initialize LRU cache: %w", err)
 	}
 
-	// Channels are used for communication between different parts of the engine,
-	// ensuring that data flows smoothly without race conditions.
-	e.detectableChunksChan = make(chan detectableChunk, defaultChannelBuffer)
-	e.results = make(chan detectors.ResultWithMetadata, defaultChannelBuffer)
-	e.dedupeCache = cache
-	e.printer = new(output.PlainPrinter)
-	e.metrics = runtimeMetrics{Metrics: Metrics{scanStartTime: time.Now()}}
+	e := &Engine{
+		detectableChunksChan: make(chan detectableChunk, defaultChannelBuffer),
+		results:              make(chan detectors.ResultWithMetadata, defaultChannelBuffer),
+		dedupeCache:          cache,
+		printer:              new(output.PlainPrinter), // default printer
+		metrics:              runtimeMetrics{Metrics: Metrics{scanStartTime: time.Now()}},
+	}
 
 	for _, option := range options {
 		option(e)
 	}
-	ctx.Logger().V(4).Info("engine initialized")
-	e.ahoCorasickCore = NewAhoCorasickCore(e.detectors)
 
-	return nil
-}
-
-// setDefaults ensures that if specific engine properties aren't provided,
-// they're set to reasonable default values. It makes the engine robust to
-// incomplete configuration.
-func (e *Engine) setDefaults(ctx context.Context) {
+	// Set defaults.
 	if e.concurrency == 0 {
 		numCPU := runtime.NumCPU()
 		ctx.Logger().Info("No concurrency specified, defaulting to max", "cpu", numCPU)
@@ -327,13 +302,12 @@ func (e *Engine) setDefaults(ctx context.Context) {
 	}
 	ctx.Logger().V(3).Info("engine started", "workers", e.concurrency)
 
+	// Create SourceManager.
 	e.sourceManager = sources.NewManager(
 		sources.WithConcurrentSources(int(e.concurrency)),
 		sources.WithConcurrentUnits(int(e.concurrency)),
-		sources.WithBufferedOutput(defaultChannelBuffer),
 	)
 
-	// Default decoders handle common encoding formats.
 	if len(e.decoders) == 0 {
 		e.decoders = decoders.DefaultDecoders()
 	}
@@ -343,30 +317,46 @@ func (e *Engine) setDefaults(ctx context.Context) {
 		e.detectors[true] = DefaultDetectors()
 		e.detectors[false] = []detectors.Detector{}
 	}
-	ctx.Logger().V(4).Info("default engine options set")
-}
 
-// Sanity check detectors for duplicate configuration. Only log in case
-// a detector has been configured in a way that isn't represented by
-// the DetectorID (type and version).
-func (e *Engine) sanityChecks(ctx context.Context) {
-	dets := append(e.detectors[true], e.detectors[false]...)
-	seenDetectors := make(map[config.DetectorID]struct{}, len(dets))
-	for _, det := range dets {
-		id := config.GetDetectorID(det)
-		if _, ok := seenDetectors[id]; ok && id.ID != detectorspb.DetectorType_CustomRegex {
-			ctx.Logger().Info("possible duplicate detector configured", "detector", id)
+	// build ahocorasick prefilter for efficient string matching
+	// on keywords
+	keywords := []string{}
+	for _, d := range e.detectors[false] {
+		for _, kw := range d.Keywords() {
+			keywords = append(keywords, strings.ToLower(kw))
+		}
+	}
+	for _, d := range e.detectors[true] {
+		for _, kw := range d.Keywords() {
+			keywords = append(keywords, strings.ToLower(kw))
+		}
+	}
+	e.prefilter = *ahocorasick.NewTrieBuilder().AddStrings(keywords).Build()
+
+	ctx.Logger().V(3).Info("loaded decoders", "count", len(e.decoders))
+	ctx.Logger().V(3).Info("loaded detectors",
+		"total", len(e.detectors[true])+len(e.detectors[false]),
+		"verification_enabled", len(e.detectors[true]),
+		"verification_disabled", len(e.detectors[false]),
+	)
+
+	// Sanity check detectors for duplicate configuration. Only log in case
+	// a detector has been configured in a way that isn't represented by
+	// the DetectorID (type and version).
+	{
+		dets := append(e.detectors[true], e.detectors[false]...)
+		seenDetectors := make(map[config.DetectorID]struct{}, len(dets))
+		for _, det := range dets {
+			id := config.GetDetectorID(det)
+			if _, ok := seenDetectors[id]; ok && id.ID != detectorspb.DetectorType_CustomRegex {
+				ctx.Logger().Info("possible duplicate detector configured", "detector", id)
+			}
+			seenDetectors[id] = struct{}{}
 		}
-		seenDetectors[id] = struct{}{}
 	}
-}
 
-// startWorkers initiates all necessary workers. Workers handle processing of
-// chunks concurrently. Separating the initialization of different types of
-// workers helps in scalability and makes it easier to diagnose issues.
-func (e *Engine) startWorkers(ctx context.Context) {
-	// Scanner workers process input data and extract chunks for detectors.
 	ctx.Logger().V(2).Info("starting scanner workers", "count", e.concurrency)
+	// Run the Secret scanner workers and Notifier pipelines.
 	for worker := uint64(0); worker < uint64(e.concurrency); worker++ {
 		e.workersWg.Add(1)
 		go func() {
@@ -377,7 +367,6 @@ func (e *Engine) startWorkers(ctx context.Context) {
 		}()
 	}
 
-	// Detector workers apply keyword matching, regexes and API calls to detect secrets in chunks.
 	const detectorWorkerMultiplier = 50
 	ctx.Logger().V(2).Info("starting detector workers", "count", e.concurrency*detectorWorkerMultiplier)
 	for worker := uint64(0); worker < uint64(e.concurrency*detectorWorkerMultiplier); worker++ {
@@ -390,7 +379,6 @@ func (e *Engine) startWorkers(ctx context.Context) {
 		}()
 	}
 
-	// Notifier workers communicate detected issues to the user or any downstream systems.
 	// We want 1/4th of the notifier workers as the number of scanner workers.
 	const notifierWorkerRatio = 4
 	maxNotifierWorkers := 1
@@ -407,6 +395,8 @@ func (e *Engine) startWorkers(ctx context.Context) {
 			e.notifyResults(ctx)
 		}()
 	}
+
+	return e, nil
 }
 
 // Finish waits for running sources to complete and workers to finish scanning
@@ -455,42 +445,64 @@ type detectableChunk struct {
 func (e *Engine) detectorWorker(ctx context.Context) {
 	var wgDetect sync.WaitGroup
 
-	// Reuse the same map to avoid allocations.
-	const avgDetectorsPerChunk = 2
-	chunkSpecificDetectors := make(map[detectorspb.DetectorType]detectorInfo, avgDetectorsPerChunk)
 	for originalChunk := range e.ChunksChan() {
 		for chunk := range sources.Chunker(originalChunk) {
+			matchedKeywords := make(map[string]struct{})
 			atomic.AddUint64(&e.metrics.BytesScanned, uint64(len(chunk.Data)))
 			for _, decoder := range e.decoders {
+				var decoderType detectorspb.DecoderType
+				switch decoder.(type) {
+				case *decoders.UTF8:
+					decoderType = detectorspb.DecoderType_PLAIN
+				case *decoders.Base64:
+					decoderType = detectorspb.DecoderType_BASE64
+				case *decoders.UTF16:
+					decoderType = detectorspb.DecoderType_UTF16
+				default:
+					ctx.Logger().Info("unknown decoder type", "type", reflect.TypeOf(decoder).String())
+					decoderType = detectorspb.DecoderType_UNKNOWN
+				}
+
 				decoded := decoder.FromChunk(chunk)
+
 				if decoded == nil {
-					ctx.Logger().V(4).Info("no decoder found for chunk", "chunk", chunk)
 					continue
 				}
 
-				for _, match := range e.ahoCorasickCore.MatchString(string(decoded.Chunk.Data)) {
-					if !e.ahoCorasickCore.PopulateDetectorsByMatch(match, chunkSpecificDetectors) {
-						continue
-					}
+				// build a map of all keywords that were matched in the chunk
+				for _, m := range e.prefilter.MatchString(strings.ToLower(string(decoded.Data))) {
+					matchedKeywords[strings.ToLower(m.MatchString())] = struct{}{}
 				}
 
-				for k, detector := range chunkSpecificDetectors {
-					decoded.Chunk.Verify = detector.shouldVerify
-					wgDetect.Add(1)
-					e.detectableChunksChan <- detectableChunk{
-						chunk:    *decoded.Chunk,
-						detector: detector,
-						decoder:  decoded.DecoderType,
-						wgDoneFn: wgDetect.Done,
+				for verify, detectorsSet := range e.detectors {
+					for _, detector := range detectorsSet {
+						chunkContainsKeyword := false
+						for _, kw := range detector.Keywords() {
+							if _, ok := matchedKeywords[strings.ToLower(kw)]; ok {
+								chunkContainsKeyword = true
+								break
+							}
+						}
+
+						if !chunkContainsKeyword {
+							continue
+						}
+
+						decoded.Verify = verify
+						wgDetect.Add(1)
+						e.detectableChunksChan <- detectableChunk{
+							chunk:    *decoded,
+							detector: detector,
+							decoder:  decoderType,
+							wgDoneFn: wgDetect.Done,
+						}
 					}
-					delete(chunkSpecificDetectors, k)
 				}
 			}
 		}
 		atomic.AddUint64(&e.metrics.ChunksScanned, 1)
 	}
 	wgDetect.Wait()
-	ctx.Logger().V(4).Info("finished scanning chunks")
 }
 
 func (e *Engine) detectChunks(ctx context.Context) {
@@ -623,7 +635,7 @@ func FragmentLineOffset(chunk *sources.Chunk, result *detectors.Result) (int64,
 	if !found {
 		return 0, false
 	}
-	lineNumber := int64(bytes.Count(before, []byte("\n")))
+	lineNumber := int64(bytes.Count(before, []byte("\n")) + 1)
 	// If the line contains the ignore tag, we should ignore the result.
 	endLine := bytes.Index(after, []byte("\n"))
 	if endLine == -1 {

pkg/engine/engine_test.go

@@ -32,7 +32,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret here"),
 			},
-			expectedLine: 2,
+			expectedLine: 3,
 			ignore:       true,
 		},
 		{
@@ -43,7 +43,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret here"),
 			},
-			expectedLine: 2,
+			expectedLine: 3,
 			ignore:       false,
 		},
 		{
@@ -54,7 +54,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret here"),
 			},
-			expectedLine: 4,
+			expectedLine: 5,
 			ignore:       false,
 		},
 		{
@@ -65,7 +65,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret\nhere"),
 			},
-			expectedLine: 4,
+			expectedLine: 5,
 			ignore:       false,
 		},
 		{
@@ -76,7 +76,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret\nhere"),
 			},
-			expectedLine: 3,
+			expectedLine: 4,
 			ignore:       true,
 		},
 		{
@@ -87,7 +87,7 @@ func TestFragmentLineOffset(t *testing.T) {
 			result: &detectors.Result{
 				Raw: []byte("secret here"),
 			},
-			expectedLine: 3,
+			expectedLine: 4,
 			ignore:       true,
 		},
 	}

pkg/engine/git_test.go

@@ -2,7 +2,6 @@ package engine
 
 import (
 	"os"
-	"runtime"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
@@ -121,10 +120,9 @@ func BenchmarkGitEngine(b *testing.B) {
 	defer cancel()
 
 	e, err := Start(ctx,
-		WithConcurrency(uint8(runtime.NumCPU())),
+		WithConcurrency(1),
 		WithDecoders(decoders.DefaultDecoders()...),
 		WithDetectors(false, DefaultDetectors()...),
-		WithPrinter(new(discardPrinter)),
 	)
 	assert.Nil(b, err)
 
@@ -135,7 +133,6 @@ func BenchmarkGitEngine(b *testing.B) {
 		}
 	}()
 
-	b.ResetTimer()
 	for i := 0; i < b.N; i++ {
 		// TODO: this is measuring the time it takes to initialize the source
 		// and not to do the full scan

pkg/giturl/giturl.go

@@ -158,6 +158,11 @@ func UpdateLinkLineNumber(ctx context.Context, link string, newLine int64) strin
 		return link
 	}
 
+	if newLine <= 0 {
+		// Don't change the link if the line number is 0.
+		return link
+	}
+
 	switch determineProvider(link) {
 	case providerBitbucket:
 		// For Bitbucket, it doesn't support line links (based on the GenerateLink function).

pkg/giturl/giturl_test.go

@@ -251,6 +251,14 @@ func TestUpdateLinkLineNumber(t *testing.T) {
 			},
 			want: "https://onprem.customdomain.com/org/repo/commit/xyz123#L50",
 		},
+		{
+			name: "Don't include line when it's 0",
+			args: args{
+				link:    "https://github.com/coinbase/cbpay-js/issues/181",
+				newLine: int64(0),
+			},
+			want: "https://github.com/coinbase/cbpay-js/issues/181",
+		},
 		{
 			name: "Invalid link",
 			args: args{

pkg/pb/detectorspb/detectors.pb.go

@@ -1007,6 +1007,7 @@ const (
 	DetectorType_Anthropic                     DetectorType = 933
 	DetectorType_Ramp                          DetectorType = 934
 	DetectorType_Klaviyo                       DetectorType = 935
+	DetectorType_SourcegraphCody               DetectorType = 936
 )
 
 // Enum value maps for DetectorType.
@@ -1944,6 +1945,7 @@ var (
 		933: "Anthropic",
 		934: "Ramp",
 		935: "Klaviyo",
+		936: "SourcegraphCody",
 	}
 	DetectorType_value = map[string]int32{
 		"Alibaba":                       0,
@@ -2878,6 +2880,7 @@ var (
 		"Anthropic":                     933,
 		"Ramp":                          934,
 		"Klaviyo":                       935,
+		"SourcegraphCody":               936,
 	}
 )
 
@@ -3256,7 +3259,7 @@ var file_detectors_proto_rawDesc = []byte{
 	0x44, 0x65, 0x63, 0x6f, 0x64, 0x65, 0x72, 0x54, 0x79, 0x70, 0x65, 0x12, 0x0b, 0x0a, 0x07, 0x55,
 	0x4e, 0x4b, 0x4e, 0x4f, 0x57, 0x4e, 0x10, 0x00, 0x12, 0x09, 0x0a, 0x05, 0x50, 0x4c, 0x41, 0x49,
 	0x4e, 0x10, 0x01, 0x12, 0x0a, 0x0a, 0x06, 0x42, 0x41, 0x53, 0x45, 0x36, 0x34, 0x10, 0x02, 0x12,
-	0x09, 0x0a, 0x05, 0x55, 0x54, 0x46, 0x31, 0x36, 0x10, 0x03, 0x2a, 0x90, 0x75, 0x0a, 0x0c, 0x44,
+	0x09, 0x0a, 0x05, 0x55, 0x54, 0x46, 0x31, 0x36, 0x10, 0x03, 0x2a, 0xa6, 0x75, 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,
@@ -4193,12 +4196,13 @@ var file_detectors_proto_rawDesc = []byte{
 	0x12, 0x12, 0x0a, 0x0d, 0x50, 0x6c, 0x61, 0x6e, 0x65, 0x74, 0x53, 0x63, 0x61, 0x6c, 0x65, 0x44,
 	0x62, 0x10, 0xa4, 0x07, 0x12, 0x0e, 0x0a, 0x09, 0x41, 0x6e, 0x74, 0x68, 0x72, 0x6f, 0x70, 0x69,
 	0x63, 0x10, 0xa5, 0x07, 0x12, 0x09, 0x0a, 0x04, 0x52, 0x61, 0x6d, 0x70, 0x10, 0xa6, 0x07, 0x12,
-	0x0c, 0x0a, 0x07, 0x4b, 0x6c, 0x61, 0x76, 0x69, 0x79, 0x6f, 0x10, 0xa7, 0x07, 0x42, 0x3d, 0x5a,
-	0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66,
-	0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75, 0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, 0x66,
-	0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f, 0x76, 0x33, 0x2f, 0x70, 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,
+	0x0c, 0x0a, 0x07, 0x4b, 0x6c, 0x61, 0x76, 0x69, 0x79, 0x6f, 0x10, 0xa7, 0x07, 0x12, 0x14, 0x0a,
+	0x0f, 0x53, 0x6f, 0x75, 0x72, 0x63, 0x65, 0x67, 0x72, 0x61, 0x70, 0x68, 0x43, 0x6f, 0x64, 0x79,
+	0x10, 0xa8, 0x07, 0x42, 0x3d, 0x5a, 0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f,
+	0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75, 0x72, 0x69, 0x74,
+	0x79, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f, 0x76, 0x33, 0x2f,
+	0x70, 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 (

pkg/sources/filesystem/filesystem.go

@@ -39,8 +39,7 @@ type Source struct {
 // Ensure the Source satisfies the interfaces at compile time
 var _ sources.Source = (*Source)(nil)
 var _ sources.SourceUnitUnmarshaller = (*Source)(nil)
-var _ sources.SourceUnitEnumerator = (*Source)(nil)
-var _ sources.SourceUnitChunker = (*Source)(nil)
+var _ sources.SourceUnitEnumChunker = (*Source)(nil)
 
 // Type returns the type of source.
 // It is used for matching source types in configuration and job input.

pkg/sources/filesystem/filesystem_test.go

@@ -15,6 +15,7 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/source_metadatapb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/sourcespb"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/sourcestest"
 )
 
 func TestSource_Scan(t *testing.T) {
@@ -84,23 +85,15 @@ func TestSource_Scan(t *testing.T) {
 }
 
 func TestScanFile(t *testing.T) {
-	tmpfile, err := os.CreateTemp("", "example.txt")
-	if err != nil {
-		t.Fatal(err)
-	}
-	defer os.Remove(tmpfile.Name())
-
 	chunkSize := sources.ChunkSize
 	secretPart1 := "SECRET"
 	secretPart2 := "SPLIT"
 	// Split the secret into two parts and pad the rest of the chunk with A's.
 	data := strings.Repeat("A", chunkSize-len(secretPart1)) + secretPart1 + secretPart2 + strings.Repeat("A", chunkSize-len(secretPart2))
 
-	_, err = tmpfile.Write([]byte(data))
-	assert.Nil(t, err)
-
-	err = tmpfile.Close()
+	tmpfile, cleanup, err := createTempFile("", data)
 	assert.Nil(t, err)
+	defer cleanup()
 
 	source := &Source{}
 	chunksChan := make(chan *sources.Chunk, 2)
@@ -120,3 +113,195 @@ func TestScanFile(t *testing.T) {
 
 	assert.Contains(t, foundSecret, secretPart1+secretPart2)
 }
+
+func TestEnumerate(t *testing.T) {
+	t.Parallel()
+	ctx := context.Background()
+
+	// Setup the connection to test enumeration.
+	units := []string{
+		"/one", "/two", "/three",
+		"/path/to/dir/", "/path/to/another/dir/",
+	}
+	conn, err := anypb.New(&sourcespb.Filesystem{
+		Paths:       units[0:3],
+		Directories: units[3:],
+	})
+	assert.NoError(t, err)
+
+	// Initialize the source.
+	s := Source{}
+	err = s.Init(ctx, "test enumerate", 0, 0, true, conn, 1)
+	assert.NoError(t, err)
+
+	reporter := sourcestest.TestReporter{}
+	err = s.Enumerate(ctx, &reporter)
+	assert.NoError(t, err)
+
+	assert.Equal(t, len(units), len(reporter.Units))
+	assert.Equal(t, 0, len(reporter.UnitErrs))
+	for _, unit := range reporter.Units {
+		assert.Contains(t, units, unit.SourceUnitID())
+	}
+	for _, unit := range units {
+		assert.Contains(t, reporter.Units, sources.CommonSourceUnit{ID: unit})
+	}
+}
+
+func TestChunkUnit(t *testing.T) {
+	t.Parallel()
+	ctx := context.Background()
+
+	// Setup test file to chunk.
+	fileContents := "TestChunkUnit"
+	tmpfile, cleanup, err := createTempFile("", fileContents)
+	assert.NoError(t, err)
+	defer cleanup()
+
+	tmpdir, cleanup, err := createTempDir("", "foo", "bar", "baz")
+	assert.NoError(t, err)
+	defer cleanup()
+
+	conn, err := anypb.New(&sourcespb.Filesystem{})
+	assert.NoError(t, err)
+
+	// Initialize the source.
+	s := Source{}
+	err = s.Init(ctx, "test chunk unit", 0, 0, true, conn, 1)
+	assert.NoError(t, err)
+
+	// Happy path single file.
+	reporter := sourcestest.TestReporter{}
+	err = s.ChunkUnit(ctx, sources.CommonSourceUnit{
+		ID: tmpfile.Name(),
+	}, &reporter)
+	assert.NoError(t, err)
+
+	// Happy path directory.
+	err = s.ChunkUnit(ctx, sources.CommonSourceUnit{
+		ID: tmpdir,
+	}, &reporter)
+	assert.NoError(t, err)
+
+	// Error path.
+	err = s.ChunkUnit(ctx, sources.CommonSourceUnit{
+		ID: "/file/not/found",
+	}, &reporter)
+	assert.NoError(t, err)
+
+	assert.Equal(t, 4, len(reporter.Chunks))
+	assert.Equal(t, 1, len(reporter.ChunkErrs))
+	dataFound := make(map[string]struct{}, 4)
+	for _, chunk := range reporter.Chunks {
+		dataFound[string(chunk.Data)] = struct{}{}
+	}
+	assert.Contains(t, dataFound, fileContents)
+	assert.Contains(t, dataFound, "foo")
+	assert.Contains(t, dataFound, "bar")
+	assert.Contains(t, dataFound, "baz")
+}
+
+func TestEnumerateReporterErr(t *testing.T) {
+	t.Parallel()
+	ctx := context.Background()
+
+	// Setup the connection to test enumeration.
+	units := []string{
+		"/one", "/two", "/three",
+		"/path/to/dir/", "/path/to/another/dir/",
+	}
+	conn, err := anypb.New(&sourcespb.Filesystem{
+		Paths:       units[0:3],
+		Directories: units[3:],
+	})
+	assert.NoError(t, err)
+
+	// Initialize the source.
+	s := Source{}
+	err = s.Init(ctx, "test enumerate", 0, 0, true, conn, 1)
+	assert.NoError(t, err)
+
+	// Enumerate should always return an error if the reporter returns an
+	// error.
+	reporter := sourcestest.ErrReporter{}
+	err = s.Enumerate(ctx, &reporter)
+	assert.Error(t, err)
+}
+
+func TestChunkUnitReporterErr(t *testing.T) {
+	t.Parallel()
+	ctx := context.Background()
+
+	// Setup test file to chunk.
+	tmpfile, err := os.CreateTemp("", "example.txt")
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer os.Remove(tmpfile.Name())
+
+	fileContents := []byte("TestChunkUnit")
+	_, err = tmpfile.Write(fileContents)
+	assert.NoError(t, err)
+	assert.NoError(t, tmpfile.Close())
+
+	conn, err := anypb.New(&sourcespb.Filesystem{})
+	assert.NoError(t, err)
+
+	// Initialize the source.
+	s := Source{}
+	err = s.Init(ctx, "test chunk unit", 0, 0, true, conn, 1)
+	assert.NoError(t, err)
+
+	// Happy path. ChunkUnit should always return an error if the reporter
+	// returns an error.
+	reporter := sourcestest.ErrReporter{}
+	err = s.ChunkUnit(ctx, sources.CommonSourceUnit{
+		ID: tmpfile.Name(),
+	}, &reporter)
+	assert.Error(t, err)
+
+	// Error path. ChunkUnit should always return an error if the reporter
+	// returns an error.
+	err = s.ChunkUnit(ctx, sources.CommonSourceUnit{
+		ID: "/file/not/found",
+	}, &reporter)
+	assert.Error(t, err)
+}
+
+// createTempFile is a helper function to create a temporary file in the given
+// directory with the provided contents. If dir is "", the operating system's
+// temp directory is used.
+func createTempFile(dir string, contents string) (*os.File, func(), error) {
+	tmpfile, err := os.CreateTemp(dir, "trufflehogtest")
+	if err != nil {
+		return nil, nil, err
+	}
+
+	if _, err := tmpfile.Write([]byte(contents)); err != nil {
+		_ = os.Remove(tmpfile.Name())
+		return nil, nil, err
+	}
+	if err := tmpfile.Close(); err != nil {
+		_ = os.Remove(tmpfile.Name())
+		return nil, nil, err
+	}
+	return tmpfile, func() { _ = os.Remove(tmpfile.Name()) }, nil
+}
+
+// createTempDir is a helper function to create a temporary directory in the
+// given directory with files containing the provided contents. If dir is "",
+// the operating system's temp directory is used.
+func createTempDir(dir string, contents ...string) (string, func(), error) {
+	tmpdir, err := os.MkdirTemp(dir, "trufflehogtest")
+	if err != nil {
+		return "", nil, err
+	}
+
+	for _, content := range contents {
+		if _, _, err := createTempFile(tmpdir, content); err != nil {
+			_ = os.RemoveAll(tmpdir)
+			return "", nil, err
+		}
+	}
+	return tmpdir, func() { _ = os.RemoveAll(tmpdir) }, nil
+}

pkg/sources/github/github.go

@@ -233,7 +233,7 @@ func (s *Source) Init(aCtx context.Context, name string, jobID sources.JobID, so
 			aCtx.Logger().Error(err, "invalid repository", "repo", repo)
 			continue
 		}
-		s.filteredRepoCache.Set(r, r)
+		s.filteredRepoCache.Set(repo, r)
 	}
 
 	s.includeIssueComments = s.conn.IncludeIssueComments

pkg/sourcestest/sourcestest.go

@@ -0,0 +1,61 @@
+package sourcestest
+
+import (
+	"fmt"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/context"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/sources"
+)
+
+type reporter interface {
+	sources.UnitReporter
+	sources.ChunkReporter
+}
+
+var (
+	_ reporter = (*TestReporter)(nil)
+	_ reporter = (*ErrReporter)(nil)
+)
+
+// TestReporter is a helper struct that implements both UnitReporter and
+// ChunkReporter by simply recording the values passed in the methods.
+type TestReporter struct {
+	Units     []sources.SourceUnit
+	UnitErrs  []error
+	Chunks    []sources.Chunk
+	ChunkErrs []error
+}
+
+func (t *TestReporter) UnitOk(_ context.Context, unit sources.SourceUnit) error {
+	t.Units = append(t.Units, unit)
+	return nil
+}
+func (t *TestReporter) UnitErr(_ context.Context, err error) error {
+	t.UnitErrs = append(t.UnitErrs, err)
+	return nil
+}
+func (t *TestReporter) ChunkOk(_ context.Context, chunk sources.Chunk) error {
+	t.Chunks = append(t.Chunks, chunk)
+	return nil
+}
+func (t *TestReporter) ChunkErr(_ context.Context, err error) error {
+	t.ChunkErrs = append(t.ChunkErrs, err)
+	return nil
+}
+
+// ErrReporter implements UnitReporter and ChunkReporter but always returns an
+// error.
+type ErrReporter struct{}
+
+func (ErrReporter) UnitOk(context.Context, sources.SourceUnit) error {
+	return fmt.Errorf("ErrReporter: UnitOk error")
+}
+func (ErrReporter) UnitErr(context.Context, error) error {
+	return fmt.Errorf("ErrReporter: UnitErr error")
+}
+func (ErrReporter) ChunkOk(context.Context, sources.Chunk) error {
+	return fmt.Errorf("ErrReporter: ChunkOk error")
+}
+func (ErrReporter) ChunkErr(context.Context, error) error {
+	return fmt.Errorf("ErrReporter: ChunkErr error")
+}

proto/detectors.proto

@@ -944,6 +944,7 @@ enum DetectorType {
   Anthropic = 933;
   Ramp = 934;
   Klaviyo = 935;
+  SourcegraphCody = 936;
 }
 
 message Result {