added azuresearchquerykey detector (#2349)

Co-authored-by: Dustin Decker <[email protected]>

pkg/detectors/azuresearchquerykey/azuresearchquerykey.go

@@ -0,0 +1,97 @@
+package azuresearchquerykey
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"strings"
+
+	regexp "github.com/wasilibs/go-re2"
+
+	"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(detectors.PrefixRegex([]string{"azure"}) + `\b([0-9a-zA-Z]{52})\b`)
+	urlPat = regexp.MustCompile(detectors.PrefixRegex([]string{"azure"}) + `https:\/\/([0-9a-z]{5,40})\.search\.windows\.net\/indexes\/([0-9a-z]{5,40})\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{"azure"}
+}
+
+// FromData will find and optionally verify AzureSearchQueryKey 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)
+	urlMatches := urlPat.FindAllStringSubmatch(dataStr, -1)
+
+	for _, match := range matches {
+		if len(match) != 2 {
+			continue
+		}
+		resMatch := strings.TrimSpace(match[1])
+
+		for _, urlMatch := range urlMatches {
+			resTrim := strings.Split(strings.TrimSpace(urlMatch[0]), " ")
+			resUrlMatch := resTrim[len(resTrim)-1]
+
+			s1 := detectors.Result{
+				DetectorType: detectorspb.DetectorType_AzureSearchQueryKey,
+				Raw:          []byte(resMatch + resUrlMatch),
+			}
+			if verify {
+				client := s.client
+				if client == nil {
+					client = defaultClient
+				}
+				req, err := http.NewRequestWithContext(ctx, "GET", resUrlMatch+"/docs/$count?api-version=2023-10-01-Preview", nil)
+				if err != nil {
+					continue
+				}
+				req.Header.Add("api-key", resMatch)
+
+				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 == 401 || res.StatusCode == 403 {
+						// The secret is determinately not verified (nothing to do)
+					} else {
+						err = fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
+						s1.SetVerificationError(err, resMatch)
+					}
+				} else {
+					s1.SetVerificationError(err, resMatch)
+				}
+			}
+
+			// 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_AzureSearchQueryKey
+}

pkg/detectors/azuresearchquerykey/azuresearchquerykey_test.go

@@ -0,0 +1,163 @@
+//go:build detectors
+// +build detectors
+
+package azuresearchquerykey
+
+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 TestAzuresearchquerykey_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("AZURE_SEARCH_QUERY_KEY")
+	inactiveSecret := testSecrets.MustGetField("AZURE_SEARCH_QUERY_KEY_INACTIVE")
+	url := testSecrets.MustGetField("AZURE_SEARCH_QUERY_KEY_URL")
+
+	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 azure secret %s and azure url %s within", secret, url)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_AzureSearchQueryKey,
+					Verified:     true,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a azure secret %s and azure url %s within but not valid", inactiveSecret, url)), // the secret would satisfy the regex but not pass validation
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_AzureSearchQueryKey,
+					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 azure secret %s and azure url %s within", secret, url)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_AzureSearchQueryKey,
+					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 azure secret %s and azure url %s within", secret, url)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_AzureSearchQueryKey,
+					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("Azuresearchquerykey.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("Azuresearchquerykey.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/engine/defaults.go

@@ -68,6 +68,7 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azuredevopspersonalaccesstoken"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azurefunctionkey"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azuresearchadminkey"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azuresearchquerykey"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azurestorage"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/bannerbear"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/baremetrics"
@@ -1610,6 +1611,7 @@ func DefaultDetectors() []detectors.Detector {
 		azurefunctionkey.Scanner{},
 		azuredevopspersonalaccesstoken.Scanner{},
 		azuresearchadminkey.Scanner{},
+		&azuresearchquerykey.Scanner{},
 		jiratoken_v2.Scanner{},
 		&googleoauth2.Scanner{},
 	}