fix(privacy): improve detector (#87)

Author: rgmz

pkg/detectors/privacy/privacy.go

@@ -3,8 +3,8 @@ package privacy
 import (
 	"context"
 	"fmt"
+	"io"
 	"net/http"
-	"strings"
 
 	regexp "github.com/wasilibs/go-re2"
 
@@ -20,68 +20,83 @@ type Scanner struct {
 // Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_Privacy
+}
+
+func (s Scanner) Description() string {
+	return "Privacy provides virtual cards for secure online payments. Privacy API keys can be used to manage and create these virtual cards."
+}
+
 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{"privacy"}) + `\b([a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{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{"privacy"}
+	return []string{"privacy.com"}
 }
 
 // FromData will find and optionally verify Privacy 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 {
-		resMatch := strings.TrimSpace(match[1])
+	uniqueMatches := make(map[string]struct{})
+	for _, match := range keyPat.FindAllStringSubmatch(dataStr, -1) {
+		m := match[1]
+		if detectors.StringShannonEntropy(m) < 3 {
+			continue
+		}
+		uniqueMatches[m] = struct{}{}
+	}
 
-		s1 := detectors.Result{
+	for key := range uniqueMatches {
+		r := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Privacy,
-			Raw:          []byte(resMatch),
+			Raw:          []byte(key),
 		}
 
 		if verify {
-			client := s.client
-			if client == nil {
-				client = defaultClient
-			}
-			req, err := http.NewRequestWithContext(ctx, "GET", "https://api.privacy.com/v1/card?page=1&page_size=50", nil)
-			if err != nil {
-				continue
-			}
-			req.Header.Set("Authorization", "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 {
-					// 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)
+			if s.client == nil {
+				s.client = common.SaneHttpClient()
 			}
+
+			isVerified, vErr := verifyMatch(ctx, s.client, key)
+			r.Verified = isVerified
+			r.SetVerificationError(vErr, key)
 		}
 
-		results = append(results, s1)
+		results = append(results, r)
 	}
 
 	return results, nil
 }
 
-func (s Scanner) Type() detectorspb.DetectorType {
-	return detectorspb.DetectorType_Privacy
-}
+func verifyMatch(ctx context.Context, client *http.Client, key string) (bool, error) {
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://api.privacy.com/v1/card?page=1&page_size=50", nil)
+	if err != nil {
+		return false, err
+	}
 
-func (s Scanner) Description() string {
-	return "Privacy provides virtual cards for secure online payments. Privacy API keys can be used to manage and create these virtual cards."
+	req.Header.Set("Authorization", "api-key "+key)
+	res, err := client.Do(req)
+	if err != nil {
+		return false, err
+	}
+	defer func() {
+		_, _ = io.Copy(io.Discard, res.Body)
+		_ = res.Body.Close()
+	}()
+
+	switch res.StatusCode {
+	case http.StatusOK:
+		return true, nil
+	case http.StatusUnauthorized:
+		return false, nil
+	default:
+		body, _ := io.ReadAll(res.Body)
+		return false, fmt.Errorf("unexpected HTTP response status %d, %q", res.StatusCode, string(body))
+	}
 }

pkg/detectors/privacy/privacy_test.go

@@ -14,7 +14,7 @@ import (
 var (
 	validPattern   = "sccqe60k-nsak-h1m1-vzq8-vr7u4gxihiwl"
 	invalidPattern = "sccqe60k?nsak-h1m1-vzq8-vr7u4gxihiwl"
-	keyword        = "privacy"
+	keyword        = "privacy.com"
 )
 
 func TestPrivacy_Pattern(t *testing.T) {
@@ -34,7 +34,7 @@ func TestPrivacy_Pattern(t *testing.T) {
 			name:  "valid pattern - ignore duplicate",
 			input: fmt.Sprintf("%s token = '%s' | '%s'", keyword, validPattern, validPattern),
 			want:  []string{validPattern},
-},
+		},
 		{
 			name:  "invalid pattern",
 			input: fmt.Sprintf("%s = '%s'", keyword, invalidPattern),