feat(azure): improve connstring matching

Author: rgmz

pkg/detectors/azurestorage/azurestorage.go

@@ -6,9 +6,9 @@ import (
 	"crypto/sha256"
 	"encoding/base64"
 	"fmt"
+	"io"
 	"net/http"
 	"regexp"
-	"strings"
 	"time"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -23,72 +23,113 @@ var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	defaultClient = http.DefaultClient
-	keyPat        = regexp.MustCompile(`DefaultEndpointsProtocol=https;AccountName=(?P<account_name>[^;]+);AccountKey=(?P<account_key>[^;]+);EndpointSuffix=core\.windows\.net`)
+	namePat       = regexp.MustCompile(`AccountName=([a-z0-9]{3,24})`) // Names can only be lowercase alphanumeric.
+	keyPat        = regexp.MustCompile(`AccountKey=([a-zA-Z0-9+/-]{86,88}={0,2})`)
+	key1Pat       = regexp.MustCompile(`DefaultEndpointsProtocol=https;AccountName=(?P<account_name>[^;]+);AccountKey=(?P<account_key>[^;]+);EndpointSuffix=core\.windows\.net`)
+
+	// https://learn.microsoft.com/en-us/azure/storage/common/storage-use-emulator
+	testNames = map[string]struct{}{
+		"devstoreaccount1": {},
+		"storagesample":    {},
+	}
+	testKeys = map[string]struct{}{
+		"Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==": {},
+	}
 )
 
 func (s Scanner) Keywords() []string {
-	return []string{"DefaultEndpointsProtocol=https;AccountName="}
+	return []string{"DefaultEndpointsProtocol=http", "EndpointSuffix="}
 }
 
 func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (results []detectors.Result, err error) {
 	dataStr := string(data)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	// Deduplicate results.
+	names := make(map[string]struct{})
+	for _, matches := range namePat.FindAllStringSubmatch(dataStr, -1) {
+		name := matches[1]
+		if _, ok := testNames[name]; ok {
+			continue
+		}
+		names[name] = struct{}{}
+	}
+	if len(names) == 0 {
+		return results, nil
+	}
 
-	for _, match := range matches {
-		if len(match) != 3 {
+	keys := make(map[string]struct{})
+	for _, matches := range keyPat.FindAllStringSubmatch(dataStr, -1) {
+		key := matches[1]
+		if _, ok := testKeys[key]; ok {
 			continue
 		}
-		accountName := strings.TrimSpace(match[1])
-		accountKey := strings.TrimSpace(match[2])
+		keys[key] = struct{}{}
+	}
+	if len(keys) == 0 {
+		return results, nil
+	}
 
+	// Check results.
+	for name := range names {
 		s1 := detectors.Result{
-			DetectorType: detectorspb.DetectorType_AzureStorage,
-			Raw:          []byte(accountName),
+			DetectorType: s.Type(),
+			Raw:          []byte(name),
 		}
 
-		if verify {
-			client := s.client
-			if client == nil {
-				client = defaultClient
+		for key := range keys {
+			if verify {
+				isVerified, verificationErr := s.verifyMatch(ctx, name, key)
+				s1.Verified = isVerified
+				s1.VerificationError = verificationErr
 			}
 
-			now := time.Now().UTC().Format(http.TimeFormat)
-			stringToSign := "GET\n\n\n\n\n\n\n\n\n\n\n\nx-ms-date:" + now + "\nx-ms-version:2019-12-12\n/" + accountName + "/\ncomp:list"
-			accountKeyBytes, _ := base64.StdEncoding.DecodeString(accountKey)
-			h := hmac.New(sha256.New, accountKeyBytes)
-			h.Write([]byte(stringToSign))
-			signature := base64.StdEncoding.EncodeToString(h.Sum(nil))
-
-			url := "https://" + accountName + ".blob.core.windows.net/?comp=list"
-			req, err := http.NewRequestWithContext(ctx, "GET", url, nil)
-			req.Header.Set("x-ms-date", now)
-			req.Header.Set("x-ms-version", "2019-12-12")
-			req.Header.Set("Authorization", "SharedKey "+accountName+":"+signature)
-
-			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 {
-				} else {
-					s1.VerificationError = fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
-				}
-			} else {
-				s1.VerificationError = err
-			}
+			results = append(results, s1)
 		}
-
-		results = append(results, s1)
 	}
 
 	return results, nil
 }
 
+func (s Scanner) verifyMatch(ctx context.Context, name string, key string) (bool, error) {
+	client := s.client
+	if client == nil {
+		client = defaultClient
+	}
+
+	now := time.Now().UTC().Format(http.TimeFormat)
+	stringToSign := "GET\n\n\n\n\n\n\n\n\n\n\n\nx-ms-date:" + now + "\nx-ms-version:2019-12-12\n/" + name + "/\ncomp:list"
+	accountKeyBytes, _ := base64.StdEncoding.DecodeString(key)
+	h := hmac.New(sha256.New, accountKeyBytes)
+	h.Write([]byte(stringToSign))
+	signature := base64.StdEncoding.EncodeToString(h.Sum(nil))
+
+	url := "https://" + name + ".blob.core.windows.net/?comp=list"
+	req, err := http.NewRequestWithContext(ctx, "GET", url, nil)
+	if err != nil {
+		return false, err
+	}
+
+	req.Header.Set("x-ms-date", now)
+	req.Header.Set("x-ms-version", "2019-12-12")
+	req.Header.Set("Authorization", "SharedKey "+name+":"+signature)
+
+	res, err := client.Do(req)
+	if err == nil {
+		_, _ = io.Copy(io.Discard, res.Body)
+		_ = res.Body.Close()
+
+		if res.StatusCode >= 200 && res.StatusCode < 300 {
+			return true, nil
+		} else if res.StatusCode == 403 {
+			return false, nil
+		} else {
+			return false, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
+		}
+	} else {
+		return false, err
+	}
+}
+
 func (s Scanner) Type() detectorspb.DetectorType {
 	return detectorspb.DetectorType_AzureStorage
 }

pkg/engine/defaults.go

@@ -65,6 +65,7 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azure"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azurebatch"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azurecontainerregistry"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/azurestorage"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/bannerbear"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/baremetrics"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/beamer"
@@ -1575,6 +1576,7 @@ func DefaultDetectors() []detectors.Detector {
 		grafana.Scanner{},
 		logzio.Scanner{},
 		eventbrite.Scanner{},
+		&azurestorage.Scanner{},
 	}
 
 }