modified Alibaba detector to use standard library

pkg/detectors/alibaba/alibaba.go

@@ -2,11 +2,18 @@ package alibaba
 
 import (
 	"context"
+	"crypto/hmac"
+	"crypto/sha1"
+	"encoding/base64"
+	"math/rand"
+	"net/http"
+	"net/url"
 	"regexp"
+	"strconv"
+	"strings"
 	"time"
 
-	"github.com/aliyun/alibaba-cloud-sdk-go/services/ecs"
-	log "github.com/sirupsen/logrus"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
@@ -17,68 +24,109 @@ type Scanner struct{}
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
-	keyPat    = regexp.MustCompile(`\b(LTAI[a-zA-Z0-9]{17,21})[\"' ;\s]*`)
-	secretPat = regexp.MustCompile(`\b([a-zA-Z0-9]{30})\b`)
+	client = common.SaneHttpClient()
+
+	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
+	keyPat = regexp.MustCompile(`\b([a-zA-Z0-9]{30})\b`)
+	idPat  = regexp.MustCompile(`\b(LTAI[a-zA-Z0-9]{17,21})[\"';\s]*`)
 )
 
 // 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{"LTAI"}
+	return []string{"alibaba"}
+}
+
+func randString(n int) string {
+	rand.Seed(time.Now().UnixNano())
+	const alphanum = "0123456789abcdefghijklmnopqrstuvwxyz"
+	var bytes = make([]byte, n)
+	rand.Read(bytes)
+	for i, b := range bytes {
+		bytes[i] = alphanum[b%byte(len(alphanum))]
+	}
+	return string(bytes)
+}
+
+func GetSignature(input, key string) string {
+	key_for_sign := []byte(key)
+	h := hmac.New(sha1.New, key_for_sign)
+	h.Write([]byte(input))
+	return base64.StdEncoding.EncodeToString(h.Sum(nil))
+}
+func buildStringToSign(input string) string {
+	filter := strings.Replace(input, "+", "%20", -1)
+	filter = strings.Replace(filter, "*", "%2A", -1)
+	filter = strings.Replace(filter, "%7E", "~", -1)
+	method := "GET"
+	filter = method + "&%2F&" + url.QueryEscape(filter)
+	return filter
 }
 
 // FromData will find and optionally verify Alibaba 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 {
-		//Plausible key pat found, look for secrets match
-		secMatches := secretPat.FindAllStringSubmatch(dataStr, -1)
+	idMatches := idPat.FindAllStringSubmatch(dataStr, -1)
 
-		for _, secMatch := range secMatches {
+	for _, match := range matches {
+		if len(match) != 2 {
+			continue
+		}
+		resMatch := strings.TrimSpace(match[1])
 
-			if len(match) != 2 {
+		for _, idMatch := range idMatches {
+			if len(idMatch) != 2 {
 				continue
 			}
 
-			s := detectors.Result{
+			resIdMatch := strings.TrimSpace(idMatch[1])
+
+			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Alibaba,
-				Raw:          []byte(match[1]),
-				Redacted:     match[1],
+				Raw:          []byte(resMatch),
 			}
 
 			if verify {
-				ecsClient, err := ecs.NewClientWithAccessKey(
-					"us-east-1", // your region ID
-					match[1],    // your AccessKey ID
-					secMatch[1]) // your AccessKey Secret
+				req, err := http.NewRequestWithContext(ctx, "GET", "http://ecs.aliyuncs.com/?", nil)
 				if err != nil {
-					log.WithError(err).Debug("error creating alibaba client, skipping")
 					continue
 				}
-				// Create an API request and set parameters
-				request := ecs.CreateDescribeInstancesRequest()
-				request.ConnectTimeout = time.Duration(5) * time.Second
-				request.Scheme = "https"
-				request.Domain = "ecs.aliyuncs.com"
-				// Initiate the request and handle exceptions
-				_, err = ecsClient.DescribeInstances(request)
-				if err != nil {
-					s.Verified = false
+				dateISO := time.Now().UTC().Format("2006-01-02T15:04:05Z07:00")
+				params := req.URL.Query()
+				params.Add("AccessKeyId", resIdMatch)
+				params.Add("Action", "DescribeRegions")
+				params.Add("Format", "JSON")
+				params.Add("SignatureMethod", "HMAC-SHA1")
+				params.Add("SignatureNonce", randString(16))
+				params.Add("SignatureVersion", "1.0")
+				params.Add("Timestamp", dateISO)
+				params.Add("Version", "2014-05-26")
 
-				} else {
-					s.Verified = true
-				}
-			}
+				stringToSign := buildStringToSign(params.Encode())
+				signature := GetSignature(stringToSign, resMatch+"&") //Get Signature HMAC SHA1
+				params.Add("Signature", signature)
+				req.URL.RawQuery = params.Encode()
 
-			if !s.Verified {
-				if detectors.IsKnownFalsePositive(string(s.Raw), detectors.DefaultFalsePositives, true) {
-					continue
+				req.Header.Add("Content-Type", "text/xml;charset=utf-8")
+				req.Header.Add("Content-Length", strconv.Itoa(len(params.Encode())))
+				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
+						}
+					}
 				}
 			}
 
-			results = append(results, s)
+			results = append(results, s1)
 		}
 	}
-	return
+
+	return detectors.CleanResults(results), nil
 }

pkg/detectors/alibaba/alibaba_test.go

@@ -7,20 +7,21 @@ import (
 	"time"
 
 	"github.com/kylelemons/godebug/pretty"
-	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
 
 func TestAlibaba_FromChunk(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
 	defer cancel()
-	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors2")
+	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors4")
 	if err != nil {
 		t.Fatalf("could not get test secrets from GCP: %s", err)
 	}
 	secret := testSecrets.MustGetField("ALIBABA_SECRET")
-	secretInactive := testSecrets.MustGetField("ALIBABA_SECRET_INACTIVE")
+	inactiveSecret := testSecrets.MustGetField("ALIBABA_SECRET_INACTIVE")
 	id := testSecrets.MustGetField("ALIBABA_ID")
 
 	type args struct {
@@ -40,14 +41,13 @@ func TestAlibaba_FromChunk(t *testing.T) {
 			s:    Scanner{},
 			args: args{
 				ctx:    context.Background(),
-				data:   []byte(fmt.Sprintf("Alibaba keys are here: %s\n %s within", id, secret)),
+				data:   []byte(fmt.Sprintf("You can find a alibaba secret %s within alibaba %s", secret, id)),
 				verify: true,
 			},
 			want: []detectors.Result{
 				{
 					DetectorType: detectorspb.DetectorType_Alibaba,
 					Verified:     true,
-					Redacted:     id,
 				},
 			},
 			wantErr: false,
@@ -57,14 +57,13 @@ func TestAlibaba_FromChunk(t *testing.T) {
 			s:    Scanner{},
 			args: args{
 				ctx:    context.Background(),
-				data:   []byte(fmt.Sprintf("Alibaba keys are here: %s\n %s within", id, secretInactive)),
+				data:   []byte(fmt.Sprintf("You can find a alibaba secret %s within alibaba %s but not valid", inactiveSecret, id)), // the secret would satisfy the regex but not pass validation
 				verify: true,
 			},
 			want: []detectors.Result{
 				{
 					DetectorType: detectorspb.DetectorType_Alibaba,
 					Verified:     false,
-					Redacted:     id,
 				},
 			},
 			wantErr: false,
@@ -91,7 +90,7 @@ func TestAlibaba_FromChunk(t *testing.T) {
 			}
 			for i := range got {
 				if len(got[i].Raw) == 0 {
-					t.Fatal("no raw secret present")
+					t.Fatalf("no raw secret present: \n %+v", got[i])
 				}
 				got[i].Raw = nil
 			}
@@ -101,3 +100,18 @@ func TestAlibaba_FromChunk(t *testing.T) {
 		})
 	}
 }
+
+func BenchmarkFromData(benchmark *testing.B) {
+	ctx := context.Background()
+	s := Scanner{}
+	for name, data := range detectors.MustGetBenchmarkData() {
+		benchmark.Run(name, func(b *testing.B) {
+			for n := 0; n < b.N; n++ {
+				_, err := s.FromData(ctx, false, data)
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
+	}
+}