Added azure COSMOSDB detector

pkg/detectors/azure_cosmosdb/azure_cosmosdb.go

@@ -0,0 +1,169 @@
+package azure_cosmosdb
+
+import (
+	"context"
+	"crypto/hmac"
+	"crypto/sha256"
+	"encoding/base64"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"regexp"
+	"strings"
+	"time"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/cache/simple"
+	"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
+}
+
+var (
+	defaultClient = common.SaneHttpClient()
+
+	dbKeyPattern = regexp.MustCompile(`([A-Za-z0-9+/]{86}==)`)
+	// account name can contain only lowercase letters, numbers and the `-` character, must be between 3 and 44 characters long.
+	accountUrlPattern = regexp.MustCompile(`([a-z0-9-]{3,44}.documents\.azure\.com)`)
+
+	invalidHosts = simple.NewCache[struct{}]()
+
+	noHostErr = errors.New("no such host")
+)
+
+func (s Scanner) getClient() *http.Client {
+	if s.client != nil {
+		return s.client
+	}
+
+	return defaultClient
+}
+
+// Ensure the Scanner satisfies the interface at compile time.
+var _ detectors.Detector = (*Scanner)(nil)
+
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_AzureCosmosDB
+}
+
+func (s Scanner) Description() string {
+	return "Azure Cosmos DB is a globally distributed, multi-model database service offered by Microsoft. CosmosDB keys and connection string are used to connect with Cosmos DB."
+}
+
+func (s Scanner) Keywords() []string {
+	return []string{".documents.azure.com"}
+}
+
+func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (results []detectors.Result, err error) {
+	dataStr := string(data)
+
+	var uniqueKeyMatches, uniqueAccountMatches = make(map[string]struct{}), make(map[string]struct{})
+
+	for _, match := range dbKeyPattern.FindAllStringSubmatch(dataStr, -1) {
+		uniqueKeyMatches[match[1]] = struct{}{}
+	}
+
+	for _, match := range accountUrlPattern.FindAllStringSubmatch(dataStr, -1) {
+		uniqueAccountMatches[match[1]] = struct{}{}
+	}
+
+	for key := range uniqueKeyMatches {
+		for accountUrl := range uniqueAccountMatches {
+			if invalidHosts.Exists(accountUrl) {
+				delete(uniqueAccountMatches, accountUrl)
+				continue
+			}
+
+			s1 := detectors.Result{
+				DetectorType: detectorspb.DetectorType_AzureCosmosDB,
+				Raw:          []byte(key),
+				ExtraData:    make(map[string]string),
+			}
+
+			if verify {
+				verified, verificationErr := verifyCosmosDB(s.getClient(), accountUrl, key)
+				s1.Verified = verified
+				if verificationErr != nil {
+					if errors.Is(verificationErr, noHostErr) {
+						invalidHosts.Set(accountUrl, struct{}{})
+						continue
+					}
+
+					s1.SetVerificationError(verificationErr)
+				}
+			}
+
+			results = append(results, s1)
+		}
+	}
+
+	return results, nil
+}
+
+// documentation: https://learn.microsoft.com/en-us/rest/api/cosmos-db/list-databases
+func verifyCosmosDB(client *http.Client, accountUrl, key string) (bool, error) {
+	// decode the base64 encoded key
+	decodedKey, err := base64.StdEncoding.DecodeString(key)
+	if err != nil {
+		return false, fmt.Errorf("failed to decode key: %v", err)
+	}
+
+	req, err := http.NewRequest(http.MethodGet, fmt.Sprintf("https://%s:443/dbs", accountUrl), nil)
+	if err != nil {
+		return false, fmt.Errorf("failed to create request: %v", err)
+	}
+
+	dateRFC1123 := time.Now().UTC().Format("Mon, 02 Jan 2006 15:04:05 GMT")
+	authHeader := fmt.Sprintf("type=master&ver=1.0&sig=%s", url.QueryEscape(createSignature(decodedKey, dateRFC1123)))
+
+	// required headers
+	// docs: https://learn.microsoft.com/en-us/rest/api/cosmos-db/common-cosmosdb-rest-request-headers
+	req.Header.Set("Authorization", authHeader)
+	req.Header.Set("x-ms-date", dateRFC1123)
+	req.Header.Set("x-ms-version", "2018-12-31")
+
+	resp, err := client.Do(req)
+	if err != nil {
+		// lookup foo.documents.azure.com: no such host
+		if strings.Contains(err.Error(), "no such host") {
+			return false, noHostErr
+		}
+
+		return false, err
+	}
+	defer func() {
+		_, _ = io.Copy(io.Discard, resp.Body)
+		_ = resp.Body.Close()
+	}()
+
+	// Check response status code
+	switch resp.StatusCode {
+	case http.StatusOK:
+		return true, nil
+	case http.StatusUnauthorized:
+		return false, nil
+	default:
+		return false, fmt.Errorf("unexpected status code: %d", resp.StatusCode)
+	}
+}
+
+func createSignature(decodedKey []byte, dateRFC1123 string) string {
+	stringToSign := fmt.Sprintf(
+		"%s\n%s\n%s\n%s\n\n",
+		strings.ToLower(http.MethodGet),
+		strings.ToLower("dbs"),
+		"",
+		strings.ToLower(dateRFC1123),
+	)
+
+	// compute HMAC-SHA256 signature
+	mac := hmac.New(sha256.New, decodedKey)
+	mac.Write([]byte(stringToSign))
+
+	return base64.StdEncoding.EncodeToString(mac.Sum(nil))
+}

pkg/detectors/azure_cosmosdb/azure_cosmosdb_integration_test.go

@@ -0,0 +1,129 @@
+//go:build detectors
+// +build detectors
+
+package azure_cosmosdb
+
+import (
+	"context"
+	"fmt"
+	"testing"
+	"time"
+
+	"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/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+func TestCosmosDB_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)
+	}
+
+	key := testSecrets.MustGetField("COSMOSDB_KEY")
+	accountUrl := testSecrets.MustGetField("COSMOSDB_ACCOUNT")
+	inactiveKey := testSecrets.MustGetField("COSMOSDB_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:    ctx,
+				data:   []byte(fmt.Sprintf("You can find a cosmosdb key: %s and account url: %s within", key, accountUrl)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_Meraki,
+					Verified:     true,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    ctx,
+				data:   []byte(fmt.Sprintf("You can find a cosmosdb key: %s and accounturl: %s within but not valid", inactiveKey, accountUrl)), // the secret would satisfy the regex but not pass validation
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_Meraki,
+					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,
+		},
+	}
+	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("CosmosDB.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("CosmosDB.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/azure_cosmosdb/azure_cosmosdb_test.go

@@ -0,0 +1,91 @@
+package azure_cosmosdb
+
+import (
+	"context"
+	"testing"
+
+	"github.com/google/go-cmp/cmp"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/engine/ahocorasick"
+)
+
+var (
+	validPattern = `
+	Cluster name: Cluster name must be at least 3 characters and at most 40 characters.
+	Cluster name must only contain lowercase letters, numbers, and hyphens.
+	The cluster name must not start or end in a hyphen.
+	// config
+	cosmosKey: FakeeP35zYGPXaEUfakeU7S8kcOY7NI7id8ddbHfakeAifake8Bbql1mXhMF2t0wQ0FAKEPQrwZZACDb3msoAg==
+	https://trufflesecurity-fake.documents.azure.com:443
+	`
+
+	invalidPattern = `
+	FakeeP35zYGPXaEUfakeU7S8kcOY7I7id8ddbHfakeAifake8Bbql1mXhMF2t0wQ0FAKEPQrwZZACDb3msoAg==
+	https://not-a-host.documents.azure.com:443
+	`
+)
+
+func TestCosmosDB_Pattern(t *testing.T) {
+	d := Scanner{}
+	ahoCorasickCore := ahocorasick.NewAhoCorasickCore([]detectors.Detector{d})
+
+	tests := []struct {
+		name  string
+		input string
+		want  []string
+	}{
+		{
+			name:  "valid pattern",
+			input: validPattern,
+			want:  []string{"FakeeP35zYGPXaEUfakeU7S8kcOY7NI7id8ddbHfakeAifake8Bbql1mXhMF2t0wQ0FAKEPQrwZZACDb3msoAg=="},
+		},
+		{
+			name:  "invalid pattern",
+			input: invalidPattern,
+			want:  nil,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			matchedDetectors := ahoCorasickCore.FindDetectorMatches([]byte(test.input))
+			if len(matchedDetectors) == 0 {
+				t.Errorf("keywords '%v' not matched by: %s", d.Keywords(), test.input)
+				return
+			}
+
+			results, err := d.FromData(context.Background(), false, []byte(test.input))
+			if err != nil {
+				t.Errorf("error = %v", err)
+				return
+			}
+
+			if len(results) != len(test.want) {
+				if len(results) == 0 {
+					t.Errorf("did not receive result")
+				} else {
+					t.Errorf("expected %d results, only received %d", len(test.want), len(results))
+				}
+				return
+			}
+
+			actual := make(map[string]struct{}, len(results))
+			for _, r := range results {
+				if len(r.RawV2) > 0 {
+					actual[string(r.RawV2)] = struct{}{}
+				} else {
+					actual[string(r.Raw)] = struct{}{}
+				}
+			}
+			expected := make(map[string]struct{}, len(test.want))
+			for _, v := range test.want {
+				expected[v] = struct{}{}
+			}
+
+			if diff := cmp.Diff(expected, actual); diff != "" {
+				t.Errorf("%s diff: (-want +got)\n%s", test.name, diff)
+			}
+		})
+	}
+}