feat(mailgun): update detector (trufflesecurity#2679)

pkg/detectors/mailgun/mailgun.go

@@ -2,7 +2,9 @@ package mailgun
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
+	"io"
 	"net/http"
 	"strings"
 
@@ -15,16 +17,17 @@ import (
 
 type Scanner struct {
 	detectors.DefaultMultiPartCredentialProvider
+	client *http.Client
 }
 
 // Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
-	client = common.SaneHttpClient()
+	defaultClient = common.SaneHttpClient()
 
 	tokenPats = map[string]*regexp.Regexp{
-		"Original MailGun Token": regexp.MustCompile(detectors.PrefixRegex([]string{"mailgun"}) + `\b([a-zA-Z-0-9]{72})\b`),
+		"Original MailGun Token": regexp.MustCompile(detectors.PrefixRegex([]string{"mailgun"}) + `\b([a-zA-Z0-9-]{72})\b`),
 		"Key-MailGun Token":      regexp.MustCompile(`\b(key-[a-z0-9]{32})\b`),
 		"Hex MailGun Token":      regexp.MustCompile(`\b([a-f0-9]{32}-[a-f0-9]{8}-[a-f0-9]{8})\b`),
 	}
@@ -33,55 +36,120 @@ var (
 // 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{"mailgun"}
+	return []string{"mailgun", "key-"}
+}
+
+func (s Scanner) getClient() *http.Client {
+	if s.client != nil {
+		return s.client
+	}
+	return defaultClient
 }
 
 // FromData will find and optionally verify Mailgun 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)
 
+	uniqueMatches := make(map[string]struct{})
 	for _, tokenPat := range tokenPats {
-		matches := tokenPat.FindAllStringSubmatch(dataStr, -1)
-		for _, match := range matches {
-			if len(match) != 2 {
-				continue
-			}
-			resMatch := strings.TrimSpace(match[1])
+		for _, match := range tokenPat.FindAllStringSubmatch(dataStr, -1) {
+			uniqueMatches[match[1]] = struct{}{}
+		}
+	}
 
-			s1 := detectors.Result{
-				DetectorType: detectorspb.DetectorType_Mailgun,
-				Raw:          []byte(resMatch),
-			}
+	for match := range uniqueMatches {
+		s1 := detectors.Result{
+			DetectorType: s.Type(),
+			Raw:          []byte(match),
+			AnalysisInfo: map[string]string{"key": match},
+		}
 
-			if verify {
-				req, err := http.NewRequestWithContext(ctx, "GET", "https://api.mailgun.net/v3/domains", nil)
-				if err != nil {
-					continue
-				}
+		if verify {
+			client := s.getClient()
+			isVerified, extraData, verificationErr := verifyMatch(ctx, client, match)
+			s1.Verified = isVerified
+			s1.ExtraData = extraData
+			s1.SetVerificationError(verificationErr)
+		}
 
-				// If resMatch has "key" prefix, use it as the username for basic auth.
-				if strings.HasPrefix(resMatch, "key-") {
-					req.SetBasicAuth("api", resMatch)
-				} else {
-					req.Header.Add("Authorization", fmt.Sprintf("Basic %s", resMatch))
-				}
+		results = append(results, s1)
+	}
 
-				res, err := client.Do(req)
-				if err == nil {
-					defer res.Body.Close()
-					if res.StatusCode >= 200 && res.StatusCode < 300 {
-						s1.Verified = true
-					}
-				}
-				s1.AnalysisInfo = map[string]string{"key": resMatch}
+	return
+}
 
-			}
+func verifyMatch(ctx context.Context, client *http.Client, token string) (bool, map[string]string, error) {
+	// https://documentation.mailgun.com/docs/mailgun/api-reference/openapi-final/tag/Domains/
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://api.mailgun.net/v3/domains", nil)
+	if err != nil {
+		return false, nil, err
+	}
+
+	if len(token) == 72 {
+		// This matches prior logic, but may not be correct.
+		req.Header.Add("Authorization", fmt.Sprintf("Basic %s", token))
+	} else {
+		// https://documentation.mailgun.com/docs/mailgun/api-reference/authentication/
+		req.SetBasicAuth("api", token)
+	}
+	req.Header.Add("Content-Type", "application/json")
 
-			results = append(results, s1)
+	res, err := client.Do(req)
+	if err != nil {
+		return false, nil, err
+	}
+	defer func() {
+		_, _ = io.Copy(io.Discard, res.Body)
+		_ = res.Body.Close()
+	}()
+
+	if res.StatusCode == http.StatusOK {
+		var domains domainResponse
+		if err := json.NewDecoder(res.Body).Decode(&domains); err != nil {
+			return false, nil, fmt.Errorf("error decoding response body: %w", err)
+		}
+
+		var extraData map[string]string
+		if len(domains.Items) > 0 {
+			sb := strings.Builder{}
+			for i, item := range domains.Items {
+				if i != 0 {
+					sb.WriteString(", ")
+				}
+				sb.WriteString(item.Name)
+				sb.WriteString(" (")
+				sb.WriteString(item.State)
+				sb.WriteString(",")
+				sb.WriteString(item.Type)
+				if item.IsDisabled {
+					sb.WriteString(",disabled")
+				}
+				sb.WriteString(")")
+			}
+			extraData = map[string]string{
+				"Domains": sb.String(),
+			}
 		}
+
+		return true, extraData, nil
+	} else if res.StatusCode == http.StatusUnauthorized {
+		return false, nil, nil
+	} else {
+		return false, nil, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
 	}
+}
+
+type domainResponse struct {
+	TotalCount int    `json:"total_count"`
+	Items      []item `json:"items"`
+}
 
-	return results, nil
+type item struct {
+	ID         string `json:"id"`
+	IsDisabled bool   `json:"is_disabled"`
+	Name       string `json:"name"`
+	State      string `json:"state"`
+	Type       string `json:"type"`
 }
 
 func (s Scanner) Type() detectorspb.DetectorType {

pkg/detectors/mailgun/mailgun_test.go

@@ -9,13 +9,111 @@ import (
 	"testing"
 	"time"
 
+	"github.com/google/go-cmp/cmp"
 	"github.com/kylelemons/godebug/pretty"
+
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/engine/ahocorasick"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
 
+func TestMailgun_Pattern(t *testing.T) {
+	d := Scanner{}
+	ahoCorasickCore := ahocorasick.NewAhoCorasickCore([]detectors.Detector{d})
+	tests := []struct {
+		name  string
+		input string
+		want  []string
+	}{
+		// TODO: Confirm that this is actually an "original token".
+		// It's just a hex token encoded as basic auth.
+		{
+			name: "original token",
+			input: `- request:
+    method: get
+    uri: https://api.mailgun.net/v3/integration-test.domain.invalid/templates/test.template
+    body:
+      encoding: US-ASCII
+      string: ''
+    headers:
+      Accept:
+      - "*/*"
+      User-Agent:
+      - rest-client/2.1.0 (darwin21.6.0 x86_64) ruby/2.5.1p57
+      Accept-Encoding:
+      - gzip;q=1.0,deflate;q=0.6,identity;q=0.3
+      Host:
+      - api.mailgun.net
+      Authorization:
+      - Basic YXBpOmFjZWM0YzA1YjFmMmZjZWJjZmE4ZGE2NDVkYTEwMjMxLTQxM2UzNzNjLTBhYWQzYzM3`,
+			want: []string{"YXBpOmFjZWM0YzA1YjFmMmZjZWJjZmE4ZGE2NDVkYTEwMjMxLTQxM2UzNzNjLTBhYWQzYzM3"},
+		},
+		{
+			name: "key- token",
+			input: `public static ClientResponse GetBounce() {
+   Client client = new Client();
+   client.addFilter(new HTTPBasicAuthFilter("api",
+           "key-3ax63njp29jz6fds4gc373sgvjxteol1"));
+   WebResource webResource =
+       client.resource("https://api.mailgun.net/v2/samples.mailgun.org/" +
+               "bounces/[email protected]");
+   return webResource.get(ClientResponse.class);
+}`,
+			want: []string{"key-3ax63njp29jz6fds4gc373sgvjxteol1"},
+		},
+		{
+			name:  "hex token",
+			input: `curl -X POST https://api.mailgun.net/v3/DOMAIN.TEST/messages -u "api:e915b5cdb9a582685d8f3fb1bea0f20f-07bc7b05-f14816a1"`,
+			want:  []string{"e915b5cdb9a582685d8f3fb1bea0f20f-07bc7b05-f14816a1"},
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			chunkSpecificDetectors := make(map[ahocorasick.DetectorKey]detectors.Detector, 2)
+			ahoCorasickCore.PopulateMatchingDetectors(test.input, chunkSpecificDetectors)
+			if len(chunkSpecificDetectors) == 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)
+			}
+		})
+	}
+}
+
 func TestMailgun_FromChunk(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
 	defer cancel()