diff --git a/pkg/detectors/wakatime/wakatime.go b/pkg/detectors/wakatime/wakatime.go new file mode 100644 index 000000000..193fee352 --- /dev/null +++ b/pkg/detectors/wakatime/wakatime.go @@ -0,0 +1,104 @@ +package wakatime + +import ( + "context" + "fmt" + "io" + "net/http" + + regexp "github.com/wasilibs/go-re2" + + "github.com/trufflesecurity/trufflehog/v3/pkg/common" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors" + "github.com/trufflesecurity/trufflehog/v3/pkg/pb/detector_typepb" +) + +type Scanner struct { + client *http.Client +} + +// Ensure the Scanner satisfies the interface at compile time. +var _ detectors.Detector = (*Scanner)(nil) + +var ( + defaultClient = common.SaneHttpClient() + // Make sure that your group is surrounded in boundary characters such as below to reduce false positives. + keyPat = regexp.MustCompile(`\b(waka_[0-9a-f]{8}-[0-9a-f]{4}-4[0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{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{"waka_"} +} + +// FromData will find and optionally verify Wakatime 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 _, match := range keyPat.FindAllStringSubmatch(dataStr, -1) { + uniqueMatches[match[1]] = struct{}{} + } + + for match := range uniqueMatches { + s1 := detectors.Result{ + DetectorType: detector_typepb.DetectorType_Wakatime, + Raw: []byte(match), + SecretParts: map[string]string{"key": match}, + } + + if verify { + client := s.client + if client == nil { + client = defaultClient + } + + isVerified, extraData, verificationErr := verifyMatch(ctx, client, match) + s1.Verified = isVerified + s1.ExtraData = extraData + s1.SetVerificationError(verificationErr, match) + } + + results = append(results, s1) + } + + return +} + +func verifyMatch(ctx context.Context, client *http.Client, token string) (bool, map[string]string, error) { + req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://wakatime.com/api/v1/users/current", nil) + if err != nil { + return false, nil, err + } + + req.SetBasicAuth(token, "") + + res, err := client.Do(req) + if err != nil { + return false, nil, err + } + defer func() { + _, _ = io.Copy(io.Discard, res.Body) + _ = res.Body.Close() + }() + + switch res.StatusCode { + case http.StatusOK: + // If the endpoint returns useful information, we can return it as a map. + return true, nil, nil + case http.StatusUnauthorized: + // The secret is determinately not verified (nothing to do) + return false, nil, nil + default: + return false, nil, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode) + } +} + +func (s Scanner) Type() detector_typepb.DetectorType { + return detector_typepb.DetectorType_Wakatime +} + +func (s Scanner) Description() string { + return "WakaTime is an automated time-tracking and metrics platform designed for software developers." +} diff --git a/pkg/detectors/wakatime/wakatime_integration_test.go b/pkg/detectors/wakatime/wakatime_integration_test.go new file mode 100644 index 000000000..0e3e92b9a --- /dev/null +++ b/pkg/detectors/wakatime/wakatime_integration_test.go @@ -0,0 +1,168 @@ +//go:build detectors +// +build detectors + +package wakatime + +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/detector_typepb" +) + +func TestWakatime_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) + } + secret := testSecrets.MustGetField("WAKATIME") + inactiveSecret := testSecrets.MustGetField("WAKATIME_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: context.Background(), + data: []byte(fmt.Sprintf("You can find a wakatime secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detector_typepb.DetectorType_Wakatime, + Verified: true, + }, + }, + wantErr: false, + wantVerificationErr: false, + }, + { + name: "found, unverified", + s: Scanner{}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a wakatime secret %s within but not valid", inactiveSecret)), // the secret would satisfy the regex but not pass validation + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detector_typepb.DetectorType_Wakatime, + 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, + }, + { + name: "found, would be verified if not for timeout", + s: Scanner{client: common.SaneHttpClientTimeOut(1 * time.Microsecond)}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a wakatime secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detector_typepb.DetectorType_Wakatime, + Verified: false, + }, + }, + wantErr: false, + wantVerificationErr: true, + }, + { + name: "found, verified but unexpected api surface", + s: Scanner{client: common.ConstantResponseHttpClient(404, "")}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a wakatime secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detector_typepb.DetectorType_Wakatime, + Verified: false, + }, + }, + wantErr: false, + wantVerificationErr: true, + }, + } + 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("Wakatime.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", + "primarySecret", + "SecretParts", + "chunkOffset", + "chunkOffsetSet", + ) + if diff := cmp.Diff(got, tt.want, ignoreOpts); diff != "" { + t.Errorf("Wakatime.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) + } + } + }) + } +} diff --git a/pkg/detectors/wakatime/wakatime_test.go b/pkg/detectors/wakatime/wakatime_test.go new file mode 100644 index 000000000..406454a82 --- /dev/null +++ b/pkg/detectors/wakatime/wakatime_test.go @@ -0,0 +1,106 @@ +package wakatime + +import ( + "context" + "fmt" + "testing" + + "github.com/google/go-cmp/cmp" + "github.com/stretchr/testify/require" + + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors" + "github.com/trufflesecurity/trufflehog/v3/pkg/engine/ahocorasick" +) + +func TestWakatime_Pattern(t *testing.T) { + d := Scanner{} + ahoCorasickCore := ahocorasick.NewAhoCorasickCore([]detectors.Detector{d}) + key1 := "waka_" + "cd1ff59d-4108-4b5f-bdb8-46da1d95c3d5" + key2 := "waka_" + "cd1ff59c-4108-4b5f-bdb8-46da1d95c3d5" + + tests := []struct { + name string + input string + want []string + }{ + { + name: "valid pattern", + input: fmt.Sprintf(` + [INFO] Sending request to the wakatime API + [DEBUG] Using Key=%s + [INFO] Response received: 200 OK + `, key1), + want: []string{key1}, + }, + { + name: "valid pattern - xml", + input: fmt.Sprintf(` + + GLOBAL + {wakatime} + {%s} + configuration for production + 2023-05-18T14:32:10Z + jenkins-admin + + `, key1), + want: []string{key1}, + }, + { + name: "finds all matches", + input: fmt.Sprintf(` + [INFO] Sending request to the wakatime API + [DEBUG] Using Key=%s + [ERROR] Response received 401 UnAuthorized + [DEBUG] Using wakatime Key=%s + [INFO] Response received: 200 OK + `, key1, key2), + want: []string{key1, key2}, + }, + { + name: "invalid pattern", + input: ` + [INFO] Sending request to the wakatime API + [DEBUG] Using Key=waka_cd1ff59d-4108-4b5f-bb8-46da1d95c3d5 + [ERROR] Response received: 401 UnAuthorized + `, + want: []string{}, + }, + } + + for _, test := range tests { + t.Run(test.name, func(t *testing.T) { + matchedDetectors := ahoCorasickCore.FindDetectorMatches([]byte(test.input)) + if len(matchedDetectors) == 0 { + t.Errorf("test %q failed: expected keywords %v to be found in the input", test.name, d.Keywords()) + return + } + + results, err := d.FromData(context.Background(), false, []byte(test.input)) + require.NoError(t, err) + + if len(results) != len(test.want) { + t.Errorf("mismatch in result count: expected %d, got %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) + } + }) + } +} diff --git a/pkg/engine/defaults/defaults.go b/pkg/engine/defaults/defaults.go index c40abc463..0ed415ca1 100644 --- a/pkg/engine/defaults/defaults.go +++ b/pkg/engine/defaults/defaults.go @@ -854,6 +854,7 @@ import ( "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/vpnapi" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/vultrapikey" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/vyte" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/wakatime" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/walkscore" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/weatherbit" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/weatherstack" @@ -1781,6 +1782,7 @@ func buildDetectorList() []detectors.Detector { &vultrapikey.Scanner{}, &vyte.Scanner{}, &walkscore.Scanner{}, + &wakatime.Scanner{}, &weatherbit.Scanner{}, &weatherstack.Scanner{}, &web3storage.Scanner{}, diff --git a/pkg/engine/defaults/defaults_test.go b/pkg/engine/defaults/defaults_test.go index f64bebfe2..7eafc1551 100644 --- a/pkg/engine/defaults/defaults_test.go +++ b/pkg/engine/defaults/defaults_test.go @@ -351,7 +351,6 @@ var excludedFromDefaultList = map[detector_typepb.DetectorType]struct{}{ detector_typepb.DetectorType_Uproc: {}, detector_typepb.DetectorType_Veevavault: {}, detector_typepb.DetectorType_Vonage: {}, - detector_typepb.DetectorType_Wakatime: {}, detector_typepb.DetectorType_Webengage: {}, detector_typepb.DetectorType_WeChatAppKey: {}, detector_typepb.DetectorType_Woopra: {}, diff --git a/proto/detector_type.proto b/proto/detector_type.proto index 2982e7056..2951a2693 100644 --- a/proto/detector_type.proto +++ b/proto/detector_type.proto @@ -143,7 +143,7 @@ enum DetectorType { D7Network = 138; ScrapingBee = 139; KeenIO = 140; - Wakatime = 141; // Not yet implemented + Wakatime = 141; Buildkite = 142; Verimail = 143; Zerobounce = 144;