Mirrors/trufflehog
2
0
Fork 0
mirror of https://github.com/trufflesecurity/trufflehog.git synced 2024-11-10 07:04:24 +00:00
Code Issues Projects Releases Packages Wiki Activity

Detector-Competition-Fix: Fix ScraperSite (deprecated) (#2074)

Browse source 
Co-authored-by: āh̳̕mͭͭͨͩ̐e̘ͬ́͋ͬ̊̓͂d <13666360+0x1@users.noreply.github.com>
This commit is contained in:
Corben Leo 2023-11-03 10:15:53 -05:00 committed by GitHub
parent 41e9cc59e2
commit 3b9ecaa704
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 27 additions and 237 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

89
pkg/detectors/scrapersite/scrapersite.go
View file

@ -1,89 +0,0 @@
package scrapersite
import (
"context"
"fmt"
"io"
"net/http"
"regexp"
"strings"
"time"
"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{}
// Ensure the Scanner satisfies the interface at compile time.
var _ detectors.Detector = (*Scanner)(nil)
var (
client = common.SaneHttpClientTimeOut(10 * time.Second)
// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"scrapersite"}) + `\b([a-zA-Z0-9]{45})\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{"scrapersite"}
}
// FromData will find and optionally verify ScraperSite 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 {
if len(match) != 2 {
continue
}
resMatch := strings.TrimSpace(match[1])
s1 := detectors.Result{
DetectorType: detectorspb.DetectorType_ScraperSite,
Raw: []byte(resMatch),
}
if verify {
req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://scrapersite.com/api-v1?api_key=%s&url=https://google.com", resMatch), nil)
if err != nil {
continue
}
res, err := client.Do(req)
if err == nil {
bodyBytes, err := io.ReadAll(res.Body)
if err != nil {
continue
}
bodyString := string(bodyBytes)
validResponse := strings.Contains(bodyString, `"status":true`)
defer res.Body.Close()
if res.StatusCode >= 200 && res.StatusCode < 300 {
if validResponse {
s1.Verified = true
} else {
s1.Verified = false
}
} 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, s1)
}
return results, nil
}
func (s Scanner) Type() detectorspb.DetectorType {
return detectorspb.DetectorType_ScraperSite
}

120
pkg/detectors/scrapersite/scrapersite_test.go
View file

@ -1,120 +0,0 @@
//go:build detectors
// +build detectors
package scrapersite
import (
"context"
"fmt"
"testing"
"time"
"github.com/kylelemons/godebug/pretty"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
"github.com/trufflesecurity/trufflehog/v3/pkg/common"
"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
)
func TestScraperSite_FromChunk(t *testing.T) {
ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
defer cancel()
testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors2")
if err != nil {
t.Fatalf("could not get test secrets from GCP: %s", err)
}
secret := testSecrets.MustGetField("SCRAPERSITE")
inactiveSecret := testSecrets.MustGetField("SCRAPERSITE_INACTIVE")
type args struct {
ctx context.Context
data []byte
verify bool
}
tests := []struct {
name string
s Scanner
args args
want []detectors.Result
wantErr bool
}{
{
name: "found, verified",
s: Scanner{},
args: args{
ctx: context.Background(),
data: []byte(fmt.Sprintf("You can find a scrapersite secret %s within", secret)),
verify: true,
},
want: []detectors.Result{
{
DetectorType: detectorspb.DetectorType_ScraperSite,
Verified: true,
},
},
wantErr: false,
},
{
name: "found, unverified",
s: Scanner{},
args: args{
ctx: context.Background(),
data: []byte(fmt.Sprintf("You can find a scrapersite secret %s within but not valid", inactiveSecret)), // the secret would satisfy the regex but not pass validation
verify: true,
},
want: []detectors.Result{
{
DetectorType: detectorspb.DetectorType_ScraperSite,
Verified: false,
},
},
wantErr: 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,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
s := Scanner{}
got, err := s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
if (err != nil) != tt.wantErr {
t.Errorf("ScraperSite.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])
}
got[i].Raw = nil
}
if diff := pretty.Compare(got, tt.want); diff != "" {
t.Errorf("ScraperSite.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)
}
}
})
}
}

2
pkg/engine/defaults.go
View file

@ -569,7 +569,6 @@ import (
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scrapeowl"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scraperapi"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scraperbox"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scrapersite"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scrapestack"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scrapfly"
"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/scrapingant"
@ -1315,7 +1314,6 @@ func DefaultDetectors() []detectors.Detector {
zenkitapi.Scanner{},
sherpadesk.Scanner{},
shotstack.Scanner{},
scrapersite.Scanner{},
luno.Scanner{},
apacta.Scanner{},
fmfw.Scanner{},

51
pkg/pb/detectorspb/detectors.pb.go
View file

@ -713,31 +713,32 @@ const (
DetectorType_Apacta DetectorType = 632
DetectorType_GetSandbox DetectorType = 633
// Deprecated: Do not use.
DetectorType_Happi DetectorType = 634
DetectorType_Oanda DetectorType = 635
DetectorType_FastForex DetectorType = 636
DetectorType_APIMatic DetectorType = 637
DetectorType_VersionEye DetectorType = 638
DetectorType_EagleEyeNetworks DetectorType = 639
DetectorType_ThousandEyes DetectorType = 640
DetectorType_SelectPDF DetectorType = 641
DetectorType_Flightstats DetectorType = 642
DetectorType_ChecIO DetectorType = 643
DetectorType_Manifest DetectorType = 644
DetectorType_ApiScience DetectorType = 645
DetectorType_AppSynergy DetectorType = 646
DetectorType_Caflou DetectorType = 647
DetectorType_Caspio DetectorType = 648
DetectorType_ChecklyHQ DetectorType = 649
DetectorType_CloudElements DetectorType = 650
DetectorType_DronaHQ DetectorType = 651
DetectorType_Enablex DetectorType = 652
DetectorType_Fmfw DetectorType = 653
DetectorType_GoodDay DetectorType = 654
DetectorType_Luno DetectorType = 655
DetectorType_Meistertask DetectorType = 656
DetectorType_Mindmeister DetectorType = 657
DetectorType_PeopleDataLabs DetectorType = 658
DetectorType_Happi DetectorType = 634
DetectorType_Oanda DetectorType = 635
DetectorType_FastForex DetectorType = 636
DetectorType_APIMatic DetectorType = 637
DetectorType_VersionEye DetectorType = 638
DetectorType_EagleEyeNetworks DetectorType = 639
DetectorType_ThousandEyes DetectorType = 640
DetectorType_SelectPDF DetectorType = 641
DetectorType_Flightstats DetectorType = 642
DetectorType_ChecIO DetectorType = 643
DetectorType_Manifest DetectorType = 644
DetectorType_ApiScience DetectorType = 645
DetectorType_AppSynergy DetectorType = 646
DetectorType_Caflou DetectorType = 647
DetectorType_Caspio DetectorType = 648
DetectorType_ChecklyHQ DetectorType = 649
DetectorType_CloudElements DetectorType = 650
DetectorType_DronaHQ DetectorType = 651
DetectorType_Enablex DetectorType = 652
DetectorType_Fmfw DetectorType = 653
DetectorType_GoodDay DetectorType = 654
DetectorType_Luno DetectorType = 655
DetectorType_Meistertask DetectorType = 656
DetectorType_Mindmeister DetectorType = 657
DetectorType_PeopleDataLabs DetectorType = 658
// Deprecated: Do not use.
DetectorType_ScraperSite DetectorType = 659
DetectorType_Scrapfly DetectorType = 660
DetectorType_SimplyNoted DetectorType = 661

2
proto/detectors.proto
View file

@ -667,7 +667,7 @@ enum DetectorType {
Meistertask = 656;
Mindmeister = 657;
PeopleDataLabs = 658;
ScraperSite = 659;
ScraperSite = 659 [deprecated = true];
Scrapfly = 660;
SimplyNoted = 661;
TravelPayouts = 662;