fixed uri regex issue (#3815)

* fixed uri regex issue

* extended password special char set

* rebased

* updated test case

* added rgmz suggested regex

* updated regex

Author: kashifkhan0771

pkg/detectors/privacy/privacy.go

@@ -3,10 +3,11 @@ package privacy
 import (
 	"context"
 	"fmt"
-	regexp "github.com/wasilibs/go-re2"
 	"net/http"
 	"strings"
 
+	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/detectorspb"

pkg/detectors/uri/uri.go

@@ -30,7 +30,7 @@ var _ interface {
 } = (*Scanner)(nil)
 
 var (
-	keyPat = regexp.MustCompile(`\b(?:https?:)?\/\/[\S]{3,50}:([\S]{3,50})@[-.%\w\/:]+\b`)
+	keyPat = regexp.MustCompile(`\bhttps?:\/\/[\w!#$%&()*+,\-./;<=>?@[\\\]^_{|}~]{0,50}:([\w!#$%&()*+,\-./:;<=>?[\\\]^_{|}~]{3,50})@[a-zA-Z0-9.-]+(?:\.[a-zA-Z]{2,})?(?::\d{1,5})?[\w/]+\b`)
 
 	// TODO: make local addr opt-out
 	defaultClient = detectors.DetectorHttpClientWithNoLocalAddresses
@@ -131,6 +131,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				continue
 			}
 		}
+
 		results = append(results, r)
 	}
 

pkg/detectors/uri/uri_test.go

@@ -13,6 +13,7 @@ import (
 
 var (
 	validPattern   = "https://kaNydBSAodo87dsm9asuiSAFtsd7.com:1234@qYY3SylY7fHP"
+	validPattern2  = `<p><a href="http://username:[email protected]">http://username:[email protected]</a></p>`
 	invalidPattern = "https://kaNydBSAodo87dsm9asuiSAFtsd7.com.1234@qYY3SylY7fHP"
 	keyword        = "uri"
 )
@@ -30,6 +31,11 @@ func TestURI_Pattern(t *testing.T) {
 			input: fmt.Sprintf("%s token = '%s'", keyword, validPattern),
 			want:  []string{validPattern},
 		},
+		{
+			name:  "valid pattern - do not process duplicate",
+			input: fmt.Sprintf("%s token = '%s'", keyword, validPattern2),
+			want:  []string{"http://username:[email protected]"},
+		},
 		{
 			name:  "invalid pattern",
 			input: fmt.Sprintf("%s = '%s'", keyword, invalidPattern),