🐛 Fix .lib false positives in binary artifacts (#1879)

* ignore printable files

* updates

* e2e tests

* e2e fix

* comments

Author: laurentsimon

checks/raw/binary_artifact.go

@@ -18,6 +18,7 @@ import (
 	"fmt"
 	"path/filepath"
 	"strings"
+	"unicode"
 
 	"github.com/h2non/filetype"
 	"github.com/h2non/filetype/types"
@@ -92,8 +93,17 @@ var checkBinaryFileContent fileparser.DoWhileTrueOnFileContent = func(path strin
 	}
 
 	exists1 := binaryFileTypes[t.Extension]
+	if exists1 {
+		*pfiles = append(*pfiles, checker.File{
+                        Path:   path,
+                        Type:   checker.FileTypeBinary,
+                        Offset: checker.OffsetDefault,
+                })
+		return true, nil
+	}
+
 	exists2 := binaryFileTypes[strings.ReplaceAll(filepath.Ext(path), ".", "")]
-	if exists1 || exists2 {
+	if !isText(content) && exists2 {
 		*pfiles = append(*pfiles, checker.File{
 			Path:   path,
 			Type:   checker.FileTypeBinary,
@@ -103,3 +113,16 @@ var checkBinaryFileContent fileparser.DoWhileTrueOnFileContent = func(path strin
 
 	return true, nil
 }
+
+// TODO: refine this function.
+func isText(content []byte) bool {
+	for _, c := range string(content) {
+		if c == '\t' || c == '\n' || c == '\r' {
+			continue
+		}
+		if !unicode.IsPrint(c) {
+			return false
+		}
+	}
+	return true
+}

checks/raw/binary_artifact_test.go

@@ -16,7 +16,6 @@ package raw
 
 import (
 	"fmt"
-	"log"
 	"os"
 	"testing"
 
@@ -29,37 +28,40 @@ import (
 func TestBinaryArtifacts(t *testing.T) {
 	t.Parallel()
 	tests := []struct {
-		name      string
-		inputFile string
-		err       error
-		files     []string
-		expect    int
+		name   string
+		err    error
+		files  []string
+		expect int
 	}{
 		{
-			name:      "Jar file",
-			inputFile: "../testdata/binaryartifacts/jars/aws-java-sdk-core-1.11.571.jar",
-			err:       nil,
+			name: "Jar file",
+			err:  nil,
 			files: []string{
 				"../testdata/binaryartifacts/jars/aws-java-sdk-core-1.11.571.jar",
 			},
 			expect: 1,
 		},
 		{
-			name:      "non binary file",
-			inputFile: "../testdata/licensedir/withlicense/LICENSE",
-			err:       nil,
+			name: "non binary file",
+			err:  nil,
 			files: []string{
 				"../testdata/licensedir/withlicense/LICENSE",
 			},
 		},
 		{
-			name:      "non binary file",
-			inputFile: "../doesnotexist",
-			err:       nil,
+			name: "non binary file",
+			err:  nil,
 			files: []string{
 				"../doesnotexist",
 			},
 		},
+		{
+			name: "printable character .lib",
+			err:  nil,
+			files: []string{
+				"../testdata/binaryartifacts/printable.lib",
+			},
+		},
 	}
 	for _, tt := range tests {
 		tt := tt // Re-initializing variable so it is not changed while executing the closure below
@@ -72,7 +74,6 @@ func TestBinaryArtifacts(t *testing.T) {
 			mockRepoClient.EXPECT().GetFileContent(gomock.Any()).DoAndReturn(func(file string) ([]byte, error) {
 				// This will read the file and return the content
 				content, err := os.ReadFile(file)
-				log.Println(os.Getwd())
 				if err != nil {
 					return content, fmt.Errorf("%w", err)
 				}