Merge pull request #58 from kubescape/sourceInfoFiltering

Source info filtering
kubescape · Apr 30, 2023 · 548ba57 · 548ba57
2 parents 9dd48dd + decadf1
commit 548ba57
Show file tree

Hide file tree

Showing 3 changed files with 52 additions and 4 deletions.
diff --git a/main.go b/main.go
@@ -27,9 +27,7 @@ func main() {
 	if err != nil {
 		logger.L().Fatal("error during validation", helpers.Error(err))
 	}
-
 	context.SetBackgroundContext()
-	// after this line we can use logger.L().Ctx() to attach events to spans
 
 	accumulatorChannelError := make(chan error, 10)
 	acc := accumulator.GetAccumulator()

diff --git a/pkg/sbom/v1/sbom_spdx_storage_format.go b/pkg/sbom/v1/sbom_spdx_storage_format.go
@@ -29,9 +29,22 @@ const (
 	KubescapeNodeAgentName    = "KubescapeNodeAgent"
 	RelationshipContainType   = "CONTAINS"
 	directorySBOM             = "SBOM"
+	sourceInfoDotnet = "acquired package info from dotnet project assets file"
+	sourceInfoNodeModule = "acquired package info from installed node module manifest file"
+	sourceInfoPythonPackage = "acquired package info from installed python package manifest file"
+	sourceInfoJava = "acquired package info from installed java archive"
+	sourceInfoGemFile = "acquired package info from installed gem metadata file"
+	sourceInfoGoModule = "acquired package info from go module information"
+	sourceInfoRustCargo = "acquired package info from rust cargo manifest"
+	sourceInfoPHPComposer = "acquired package info from PHP composer manifest"
+	sourceInfoCabal = "acquired package info from cabal or stack manifest files"
+	sourceInfoRebar = "acquired package info from rebar3 or mix manifest file"
+	sourceInfoLinuxKernel = "acquired package info from linux kernel archive"
+	sourceInfoLinuxKernelModule = "acquired package info from linux kernel module files"
 )
 
 var spdxDataDirPath string
+var sourceInfoRequiredPrefix []string
 
 type SBOMData struct {
 	spdxDataPath                             string
@@ -62,6 +75,8 @@ func createSBOMDir() {
 
 func init() {
 	createSBOMDir()
+	sourceInfoPrefixData := []string{sourceInfoDotnet, sourceInfoNodeModule, sourceInfoPythonPackage, sourceInfoJava, sourceInfoGemFile, sourceInfoGoModule, sourceInfoRustCargo, sourceInfoPHPComposer, sourceInfoCabal, sourceInfoRebar, sourceInfoLinuxKernel, sourceInfoLinuxKernelModule}
+	sourceInfoRequiredPrefix = append(sourceInfoRequiredPrefix,sourceInfoPrefixData...) 
 }
 
 func CreateSBOMDataSPDXVersionV040(instanceID instanceidhandler.IInstanceID) SBOMFormat {
@@ -98,8 +113,18 @@ func (sbom *SBOMData) saveSBOM(spdxData *spdxv1beta1.SBOMSPDXv2p3) error {
 }
 
 func parsedFilesBySourceInfo(packageSourceInfo string) []string {
-	fileListInString := utils.After(packageSourceInfo, ": ")
-	return strings.Split(fileListInString, ", ")
+	needToMonitor := false
+	for i := range sourceInfoRequiredPrefix {
+		if strings.Contains(packageSourceInfo, sourceInfoRequiredPrefix[i]) {
+			needToMonitor = true
+			break
+		}
+	}
+	if needToMonitor {
+		fileListInString := utils.After(packageSourceInfo, ": ")
+		return strings.Split(fileListInString, ", ")
+	}
+	return []string{}
 }
 
 func (sbom *SBOMData) StoreSBOM(sbomData any) error {

diff --git a/pkg/sbom/v1/sbom_spdx_storage_format_test.go b/pkg/sbom/v1/sbom_spdx_storage_format_test.go
@@ -481,3 +481,28 @@ func TestParsedFilesBySourceInfo(t *testing.T) {
 		t.Fatalf("list[2] should be %s, not %s", "/usr/local/lib/python3.10/site-packages/Deprecated-1.2.13.dist-info/top_level.txt", list[2])
 	}
 }
+
+func TestParsedFilesBySourceInfoFiltered(t *testing.T) {
+	shouldBeSourcesInfo := []string{"acquired package info from dotnet project assets file: 123, 456", "acquired package info from installed node module manifest file: 123, 456", "acquired package info from installed python package manifest file: 123, 456", "acquired package info from installed java archive: 123, 456", "acquired package info from installed gem metadata file: 123, 456", "acquired package info from go module information: 123, 456", "acquired package info from rust cargo manifest: 123, 456", "acquired package info from PHP composer manifest: 123, 456", "acquired package info from cabal or stack manifest files: 123, 456", "acquired package info from rebar3 or mix manifest files: 123, 456", "acquired package info from linux kernel archive: 123, 456", "acquired package info from linux kernel module files: 123, 456"}
+	for i := range shouldBeSourcesInfo {
+		list := parsedFilesBySourceInfo(shouldBeSourcesInfo[i])
+		if len(list) != 2 {
+			t.Fatalf("source Info %s: parsed source Info list must be equal to 2", shouldBeSourcesInfo[i])
+		}
+		if list[0] != "123" {
+			t.Fatalf("list[0] should be %s, not %s", "/usr/local/lib/python3.10/site-packages/Deprecated-1.2.13.dist-info/METADATA", list[0])
+		}
+		if list[1] != "456" {
+			t.Fatalf("list[1] should be %s, not %s", "/usr/local/lib/python3.10/site-packages/Deprecated-1.2.13.dist-info/RECORD", list[1])
+		}
+	}
+
+
+	shouldNotBeSourcesInfo := []string{"acquired package info from ALPM DB: 1234, 456", "acquired package info from RPM DB: 1234, 456", "acquired package info from APK DB: 1234, 456", "acquired package info from DPKG DB: 1234, 456", "acquired package info from installed cocoapods manifest file: 1234, 456", "acquired package info from conan manifest: 1234, 456", "acquired package info from portage DB: 1234, 456", "acquired package info from nix store path: 123, 456"}
+	for i := range shouldNotBeSourcesInfo {
+		list := parsedFilesBySourceInfo(shouldNotBeSourcesInfo[i])
+		if len(list) != 0 {
+			t.Fatalf("source Info %s: parsed source Info list must be equal to 0", shouldNotBeSourcesInfo[i])
+		}
+	}
+}