panic parsing cyclonedx (#892)

Signed-off-by: houdini91 <mdstrauss91@gmail.com>

.github/scripts/apple-signing/notarize.sh

@@ -29,7 +29,7 @@ notarize() {
   fi
 
   # install gon
-  which gon || (brew tap mitchellh/gon && brew install mitchellh/gon/gon)
+  which gon || (go install github.com/mitchellh/gon/cmd/gon@latest)
 
   # create config (note: json via stdin with gon is broken, can only use HCL from file)
   hcl_file=$(mktemp).hcl

README.md

@@ -11,6 +11,14 @@
 
 A CLI tool and Go library for generating a Software Bill of Materials (SBOM) from container images and filesystems. Exceptional for vulnerability detection when used with a scanner tool like [Grype](https://github.com/anchore/grype).
 
+### Join our Virtual OSS Meetup!
+
+You are invited to join us on March 23rd, 11AM-Noon PT for our virtual open source meetup.  
+
+Hosts Priyanka “Pinky” Ravi of Weaveworks and Dan Luhring from Anchore OSS will explore how open source project Flux is utilizing Syft-generated SBOMs to secure the software supply chain.
+
+[Register here!](https://get.anchore.com/anchore-oss-meetup-mar-23-2022/)
+
 ### Join our community meetings!
 
 - Calendar: https://calendar.google.com/calendar/u/0/r?cid=Y182OTM4dGt0MjRtajI0NnNzOThiaGtnM29qNEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t

cmd/format_aliases.go

@@ -22,6 +22,8 @@ func formatAliases(ids ...sbom.FormatID) (aliases []string) {
 			aliases = append(aliases, "cyclonedx-xml")
 		case syft.CycloneDxJSONFormatID:
 			aliases = append(aliases, "cyclonedx-json")
+		case syft.GitHubID:
+			aliases = append(aliases, "github", "github-json")
 		default:
 			aliases = append(aliases, string(id))
 		}

go.mod

@@ -12,7 +12,7 @@ require (
 	github.com/anchore/go-rpmdb v0.0.0-20210914181456-a9c52348da63
 	github.com/anchore/go-testutils v0.0.0-20200925183923-d5f45b0d3c04
 	github.com/anchore/go-version v1.2.2-0.20200701162849-18adb9c92b9b
-	github.com/anchore/packageurl-go v0.0.0-20210922164639-b3fa992ebd29
+	github.com/anchore/packageurl-go v0.1.1-0.20220314153042-1bcd40e5206b
 	github.com/anchore/stereoscope v0.0.0-20220307154759-8a5a70c227d3
 	github.com/antihax/optional v1.0.0
 	github.com/bmatcuk/doublestar/v4 v4.0.2
@@ -171,7 +171,7 @@ require (
 	github.com/hashicorp/go-uuid v1.0.2 // indirect
 	github.com/hashicorp/go-version v1.4.0 // indirect
 	github.com/hashicorp/golang-lru v0.5.4 // indirect
-	github.com/hashicorp/vault/api v1.3.1 // indirect
+	github.com/hashicorp/vault/api v1.4.1 // indirect
 	github.com/hashicorp/vault/sdk v0.4.1 // indirect
 	github.com/hashicorp/yamux v0.0.0-20211028200310-0bc27b27de87 // indirect
 	github.com/imdario/mergo v0.3.12 // indirect

go.sum

@@ -282,8 +282,8 @@ github.com/anchore/go-testutils v0.0.0-20200925183923-d5f45b0d3c04 h1:VzprUTpc0v
 github.com/anchore/go-testutils v0.0.0-20200925183923-d5f45b0d3c04/go.mod h1:6dK64g27Qi1qGQZ67gFmBFvEHScy0/C8qhQhNe5B5pQ=
 github.com/anchore/go-version v1.2.2-0.20200701162849-18adb9c92b9b h1:e1bmaoJfZVsCYMrIZBpFxwV26CbsuoEh5muXD5I1Ods=
 github.com/anchore/go-version v1.2.2-0.20200701162849-18adb9c92b9b/go.mod h1:Bkc+JYWjMCF8OyZ340IMSIi2Ebf3uwByOk6ho4wne1E=
-github.com/anchore/packageurl-go v0.0.0-20210922164639-b3fa992ebd29 h1:K9LfnxwhqvihqU0+MF325FNy7fsKV9EGaUxdfR4gnWk=
-github.com/anchore/packageurl-go v0.0.0-20210922164639-b3fa992ebd29/go.mod h1:Oc1UkGaJwY6ND6vtAqPSlYrptKRJngHwkwB6W7l1uP0=
+github.com/anchore/packageurl-go v0.1.1-0.20220314153042-1bcd40e5206b h1:YJWYt/6KQXR9JR46lLHrTTYi8rcye42tKcyjREA/hvA=
+github.com/anchore/packageurl-go v0.1.1-0.20220314153042-1bcd40e5206b/go.mod h1:Blo6OgJNiYF41ufcgHKkbCKF2MDOMlrqhXv/ij6ocR4=
 github.com/anchore/stereoscope v0.0.0-20220307154759-8a5a70c227d3 h1:Kx2jlMdENAf4cVjYGYLI+fiavVhzhtmU89GUYDITJ1w=
 github.com/anchore/stereoscope v0.0.0-20220307154759-8a5a70c227d3/go.mod h1:XESZQTgFETDBatmyoet6XZ0zVknoIMDSAhj2INj2a5w=
 github.com/andreyvit/diff v0.0.0-20170406064948-c7f18ee00883/go.mod h1:rCTlJbsFo29Kk6CurOXKm700vrz8f0KW0JNfpkRJY/8=
@@ -1349,8 +1349,9 @@ github.com/hashicorp/serf v0.8.2/go.mod h1:6hOLApaqBFA1NXqRQAsxw9QxuDEvNxSQRwA/J
 github.com/hashicorp/serf v0.9.5/go.mod h1:UWDWwZeL5cuWDJdl0C6wrvrUwEqtQ4ZKBKKENpqIUyk=
 github.com/hashicorp/serf v0.9.6/go.mod h1:TXZNMjZQijwlDvp+r0b63xZ45H7JmCmgg4gpTwn9UV4=
 github.com/hashicorp/vault/api v1.3.0/go.mod h1:EabNQLI0VWbWoGlA+oBLC8PXmR9D60aUVgQGvangFWQ=
-github.com/hashicorp/vault/api v1.3.1 h1:pkDkcgTh47PRjY1NEFeofqR4W/HkNUi9qIakESO2aRM=
 github.com/hashicorp/vault/api v1.3.1/go.mod h1:QeJoWxMFt+MsuWcYhmwRLwKEXrjwAFFywzhptMsTIUw=
+github.com/hashicorp/vault/api v1.4.1 h1:mWLfPT0RhxBitjKr6swieCEP2v5pp/M//t70S3kMLRo=
+github.com/hashicorp/vault/api v1.4.1/go.mod h1:LkMdrZnWNrFaQyYYazWVn7KshilfDidgVBq6YiTq/bM=
 github.com/hashicorp/vault/sdk v0.3.0/go.mod h1:aZ3fNuL5VNydQk8GcLJ2TV8YCRVvyaakYkhZRoVuhj0=
 github.com/hashicorp/vault/sdk v0.4.1 h1:3SaHOJY687jY1fnB61PtL0cOkKItphrbLmux7T92HBo=
 github.com/hashicorp/vault/sdk v0.4.1/go.mod h1:aZ3fNuL5VNydQk8GcLJ2TV8YCRVvyaakYkhZRoVuhj0=

internal/constants.go

@@ -6,5 +6,5 @@ const (
 
 	// JSONSchemaVersion is the current schema version output by the JSON encoder
 	// This is roughly following the "SchemaVer" guidelines for versioning the JSON schema. Please see schema/json/README.md for details on how to increment.
-	JSONSchemaVersion = "3.0.1"
+	JSONSchemaVersion = "3.1.0"
 )

internal/formats/common/cyclonedxhelpers/component.go

@@ -47,8 +47,10 @@ func hasMetadata(p pkg.Package) bool {
 
 func decodeComponent(c *cyclonedx.Component) *pkg.Package {
 	values := map[string]string{}
-	for _, p := range *c.Properties {
-		values[p.Name] = p.Value
+	if c.Properties != nil {
+		for _, p := range *c.Properties {
+			values[p.Name] = p.Value
+		}
 	}
 
 	p := &pkg.Package{

internal/formats/common/cyclonedxhelpers/decoder.go

@@ -5,6 +5,8 @@ import (
 	"io"
 
 	"github.com/CycloneDX/cyclonedx-go"
+
+	"github.com/anchore/syft/internal/formats/common"
 	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/linux"
 	"github.com/anchore/syft/syft/pkg"
@@ -157,6 +159,14 @@ func linuxReleaseFromOSComponent(component *cyclonedx.Component) *linux.Release
 		}
 	}
 
+	if component.Properties != nil {
+		values := map[string]string{}
+		for _, p := range *component.Properties {
+			values[p.Name] = p.Value
+		}
+		common.DecodeInto(&rel, values, "syft:distro", CycloneDXFields)
+	}
+
 	return rel
 }
 

internal/formats/github/encoder.go

@@ -0,0 +1,183 @@
+package github
+
+import (
+	"fmt"
+	"strings"
+	"time"
+
+	"github.com/mholt/archiver/v3"
+
+	"github.com/anchore/packageurl-go"
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/version"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+)
+
+// toGithubModel converts the provided SBOM to a GitHub dependency model
+func toGithubModel(s *sbom.SBOM) DependencySnapshot {
+	scanTime := time.Now().Format(time.RFC3339) // TODO is there a record of this somewhere?
+	v := version.FromBuild().Version
+	if v == "[not provided]" {
+		v = "0.0.0-dev"
+	}
+	return DependencySnapshot{
+		Version: 0,
+		// TODO allow property input to specify the Job, Sha, and Ref
+		Detector: DetectorMetadata{
+			Name:    internal.ApplicationName,
+			URL:     "https://github.com/anchore/syft",
+			Version: v,
+		},
+		Metadata:  toSnapshotMetadata(s),
+		Manifests: toGithubManifests(s),
+		Scanned:   scanTime,
+	}
+}
+
+// toSnapshotMetadata captures the linux distribution information and other metadata
+func toSnapshotMetadata(s *sbom.SBOM) Metadata {
+	out := Metadata{}
+
+	if s.Artifacts.LinuxDistribution != nil {
+		d := s.Artifacts.LinuxDistribution
+		qualifiers := packageurl.Qualifiers{}
+		if len(d.IDLike) > 0 {
+			qualifiers = append(qualifiers, packageurl.Qualifier{
+				Key:   "like",
+				Value: strings.Join(d.IDLike, ","),
+			})
+		}
+		purl := packageurl.NewPackageURL("generic", "", d.ID, d.VersionID, qualifiers, "")
+		out["syft:distro"] = purl.ToString()
+	}
+
+	return out
+}
+
+func filesystem(p pkg.Package) string {
+	if len(p.Locations) > 0 {
+		return p.Locations[0].FileSystemID
+	}
+	return ""
+}
+
+// isArchive returns true if the path appears to be an archive
+func isArchive(path string) bool {
+	_, err := archiver.ByExtension(path)
+	return err == nil
+}
+
+// toPath Generates a string representation of the package location, optionally including the layer hash
+func toPath(s source.Metadata, p pkg.Package) string {
+	inputPath := strings.TrimPrefix(s.Path, "./")
+	if inputPath == "." {
+		inputPath = ""
+	}
+	if len(p.Locations) > 0 {
+		location := p.Locations[0]
+		packagePath := location.RealPath
+		if location.VirtualPath != "" {
+			packagePath = location.VirtualPath
+		}
+		packagePath = strings.TrimPrefix(packagePath, "/")
+		switch s.Scheme {
+		case source.ImageScheme:
+			image := strings.ReplaceAll(s.ImageMetadata.UserInput, ":/", "//")
+			return fmt.Sprintf("%s:/%s", image, packagePath)
+		case source.FileScheme:
+			if isArchive(inputPath) {
+				return fmt.Sprintf("%s:/%s", inputPath, packagePath)
+			}
+			return inputPath
+		case source.DirectoryScheme:
+			if inputPath != "" {
+				return fmt.Sprintf("%s/%s", inputPath, packagePath)
+			}
+			return packagePath
+		}
+	}
+	return fmt.Sprintf("%s%s", inputPath, s.ImageMetadata.UserInput)
+}
+
+// toGithubManifests manifests, each of which represents a specific location that has dependencies
+func toGithubManifests(s *sbom.SBOM) Manifests {
+	manifests := map[string]*Manifest{}
+
+	for _, p := range s.Artifacts.PackageCatalog.Sorted() {
+		path := toPath(s.Source, p)
+		manifest, ok := manifests[path]
+		if !ok {
+			manifest = &Manifest{
+				Name: path,
+				File: FileInfo{
+					SourceLocation: path,
+				},
+				Resolved: DependencyGraph{},
+			}
+			fs := filesystem(p)
+			if fs != "" {
+				manifest.Metadata = Metadata{
+					"syft:filesystem": fs,
+				}
+			}
+			manifests[path] = manifest
+		}
+
+		name := dependencyName(p)
+		manifest.Resolved[name] = DependencyNode{
+			Purl:         p.PURL,
+			Metadata:     toDependencyMetadata(p),
+			Relationship: toDependencyRelationshipType(p),
+			Scope:        toDependencyScope(p),
+			Dependencies: toDependencies(s, p),
+		}
+	}
+
+	out := Manifests{}
+	for k, v := range manifests {
+		out[k] = *v
+	}
+	return out
+}
+
+// dependencyName to make things a little nicer to read; this might end up being lossy
+func dependencyName(p pkg.Package) string {
+	purl, err := packageurl.FromString(p.PURL)
+	if err != nil {
+		log.Warnf("Invalid PURL for package: '%s' PURL: '%s' (%w)", p.Name, p.PURL, err)
+		return ""
+	}
+	// don't use qualifiers for this
+	purl.Qualifiers = nil
+	return purl.ToString()
+}
+
+func toDependencyScope(_ pkg.Package) DependencyScope {
+	return DependencyScopeRuntime
+}
+
+func toDependencyRelationshipType(_ pkg.Package) DependencyRelationship {
+	return DependencyRelationshipDirect
+}
+
+func toDependencyMetadata(_ pkg.Package) Metadata {
+	// We have limited properties: up to 8 with reasonably small values
+	// For now, we are encoding the location as part of the key, we are encoding PURLs with most
+	// of the other information Grype might need; and the distro information at the top level
+	// so we don't need anything here yet
+	return Metadata{}
+}
+
+func toDependencies(s *sbom.SBOM, p pkg.Package) (out []string) {
+	for _, r := range s.Relationships {
+		if r.From.ID() == p.ID() {
+			if p, ok := r.To.(pkg.Package); ok {
+				out = append(out, dependencyName(p))
+			}
+		}
+	}
+	return
+}

internal/formats/github/encoder_test.go

@@ -0,0 +1,161 @@
+package github
+
+import (
+	"encoding/json"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+
+	"github.com/anchore/packageurl-go"
+	"github.com/anchore/syft/syft/linux"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+)
+
+func Test_toGithubModel(t *testing.T) {
+	s := sbom.SBOM{
+		Source: source.Metadata{
+			Scheme: source.ImageScheme,
+			ImageMetadata: source.ImageMetadata{
+				UserInput:    "ubuntu:18.04",
+				Architecture: "amd64",
+			},
+		},
+		Artifacts: sbom.Artifacts{
+			LinuxDistribution: &linux.Release{
+				ID:        "ubuntu",
+				VersionID: "18.04",
+				IDLike:    []string{"debian"},
+			},
+			PackageCatalog: pkg.NewCatalog(),
+		},
+	}
+	for _, p := range []pkg.Package{
+		{
+			Name:    "pkg-1",
+			Version: "1.0.1",
+			Locations: []source.Location{{
+				Coordinates: source.Coordinates{
+					RealPath:     "/usr/lib",
+					FileSystemID: "fsid-1",
+				},
+			}},
+		},
+		{
+			Name:    "pkg-2",
+			Version: "2.0.2",
+			Locations: []source.Location{{
+				Coordinates: source.Coordinates{
+					RealPath:     "/usr/lib",
+					FileSystemID: "fsid-1",
+				},
+			}},
+		},
+		{
+			Name:    "pkg-3",
+			Version: "3.0.3",
+			Locations: []source.Location{{
+				Coordinates: source.Coordinates{
+					RealPath:     "/etc",
+					FileSystemID: "fsid-1",
+				},
+			}},
+		},
+	} {
+		p.PURL = packageurl.NewPackageURL(
+			"generic",
+			"",
+			p.Name,
+			p.Version,
+			nil,
+			"",
+		).ToString()
+		s.Artifacts.PackageCatalog.Add(p)
+	}
+
+	actual := toGithubModel(&s)
+
+	expected := DependencySnapshot{
+		Version: 0,
+		Detector: DetectorMetadata{
+			Name:    "syft",
+			Version: "0.0.0-dev",
+			URL:     "https://github.com/anchore/syft",
+		},
+		Metadata: Metadata{
+			"syft:distro": "pkg:generic/ubuntu@18.04?like=debian",
+		},
+		Scanned: actual.Scanned,
+		Manifests: Manifests{
+			"ubuntu:18.04:/usr/lib": Manifest{
+				Name: "ubuntu:18.04:/usr/lib",
+				File: FileInfo{
+					SourceLocation: "ubuntu:18.04:/usr/lib",
+				},
+				Metadata: Metadata{
+					"syft:filesystem": "fsid-1",
+				},
+				Resolved: DependencyGraph{
+					"pkg:generic/pkg-1@1.0.1": DependencyNode{
+						Purl:         "pkg:generic/pkg-1@1.0.1",
+						Scope:        DependencyScopeRuntime,
+						Relationship: DependencyRelationshipDirect,
+					},
+					"pkg:generic/pkg-2@2.0.2": DependencyNode{
+						Purl:         "pkg:generic/pkg-2@2.0.2",
+						Scope:        DependencyScopeRuntime,
+						Relationship: DependencyRelationshipDirect,
+					},
+				},
+			},
+			"ubuntu:18.04:/etc": Manifest{
+				Name: "ubuntu:18.04:/etc",
+				File: FileInfo{
+					SourceLocation: "ubuntu:18.04:/etc",
+				},
+				Metadata: Metadata{
+					"syft:filesystem": "fsid-1",
+				},
+				Resolved: DependencyGraph{
+					"pkg:generic/pkg-3@3.0.3": DependencyNode{
+						Purl:         "pkg:generic/pkg-3@3.0.3",
+						Scope:        DependencyScopeRuntime,
+						Relationship: DependencyRelationshipDirect,
+					},
+				},
+			},
+		},
+	}
+
+	// just using JSONEq because it gives a comprehensible diff
+	s1, _ := json.Marshal(expected)
+	s2, _ := json.Marshal(actual)
+	assert.JSONEq(t, string(s1), string(s2))
+
+	// Just test the other schemes:
+	s.Source.Path = "."
+	s.Source.Scheme = source.DirectoryScheme
+	actual = toGithubModel(&s)
+	assert.Equal(t, "etc", actual.Manifests["etc"].Name)
+
+	s.Source.Path = "./artifacts"
+	s.Source.Scheme = source.DirectoryScheme
+	actual = toGithubModel(&s)
+	assert.Equal(t, "artifacts/etc", actual.Manifests["artifacts/etc"].Name)
+
+	s.Source.Path = "/artifacts"
+	s.Source.Scheme = source.DirectoryScheme
+	actual = toGithubModel(&s)
+	assert.Equal(t, "/artifacts/etc", actual.Manifests["/artifacts/etc"].Name)
+
+	s.Source.Path = "./executable"
+	s.Source.Scheme = source.FileScheme
+	actual = toGithubModel(&s)
+	assert.Equal(t, "executable", actual.Manifests["executable"].Name)
+
+	s.Source.Path = "./archive.tar.gz"
+	s.Source.Scheme = source.FileScheme
+	actual = toGithubModel(&s)
+	assert.Equal(t, "archive.tar.gz:/etc", actual.Manifests["archive.tar.gz:/etc"].Name)
+}

internal/formats/github/format.go

@@ -0,0 +1,29 @@
+package github
+
+import (
+	"encoding/json"
+	"io"
+
+	"github.com/anchore/syft/syft/sbom"
+)
+
+const ID sbom.FormatID = "github-0-json"
+
+func Format() sbom.Format {
+	return sbom.NewFormat(
+		ID,
+		func(writer io.Writer, sbom sbom.SBOM) error {
+			bom := toGithubModel(&sbom)
+
+			bytes, err := json.MarshalIndent(bom, "", "  ")
+			if err != nil {
+				return err
+			}
+			_, err = writer.Write(bytes)
+
+			return err
+		},
+		nil,
+		nil,
+	)
+}

internal/formats/github/github_dependency_api.go

@@ -0,0 +1,78 @@
+package github
+
+// Derived from: https://gist.github.com/reiddraper/fdab2883db0f372c146d1a750fc1c43f
+
+type DependencySnapshot struct {
+	Version   int              `json:"version"`
+	Job       Job              `json:"job,omitempty"` // !omitempty
+	Sha       string           `json:"sha,omitempty"` // !omitempty sha of the Git commit
+	Ref       string           `json:"ref,omitempty"` // !omitempty ref of the Git commit example "refs/heads/main"
+	Detector  DetectorMetadata `json:"detector,omitempty"`
+	Metadata  Metadata         `json:"metadata,omitempty"`
+	Manifests Manifests        `json:"manifests,omitempty"`
+	Scanned   ISO8601Date      `json:"scanned,omitempty"`
+}
+
+type Job struct {
+	Name    string `json:"name,omitempty"` // !omitempty
+	ID      string `json:"id,omitempty"`   // !omitempty
+	HTMLURL string `json:"html_url,omitempty"`
+}
+
+type DetectorMetadata struct {
+	Name    string `json:"name,omitempty"`
+	URL     string `json:"url,omitempty"`
+	Version string `json:"version,omitempty"`
+}
+
+type Manifests map[string]Manifest
+
+// Manifest A collection of related dependencies, either declared in a file,
+// or representing a logical group of dependencies.
+type Manifest struct {
+	Name     string          `json:"name"`
+	File     FileInfo        `json:"file"`
+	Metadata Metadata        `json:"metadata,omitempty"`
+	Resolved DependencyGraph `json:"resolved,omitempty"`
+}
+
+type FileInfo struct {
+	SourceLocation string `json:"source_location,omitempty"`
+}
+
+// DependencyRelationship A notation of whether a dependency is requested directly
+// by this manifest, or is a dependency of another dependency.
+type DependencyRelationship string
+
+const (
+	DependencyRelationshipDirect   DependencyRelationship = "direct"
+	DependencyRelationshipIndirect DependencyRelationship = "indirect"
+)
+
+// DependencyScope A notation of whether the dependency is required for the primary
+// build artifact (runtime), or is only used for development.
+// Future versions of this specification may allow for more granular
+// scopes, like `runtimeserver`, `runtimeshipped`,
+// `developmenttest`, `developmentbenchmark`.
+type DependencyScope string
+
+const (
+	DependencyScopeRuntime     DependencyScope = "runtime"
+	DependencyScopeDevelopment DependencyScope = "development"
+)
+
+type DependencyNode struct {
+	Purl         string                 `json:"purl,omitempty"`
+	Metadata     Metadata               `json:"metadata,omitempty"`
+	Relationship DependencyRelationship `json:"relationship,omitempty"`
+	Scope        DependencyScope        `json:"scope,omitempty"`
+	Dependencies []string               `json:"dependencies,omitempty"`
+}
+
+type DependencyGraph map[string]DependencyNode
+
+type ISO8601Date = string
+
+type Scalar interface{} // should be: null | boolean | string | number
+
+type Metadata map[string]Scalar

internal/formats/syftjson/test-fixtures/snapshot/TestDirectoryEncoder.golden

@@ -88,7 +88,7 @@
   }
  },
  "schema": {
-  "version": "3.0.1",
-  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.0.1.json"
+  "version": "3.1.0",
+  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.1.0.json"
  }
 }

internal/formats/syftjson/test-fixtures/snapshot/TestEncodeFullJSONDocument.golden

@@ -184,7 +184,7 @@
   }
  },
  "schema": {
-  "version": "3.0.1",
-  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.0.1.json"
+  "version": "3.1.0",
+  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.1.0.json"
  }
 }

internal/formats/syftjson/test-fixtures/snapshot/TestImageEncoder.golden

@@ -111,7 +111,7 @@
   }
  },
  "schema": {
-  "version": "3.0.1",
-  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.0.1.json"
+  "version": "3.1.0",
+  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-3.1.0.json"
  }
 }

syft/formats.go

@@ -6,6 +6,7 @@ import (
 
 	"github.com/anchore/syft/internal/formats/cyclonedxjson"
 	"github.com/anchore/syft/internal/formats/cyclonedxxml"
+	"github.com/anchore/syft/internal/formats/github"
 	"github.com/anchore/syft/internal/formats/spdx22json"
 	"github.com/anchore/syft/internal/formats/spdx22tagvalue"
 	"github.com/anchore/syft/internal/formats/syftjson"
@@ -21,6 +22,7 @@ const (
 	TableFormatID         = table.ID
 	CycloneDxXMLFormatID  = cyclonedxxml.ID
 	CycloneDxJSONFormatID = cyclonedxjson.ID
+	GitHubID              = github.ID
 	SPDXTagValueFormatID  = spdx22tagvalue.ID
 	SPDXJSONFormatID      = spdx22json.ID
 )
@@ -32,6 +34,7 @@ func init() {
 		syftjson.Format(),
 		cyclonedxxml.Format(),
 		cyclonedxjson.Format(),
+		github.Format(),
 		spdx22tagvalue.Format(),
 		spdx22json.Format(),
 		table.Format(),
@@ -71,6 +74,8 @@ func FormatByName(name string) sbom.Format {
 		return FormatByID(cyclonedxxml.ID)
 	case "cyclonedxjson":
 		return FormatByID(cyclonedxjson.ID)
+	case "github":
+		return FormatByID(github.ID)
 	case "spdx", "spdxtv", "spdxtagvalue":
 		return FormatByID(spdx22tagvalue.ID)
 	case "spdxjson":

syft/pkg/cataloger/common/cpe/java.go

@@ -17,6 +17,7 @@ var (
 		"org",
 		"net",
 		"io",
+		"be",
 	}
 
 	primaryJavaManifestGroupIDFields = []string{

syft/pkg/cataloger/golang/parse_go_bin.go

@@ -70,18 +70,26 @@ func buildGoPkgInfo(location source.Location, mod, goVersion, arch string) []pkg
 	for scanner.Scan() {
 		fields := strings.Fields(scanner.Text())
 
-		// must have dep, name, version, sha
-		if len(fields) < 4 {
+		// must have dep, name, version
+		if len(fields) < 3 {
 			continue
 		}
 
-		if fields[0] == packageIdentifier || fields[0] == replaceIdentifier {
-			name := fields[1]
-			version := fields[2]
-			h1Digest := fields[3]
+		name := fields[1]
+		version := fields[2]
+		h1Digest := ""
+		// if dep is *not* vendored, it'll also have h1digest
+		if len(fields) >= 4 {
+			h1Digest = fields[3]
+		}
+
+		if fields[0] == packageIdentifier {
 			pkgsSlice = append(pkgsSlice, newGoBinaryPackage(name, version, h1Digest, goVersion, arch, location))
 		}
+		if fields[0] == replaceIdentifier {
+			// replace the previous entry in the package slice
+			pkgsSlice[len(pkgsSlice)-1] = newGoBinaryPackage(name, version, h1Digest, goVersion, arch, location)
+		}
 	}
-
 	return pkgsSlice
 }

syft/pkg/cataloger/golang/parse_go_bin_test.go

@@ -28,7 +28,8 @@ func TestBuildGoPkgInfo(t *testing.T) {
 			name: "buildGoPkgInfo parses a populated mod string and returns packages but no source info",
 			mod: `path    github.com/anchore/syft mod     github.com/anchore/syft (devel)
 				  dep     github.com/adrg/xdg     v0.2.1  h1:VSVdnH7cQ7V+B33qSJHTCRlNgra1607Q8PzEmnvb2Ic=
-				  dep     github.com/anchore/client-go    v0.0.0-20210222170800-9c70f9b80bcf      h1:DYssiUV1pBmKqzKsm4mqXx8artqC0Q8HgZsVI3lMsAg=`,
+				  dep     github.com/anchore/client-go    v0.0.0-20210222170800-9c70f9b80bcf      h1:DYssiUV1pBmKqzKsm4mqXx8artqC0Q8HgZsVI3lMsAg=
+				  dep     github.com/anchore/client-go    v1.2.3`,
 			expected: []pkg.Package{
 				{
 					Name:     "github.com/adrg/xdg",
@@ -70,6 +71,25 @@ func TestBuildGoPkgInfo(t *testing.T) {
 						H1Digest:          "h1:DYssiUV1pBmKqzKsm4mqXx8artqC0Q8HgZsVI3lMsAg=",
 					},
 				},
+				{
+					Name:     "github.com/anchore/client-go",
+					Version:  "v1.2.3",
+					Language: pkg.Go,
+					Type:     pkg.GoModulePkg,
+					Locations: []source.Location{
+						{
+							Coordinates: source.Coordinates{
+								RealPath:     "/a-path",
+								FileSystemID: "layer-id",
+							},
+						},
+					},
+					MetadataType: pkg.GolangBinMetadataType,
+					Metadata: pkg.GolangBinMetadata{
+						GoCompiledVersion: goCompiledVersion,
+						Architecture:      archDetails,
+					},
+				},
 			},
 		},
 		{
@@ -79,7 +99,8 @@ func TestBuildGoPkgInfo(t *testing.T) {
 				  dep     golang.org/x/net        v0.0.0-20211006190231-62292e806868      h1:KlOXYy8wQWTUJYFgkUI40Lzr06ofg5IRXUK5C7qZt1k=
 				  dep     golang.org/x/sys        v0.0.0-20211006194710-c8a6f5223071      h1:PjhxBct4MZii8FFR8+oeS7QOvxKOTZXgk63EU2XpfJE=
 				  dep     golang.org/x/term       v0.0.0-20210927222741-03fcf44c2211
-				  =>      golang.org/x/term       v0.0.0-20210916214954-140adaaadfaf      h1:Ihq/mm/suC88gF8WFcVwk+OV6Tq+wyA1O0E5UEvDglI=`,
+				  =>      golang.org/x/term       v0.0.0-20210916214954-140adaaadfaf      h1:Ihq/mm/suC88gF8WFcVwk+OV6Tq+wyA1O0E5UEvDglI=
+				  dep     github.com/anchore/client-go    v1.2.3`,
 			expected: []pkg.Package{
 				{
 					Name:     "golang.org/x/net",
@@ -141,6 +162,25 @@ func TestBuildGoPkgInfo(t *testing.T) {
 						H1Digest:          "h1:Ihq/mm/suC88gF8WFcVwk+OV6Tq+wyA1O0E5UEvDglI=",
 					},
 				},
+				{
+					Name:     "github.com/anchore/client-go",
+					Version:  "v1.2.3",
+					Language: pkg.Go,
+					Type:     pkg.GoModulePkg,
+					Locations: []source.Location{
+						{
+							Coordinates: source.Coordinates{
+								RealPath:     "/a-path",
+								FileSystemID: "layer-id",
+							},
+						},
+					},
+					MetadataType: pkg.GolangBinMetadataType,
+					Metadata: pkg.GolangBinMetadata{
+						GoCompiledVersion: goCompiledVersion,
+						Architecture:      archDetails,
+					},
+				},
 			},
 		},
 	}

syft/pkg/dpkg_metadata_test.go

@@ -75,7 +75,7 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 				Version:       "v",
 				SourceVersion: "2.3",
 			},
-			expected: "pkg:deb/debian/p@v?upstream=s@2.3&distro=debian-11",
+			expected: "pkg:deb/debian/p@v?upstream=s%402.3&distro=debian-11",
 		},
 	}
 

syft/pkg/npm_package_json_metadata_test.go

@@ -1,18 +1,21 @@
 package pkg
 
 import (
+	"fmt"
+	"testing"
+
 	"github.com/anchore/packageurl-go"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"testing"
 )
 
 func TestNpmPackageJSONMetadata_PackageURL(t *testing.T) {
 
 	tests := []struct {
-		name     string
-		metadata NpmPackageJSONMetadata
-		expected string
+		name      string
+		metadata  NpmPackageJSONMetadata
+		expected  string
+		namespace string
 	}{
 		{
 			name: "no namespace",
@@ -24,19 +27,36 @@ func TestNpmPackageJSONMetadata_PackageURL(t *testing.T) {
 		},
 		{
 			name: "split by namespace",
+			metadata: NpmPackageJSONMetadata{
+				Name:    "npmcli/arborist",
+				Version: "2.6.2",
+			},
+			expected:  "pkg:npm/npmcli/arborist@2.6.2",
+			namespace: "npmcli",
+		},
+		{
+			name: "encoding @ symobl",
 			metadata: NpmPackageJSONMetadata{
 				Name:    "@npmcli/arborist",
 				Version: "2.6.2",
 			},
-			expected: "pkg:npm/@npmcli/arborist@2.6.2",
+			expected:  "pkg:npm/%40npmcli/arborist@2.6.2",
+			namespace: "@npmcli",
 		},
 	}
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
 			actual := tt.metadata.PackageURL(nil)
 			assert.Equal(t, tt.expected, actual)
-			_, err := packageurl.FromString(actual)
+			decoded, err := packageurl.FromString(actual)
 			require.NoError(t, err)
+			assert.Equal(t, tt.namespace, decoded.Namespace)
+			if decoded.Namespace != "" {
+				assert.Equal(t, tt.metadata.Name, fmt.Sprintf("%s/%s", decoded.Namespace, decoded.Name))
+			} else {
+				assert.Equal(t, tt.metadata.Name, decoded.Name)
+			}
+			assert.Equal(t, tt.metadata.Version, decoded.Version)
 		})
 	}
 }

syft/pkg/python_package_metadata_test.go

@@ -27,7 +27,7 @@ func TestPythonPackageMetadata_pURL(t *testing.T) {
 					CommitID: "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
 				},
 			},
-			expected: "pkg:pypi/name@v0.1.0?vcs_url=git+https:%2F%2Fgithub.com%2Ftest%2Ftest.git@aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+			expected: "pkg:pypi/name@v0.1.0?vcs_url=git+https://github.com/test/test.git%40aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
 		},
 		{
 			name: "should not respond to release info",

syft/pkg/rpmdb_metadata.go

@@ -5,12 +5,10 @@ import (
 	"sort"
 	"strconv"
 
-	"github.com/anchore/syft/syft/file"
-
-	"github.com/scylladb/go-set/strset"
-
 	"github.com/anchore/packageurl-go"
+	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/linux"
+	"github.com/scylladb/go-set/strset"
 )
 
 const RpmDBGlob = "**/var/lib/rpm/Packages"
@@ -24,7 +22,7 @@ var (
 type RpmdbMetadata struct {
 	Name      string            `json:"name"`
 	Version   string            `json:"version"`
-	Epoch     *int              `json:"epoch"  cyclonedx:"epoch"`
+	Epoch     *int              `json:"epoch"  cyclonedx:"epoch" jsonschema:"nullable"`
 	Arch      string            `json:"architecture"`
 	Release   string            `json:"release" cyclonedx:"release"`
 	SourceRpm string            `json:"sourceRpm" cyclonedx:"sourceRpm"`

syft/pkg/url.go

@@ -50,8 +50,13 @@ func URL(p Package, release *linux.Release) string {
 		fields := re.Split(p.Name, -1)
 		namespace = fields[0]
 		name = strings.TrimPrefix(p.Name, namespace+"/")
+	case p.Type == NpmPkg:
+		fields := strings.SplitN(p.Name, "/", 2)
+		if len(fields) > 1 {
+			namespace = fields[0]
+			name = fields[1]
+		}
 	}
-
 	// generate a purl from the package data
 	return packageurl.NewPackageURL(
 		purlType,

test/integration/encode_decode_cycle_test.go

@@ -54,35 +54,38 @@ func TestEncodeDecodeEncodeCycleComparison(t *testing.T) {
 	for _, test := range tests {
 		t.Run(string(test.formatOption), func(t *testing.T) {
 
-			originalSBOM, _ := catalogFixtureImage(t, "image-pkg-coverage")
+			// use second image for relationships
+			for _, image := range []string{"image-pkg-coverage", "image-owning-package"} {
+				originalSBOM, _ := catalogFixtureImage(t, image)
 
-			format := syft.FormatByID(test.formatOption)
-			require.NotNil(t, format)
+				format := syft.FormatByID(test.formatOption)
+				require.NotNil(t, format)
 
-			by1, err := syft.Encode(originalSBOM, format)
-			assert.NoError(t, err)
+				by1, err := syft.Encode(originalSBOM, format)
+				assert.NoError(t, err)
 
-			newSBOM, newFormat, err := syft.Decode(bytes.NewReader(by1))
-			assert.NoError(t, err)
-			assert.Equal(t, format.ID(), newFormat.ID())
+				newSBOM, newFormat, err := syft.Decode(bytes.NewReader(by1))
+				assert.NoError(t, err)
+				assert.Equal(t, format.ID(), newFormat.ID())
 
-			by2, err := syft.Encode(*newSBOM, format)
-			assert.NoError(t, err)
+				by2, err := syft.Encode(*newSBOM, format)
+				assert.NoError(t, err)
 
-			if test.redactor != nil {
-				by1 = test.redactor(by1)
-				by2 = test.redactor(by2)
-			}
+				if test.redactor != nil {
+					by1 = test.redactor(by1)
+					by2 = test.redactor(by2)
+				}
 
-			if test.json {
-				s1 := string(by1)
-				s2 := string(by2)
-				assert.JSONEq(t, s1, s2)
-			} else {
-				if !assert.True(t, bytes.Equal(by1, by2)) {
-					dmp := diffmatchpatch.New()
-					diffs := dmp.DiffMain(string(by1), string(by2), true)
-					t.Errorf("diff: %s", dmp.DiffPrettyText(diffs))
+				if test.json {
+					s1 := string(by1)
+					s2 := string(by2)
+					assert.JSONEq(t, s1, s2)
+				} else {
+					if !assert.True(t, bytes.Equal(by1, by2)) {
+						dmp := diffmatchpatch.New()
+						diffs := dmp.DiffMain(string(by1), string(by2), true)
+						t.Errorf("diff: %s", dmp.DiffPrettyText(diffs))
+					}
 				}
 			}
 		})