Introduce new format pattern + port json processing (#550)

* add new format pattern Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * add syftjson format Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * add internal formats helper Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * add SBOM encode/decode to lib API Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * remove json presenter + update presenter tests to use common utils Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * remove presenter format enum type + add formats shim in presenter helper Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * add MustCPE helper for tests Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * update usage of format enum Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * add test fixtures for encode/decode tests Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * fix integration test Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * migrate format detection to use reader Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * address review comments Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
2026-11-01 23:16:30 +01:00 · 2021-10-20 21:36:34 +00:00
parent 5e315c0f17
commit 560b05c2c9
57 changed files with 3139 additions and 421 deletions
@@ -0,0 +1,200 @@
+package testutils
+
+import (
+	"bytes"
+	"testing"
+
+	"github.com/anchore/syft/syft/presenter"
+
+	"github.com/anchore/go-testutils"
+	"github.com/anchore/stereoscope/pkg/filetree"
+	"github.com/anchore/stereoscope/pkg/imagetest"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+	"github.com/sergi/go-diff/diffmatchpatch"
+	"github.com/stretchr/testify/assert"
+)
+
+type redactor func(s []byte) []byte
+
+func AssertPresenterAgainstGoldenImageSnapshot(t *testing.T, pres presenter.Presenter, testImage string, updateSnapshot bool, redactors ...redactor) {
+	var buffer bytes.Buffer
+
+	// grab the latest image contents and persist
+	if updateSnapshot {
+		imagetest.UpdateGoldenFixtureImage(t, testImage)
+	}
+
+	err := pres.Present(&buffer)
+	assert.NoError(t, err)
+	actual := buffer.Bytes()
+
+	// replace the expected snapshot contents with the current presenter contents
+	if updateSnapshot {
+		testutils.UpdateGoldenFileContents(t, actual)
+	}
+
+	var expected = testutils.GetGoldenFileContents(t)
+
+	// remove dynamic values, which should be tested independently
+	for _, r := range redactors {
+		actual = r(actual)
+		expected = r(expected)
+	}
+
+	// assert that the golden file snapshot matches the actual contents
+	if !bytes.Equal(expected, actual) {
+		dmp := diffmatchpatch.New()
+		diffs := dmp.DiffMain(string(expected), string(actual), true)
+		t.Errorf("mismatched output:\n%s", dmp.DiffPrettyText(diffs))
+	}
+}
+
+func AssertPresenterAgainstGoldenSnapshot(t *testing.T, pres presenter.Presenter, updateSnapshot bool, redactors ...redactor) {
+	var buffer bytes.Buffer
+
+	err := pres.Present(&buffer)
+	assert.NoError(t, err)
+	actual := buffer.Bytes()
+
+	// replace the expected snapshot contents with the current presenter contents
+	if updateSnapshot {
+		testutils.UpdateGoldenFileContents(t, actual)
+	}
+
+	var expected = testutils.GetGoldenFileContents(t)
+
+	// remove dynamic values, which should be tested independently
+	for _, r := range redactors {
+		actual = r(actual)
+		expected = r(expected)
+	}
+
+	if !bytes.Equal(expected, actual) {
+		dmp := diffmatchpatch.New()
+		diffs := dmp.DiffMain(string(expected), string(actual), true)
+		t.Errorf("mismatched output:\n%s", dmp.DiffPrettyText(diffs))
+	}
+}
+
+func ImageInput(t testing.TB, testImage string) (*pkg.Catalog, source.Metadata, *distro.Distro) {
+	t.Helper()
+	catalog := pkg.NewCatalog()
+	img := imagetest.GetGoldenFixtureImage(t, testImage)
+
+	_, ref1, _ := img.SquashedTree().File("/somefile-1.txt", filetree.FollowBasenameLinks)
+	_, ref2, _ := img.SquashedTree().File("/somefile-2.txt", filetree.FollowBasenameLinks)
+
+	// populate catalog with test data
+	catalog.Add(pkg.Package{
+		ID:      "package-1-id",
+		Name:    "package-1",
+		Version: "1.0.1",
+		Locations: []source.Location{
+			source.NewLocationFromImage(string(ref1.RealPath), *ref1, img),
+		},
+		Type:         pkg.PythonPkg,
+		FoundBy:      "the-cataloger-1",
+		Language:     pkg.Python,
+		MetadataType: pkg.PythonPackageMetadataType,
+		Licenses:     []string{"MIT"},
+		Metadata: pkg.PythonPackageMetadata{
+			Name:    "package-1",
+			Version: "1.0.1",
+		},
+		PURL: "a-purl-1",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:1:*:*:*:*:*:*:*"),
+		},
+	})
+	catalog.Add(pkg.Package{
+		ID:      "package-2-id",
+		Name:    "package-2",
+		Version: "2.0.1",
+		Locations: []source.Location{
+			source.NewLocationFromImage(string(ref2.RealPath), *ref2, img),
+		},
+		Type:         pkg.DebPkg,
+		FoundBy:      "the-cataloger-2",
+		MetadataType: pkg.DpkgMetadataType,
+		Metadata: pkg.DpkgMetadata{
+			Package: "package-2",
+			Version: "2.0.1",
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+
+	// this is a hard coded value that is not given by the fixture helper and must be provided manually
+	img.Metadata.ManifestDigest = "sha256:2731251dc34951c0e50fcc643b4c5f74922dad1a5d98f302b504cf46cd5d9368"
+
+	src, err := source.NewFromImage(img, "user-image-input")
+	assert.NoError(t, err)
+
+	dist, err := distro.NewDistro(distro.Debian, "1.2.3", "like!")
+	assert.NoError(t, err)
+
+	return catalog, src.Metadata, &dist
+}
+
+func DirectoryInput(t testing.TB) (*pkg.Catalog, source.Metadata, *distro.Distro) {
+	catalog := pkg.NewCatalog()
+
+	// populate catalog with test data
+	catalog.Add(pkg.Package{
+		ID:      "package-1-id",
+		Name:    "package-1",
+		Version: "1.0.1",
+		Type:    pkg.PythonPkg,
+		FoundBy: "the-cataloger-1",
+		Locations: []source.Location{
+			{RealPath: "/some/path/pkg1"},
+		},
+		Language:     pkg.Python,
+		MetadataType: pkg.PythonPackageMetadataType,
+		Licenses:     []string{"MIT"},
+		Metadata: pkg.PythonPackageMetadata{
+			Name:    "package-1",
+			Version: "1.0.1",
+			Files: []pkg.PythonFileRecord{
+				{
+					Path: "/some/path/pkg1/dependencies/foo",
+				},
+			},
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+	catalog.Add(pkg.Package{
+		ID:      "package-2-id",
+		Name:    "package-2",
+		Version: "2.0.1",
+		Type:    pkg.DebPkg,
+		FoundBy: "the-cataloger-2",
+		Locations: []source.Location{
+			{RealPath: "/some/path/pkg1"},
+		},
+		MetadataType: pkg.DpkgMetadataType,
+		Metadata: pkg.DpkgMetadata{
+			Package: "package-2",
+			Version: "2.0.1",
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+
+	dist, err := distro.NewDistro(distro.Debian, "1.2.3", "like!")
+	assert.NoError(t, err)
+
+	src, err := source.NewFromDirectory("/some/path")
+	assert.NoError(t, err)
+
+	return catalog, src.Metadata, &dist
+}
@@ -0,0 +1,34 @@
+package formats
+
+import (
+	"bytes"
+
+	"github.com/anchore/syft/internal/formats/syftjson"
+	"github.com/anchore/syft/syft/format"
+)
+
+// TODO: eventually this is the source of truth for all formatters
+func All() []format.Format {
+	return []format.Format{
+		syftjson.Format(),
+	}
+}
+
+func Identify(by []byte) (*format.Format, error) {
+	for _, f := range All() {
+		if err := f.Validate(bytes.NewReader(by)); err != nil {
+			continue
+		}
+		return &f, nil
+	}
+	return nil, nil
+}
+
+func ByOption(option format.Option) *format.Format {
+	for _, f := range All() {
+		if f.Option == option {
+			return &f
+		}
+	}
+	return nil
+}
@@ -0,0 +1,34 @@
+package formats
+
+import (
+	"io"
+	"os"
+	"testing"
+
+	"github.com/anchore/syft/syft/format"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestIdentify(t *testing.T) {
+	tests := []struct {
+		fixture  string
+		expected format.Option
+	}{
+		{
+			fixture:  "test-fixtures/alpine-syft.json",
+			expected: format.JSONOption,
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.fixture, func(t *testing.T) {
+			f, err := os.Open(test.fixture)
+			assert.NoError(t, err)
+			by, err := io.ReadAll(f)
+			assert.NoError(t, err)
+			frmt, err := Identify(by)
+			assert.NoError(t, err)
+			assert.NotNil(t, frmt)
+			assert.Equal(t, test.expected, frmt.Option)
+		})
+	}
+}
@@ -0,0 +1,24 @@
+package syftjson
+
+import (
+	"encoding/json"
+	"fmt"
+	"io"
+
+	"github.com/anchore/syft/internal/formats/syftjson/model"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+func decoder(reader io.Reader) (*pkg.Catalog, *source.Metadata, *distro.Distro, source.Scope, error) {
+	dec := json.NewDecoder(reader)
+
+	var doc model.Document
+	err := dec.Decode(&doc)
+	if err != nil {
+		return nil, nil, nil, source.UnknownScope, fmt.Errorf("unable to decode syft-json: %w", err)
+	}
+
+	return toSyftModel(doc)
+}
@@ -0,0 +1,52 @@
+package syftjson
+
+import (
+	"bytes"
+	"strings"
+	"testing"
+
+	"github.com/anchore/syft/syft/source"
+
+	"github.com/anchore/syft/internal/formats/common/testutils"
+	"github.com/go-test/deep"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestEncodeDecodeCycle(t *testing.T) {
+	testImage := "image-simple"
+	originalCatalog, originalMetadata, _ := testutils.ImageInput(t, testImage)
+
+	var buf bytes.Buffer
+	assert.NoError(t, encoder(&buf, originalCatalog, &originalMetadata, nil, source.SquashedScope))
+
+	actualCatalog, actualMetadata, _, _, err := decoder(bytes.NewReader(buf.Bytes()))
+	assert.NoError(t, err)
+
+	for _, d := range deep.Equal(originalMetadata, *actualMetadata) {
+		t.Errorf("metadata difference: %+v", d)
+	}
+
+	actualPackages := actualCatalog.Sorted()
+	for idx, p := range originalCatalog.Sorted() {
+		if !assert.Equal(t, p.Name, actualPackages[idx].Name) {
+			t.Errorf("different package at idx=%d: %s vs %s", idx, p.Name, actualPackages[idx].Name)
+			continue
+		}
+
+		// ids will never be equal
+		p.ID = ""
+		actualPackages[idx].ID = ""
+
+		for _, d := range deep.Equal(*p, *actualPackages[idx]) {
+			if strings.Contains(d, ".VirtualPath: ") {
+				// location.Virtual path is not exposed in the json output
+				continue
+			}
+			if strings.HasSuffix(d, "<nil slice> != []") {
+				// semantically the same
+				continue
+			}
+			t.Errorf("package difference (%s): %+v", p.Name, d)
+		}
+	}
+}
@@ -0,0 +1,23 @@
+package syftjson
+
+import (
+	"encoding/json"
+	"io"
+
+	"github.com/anchore/syft/syft/distro"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+func encoder(output io.Writer, catalog *pkg.Catalog, srcMetadata *source.Metadata, d *distro.Distro, scope source.Scope) error {
+	// TODO: application config not available yet
+	doc := ToFormatModel(catalog, srcMetadata, d, scope, nil)
+
+	enc := json.NewEncoder(output)
+	// prevent > and < from being escaped in the payload
+	enc.SetEscapeHTML(false)
+	enc.SetIndent("", " ")
+
+	return enc.Encode(&doc)
+}
@@ -0,0 +1,31 @@
+package syftjson
+
+import (
+	"flag"
+	"testing"
+
+	"github.com/anchore/syft/syft/source"
+
+	"github.com/anchore/syft/internal/formats/common/testutils"
+	"github.com/anchore/syft/syft/format"
+)
+
+var updateJson = flag.Bool("update-json", false, "update the *.golden files for json presenters")
+
+func TestDirectoryPresenter(t *testing.T) {
+	catalog, metadata, distro := testutils.DirectoryInput(t)
+	testutils.AssertPresenterAgainstGoldenSnapshot(t,
+		format.NewPresenter(encoder, catalog, &metadata, distro, source.SquashedScope),
+		*updateJson,
+	)
+}
+
+func TestImagePresenter(t *testing.T) {
+	testImage := "image-simple"
+	catalog, metadata, distro := testutils.ImageInput(t, testImage)
+	testutils.AssertPresenterAgainstGoldenImageSnapshot(t,
+		format.NewPresenter(encoder, catalog, &metadata, distro, source.SquashedScope),
+		testImage,
+		*updateJson,
+	)
+}
@@ -0,0 +1,12 @@
+package syftjson
+
+import "github.com/anchore/syft/syft/format"
+
+func Format() format.Format {
+	return format.NewFormat(
+		format.JSONOption,
+		encoder,
+		decoder,
+		validator,
+	)
+}
@@ -0,0 +1,8 @@
+package model
+
+// Distro provides information about a detected Linux Distro.
+type Distro struct {
+	Name    string `json:"name"`    // Name of the Linux distribution
+	Version string `json:"version"` // Version of the Linux distribution (major or major.minor version)
+	IDLike  string `json:"idLike"`  // the ID_LIKE field found within the /etc/os-release file
+}
@@ -0,0 +1,23 @@
+package model
+
+// Document represents the syft cataloging findings as a JSON document
+type Document struct {
+	Artifacts             []Package      `json:"artifacts"` // Artifacts is the list of packages discovered and placed into the catalog
+	ArtifactRelationships []Relationship `json:"artifactRelationships"`
+	Source                Source         `json:"source"`     // Source represents the original object that was cataloged
+	Distro                Distro         `json:"distro"`     // Distro represents the Linux distribution that was detected from the source
+	Descriptor            Descriptor     `json:"descriptor"` // Descriptor is a block containing self-describing information about syft
+	Schema                Schema         `json:"schema"`     // Schema is a block reserved for defining the version for the shape of this JSON document and where to find the schema document to validate the shape
+}
+
+// Descriptor describes what created the document as well as surrounding metadata
+type Descriptor struct {
+	Name          string      `json:"name"`
+	Version       string      `json:"version"`
+	Configuration interface{} `json:"configuration,omitempty"`
+}
+
+type Schema struct {
+	Version string `json:"version"`
+	URL     string `json:"url"`
+}
@@ -0,0 +1,123 @@
+package model
+
+import (
+	"encoding/json"
+	"fmt"
+
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+// Package represents a pkg.Package object specialized for JSON marshaling and unmarshaling.
+type Package struct {
+	PackageBasicData
+	PackageCustomData
+}
+
+// PackageBasicData contains non-ambiguous values (type-wise) from pkg.Package.
+type PackageBasicData struct {
+	ID        string            `json:"id"`
+	Name      string            `json:"name"`
+	Version   string            `json:"version"`
+	Type      pkg.Type          `json:"type"`
+	FoundBy   string            `json:"foundBy"`
+	Locations []source.Location `json:"locations"`
+	Licenses  []string          `json:"licenses"`
+	Language  pkg.Language      `json:"language"`
+	CPEs      []string          `json:"cpes"`
+	PURL      string            `json:"purl"`
+}
+
+// PackageCustomData contains ambiguous values (type-wise) from pkg.Package.
+type PackageCustomData struct {
+	MetadataType pkg.MetadataType `json:"metadataType"`
+	Metadata     interface{}      `json:"metadata"`
+}
+
+// packageMetadataUnpacker is all values needed from Package to disambiguate ambiguous fields during json unmarshaling.
+type packageMetadataUnpacker struct {
+	MetadataType pkg.MetadataType `json:"metadataType"`
+	Metadata     json.RawMessage  `json:"metadata"`
+}
+
+func (p *packageMetadataUnpacker) String() string {
+	return fmt.Sprintf("metadataType: %s, metadata: %s", p.MetadataType, string(p.Metadata))
+}
+
+// UnmarshalJSON is a custom unmarshaller for handling basic values and values with ambiguous types.
+// nolint:funlen
+func (p *Package) UnmarshalJSON(b []byte) error {
+	var basic PackageBasicData
+	if err := json.Unmarshal(b, &basic); err != nil {
+		return err
+	}
+	p.PackageBasicData = basic
+
+	var unpacker packageMetadataUnpacker
+	if err := json.Unmarshal(b, &unpacker); err != nil {
+		log.Warnf("failed to unmarshall into packageMetadataUnpacker: %v", err)
+		return err
+	}
+
+	p.MetadataType = unpacker.MetadataType
+
+	switch p.MetadataType {
+	case pkg.ApkMetadataType:
+		var payload pkg.ApkMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.RpmdbMetadataType:
+		var payload pkg.RpmdbMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.DpkgMetadataType:
+		var payload pkg.DpkgMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.JavaMetadataType:
+		var payload pkg.JavaMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.RustCargoPackageMetadataType:
+		var payload pkg.CargoPackageMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.GemMetadataType:
+		var payload pkg.GemMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.KbPackageMetadataType:
+		var payload pkg.KbPackageMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.PythonPackageMetadataType:
+		var payload pkg.PythonPackageMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	case pkg.NpmPackageJSONMetadataType:
+		var payload pkg.NpmPackageJSONMetadata
+		if err := json.Unmarshal(unpacker.Metadata, &payload); err != nil {
+			return err
+		}
+		p.Metadata = payload
+	}
+
+	return nil
+}
@@ -0,0 +1,8 @@
+package model
+
+type Relationship struct {
+	Parent   string      `json:"parent"`
+	Child    string      `json:"child"`
+	Type     string      `json:"type"`
+	Metadata interface{} `json:"metadata"`
+}
@@ -0,0 +1,57 @@
+package model
+
+import (
+	"encoding/json"
+	"fmt"
+	"strconv"
+
+	"github.com/anchore/syft/syft/source"
+)
+
+// Source object represents the thing that was cataloged
+type Source struct {
+	Type   string      `json:"type"`
+	Target interface{} `json:"target"`
+}
+
+// sourceUnpacker is used to unmarshal Source objects
+type sourceUnpacker struct {
+	Type   string          `json:"type"`
+	Target json.RawMessage `json:"target"`
+}
+
+type ImageSource struct {
+	source.ImageMetadata
+	Scope source.Scope `json:"scope"`
+}
+
+// UnmarshalJSON populates a source object from JSON bytes.
+func (s *Source) UnmarshalJSON(b []byte) error {
+	var unpacker sourceUnpacker
+	if err := json.Unmarshal(b, &unpacker); err != nil {
+		return err
+	}
+
+	s.Type = unpacker.Type
+
+	switch s.Type {
+	case "directory":
+		if target, err := strconv.Unquote(string(unpacker.Target)); err == nil {
+			s.Target = target
+		} else {
+			s.Target = string(unpacker.Target[:])
+		}
+
+	case "image":
+		var payload ImageSource
+		if err := json.Unmarshal(unpacker.Target, &payload); err != nil {
+			return err
+		}
+		s.Target = payload
+
+	default:
+		return fmt.Errorf("unsupported package metadata type: %+v", s.Type)
+	}
+
+	return nil
+}
@@ -0,0 +1,4 @@
+# Note: changes to this file will result in updating several test values. Consider making a new image fixture instead of editing this one.
+FROM scratch
+ADD file-1.txt /somefile-1.txt
+ADD file-2.txt /somefile-2.txt
@@ -0,0 +1 @@
+this file has contents
@@ -0,0 +1 @@
+file-2 contents!
@@ -0,0 +1,86 @@
+{
+ "artifacts": [
+  {
+   "id": "package-1-id",
+   "name": "package-1",
+   "version": "1.0.1",
+   "type": "python",
+   "foundBy": "the-cataloger-1",
+   "locations": [
+    {
+     "path": "/some/path/pkg1"
+    }
+   ],
+   "licenses": [
+    "MIT"
+   ],
+   "language": "python",
+   "cpes": [
+    "cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"
+   ],
+   "purl": "a-purl-2",
+   "metadataType": "PythonPackageMetadata",
+   "metadata": {
+    "name": "package-1",
+    "version": "1.0.1",
+    "license": "",
+    "author": "",
+    "authorEmail": "",
+    "platform": "",
+    "files": [
+     {
+      "path": "/some/path/pkg1/dependencies/foo"
+     }
+    ],
+    "sitePackagesRootPath": ""
+   }
+  },
+  {
+   "id": "package-2-id",
+   "name": "package-2",
+   "version": "2.0.1",
+   "type": "deb",
+   "foundBy": "the-cataloger-2",
+   "locations": [
+    {
+     "path": "/some/path/pkg1"
+    }
+   ],
+   "licenses": [],
+   "language": "",
+   "cpes": [
+    "cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"
+   ],
+   "purl": "a-purl-2",
+   "metadataType": "DpkgMetadata",
+   "metadata": {
+    "package": "package-2",
+    "source": "",
+    "version": "2.0.1",
+    "sourceVersion": "",
+    "architecture": "",
+    "maintainer": "",
+    "installedSize": 0,
+    "files": null
+   }
+  }
+ ],
+ "artifactRelationships": [],
+ "source": {
+  "type": "directory",
+  "target": "/some/path"
+ },
+ "distro": {
+  "name": "debian",
+  "version": "1.2.3",
+  "idLike": "like!"
+ },
+ "descriptor": {
+  "name": "syft",
+  "version": "[not provided]"
+ },
+ "schema": {
+  "version": "1.1.0",
+  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-1.1.0.json"
+ }
+}
@@ -0,0 +1,108 @@
+{
+ "artifacts": [
+  {
+   "id": "package-1-id",
+   "name": "package-1",
+   "version": "1.0.1",
+   "type": "python",
+   "foundBy": "the-cataloger-1",
+   "locations": [
+    {
+     "path": "/somefile-1.txt",
+     "layerID": "sha256:fb6beecb75b39f4bb813dbf177e501edd5ddb3e69bb45cedeb78c676ee1b7a59"
+    }
+   ],
+   "licenses": [
+    "MIT"
+   ],
+   "language": "python",
+   "cpes": [
+    "cpe:2.3:*:some:package:1:*:*:*:*:*:*:*"
+   ],
+   "purl": "a-purl-1",
+   "metadataType": "PythonPackageMetadata",
+   "metadata": {
+    "name": "package-1",
+    "version": "1.0.1",
+    "license": "",
+    "author": "",
+    "authorEmail": "",
+    "platform": "",
+    "sitePackagesRootPath": ""
+   }
+  },
+  {
+   "id": "package-2-id",
+   "name": "package-2",
+   "version": "2.0.1",
+   "type": "deb",
+   "foundBy": "the-cataloger-2",
+   "locations": [
+    {
+     "path": "/somefile-2.txt",
+     "layerID": "sha256:319b588ce64253a87b533c8ed01cf0025e0eac98e7b516e12532957e1244fdec"
+    }
+   ],
+   "licenses": [],
+   "language": "",
+   "cpes": [
+    "cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"
+   ],
+   "purl": "a-purl-2",
+   "metadataType": "DpkgMetadata",
+   "metadata": {
+    "package": "package-2",
+    "source": "",
+    "version": "2.0.1",
+    "sourceVersion": "",
+    "architecture": "",
+    "maintainer": "",
+    "installedSize": 0,
+    "files": null
+   }
+  }
+ ],
+ "artifactRelationships": [],
+ "source": {
+  "type": "image",
+  "target": {
+   "userInput": "user-image-input",
+   "imageID": "sha256:2480160b55bec40c44d3b145c7b2c1c47160db8575c3dcae086d76b9370ae7ca",
+   "manifestDigest": "sha256:2731251dc34951c0e50fcc643b4c5f74922dad1a5d98f302b504cf46cd5d9368",
+   "mediaType": "application/vnd.docker.distribution.manifest.v2+json",
+   "tags": [
+    "stereoscope-fixture-image-simple:85066c51088bdd274f7a89e99e00490f666c49e72ffc955707cd6e18f0e22c5b"
+   ],
+   "imageSize": 38,
+   "layers": [
+    {
+     "mediaType": "application/vnd.docker.image.rootfs.diff.tar.gzip",
+     "digest": "sha256:fb6beecb75b39f4bb813dbf177e501edd5ddb3e69bb45cedeb78c676ee1b7a59",
+     "size": 22
+    },
+    {
+     "mediaType": "application/vnd.docker.image.rootfs.diff.tar.gzip",
+     "digest": "sha256:319b588ce64253a87b533c8ed01cf0025e0eac98e7b516e12532957e1244fdec",
+     "size": 16
+    }
+   ],
+   "manifest": "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",
+   "config": "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",
+   "repoDigests": [],
+   "scope": "Squashed"
+  }
+ },
+ "distro": {
+  "name": "debian",
+  "version": "1.2.3",
+  "idLike": "like!"
+ },
+ "descriptor": {
+  "name": "syft",
+  "version": "[not provided]"
+ },
+ "schema": {
+  "version": "1.1.0",
+  "url": "https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-1.1.0.json"
+ }
+}
@@ -0,0 +1,133 @@
+package syftjson
+
+import (
+	"fmt"
+
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/formats/syftjson/model"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/version"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+// TODO: this is export4ed for the use of the power-user command (temp)
+func ToFormatModel(catalog *pkg.Catalog, srcMetadata *source.Metadata, d *distro.Distro, scope source.Scope, applicationConfig interface{}) model.Document {
+	src, err := toSourceModel(srcMetadata, scope)
+	if err != nil {
+		log.Warnf("unable to create syft-json source object: %+v", err)
+	}
+
+	return model.Document{
+		Artifacts:             toPackageModels(catalog),
+		ArtifactRelationships: toRelationshipModel(pkg.NewRelationships(catalog)),
+		Source:                src,
+		Distro:                toDistroModel(d),
+		Descriptor: model.Descriptor{
+			Name:          internal.ApplicationName,
+			Version:       version.FromBuild().Version,
+			Configuration: applicationConfig,
+		},
+		Schema: model.Schema{
+			Version: internal.JSONSchemaVersion,
+			URL:     fmt.Sprintf("https://raw.githubusercontent.com/anchore/syft/main/schema/json/schema-%s.json", internal.JSONSchemaVersion),
+		},
+	}
+}
+
+func toPackageModels(catalog *pkg.Catalog) []model.Package {
+	artifacts := make([]model.Package, 0)
+	if catalog == nil {
+		return artifacts
+	}
+	for _, p := range catalog.Sorted() {
+		artifacts = append(artifacts, toPackageModel(p))
+	}
+	return artifacts
+}
+
+// toPackageModel crates a new Package from the given pkg.Package.
+func toPackageModel(p *pkg.Package) model.Package {
+	var cpes = make([]string, len(p.CPEs))
+	for i, c := range p.CPEs {
+		cpes[i] = c.BindToFmtString()
+	}
+
+	// ensure collections are never nil for presentation reasons
+	var locations = make([]source.Location, 0)
+	if p.Locations != nil {
+		locations = p.Locations
+	}
+
+	var licenses = make([]string, 0)
+	if p.Licenses != nil {
+		licenses = p.Licenses
+	}
+
+	return model.Package{
+		PackageBasicData: model.PackageBasicData{
+			ID:        string(p.ID),
+			Name:      p.Name,
+			Version:   p.Version,
+			Type:      p.Type,
+			FoundBy:   p.FoundBy,
+			Locations: locations,
+			Licenses:  licenses,
+			Language:  p.Language,
+			CPEs:      cpes,
+			PURL:      p.PURL,
+		},
+		PackageCustomData: model.PackageCustomData{
+			MetadataType: p.MetadataType,
+			Metadata:     p.Metadata,
+		},
+	}
+}
+
+func toRelationshipModel(relationships []pkg.Relationship) []model.Relationship {
+	result := make([]model.Relationship, len(relationships))
+	for i, r := range relationships {
+		result[i] = model.Relationship{
+			Parent:   string(r.Parent),
+			Child:    string(r.Child),
+			Type:     string(r.Type),
+			Metadata: r.Metadata,
+		}
+	}
+	return result
+}
+
+// toSourceModel creates a new source object to be represented into JSON.
+func toSourceModel(src *source.Metadata, scope source.Scope) (model.Source, error) {
+	switch src.Scheme {
+	case source.ImageScheme:
+		return model.Source{
+			Type: "image",
+			Target: model.ImageSource{
+				ImageMetadata: src.ImageMetadata,
+				Scope:         scope,
+			},
+		}, nil
+	case source.DirectoryScheme:
+		return model.Source{
+			Type:   "directory",
+			Target: src.Path,
+		}, nil
+	default:
+		return model.Source{}, fmt.Errorf("unsupported source: %q", src.Scheme)
+	}
+}
+
+// toDistroModel creates a struct with the Linux distribution to be represented in JSON.
+func toDistroModel(d *distro.Distro) model.Distro {
+	if d == nil {
+		return model.Distro{}
+	}
+
+	return model.Distro{
+		Name:    d.Name(),
+		Version: d.FullVersion(),
+		IDLike:  d.IDLike,
+	}
+}
@@ -0,0 +1,73 @@
+package syftjson
+
+import (
+	"github.com/anchore/syft/internal/formats/syftjson/model"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+func toSyftModel(doc model.Document) (*pkg.Catalog, *source.Metadata, *distro.Distro, source.Scope, error) {
+	dist, err := distro.NewDistro(distro.Type(doc.Distro.Name), doc.Distro.Version, doc.Distro.IDLike)
+	if err != nil {
+		return nil, nil, nil, source.UnknownScope, err
+	}
+
+	srcMetadata, scope := toSyftSourceData(doc.Source)
+
+	return toSyftCatalog(doc.Artifacts), srcMetadata, &dist, scope, nil
+}
+
+func toSyftSourceData(s model.Source) (*source.Metadata, source.Scope) {
+	switch s.Type {
+	case "directory":
+		return &source.Metadata{
+			Scheme: source.DirectoryScheme,
+			Path:   s.Target.(string),
+		}, source.UnknownScope
+	case "image":
+		parsedSource := s.Target.(model.ImageSource)
+		return &source.Metadata{
+			Scheme:        source.ImageScheme,
+			ImageMetadata: parsedSource.ImageMetadata,
+		}, parsedSource.Scope
+	}
+	return nil, source.UnknownScope
+}
+
+func toSyftCatalog(pkgs []model.Package) *pkg.Catalog {
+	catalog := pkg.NewCatalog()
+	for _, p := range pkgs {
+		catalog.Add(toSyftPackage(p))
+	}
+	return catalog
+}
+
+func toSyftPackage(p model.Package) pkg.Package {
+	var cpes []pkg.CPE
+	for _, c := range p.CPEs {
+		value, err := pkg.NewCPE(c)
+		if err != nil {
+			log.Warnf("excluding invalid CPE %q: %v", c, err)
+			continue
+		}
+
+		cpes = append(cpes, value)
+	}
+
+	return pkg.Package{
+		ID:           pkg.ID(p.ID),
+		Name:         p.Name,
+		Version:      p.Version,
+		FoundBy:      p.FoundBy,
+		Locations:    p.Locations,
+		Licenses:     p.Licenses,
+		Language:     p.Language,
+		Type:         p.Type,
+		CPEs:         cpes,
+		PURL:         p.PURL,
+		MetadataType: p.MetadataType,
+		Metadata:     p.Metadata,
+	}
+}
@@ -0,0 +1,31 @@
+package syftjson
+
+import (
+	"encoding/json"
+	"fmt"
+	"io"
+	"strings"
+
+	"github.com/anchore/syft/internal/formats/syftjson/model"
+)
+
+func validator(reader io.Reader) error {
+	type Document struct {
+		Schema model.Schema `json:"schema"`
+	}
+
+	dec := json.NewDecoder(reader)
+
+	var doc Document
+	err := dec.Decode(&doc)
+	if err != nil {
+		return fmt.Errorf("unable to decode: %w", err)
+	}
+
+	// note: we accept all schema versions
+	// TODO: add per-schema version parsing
+	if strings.Contains(doc.Schema.URL, "anchore/syft") {
+		return nil
+	}
+	return fmt.Errorf("could not extract syft schema")
+}