port deb/dpkg cataloger to new generic cataloger pattern (#1288)

Signed-off-by: Alex Goodman <alex.goodman@anchore.com>

Signed-off-by: Alex Goodman <alex.goodman@anchore.com>

internal/string_helpers.go

@@ -19,3 +19,12 @@ func TruncateMiddleEllipsis(input string, maxLen int) string {
 	}
 	return input[:maxLen/2] + "..." + input[len(input)-(maxLen/2):]
 }
+
+func StringInSlice(a string, list []string) bool {
+	for _, b := range list {
+		if b == a {
+			return true
+		}
+	}
+	return false
+}

syft/lib.go

@@ -49,7 +49,7 @@ func CatalogPackages(src *source.Source, cfg cataloger.Config) (*pkg.Catalog, []
 	}
 
 	// if the catalogers have been configured, use them regardless of input type
-	var catalogers []cataloger.Cataloger
+	var catalogers []pkg.Cataloger
 	if len(cfg.Catalogers) > 0 {
 		catalogers = cataloger.AllCatalogers(cfg)
 	} else {

syft/pkg/cataloger.go

@@ -0,0 +1,16 @@
+package pkg
+
+import (
+	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/source"
+)
+
+// Cataloger describes behavior for an object to participate in parsing container image or file system
+// contents for the purpose of discovering Packages. Each concrete implementation should focus on discovering Packages
+// for a specific Package Type or ecosystem.
+type Cataloger interface {
+	// Name returns a string that uniquely describes a cataloger
+	Name() string
+	// Catalog is given an object to resolve file references and content, this function returns any discovered Packages after analyzing the catalog source.
+	Catalog(resolver source.FileResolver) ([]Package, []artifact.Relationship, error)
+}

syft/pkg/cataloger/apkdb/parse_apk_db_test.go

@@ -754,7 +754,7 @@ func TestMultiplePackages(t *testing.T) {
 		VersionID: "3.12",
 	}}
 
-	pkgtest.TestGenericParserWithEnv(t, fixture, parseApkDB, &env, expected, expectedRelationships)
+	pkgtest.TestFileParserWithEnv(t, fixture, parseApkDB, &env, expected, expectedRelationships)
 
 }
 

syft/pkg/cataloger/catalog.go

@@ -42,7 +42,7 @@ func newMonitor() (*progress.Manual, *progress.Manual) {
 // In order to efficiently retrieve contents from a underlying container image the content fetch requests are
 // done in bulk. Specifically, all files of interest are collected from each catalogers and accumulated into a single
 // request.
-func Catalog(resolver source.FileResolver, release *linux.Release, catalogers ...Cataloger) (*pkg.Catalog, []artifact.Relationship, error) {
+func Catalog(resolver source.FileResolver, release *linux.Release, catalogers ...pkg.Cataloger) (*pkg.Catalog, []artifact.Relationship, error) {
 	catalog := pkg.NewCatalog()
 	var allRelationships []artifact.Relationship
 

syft/pkg/cataloger/cataloger.go

@@ -9,7 +9,6 @@ import (
 	"strings"
 
 	"github.com/anchore/syft/internal/log"
-	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/pkg"
 	"github.com/anchore/syft/syft/pkg/cataloger/alpm"
 	"github.com/anchore/syft/syft/pkg/cataloger/apkdb"
@@ -28,24 +27,13 @@ import (
 	"github.com/anchore/syft/syft/pkg/cataloger/ruby"
 	"github.com/anchore/syft/syft/pkg/cataloger/rust"
 	"github.com/anchore/syft/syft/pkg/cataloger/swift"
-	"github.com/anchore/syft/syft/source"
 )
 
 const AllCatalogersPattern = "all"
 
-// Cataloger describes behavior for an object to participate in parsing container image or file system
-// contents for the purpose of discovering Packages. Each concrete implementation should focus on discovering Packages
-// for a specific Package Type or ecosystem.
-type Cataloger interface {
-	// Name returns a string that uniquely describes a cataloger
-	Name() string
-	// Catalog is given an object to resolve file references and content, this function returns any discovered Packages after analyzing the catalog source.
-	Catalog(resolver source.FileResolver) ([]pkg.Package, []artifact.Relationship, error)
-}
-
 // ImageCatalogers returns a slice of locally implemented catalogers that are fit for detecting installations of packages.
-func ImageCatalogers(cfg Config) []Cataloger {
+func ImageCatalogers(cfg Config) []pkg.Cataloger {
-	return filterCatalogers([]Cataloger{
+	return filterCatalogers([]pkg.Cataloger{
 		alpm.NewAlpmdbCataloger(),
 		ruby.NewGemSpecCataloger(),
 		python.NewPythonPackageCataloger(),
@@ -62,8 +50,8 @@ func ImageCatalogers(cfg Config) []Cataloger {
 }
 
 // DirectoryCatalogers returns a slice of locally implemented catalogers that are fit for detecting packages from index files (and select installations)
-func DirectoryCatalogers(cfg Config) []Cataloger {
+func DirectoryCatalogers(cfg Config) []pkg.Cataloger {
-	return filterCatalogers([]Cataloger{
+	return filterCatalogers([]pkg.Cataloger{
 		alpm.NewAlpmdbCataloger(),
 		ruby.NewGemFileLockCataloger(),
 		python.NewPythonIndexCataloger(),
@@ -89,8 +77,8 @@ func DirectoryCatalogers(cfg Config) []Cataloger {
 }
 
 // AllCatalogers returns all implemented catalogers
-func AllCatalogers(cfg Config) []Cataloger {
+func AllCatalogers(cfg Config) []pkg.Cataloger {
-	return filterCatalogers([]Cataloger{
+	return filterCatalogers([]pkg.Cataloger{
 		alpm.NewAlpmdbCataloger(),
 		ruby.NewGemFileLockCataloger(),
 		ruby.NewGemSpecCataloger(),
@@ -128,7 +116,7 @@ func RequestedAllCatalogers(cfg Config) bool {
 	return false
 }
 
-func filterCatalogers(catalogers []Cataloger, enabledCatalogerPatterns []string) []Cataloger {
+func filterCatalogers(catalogers []pkg.Cataloger, enabledCatalogerPatterns []string) []pkg.Cataloger {
 	// if cataloger is not set, all applicable catalogers are enabled by default
 	if len(enabledCatalogerPatterns) == 0 {
 		return catalogers
@@ -138,7 +126,7 @@ func filterCatalogers(catalogers []Cataloger, enabledCatalogerPatterns []string)
 			return catalogers
 		}
 	}
-	var keepCatalogers []Cataloger
+	var keepCatalogers []pkg.Cataloger
 	for _, cataloger := range catalogers {
 		if contains(enabledCatalogerPatterns, cataloger.Name()) {
 			keepCatalogers = append(keepCatalogers, cataloger)

syft/pkg/cataloger/cataloger_test.go

@@ -10,7 +10,7 @@ import (
 	"github.com/anchore/syft/syft/source"
 )
 
-var _ Cataloger = (*dummy)(nil)
+var _ pkg.Cataloger = (*dummy)(nil)
 
 type dummy struct {
 	name string
@@ -147,7 +147,7 @@ func Test_filterCatalogers(t *testing.T) {
 	}
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			var catalogers []Cataloger
+			var catalogers []pkg.Cataloger
 			for _, n := range tt.catalogers {
 				catalogers = append(catalogers, dummy{name: n})
 			}

syft/pkg/cataloger/cpp/parse_conanfile_test.go

@@ -90,5 +90,5 @@ func TestParseConanfile(t *testing.T) {
 	// TODO: relationships are not under test
 	var expectedRelationships []artifact.Relationship
 
-	pkgtest.TestGenericParser(t, fixture, parseConanfile, expected, expectedRelationships)
+	pkgtest.TestFileParser(t, fixture, parseConanfile, expected, expectedRelationships)
 }

syft/pkg/cataloger/cpp/parse_conanlock_test.go

@@ -35,5 +35,5 @@ func TestParseConanlock(t *testing.T) {
 	// TODO: relationships are not under test
 	var expectedRelationships []artifact.Relationship
 
-	pkgtest.TestGenericParser(t, fixture, parseConanlock, expected, expectedRelationships)
+	pkgtest.TestFileParser(t, fixture, parseConanlock, expected, expectedRelationships)
 }

syft/pkg/cataloger/dart/parse_pubspec_lock_test.go

@@ -98,5 +98,5 @@ func TestParsePubspecLock(t *testing.T) {
 	// TODO: relationships are not under test
 	var expectedRelationships []artifact.Relationship
 
-	pkgtest.TestGenericParser(t, fixture, parsePubspecLock, expected, expectedRelationships)
+	pkgtest.TestFileParser(t, fixture, parsePubspecLock, expected, expectedRelationships)
 }

syft/pkg/cataloger/deb/cataloger.go

@@ -4,230 +4,14 @@ Package dpkg provides a concrete Cataloger implementation for Debian package DB
 package deb
 
 import (
-	"fmt"
-	"io"
-	"path"
-	"path/filepath"
-	"sort"
-
-	"github.com/anchore/syft/internal"
-	"github.com/anchore/syft/internal/log"
-	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/source"
+	"github.com/anchore/syft/syft/pkg/cataloger/generic"
 )
 
-const (
+const catalogerName = "dpkgdb-cataloger"
-	md5sumsExt   = ".md5sums"
-	conffilesExt = ".conffiles"
-	docsPath     = "/usr/share/doc"
-)
 
-type Cataloger struct{}
+// NewDpkgdbCataloger returns a new Deb package cataloger capable of parsing DPKG status DB files.
-
+func NewDpkgdbCataloger() *generic.Cataloger {
-// NewDpkgdbCataloger returns a new Deb package cataloger object.
+	return generic.NewCataloger(catalogerName).
-func NewDpkgdbCataloger() *Cataloger {
+		WithParserByGlobs(parseDpkgDB, pkg.DpkgDBGlob)
-	return &Cataloger{}
-}
-
-// Name returns a string that uniquely describes a cataloger
-func (c *Cataloger) Name() string {
-	return "dpkgdb-cataloger"
-}
-
-// Catalog is given an object to resolve file references and content, this function returns any discovered Packages after analyzing dpkg support files.
-func (c *Cataloger) Catalog(resolver source.FileResolver) ([]pkg.Package, []artifact.Relationship, error) {
-	dbFileMatches, err := resolver.FilesByGlob(pkg.DpkgDBGlob)
-	if err != nil {
-		return nil, nil, fmt.Errorf("failed to find dpkg status files's by glob: %w", err)
-	}
-
-	var allPackages []pkg.Package
-	for _, dbLocation := range dbFileMatches {
-		dbContents, err := resolver.FileContentsByLocation(dbLocation)
-		if err != nil {
-			return nil, nil, err
-		}
-
-		pkgs, err := parseDpkgStatus(dbContents)
-		internal.CloseAndLogError(dbContents, dbLocation.VirtualPath)
-		if err != nil {
-			log.Warnf("dpkg cataloger: unable to catalog package=%+v: %w", dbLocation.RealPath, err)
-			continue
-		}
-
-		for i := range pkgs {
-			p := &pkgs[i]
-			p.FoundBy = c.Name()
-			p.Locations.Add(dbLocation)
-
-			// the current entry only has what may have been listed in the status file, however, there are additional
-			// files that are listed in multiple other locations. We should retrieve them all and merge the file lists
-			// together.
-			mergeFileListing(resolver, dbLocation, p)
-
-			// fetch additional data from the copyright file to derive the license information
-			addLicenses(resolver, dbLocation, p)
-
-			p.SetID()
-		}
-
-		allPackages = append(allPackages, pkgs...)
-	}
-	return allPackages, nil, nil
-}
-
-func addLicenses(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) {
-	// get license information from the copyright file
-	copyrightReader, copyrightLocation := fetchCopyrightContents(resolver, dbLocation, p)
-
-	if copyrightReader != nil && copyrightLocation != nil {
-		defer internal.CloseAndLogError(copyrightReader, copyrightLocation.VirtualPath)
-		// attach the licenses
-		p.Licenses = parseLicensesFromCopyright(copyrightReader)
-
-		// keep a record of the file where this was discovered
-		p.Locations.Add(*copyrightLocation)
-	}
-}
-
-func mergeFileListing(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) {
-	metadata := p.Metadata.(pkg.DpkgMetadata)
-
-	// get file listing (package files + additional config files)
-	files, infoLocations := getAdditionalFileListing(resolver, dbLocation, p)
-loopNewFiles:
-	for _, newFile := range files {
-		for _, existingFile := range metadata.Files {
-			if existingFile.Path == newFile.Path {
-				// skip adding this file since it already exists
-				continue loopNewFiles
-			}
-		}
-		metadata.Files = append(metadata.Files, newFile)
-	}
-
-	// sort files by path
-	sort.SliceStable(metadata.Files, func(i, j int) bool {
-		return metadata.Files[i].Path < metadata.Files[j].Path
-	})
-
-	// persist alterations
-	p.Metadata = metadata
-
-	// persist location information from each new source of information
-	p.Locations.Add(infoLocations...)
-}
-
-func getAdditionalFileListing(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) ([]pkg.DpkgFileRecord, []source.Location) {
-	// ensure the default value for a collection is never nil since this may be shown as JSON
-	var files = make([]pkg.DpkgFileRecord, 0)
-	var locations []source.Location
-
-	md5Reader, md5Location := fetchMd5Contents(resolver, dbLocation, p)
-
-	if md5Reader != nil && md5Location != nil {
-		defer internal.CloseAndLogError(md5Reader, md5Location.VirtualPath)
-		// attach the file list
-		files = append(files, parseDpkgMD5Info(md5Reader)...)
-
-		// keep a record of the file where this was discovered
-		locations = append(locations, *md5Location)
-	}
-
-	conffilesReader, conffilesLocation := fetchConffileContents(resolver, dbLocation, p)
-
-	if conffilesReader != nil && conffilesLocation != nil {
-		defer internal.CloseAndLogError(conffilesReader, conffilesLocation.VirtualPath)
-		// attach the file list
-		files = append(files, parseDpkgConffileInfo(conffilesReader)...)
-
-		// keep a record of the file where this was discovered
-		locations = append(locations, *conffilesLocation)
-	}
-
-	return files, locations
-}
-
-func fetchMd5Contents(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) (io.ReadCloser, *source.Location) {
-	var md5Reader io.ReadCloser
-	var err error
-
-	parentPath := filepath.Dir(dbLocation.RealPath)
-
-	// look for /var/lib/dpkg/info/NAME:ARCH.md5sums
-	name := md5Key(p)
-	location := resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", name+md5sumsExt))
-
-	if location == nil {
-		// the most specific key did not work, fallback to just the name
-		// look for /var/lib/dpkg/info/NAME.md5sums
-		location = resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", p.Name+md5sumsExt))
-	}
-
-	// this is unexpected, but not a show-stopper
-	if location != nil {
-		md5Reader, err = resolver.FileContentsByLocation(*location)
-		if err != nil {
-			log.Warnf("failed to fetch deb md5 contents (package=%s): %+v", p.Name, err)
-		}
-	}
-
-	return md5Reader, location
-}
-
-func fetchConffileContents(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) (io.ReadCloser, *source.Location) {
-	var reader io.ReadCloser
-	var err error
-
-	parentPath := filepath.Dir(dbLocation.RealPath)
-
-	// look for /var/lib/dpkg/info/NAME:ARCH.conffiles
-	name := md5Key(p)
-	location := resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", name+conffilesExt))
-
-	if location == nil {
-		// the most specific key did not work, fallback to just the name
-		// look for /var/lib/dpkg/info/NAME.conffiles
-		location = resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", p.Name+conffilesExt))
-	}
-
-	// this is unexpected, but not a show-stopper
-	if location != nil {
-		reader, err = resolver.FileContentsByLocation(*location)
-		if err != nil {
-			log.Warnf("failed to fetch deb conffiles contents (package=%s): %+v", p.Name, err)
-		}
-	}
-
-	return reader, location
-}
-
-func fetchCopyrightContents(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) (io.ReadCloser, *source.Location) {
-	// look for /usr/share/docs/NAME/copyright files
-	name := p.Name
-	copyrightPath := path.Join(docsPath, name, "copyright")
-	location := resolver.RelativeFileByPath(dbLocation, copyrightPath)
-
-	// we may not have a copyright file for each package, ignore missing files
-	if location == nil {
-		return nil, nil
-	}
-
-	reader, err := resolver.FileContentsByLocation(*location)
-	if err != nil {
-		log.Warnf("failed to fetch deb copyright contents (package=%s): %w", p.Name, err)
-	}
-
-	return reader, location
-}
-
-func md5Key(p *pkg.Package) string {
-	metadata := p.Metadata.(pkg.DpkgMetadata)
-
-	contentKey := p.Name
-	if metadata.Architecture != "" && metadata.Architecture != "all" {
-		contentKey = contentKey + ":" + metadata.Architecture
-	}
-	return contentKey
 }

syft/pkg/cataloger/deb/cataloger_test.go

@@ -3,138 +3,81 @@ package deb
 import (
 	"testing"
 
-	"github.com/go-test/deep"
-	"github.com/stretchr/testify/assert"
-
-	"github.com/anchore/stereoscope/pkg/imagetest"
 	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/pkg/cataloger/internal/pkgtest"
 	"github.com/anchore/syft/syft/source"
 )
 
 func TestDpkgCataloger(t *testing.T) {
-	tests := []struct {
+	expected := []pkg.Package{
-		name     string
-		sources  map[string][]string
-		expected []pkg.Package
-	}{
 		{
-			name: "go-case",
+			Name:     "libpam-runtime",
-			sources: map[string][]string{
+			Version:  "1.1.8-3.6",
-				"libpam-runtime": {
+			FoundBy:  "dpkgdb-cataloger",
-					"/var/lib/dpkg/status",
+			Licenses: []string{"GPL-1", "GPL-2", "LGPL-2.1"},
-					"/var/lib/dpkg/info/libpam-runtime.md5sums",
+			Locations: source.NewLocationSet(
-					"/var/lib/dpkg/info/libpam-runtime.conffiles",
+				source.NewVirtualLocation("/var/lib/dpkg/status", "/var/lib/dpkg/status"),
-					"/usr/share/doc/libpam-runtime/copyright",
+				source.NewVirtualLocation("/var/lib/dpkg/info/libpam-runtime.md5sums", "/var/lib/dpkg/info/libpam-runtime.md5sums"),
-				},
+				source.NewVirtualLocation("/var/lib/dpkg/info/libpam-runtime.conffiles", "/var/lib/dpkg/info/libpam-runtime.conffiles"),
-			},
+				source.NewVirtualLocation("/usr/share/doc/libpam-runtime/copyright", "/usr/share/doc/libpam-runtime/copyright"),
-			expected: []pkg.Package{
+			),
-				{
+			Type:         pkg.DebPkg,
-					Name:         "libpam-runtime",
+			MetadataType: pkg.DpkgMetadataType,
-					Version:      "1.1.8-3.6",
+			Metadata: pkg.DpkgMetadata{
-					FoundBy:      "dpkgdb-cataloger",
+				Package:       "libpam-runtime",
-					Licenses:     []string{"GPL-1", "GPL-2", "LGPL-2.1"},
+				Source:        "pam",
-					Type:         pkg.DebPkg,
+				Version:       "1.1.8-3.6",
-					MetadataType: pkg.DpkgMetadataType,
+				Architecture:  "all",
-					Metadata: pkg.DpkgMetadata{
+				Maintainer:    "Steve Langasek <vorlon@debian.org>",
-						Package:       "libpam-runtime",
+				InstalledSize: 1016,
-						Source:        "pam",
+				Description: `Runtime support for the PAM library
-						Version:       "1.1.8-3.6",
-						Architecture:  "all",
-						Maintainer:    "Steve Langasek <vorlon@debian.org>",
-						InstalledSize: 1016,
-						Description: `Runtime support for the PAM library
  Contains configuration files and  directories required for
  authentication  to work on Debian systems.  This package is required
  on almost all installations.`,
-						Files: []pkg.DpkgFileRecord{
+				Files: []pkg.DpkgFileRecord{
-							{
+					{
-								Path: "/etc/pam.conf",
+						Path: "/etc/pam.conf",
-								Digest: &file.Digest{
+						Digest: &file.Digest{
-									Algorithm: "md5",
+							Algorithm: "md5",
-									Value:     "87fc76f18e98ee7d3848f6b81b3391e5",
+							Value:     "87fc76f18e98ee7d3848f6b81b3391e5",
-								},
-								IsConfigFile: true,
-							},
-							{
-								Path: "/etc/pam.d/other",
-								Digest: &file.Digest{
-									Algorithm: "md5",
-									Value:     "31aa7f2181889ffb00b87df4126d1701",
-								},
-								IsConfigFile: true,
-							},
-							{Path: "/lib/x86_64-linux-gnu/libz.so.1.2.11", Digest: &file.Digest{
-								Algorithm: "md5",
-								Value:     "55f905631797551d4d936a34c7e73474",
-							}},
-							{Path: "/usr/share/doc/zlib1g/changelog.Debian.gz", Digest: &file.Digest{
-								Algorithm: "md5",
-								Value:     "cede84bda30d2380217f97753c8ccf3a",
-							}},
-							{Path: "/usr/share/doc/zlib1g/changelog.gz", Digest: &file.Digest{
-								Algorithm: "md5",
-								Value:     "f3c9dafa6da7992c47328b4464f6d122",
-							}},
-							{Path: "/usr/share/doc/zlib1g/copyright", Digest: &file.Digest{
-								Algorithm: "md5",
-								Value:     "a4fae96070439a5209a62ae5b8017ab2",
-							}},
 						},
+						IsConfigFile: true,
 					},
+					{
+						Path: "/etc/pam.d/other",
+						Digest: &file.Digest{
+							Algorithm: "md5",
+							Value:     "31aa7f2181889ffb00b87df4126d1701",
+						},
+						IsConfigFile: true,
+					},
+					{Path: "/lib/x86_64-linux-gnu/libz.so.1.2.11", Digest: &file.Digest{
+						Algorithm: "md5",
+						Value:     "55f905631797551d4d936a34c7e73474",
+					}},
+					{Path: "/usr/share/doc/zlib1g/changelog.Debian.gz", Digest: &file.Digest{
+						Algorithm: "md5",
+						Value:     "cede84bda30d2380217f97753c8ccf3a",
+					}},
+					{Path: "/usr/share/doc/zlib1g/changelog.gz", Digest: &file.Digest{
+						Algorithm: "md5",
+						Value:     "f3c9dafa6da7992c47328b4464f6d122",
+					}},
+					{Path: "/usr/share/doc/zlib1g/copyright", Digest: &file.Digest{
+						Algorithm: "md5",
+						Value:     "a4fae96070439a5209a62ae5b8017ab2",
+					}},
 				},
 			},
 		},
 	}
 
-	for _, test := range tests {
+	c := NewDpkgdbCataloger()
-		t.Run(test.name, func(t *testing.T) {
-
-			img := imagetest.GetFixtureImage(t, "docker-archive", "image-dpkg")
-
-			s, err := source.NewFromImage(img, "")
-			if err != nil {
-				t.Fatal(err)
-			}
-
-			c := NewDpkgdbCataloger()
-
-			resolver, err := s.FileResolver(source.SquashedScope)
-			if err != nil {
-				t.Errorf("could not get resolver error: %+v", err)
-			}
-
-			actual, _, err := c.Catalog(resolver)
-			if err != nil {
-				t.Fatalf("failed to catalog: %+v", err)
-			}
-
-			if len(actual) != len(test.expected) {
-				for _, a := range actual {
-					t.Logf("   %+v", a)
-				}
-				t.Fatalf("unexpected package count: %d!=%d", len(actual), len(test.expected))
-			}
-
-			// test sources...
-			for idx := range actual {
-				a := &actual[idx]
-				// we will test the sources separately
-				var sourcesList = make([]string, len(a.Locations.ToSlice()))
-				for i, s := range a.Locations.ToSlice() {
-					sourcesList[i] = s.RealPath
-				}
-				a.Locations = source.NewLocationSet()
-
-				assert.ElementsMatch(t, sourcesList, test.sources[a.Name])
-			}
-
-			// test remaining fields...
-			for _, d := range deep.Equal(actual, test.expected) {
-				t.Errorf("diff: %+v", d)
-			}
-
-		})
-	}
 
+	pkgtest.NewCatalogTester().
+		WithImageResolver(t, "image-dpkg").
+		IgnoreLocationLayer(). // this fixture can be rebuilt, thus the layer ID will change
+		Expects(expected, nil).
+		TestCataloger(t, c)
 }

syft/pkg/cataloger/deb/package.go

@@ -0,0 +1,247 @@
+package deb
+
+import (
+	"fmt"
+	"io"
+	"path"
+	"path/filepath"
+	"sort"
+
+	"github.com/anchore/packageurl-go"
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/linux"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+)
+
+const (
+	md5sumsExt   = ".md5sums"
+	conffilesExt = ".conffiles"
+	docsPath     = "/usr/share/doc"
+)
+
+func newDpkgPackage(d pkg.DpkgMetadata, dbLocation source.Location, resolver source.FileResolver, release *linux.Release) pkg.Package {
+	p := pkg.Package{
+		Name:         d.Package,
+		Version:      d.Version,
+		Locations:    source.NewLocationSet(dbLocation),
+		PURL:         packageURL(d, release),
+		Type:         pkg.DebPkg,
+		MetadataType: pkg.DpkgMetadataType,
+		Metadata:     d,
+	}
+
+	// the current entry only has what may have been listed in the status file, however, there are additional
+	// files that are listed in multiple other locations. We should retrieve them all and merge the file lists
+	// together.
+	mergeFileListing(resolver, dbLocation, &p)
+
+	// fetch additional data from the copyright file to derive the license information
+	addLicenses(resolver, dbLocation, &p)
+
+	p.SetID()
+
+	return p
+}
+
+// PackageURL returns the PURL for the specific Debian package (see https://github.com/package-url/purl-spec)
+func packageURL(m pkg.DpkgMetadata, distro *linux.Release) string {
+	if distro == nil {
+		return ""
+	}
+
+	if distro.ID != "debian" && !internal.StringInSlice("debian", distro.IDLike) {
+		return ""
+	}
+
+	qualifiers := map[string]string{
+		pkg.PURLQualifierArch: m.Architecture,
+	}
+
+	if m.Source != "" {
+		if m.SourceVersion != "" {
+			qualifiers[pkg.PURLQualifierUpstream] = fmt.Sprintf("%s@%s", m.Source, m.SourceVersion)
+		} else {
+			qualifiers[pkg.PURLQualifierUpstream] = m.Source
+		}
+	}
+
+	return packageurl.NewPackageURL(
+		packageurl.TypeDebian,
+		distro.ID,
+		m.Package,
+		m.Version,
+		pkg.PURLQualifiers(
+			qualifiers,
+			distro,
+		),
+		"",
+	).ToString()
+}
+
+func addLicenses(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) {
+	metadata := p.Metadata.(pkg.DpkgMetadata)
+
+	// get license information from the copyright file
+	copyrightReader, copyrightLocation := fetchCopyrightContents(resolver, dbLocation, metadata)
+
+	if copyrightReader != nil && copyrightLocation != nil {
+		defer internal.CloseAndLogError(copyrightReader, copyrightLocation.VirtualPath)
+		// attach the licenses
+		p.Licenses = parseLicensesFromCopyright(copyrightReader)
+
+		// keep a record of the file where this was discovered
+		p.Locations.Add(*copyrightLocation)
+	}
+}
+
+func mergeFileListing(resolver source.FileResolver, dbLocation source.Location, p *pkg.Package) {
+	metadata := p.Metadata.(pkg.DpkgMetadata)
+
+	// get file listing (package files + additional config files)
+	files, infoLocations := getAdditionalFileListing(resolver, dbLocation, metadata)
+loopNewFiles:
+	for _, newFile := range files {
+		for _, existingFile := range metadata.Files {
+			if existingFile.Path == newFile.Path {
+				// skip adding this file since it already exists
+				continue loopNewFiles
+			}
+		}
+		metadata.Files = append(metadata.Files, newFile)
+	}
+
+	// sort files by path
+	sort.SliceStable(metadata.Files, func(i, j int) bool {
+		return metadata.Files[i].Path < metadata.Files[j].Path
+	})
+
+	// persist alterations
+	p.Metadata = metadata
+
+	// persist location information from each new source of information
+	p.Locations.Add(infoLocations...)
+}
+
+func getAdditionalFileListing(resolver source.FileResolver, dbLocation source.Location, m pkg.DpkgMetadata) ([]pkg.DpkgFileRecord, []source.Location) {
+	// ensure the default value for a collection is never nil since this may be shown as JSON
+	var files = make([]pkg.DpkgFileRecord, 0)
+	var locations []source.Location
+
+	md5Reader, md5Location := fetchMd5Contents(resolver, dbLocation, m)
+
+	if md5Reader != nil && md5Location != nil {
+		defer internal.CloseAndLogError(md5Reader, md5Location.VirtualPath)
+		// attach the file list
+		files = append(files, parseDpkgMD5Info(md5Reader)...)
+
+		// keep a record of the file where this was discovered
+		locations = append(locations, *md5Location)
+	}
+
+	conffilesReader, conffilesLocation := fetchConffileContents(resolver, dbLocation, m)
+
+	if conffilesReader != nil && conffilesLocation != nil {
+		defer internal.CloseAndLogError(conffilesReader, conffilesLocation.VirtualPath)
+		// attach the file list
+		files = append(files, parseDpkgConffileInfo(conffilesReader)...)
+
+		// keep a record of the file where this was discovered
+		locations = append(locations, *conffilesLocation)
+	}
+
+	return files, locations
+}
+
+func fetchMd5Contents(resolver source.FileResolver, dbLocation source.Location, m pkg.DpkgMetadata) (io.ReadCloser, *source.Location) {
+	var md5Reader io.ReadCloser
+	var err error
+
+	if resolver == nil {
+		return nil, nil
+	}
+
+	parentPath := filepath.Dir(dbLocation.RealPath)
+
+	// look for /var/lib/dpkg/info/NAME:ARCH.md5sums
+	name := md5Key(m)
+	location := resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", name+md5sumsExt))
+
+	if location == nil {
+		// the most specific key did not work, fallback to just the name
+		// look for /var/lib/dpkg/info/NAME.md5sums
+		location = resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", m.Package+md5sumsExt))
+	}
+
+	// this is unexpected, but not a show-stopper
+	if location != nil {
+		md5Reader, err = resolver.FileContentsByLocation(*location)
+		if err != nil {
+			log.Warnf("failed to fetch deb md5 contents (package=%s): %+v", m.Package, err)
+		}
+	}
+
+	return md5Reader, location
+}
+
+func fetchConffileContents(resolver source.FileResolver, dbLocation source.Location, m pkg.DpkgMetadata) (io.ReadCloser, *source.Location) {
+	var reader io.ReadCloser
+	var err error
+
+	if resolver == nil {
+		return nil, nil
+	}
+
+	parentPath := filepath.Dir(dbLocation.RealPath)
+
+	// look for /var/lib/dpkg/info/NAME:ARCH.conffiles
+	name := md5Key(m)
+	location := resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", name+conffilesExt))
+
+	if location == nil {
+		// the most specific key did not work, fallback to just the name
+		// look for /var/lib/dpkg/info/NAME.conffiles
+		location = resolver.RelativeFileByPath(dbLocation, path.Join(parentPath, "info", m.Package+conffilesExt))
+	}
+
+	// this is unexpected, but not a show-stopper
+	if location != nil {
+		reader, err = resolver.FileContentsByLocation(*location)
+		if err != nil {
+			log.Warnf("failed to fetch deb conffiles contents (package=%s): %+v", m.Package, err)
+		}
+	}
+
+	return reader, location
+}
+
+func fetchCopyrightContents(resolver source.FileResolver, dbLocation source.Location, m pkg.DpkgMetadata) (io.ReadCloser, *source.Location) {
+	if resolver == nil {
+		return nil, nil
+	}
+
+	// look for /usr/share/docs/NAME/copyright files
+	copyrightPath := path.Join(docsPath, m.Package, "copyright")
+	location := resolver.RelativeFileByPath(dbLocation, copyrightPath)
+
+	// we may not have a copyright file for each package, ignore missing files
+	if location == nil {
+		return nil, nil
+	}
+
+	reader, err := resolver.FileContentsByLocation(*location)
+	if err != nil {
+		log.Warnf("failed to fetch deb copyright contents (package=%s): %w", m.Package, err)
+	}
+
+	return reader, location
+}
+
+func md5Key(metadata pkg.DpkgMetadata) string {
+	contentKey := metadata.Package
+	if metadata.Architecture != "" && metadata.Architecture != "all" {
+		contentKey = contentKey + ":" + metadata.Architecture
+	}
+	return contentKey
+}

syft/pkg/cataloger/deb/package_test.go

@@ -1,20 +1,19 @@
-package pkg
+package deb
 
 import (
-	"strings"
 	"testing"
 
-	"github.com/go-test/deep"
+	"github.com/google/go-cmp/cmp"
-	"github.com/sergi/go-diff/diffmatchpatch"
 
 	"github.com/anchore/syft/syft/linux"
+	"github.com/anchore/syft/syft/pkg"
 )
 
-func TestDpkgMetadata_pURL(t *testing.T) {
+func Test_packageURL(t *testing.T) {
 	tests := []struct {
 		name     string
 		distro   *linux.Release
-		metadata DpkgMetadata
+		metadata pkg.DpkgMetadata
 		expected string
 	}{
 		{
@@ -22,8 +21,23 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 			distro: &linux.Release{
 				ID:        "debian",
 				VersionID: "11",
+				IDLike: []string{
+					"debian",
+				},
 			},
-			metadata: DpkgMetadata{
+			metadata: pkg.DpkgMetadata{
+				Package: "p",
+				Version: "v",
+			},
+			expected: "pkg:deb/debian/p@v?distro=debian-11",
+		},
+		{
+			name: "missing ID_LIKE",
+			distro: &linux.Release{
+				ID:        "debian",
+				VersionID: "11",
+			},
+			metadata: pkg.DpkgMetadata{
 				Package: "p",
 				Version: "v",
 			},
@@ -32,10 +46,13 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 		{
 			name: "with arch info",
 			distro: &linux.Release{
-				ID:        "ubuntu",
+				ID: "ubuntu",
+				IDLike: []string{
+					"debian",
+				},
 				VersionID: "16.04",
 			},
-			metadata: DpkgMetadata{
+			metadata: pkg.DpkgMetadata{
 				Package:      "p",
 				Version:      "v",
 				Architecture: "a",
@@ -44,19 +61,22 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 		},
 		{
 			name: "missing distro",
-			metadata: DpkgMetadata{
+			metadata: pkg.DpkgMetadata{
 				Package: "p",
 				Version: "v",
 			},
-			expected: "pkg:deb/p@v",
+			expected: "",
 		},
 		{
 			name: "with upstream qualifier with source pkg name info",
 			distro: &linux.Release{
 				ID:        "debian",
 				VersionID: "11",
+				IDLike: []string{
+					"debian",
+				},
 			},
-			metadata: DpkgMetadata{
+			metadata: pkg.DpkgMetadata{
 				Package: "p",
 				Source:  "s",
 				Version: "v",
@@ -68,8 +88,11 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 			distro: &linux.Release{
 				ID:        "debian",
 				VersionID: "11",
+				IDLike: []string{
+					"debian",
+				},
 			},
-			metadata: DpkgMetadata{
+			metadata: pkg.DpkgMetadata{
 				Package:       "p",
 				Source:        "s",
 				Version:       "v",
@@ -81,51 +104,9 @@ func TestDpkgMetadata_pURL(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
-			actual := test.metadata.PackageURL(test.distro)
+			actual := packageURL(test.metadata, test.distro)
-			if actual != test.expected {
+			if diff := cmp.Diff(test.expected, actual); diff != "" {
-				dmp := diffmatchpatch.New()
+				t.Errorf("unexpected packageURL (-want +got):\n%s", diff)
-				diffs := dmp.DiffMain(test.expected, actual, true)
-				t.Errorf("diff: %s", dmp.DiffPrettyText(diffs))
-			}
-		})
-	}
-}
-
-func TestDpkgMetadata_FileOwner(t *testing.T) {
-	tests := []struct {
-		metadata DpkgMetadata
-		expected []string
-	}{
-		{
-			metadata: DpkgMetadata{
-				Files: []DpkgFileRecord{
-					{Path: "/somewhere"},
-					{Path: "/else"},
-				},
-			},
-			expected: []string{
-				"/else",
-				"/somewhere",
-			},
-		},
-		{
-			metadata: DpkgMetadata{
-				Files: []DpkgFileRecord{
-					{Path: "/somewhere"},
-					{Path: ""},
-				},
-			},
-			expected: []string{
-				"/somewhere",
-			},
-		},
-	}
-
-	for _, test := range tests {
-		t.Run(strings.Join(test.expected, ","), func(t *testing.T) {
-			actual := test.metadata.OwnedFiles()
-			for _, d := range deep.Equal(test.expected, actual) {
-				t.Errorf("diff: %+v", d)
 			}
 		})
 	}

syft/pkg/cataloger/deb/parse_copyright_test.go

@@ -4,7 +4,8 @@ import (
 	"os"
 	"testing"
 
-	"github.com/go-test/deep"
+	"github.com/google/go-cmp/cmp"
+	"github.com/stretchr/testify/require"
 )
 
 func TestParseLicensesFromCopyright(t *testing.T) {
@@ -38,31 +39,15 @@ func TestParseLicensesFromCopyright(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.fixture, func(t *testing.T) {
-			file, err := os.Open(test.fixture)
+			f, err := os.Open(test.fixture)
-			if err != nil {
+			require.NoError(t, err)
-				t.Fatal("Unable to read: ", err)
+			t.Cleanup(func() { require.NoError(t, f.Close()) })
+
+			actual := parseLicensesFromCopyright(f)
+
+			if diff := cmp.Diff(test.expected, actual); diff != "" {
+				t.Errorf("unexpected package licenses (-want +got):\n%s", diff)
 			}
-			defer func() {
-				err := file.Close()
-				if err != nil {
-					t.Fatal("closing file failed:", err)
-				}
-			}()
-
-			actual := parseLicensesFromCopyright(file)
-
-			if len(actual) != len(test.expected) {
-				for _, a := range actual {
-					t.Logf("   %+v", a)
-				}
-				t.Fatalf("unexpected package count: %d!=%d", len(actual), len(test.expected))
-			}
-
-			diffs := deep.Equal(actual, test.expected)
-			for _, d := range diffs {
-				t.Errorf("diff: %+v", d)
-			}
-
 		})
 	}
 }

syft/pkg/cataloger/deb/parse_dpkg_db.go

@@ -13,7 +13,10 @@ import (
 
 	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/pkg/cataloger/generic"
+	"github.com/anchore/syft/syft/source"
 )
 
 var (
@@ -21,20 +24,24 @@ var (
 	sourceRegexp     = regexp.MustCompile(`(?P<name>\S+)( \((?P<version>.*)\))?`)
 )
 
-func newDpkgPackage(d pkg.DpkgMetadata) *pkg.Package {
+func parseDpkgDB(resolver source.FileResolver, env *generic.Environment, reader source.LocationReadCloser) ([]pkg.Package, []artifact.Relationship, error) {
-	return &pkg.Package{
+	metadata, err := parseDpkgStatus(reader)
-		Name:         d.Package,
+	if err != nil {
-		Version:      d.Version,
+		return nil, nil, fmt.Errorf("unable to catalog dpkg DB=%q: %w", reader.RealPath, err)
-		Type:         pkg.DebPkg,
-		MetadataType: pkg.DpkgMetadataType,
-		Metadata:     d,
 	}
+
+	var pkgs []pkg.Package
+	for _, m := range metadata {
+		pkgs = append(pkgs, newDpkgPackage(m, reader.Location, resolver, env.LinuxRelease))
+	}
+
+	return pkgs, nil, nil
 }
 
 // parseDpkgStatus is a parser function for Debian DB status contents, returning all Debian packages listed.
-func parseDpkgStatus(reader io.Reader) ([]pkg.Package, error) {
+func parseDpkgStatus(reader io.Reader) ([]pkg.DpkgMetadata, error) {
 	buffedReader := bufio.NewReader(reader)
-	var packages []pkg.Package
+	var metadata []pkg.DpkgMetadata
 
 	continueProcessing := true
 	for continueProcessing {
@@ -46,40 +53,44 @@ func parseDpkgStatus(reader io.Reader) ([]pkg.Package, error) {
 				return nil, err
 			}
 		}
-
+		if entry == nil {
-		p := newDpkgPackage(entry)
+			continue
-		if pkg.IsValid(p) {
-			packages = append(packages, *p)
 		}
+
+		metadata = append(metadata, *entry)
 	}
 
-	return packages, nil
+	return metadata, nil
 }
 
 // parseDpkgStatusEntry returns an individual Dpkg entry, or returns errEndOfPackages if there are no more packages to parse from the reader.
-func parseDpkgStatusEntry(reader *bufio.Reader) (pkg.DpkgMetadata, error) {
+func parseDpkgStatusEntry(reader *bufio.Reader) (*pkg.DpkgMetadata, error) {
 	var retErr error
 	dpkgFields, err := extractAllFields(reader)
 	if err != nil {
 		if !errors.Is(err, errEndOfPackages) {
-			return pkg.DpkgMetadata{}, err
+			return nil, err
+		}
+		if len(dpkgFields) == 0 {
+			return nil, err
 		}
 		retErr = err
 	}
 
-	entry := pkg.DpkgMetadata{
+	entry := pkg.DpkgMetadata{}
-		// ensure the default value for a collection is never nil since this may be shown as JSON
-		Files: make([]pkg.DpkgFileRecord, 0),
-	}
 	err = mapstructure.Decode(dpkgFields, &entry)
 	if err != nil {
-		return pkg.DpkgMetadata{}, err
+		return nil, err
 	}
 
-	name, version := extractSourceVersion(entry.Source)
+	sourceName, sourceVersion := extractSourceVersion(entry.Source)
-	if version != "" {
+	if sourceVersion != "" {
-		entry.SourceVersion = version
+		entry.SourceVersion = sourceVersion
-		entry.Source = name
+		entry.Source = sourceName
+	}
+
+	if entry.Package == "" {
+		return nil, retErr
 	}
 
 	// there may be an optional conffiles section that we should persist as files
@@ -89,7 +100,12 @@ func parseDpkgStatusEntry(reader *bufio.Reader) (pkg.DpkgMetadata, error) {
 		}
 	}
 
-	return entry, retErr
+	if entry.Files == nil {
+		// ensure the default value for a collection is never nil since this may be shown as JSON
+		entry.Files = make([]pkg.DpkgFileRecord, 0)
+	}
+
+	return &entry, retErr
 }
 
 func extractAllFields(reader *bufio.Reader) (map[string]interface{}, error) {

syft/pkg/cataloger/deb/parse_dpkg_db_test.go

@@ -5,41 +5,37 @@ import (
 	"errors"
 	"fmt"
 	"os"
-	"path/filepath"
-	"strings"
 	"testing"
 
-	"github.com/go-test/deep"
+	"github.com/google/go-cmp/cmp"
 	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
 
 	"github.com/anchore/syft/syft/file"
+	"github.com/anchore/syft/syft/linux"
 	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/pkg/cataloger/internal/pkgtest"
+	"github.com/anchore/syft/syft/source"
 )
 
-func compareEntries(t *testing.T, left, right pkg.DpkgMetadata) {
+func Test_parseDpkgStatus(t *testing.T) {
-	t.Helper()
-	if diff := deep.Equal(left, right); diff != nil {
-		t.Error(diff)
-	}
-}
-
-func TestSinglePackage(t *testing.T) {
 	tests := []struct {
 		name        string
-		expected    pkg.DpkgMetadata
+		expected    []pkg.DpkgMetadata
 		fixturePath string
 	}{
 		{
-			name:        "Test Single Package",
+			name:        "single package",
-			fixturePath: filepath.Join("test-fixtures", "status", "single"),
+			fixturePath: "test-fixtures/status/single",
-			expected: pkg.DpkgMetadata{
+			expected: []pkg.DpkgMetadata{
-				Package:       "apt",
+				{
-				Source:        "apt-dev",
+					Package:       "apt",
-				Version:       "1.8.2",
+					Source:        "apt-dev",
-				Architecture:  "amd64",
+					Version:       "1.8.2",
-				InstalledSize: 4064,
+					Architecture:  "amd64",
-				Maintainer:    "APT Development Team <deity@lists.debian.org>",
+					InstalledSize: 4064,
-				Description: `commandline package manager
+					Maintainer:    "APT Development Team <deity@lists.debian.org>",
+					Description: `commandline package manager
  This package provides commandline tools for searching and
  managing as well as querying information about packages
  as a low-level access to all features of the libapt-pkg library.
@@ -53,53 +49,55 @@ func TestSinglePackage(t *testing.T) {
  * apt-cdrom to use removable media as a source for packages
  * apt-config as an interface to the configuration settings
  * apt-key as an interface to manage authentication keys`,
-				Files: []pkg.DpkgFileRecord{
+					Files: []pkg.DpkgFileRecord{
-					{
+						{
-						Path: "/etc/apt/apt.conf.d/01autoremove",
+							Path: "/etc/apt/apt.conf.d/01autoremove",
-						Digest: &file.Digest{
+							Digest: &file.Digest{
-							Algorithm: "md5",
+								Algorithm: "md5",
-							Value:     "76120d358bc9037bb6358e737b3050b5",
+								Value:     "76120d358bc9037bb6358e737b3050b5",
+							},
+							IsConfigFile: true,
 						},
-						IsConfigFile: true,
+						{
-					},
+							Path: "/etc/cron.daily/apt-compat",
-					{
+							Digest: &file.Digest{
-						Path: "/etc/cron.daily/apt-compat",
+								Algorithm: "md5",
-						Digest: &file.Digest{
+								Value:     "49e9b2cfa17849700d4db735d04244f3",
-							Algorithm: "md5",
+							},
-							Value:     "49e9b2cfa17849700d4db735d04244f3",
+							IsConfigFile: true,
 						},
-						IsConfigFile: true,
+						{
-					},
+							Path: "/etc/kernel/postinst.d/apt-auto-removal",
-					{
+							Digest: &file.Digest{
-						Path: "/etc/kernel/postinst.d/apt-auto-removal",
+								Algorithm: "md5",
-						Digest: &file.Digest{
+								Value:     "4ad976a68f045517cf4696cec7b8aa3a",
-							Algorithm: "md5",
+							},
-							Value:     "4ad976a68f045517cf4696cec7b8aa3a",
+							IsConfigFile: true,
 						},
-						IsConfigFile: true,
+						{
-					},
+							Path: "/etc/logrotate.d/apt",
-					{
+							Digest: &file.Digest{
-						Path: "/etc/logrotate.d/apt",
+								Algorithm: "md5",
-						Digest: &file.Digest{
+								Value:     "179f2ed4f85cbaca12fa3d69c2a4a1c3",
-							Algorithm: "md5",
+							},
-							Value:     "179f2ed4f85cbaca12fa3d69c2a4a1c3",
+							IsConfigFile: true,
 						},
-						IsConfigFile: true,
 					},
 				},
 			},
 		},
 		{
-			name:        "parse storage notation",
+			name:        "single package with installed size",
-			fixturePath: filepath.Join("test-fixtures", "status", "installed-size-4KB"),
+			fixturePath: "test-fixtures/status/installed-size-4KB",
-			expected: pkg.DpkgMetadata{
+			expected: []pkg.DpkgMetadata{
-				Package:       "apt",
+				{
-				Source:        "apt-dev",
+					Package:       "apt",
-				Version:       "1.8.2",
+					Source:        "apt-dev",
-				Architecture:  "amd64",
+					Version:       "1.8.2",
-				InstalledSize: 4000,
+					Architecture:  "amd64",
-				Maintainer:    "APT Development Team <deity@lists.debian.org>",
+					InstalledSize: 4000,
-				Description: `commandline package manager
+					Maintainer:    "APT Development Team <deity@lists.debian.org>",
+					Description: `commandline package manager
  This package provides commandline tools for searching and
  managing as well as querying information about packages
  as a low-level access to all features of the libapt-pkg library.
@@ -113,41 +111,13 @@ func TestSinglePackage(t *testing.T) {
  * apt-cdrom to use removable media as a source for packages
  * apt-config as an interface to the configuration settings
  * apt-key as an interface to manage authentication keys`,
+					Files: []pkg.DpkgFileRecord{},
+				},
 			},
-		}}
+		},
-
-	for _, test := range tests {
-		t.Run(test.name, func(t *testing.T) {
-			file, err := os.Open(test.fixturePath)
-			if err != nil {
-				t.Fatal("Unable to read test_fixtures/single: ", err)
-			}
-			defer func() {
-				err := file.Close()
-				if err != nil {
-					t.Fatal("closing file failed:", err)
-				}
-			}()
-
-			reader := bufio.NewReader(file)
-
-			entry, err := parseDpkgStatusEntry(reader)
-			if err != nil {
-				t.Fatal("Unable to read file contents: ", err)
-			}
-
-			compareEntries(t, entry, test.expected)
-		})
-	}
-}
-
-func TestMultiplePackages(t *testing.T) {
-	tests := []struct {
-		name     string
-		expected []pkg.DpkgMetadata
-	}{
 		{
-			name: "Test Multiple Package",
+			name:        "multiple entries",
+			fixturePath: "test-fixtures/status/multiple",
 			expected: []pkg.DpkgMetadata{
 				{
 					Package: "no-version",
@@ -237,30 +207,18 @@ func TestMultiplePackages(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
-			file, err := os.Open("test-fixtures/status/multiple")
+			f, err := os.Open(test.fixturePath)
-			if err != nil {
+			require.NoError(t, err)
-				t.Fatal("Unable to read: ", err)
+			t.Cleanup(func() { require.NoError(t, f.Close()) })
-			}
-			defer func() {
-				err := file.Close()
-				if err != nil {
-					t.Fatal("closing file failed:", err)
-				}
-			}()
 
-			pkgs, err := parseDpkgStatus(file)
+			reader := bufio.NewReader(f)
-			if err != nil {
-				t.Fatal("Unable to read file contents: ", err)
-			}
 
-			if len(pkgs) != 3 {
+			entries, err := parseDpkgStatus(reader)
-				t.Fatalf("unexpected number of entries: %d", len(pkgs))
+			require.NoError(t, err)
-			}
 
-			for idx, entry := range pkgs {
+			if diff := cmp.Diff(test.expected, entries); diff != "" {
-				compareEntries(t, entry.Metadata.(pkg.DpkgMetadata), test.expected[idx])
+				t.Errorf("unexpected entry (-want +got):\n%s", diff)
 			}
-
 		})
 	}
 }
@@ -304,23 +262,23 @@ func TestSourceVersionExtract(t *testing.T) {
 	}
 }
 
-func assertAs(expected error) assert.ErrorAssertionFunc {
+func requireAs(expected error) require.ErrorAssertionFunc {
-	return func(t assert.TestingT, err error, i ...interface{}) bool {
+	return func(t require.TestingT, err error, i ...interface{}) {
-		return assert.ErrorAs(t, err, &expected)
+		require.ErrorAs(t, err, &expected)
 	}
 }
 
-func Test_parseDpkgStatus(t *testing.T) {
+func Test_parseDpkgStatus_negativeCases(t *testing.T) {
 	tests := []struct {
 		name    string
 		input   string
 		want    []pkg.Package
-		wantErr assert.ErrorAssertionFunc
+		wantErr require.ErrorAssertionFunc
 	}{
 		{
 			name:    "no more packages",
 			input:   `Package: apt`,
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name: "duplicated key",
@@ -328,14 +286,14 @@ func Test_parseDpkgStatus(t *testing.T) {
 Package: apt-get
 
 `,
-			wantErr: assertAs(errors.New("duplicate key discovered: Package")),
+			wantErr: requireAs(errors.New("duplicate key discovered: Package")),
 		},
 		{
 			name: "no match for continuation",
 			input: `  Package: apt
 
 `,
-			wantErr: assertAs(errors.New("no match for continuation: line: '  Package: apt'")),
+			wantErr: requireAs(errors.New("no match for continuation: line: '  Package: apt'")),
 		},
 		{
 			name: "find keys",
@@ -348,6 +306,8 @@ Installed-Size: 10kib
 				{
 					Name:         "apt",
 					Type:         "deb",
+					PURL:         "pkg:deb/debian/apt?distro=debian-10",
+					Locations:    source.NewLocationSet(source.NewLocation("place")),
 					MetadataType: "DpkgMetadata",
 					Metadata: pkg.DpkgMetadata{
 						Package:       "apt",
@@ -356,16 +316,18 @@ Installed-Size: 10kib
 					},
 				},
 			},
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 	}
 
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			r := bufio.NewReader(strings.NewReader(tt.input))
+			pkgtest.NewCatalogTester().
-			got, err := parseDpkgStatus(r)
+				FromString("place", tt.input).
-			tt.wantErr(t, err, fmt.Sprintf("parseDpkgStatus"))
+				WithErrorAssertion(tt.wantErr).
-			assert.Equal(t, tt.want, got)
+				WithLinuxRelease(linux.Release{ID: "debian", VersionID: "10"}).
+				Expects(tt.want, nil).
+				TestParser(t, parseDpkgDB)
 		})
 	}
 }
@@ -376,53 +338,53 @@ func Test_handleNewKeyValue(t *testing.T) {
 		line    string
 		wantKey string
 		wantVal interface{}
-		wantErr assert.ErrorAssertionFunc
+		wantErr require.ErrorAssertionFunc
 	}{
 		{
 			name:    "cannot parse field",
 			line:    "blabla",
-			wantErr: assertAs(errors.New("cannot parse field from line: 'blabla'")),
+			wantErr: requireAs(errors.New("cannot parse field from line: 'blabla'")),
 		},
 		{
 			name:    "parse field",
 			line:    "key: val",
 			wantKey: "key",
 			wantVal: "val",
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name:    "parse installed size",
 			line:    "InstalledSize: 128",
 			wantKey: "InstalledSize",
 			wantVal: 128,
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name:    "parse installed kib size",
 			line:    "InstalledSize: 1kib",
 			wantKey: "InstalledSize",
 			wantVal: 1024,
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name:    "parse installed kb size",
 			line:    "InstalledSize: 1kb",
 			wantKey: "InstalledSize",
 			wantVal: 1000,
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name:    "parse installed-size mb",
 			line:    "Installed-Size: 1 mb",
 			wantKey: "InstalledSize",
 			wantVal: 1000000,
-			wantErr: assert.NoError,
+			wantErr: require.NoError,
 		},
 		{
 			name:    "fail parsing installed-size",
 			line:    "Installed-Size: 1bla",
 			wantKey: "",
-			wantErr: assertAs(fmt.Errorf("unhandled size name: %s", "bla")),
+			wantErr: requireAs(fmt.Errorf("unhandled size name: %s", "bla")),
 		},
 	}
 	for _, tt := range tests {

syft/pkg/cataloger/deb/parse_dpkg_info_files_test.go

@@ -4,7 +4,8 @@ import (
 	"os"
 	"testing"
 
-	"github.com/go-test/deep"
+	"github.com/google/go-cmp/cmp"
+	"github.com/stretchr/testify/require"
 
 	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/pkg"
@@ -40,29 +41,14 @@ func TestMD5SumInfoParsing(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.fixture, func(t *testing.T) {
-			file, err := os.Open(test.fixture)
+			f, err := os.Open(test.fixture)
-			if err != nil {
+			require.NoError(t, err)
-				t.Fatal("Unable to read: ", err)
+			t.Cleanup(func() { require.NoError(t, f.Close()) })
-			}
-			defer func() {
-				err := file.Close()
-				if err != nil {
-					t.Fatal("closing file failed:", err)
-				}
-			}()
 
-			actual := parseDpkgMD5Info(file)
+			actual := parseDpkgMD5Info(f)
 
-			if len(actual) != len(test.expected) {
+			if diff := cmp.Diff(test.expected, actual); diff != "" {
-				for _, a := range actual {
+				t.Errorf("unexpected md5 files (-want +got):\n%s", diff)
-					t.Logf("   %+v", a)
-				}
-				t.Fatalf("unexpected package count: %d!=%d", len(actual), len(test.expected))
-			}
-
-			diffs := deep.Equal(actual, test.expected)
-			for _, d := range diffs {
-				t.Errorf("diff: %+v", d)
 			}
 
 		})
@@ -89,29 +75,14 @@ func TestConffileInfoParsing(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.fixture, func(t *testing.T) {
-			file, err := os.Open(test.fixture)
+			f, err := os.Open(test.fixture)
-			if err != nil {
+			require.NoError(t, err)
-				t.Fatal("Unable to read: ", err)
+			t.Cleanup(func() { require.NoError(t, f.Close()) })
-			}
-			defer func() {
-				err := file.Close()
-				if err != nil {
-					t.Fatal("closing file failed:", err)
-				}
-			}()
 
-			actual := parseDpkgConffileInfo(file)
+			actual := parseDpkgConffileInfo(f)
 
-			if len(actual) != len(test.expected) {
+			if diff := cmp.Diff(test.expected, actual); diff != "" {
-				for _, a := range actual {
+				t.Errorf("unexpected md5 files (-want +got):\n%s", diff)
-					t.Logf("   %+v", a)
-				}
-				t.Fatalf("unexpected package count: %d!=%d", len(actual), len(test.expected))
-			}
-
-			diffs := deep.Equal(actual, test.expected)
-			for _, d := range diffs {
-				t.Errorf("diff: %+v", d)
 			}
 
 		})

syft/pkg/cataloger/dotnet/parse_dotnet_deps_test.go

@@ -209,5 +209,5 @@ func TestParseDotnetDeps(t *testing.T) {
 	}
 
 	var expectedRelationships []artifact.Relationship
-	pkgtest.TestGenericParser(t, fixture, parseDotnetDeps, expected, expectedRelationships)
+	pkgtest.TestFileParser(t, fixture, parseDotnetDeps, expected, expectedRelationships)
 }

syft/pkg/cataloger/internal/pkgtest/assert_packages_equal.go

@@ -1,37 +0,0 @@
-package pkgtest
-
-import (
-	"testing"
-
-	"github.com/google/go-cmp/cmp"
-	"github.com/google/go-cmp/cmp/cmpopts"
-
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/source"
-)
-
-func AssertPackagesEqual(t testing.TB, expected, actual []pkg.Package) {
-	if diff := cmp.Diff(expected, actual,
-		cmpopts.IgnoreFields(pkg.Package{}, "id"), // note: ID is not deterministic for test purposes
-		cmp.Comparer(
-			func(x, y source.LocationSet) bool {
-				xs := x.ToSlice()
-				ys := y.ToSlice()
-
-				if len(xs) != len(ys) {
-					return false
-				}
-				for i, xe := range xs {
-					ye := ys[i]
-					if !(cmp.Equal(xe.Coordinates, ye.Coordinates) && cmp.Equal(xe.VirtualPath, ye.VirtualPath)) {
-						return false
-					}
-				}
-
-				return true
-			},
-		),
-	); diff != "" {
-		t.Errorf("unexpected packages from parsing (-expected +actual)\n%s", diff)
-	}
-}

syft/pkg/cataloger/internal/pkgtest/test_generic_parser.go

@@ -1,37 +1,175 @@
 package pkgtest
 
 import (
+	"io"
 	"os"
+	"strings"
 	"testing"
 
 	"github.com/google/go-cmp/cmp"
+	"github.com/google/go-cmp/cmp/cmpopts"
 	"github.com/stretchr/testify/require"
 
+	"github.com/anchore/stereoscope/pkg/imagetest"
 	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/linux"
 	"github.com/anchore/syft/syft/pkg"
 	"github.com/anchore/syft/syft/pkg/cataloger/generic"
 	"github.com/anchore/syft/syft/source"
 )
 
-func TestGenericParser(t *testing.T, fixturePath string, parser generic.Parser, expectedPkgs []pkg.Package, expectedRelationships []artifact.Relationship) {
+type locationComparer func(x, y source.Location) bool
-	t.Helper()
+
-	TestGenericParserWithEnv(t, fixturePath, parser, nil, expectedPkgs, expectedRelationships)
+type CatalogTester struct {
+	expectedPkgs          []pkg.Package
+	expectedRelationships []artifact.Relationship
+	env                   *generic.Environment
+	reader                source.LocationReadCloser
+	resolver              source.FileResolver
+	wantErr               require.ErrorAssertionFunc
+	compareOptions        []cmp.Option
+	locationComparer      locationComparer
 }
 
-func TestGenericParserWithEnv(t *testing.T, fixturePath string, parser generic.Parser, env *generic.Environment, expectedPkgs []pkg.Package, expectedRelationships []artifact.Relationship) {
+func NewCatalogTester() *CatalogTester {
+	return &CatalogTester{
+		wantErr: require.NoError,
+		locationComparer: func(x, y source.Location) bool {
+			return cmp.Equal(x.Coordinates, y.Coordinates) && cmp.Equal(x.VirtualPath, y.VirtualPath)
+		},
+	}
+}
+
+func (p *CatalogTester) FromFile(t *testing.T, path string) *CatalogTester {
 	t.Helper()
-	fixture, err := os.Open(fixturePath)
+
+	fixture, err := os.Open(path)
 	require.NoError(t, err)
 
-	actualPkgs, actualRelationships, err := parser(nil, env, source.LocationReadCloser{
+	p.reader = source.LocationReadCloser{
 		Location:   source.NewLocation(fixture.Name()),
 		ReadCloser: fixture,
-	})
+	}
+	return p
+}
+
+func (p *CatalogTester) FromString(location, data string) *CatalogTester {
+	p.reader = source.LocationReadCloser{
+		Location:   source.NewLocation(location),
+		ReadCloser: io.NopCloser(strings.NewReader(data)),
+	}
+	return p
+}
+
+func (p *CatalogTester) WithLinuxRelease(r linux.Release) *CatalogTester {
+	if p.env == nil {
+		p.env = &generic.Environment{}
+	}
+	p.env.LinuxRelease = &r
+	return p
+}
+
+func (p *CatalogTester) WithEnv(env *generic.Environment) *CatalogTester {
+	p.env = env
+	return p
+}
+
+func (p *CatalogTester) WithError() *CatalogTester {
+	p.wantErr = require.Error
+	return p
+}
+
+func (p *CatalogTester) WithErrorAssertion(a require.ErrorAssertionFunc) *CatalogTester {
+	p.wantErr = a
+	return p
+}
+
+func (p *CatalogTester) WithResolver(r source.FileResolver) *CatalogTester {
+	p.resolver = r
+	return p
+}
+
+func (p *CatalogTester) WithImageResolver(t *testing.T, fixtureName string) *CatalogTester {
+	t.Helper()
+	img := imagetest.GetFixtureImage(t, "docker-archive", fixtureName)
+
+	s, err := source.NewFromImage(img, fixtureName)
 	require.NoError(t, err)
 
-	AssertPackagesEqual(t, expectedPkgs, actualPkgs)
+	r, err := s.FileResolver(source.SquashedScope)
+	require.NoError(t, err)
+	p.resolver = r
+	return p
+}
 
-	if diff := cmp.Diff(expectedRelationships, actualRelationships); diff != "" {
+func (p *CatalogTester) IgnoreLocationLayer() *CatalogTester {
+	p.locationComparer = func(x, y source.Location) bool {
+		return cmp.Equal(x.Coordinates.RealPath, y.Coordinates.RealPath) && cmp.Equal(x.VirtualPath, y.VirtualPath)
+	}
+	return p
+}
+
+func (p *CatalogTester) Expects(pkgs []pkg.Package, relationships []artifact.Relationship) *CatalogTester {
+	p.expectedPkgs = pkgs
+	p.expectedRelationships = relationships
+	return p
+}
+
+func (p *CatalogTester) TestParser(t *testing.T, parser generic.Parser) {
+	t.Helper()
+	pkgs, relationships, err := parser(p.resolver, p.env, p.reader)
+	p.wantErr(t, err)
+	p.assertPkgs(t, pkgs, relationships)
+}
+
+func (p *CatalogTester) TestCataloger(t *testing.T, cataloger pkg.Cataloger) {
+	t.Helper()
+	pkgs, relationships, err := cataloger.Catalog(p.resolver)
+	p.wantErr(t, err)
+	p.assertPkgs(t, pkgs, relationships)
+}
+
+func (p *CatalogTester) assertPkgs(t *testing.T, pkgs []pkg.Package, relationships []artifact.Relationship) {
+	t.Helper()
+
+	p.compareOptions = append(p.compareOptions,
+		cmpopts.IgnoreFields(pkg.Package{}, "id"), // note: ID is not deterministic for test purposes
+		cmp.Comparer(
+			func(x, y source.LocationSet) bool {
+				xs := x.ToSlice()
+				ys := y.ToSlice()
+
+				if len(xs) != len(ys) {
+					return false
+				}
+				for i, xe := range xs {
+					ye := ys[i]
+					if !p.locationComparer(xe, ye) {
+						return false
+					}
+				}
+
+				return true
+			},
+		),
+	)
+
+	if diff := cmp.Diff(p.expectedPkgs, pkgs, p.compareOptions...); diff != "" {
+		t.Errorf("unexpected packages from parsing (-expected +actual)\n%s", diff)
+	}
+
+	if diff := cmp.Diff(p.expectedRelationships, relationships, p.compareOptions...); diff != "" {
 		t.Errorf("unexpected relationships from parsing (-expected +actual)\n%s", diff)
 	}
 }
+
+func TestFileParser(t *testing.T, fixturePath string, parser generic.Parser, expectedPkgs []pkg.Package, expectedRelationships []artifact.Relationship) {
+	t.Helper()
+	NewCatalogTester().FromFile(t, fixturePath).Expects(expectedPkgs, expectedRelationships).TestParser(t, parser)
+}
+
+func TestFileParserWithEnv(t *testing.T, fixturePath string, parser generic.Parser, env *generic.Environment, expectedPkgs []pkg.Package, expectedRelationships []artifact.Relationship) {
+	t.Helper()
+
+	NewCatalogTester().FromFile(t, fixturePath).WithEnv(env).Expects(expectedPkgs, expectedRelationships).TestParser(t, parser)
+}

syft/pkg/dpkg_metadata.go

@@ -1,22 +1,16 @@
 package pkg
 
 import (
-	"fmt"
 	"sort"
 
 	"github.com/scylladb/go-set/strset"
 
-	"github.com/anchore/packageurl-go"
 	"github.com/anchore/syft/syft/file"
-	"github.com/anchore/syft/syft/linux"
 )
 
 const DpkgDBGlob = "**/var/lib/dpkg/{status,status.d/**}"
 
-var (
+var _ FileOwner = (*DpkgMetadata)(nil)
-	_ FileOwner     = (*DpkgMetadata)(nil)
-	_ urlIdentifier = (*DpkgMetadata)(nil)
-)
 
 // DpkgMetadata represents all captured data for a Debian package DB entry; available fields are described
 // at http://manpages.ubuntu.com/manpages/xenial/man1/dpkg-query.1.html in the --showformat section.
@@ -39,40 +33,6 @@ type DpkgFileRecord struct {
 	IsConfigFile bool         `json:"isConfigFile"`
 }
 
-// PackageURL returns the PURL for the specific Debian package (see https://github.com/package-url/purl-spec)
-func (m DpkgMetadata) PackageURL(distro *linux.Release) string {
-	var namespace string
-	if distro != nil {
-		namespace = distro.ID
-	}
-
-	qualifiers := map[string]string{
-		PURLQualifierArch: m.Architecture,
-	}
-
-	if m.Source != "" {
-		if m.SourceVersion != "" {
-			qualifiers[PURLQualifierUpstream] = fmt.Sprintf("%s@%s", m.Source, m.SourceVersion)
-		} else {
-			qualifiers[PURLQualifierUpstream] = m.Source
-		}
-	}
-
-	return packageurl.NewPackageURL(
-		// TODO: replace with `packageurl.TypeDebian` upon merge of https://github.com/package-url/packageurl-go/pull/21
-		// TODO: or, since we're now using an Anchore fork of this module, we could do this sooner.
-		"deb",
-		namespace,
-		m.Package,
-		m.Version,
-		PURLQualifiers(
-			qualifiers,
-			distro,
-		),
-		"",
-	).ToString()
-}
-
 func (m DpkgMetadata) OwnedFiles() (result []string) {
 	s := strset.New()
 	for _, f := range m.Files {

syft/pkg/url_test.go

@@ -66,24 +66,6 @@ func TestPackageURL(t *testing.T) {
 			},
 			expected: "pkg:npm/name@v0.1.0",
 		},
-		{
-			name: "deb",
-			distro: &linux.Release{
-				ID:        "ubuntu",
-				VersionID: "20.04",
-			},
-			pkg: Package{
-				Name:    "bad-name",
-				Version: "bad-v0.1.0",
-				Type:    DebPkg,
-				Metadata: DpkgMetadata{
-					Package:      "name",
-					Version:      "v0.1.0",
-					Architecture: "amd64",
-				},
-			},
-			expected: "pkg:deb/ubuntu/name@v0.1.0?arch=amd64&distro=ubuntu-20.04",
-		},
 		{
 			name: "rpm",
 			distro: &linux.Release{
@@ -199,6 +181,7 @@ func TestPackageURL(t *testing.T) {
 	expectedTypes.Remove(string(ConanPkg))
 	expectedTypes.Remove(string(DartPubPkg))
 	expectedTypes.Remove(string(DotnetPkg))
+	expectedTypes.Remove(string(DebPkg))
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {