defer to GenericCataloger instances for pkg catalogers

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

syft/cataloger/apkdb/cataloger.go

@@ -4,39 +4,14 @@ Package apkdb provides a concrete Cataloger implementation for Alpine DB files.
 package apkdb
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.ApkPkg Package Types defined in Alpine DB files.
+// NewApkdbCataloger returns a new Alpine DB cataloger object.
-type Cataloger struct {
+func NewApkdbCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Alpine DB cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
 		"**/lib/apk/db/installed": parseApkDB,
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "apkdb-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "apkdb-cataloger"
-}
-
-// SelectFiles returns a set of discovered Alpine DB files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all Alpine DB files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/bundler/cataloger.go

@@ -4,40 +4,23 @@ Package bundler provides a concrete Cataloger implementation for Ruby Gemfile.lo
 package bundler
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.GemPkg Package Types defined in Bundler Gemfile.lock files.
+// NewGemfileLockCataloger returns a new Bundler cataloger object tailored for parsing index-oriented files (e.g. Gemfile.lock).
-type Cataloger struct {
+func NewGemfileLockCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Bundler cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
-		"**/Gemfile.lock": parseGemfileLockEntries, // valid in a dir context
+		"**/Gemfile.lock": parseGemfileLockEntries,
-		//"**/specification/*.gemspec": parseGemSpecEntries,     // valid in an image context (against installed gems)
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "ruby-gemfile-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
+}
+
+// NewGemspecCataloger returns a new Bundler cataloger object tailored for detecting installations of gems (e.g. Gemspec).
+func NewGemspecCataloger() *common.GenericCataloger {
+	globParsers := map[string]common.ParserFn{
+		"**/specification/*.gemspec": parseGemspecEntries,
 	}
-}
 
-// Name returns a string that uniquely describes this cataloger.
+	return common.NewGenericCataloger(nil, globParsers, "ruby-gemspec-cataloger")
-func (a *Cataloger) Name() string {
-	return "bundler-cataloger"
-}
-
-// SelectFiles returns a set of discovered Gemfile.lock files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all Gemfile.lock files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/cataloger.go

@@ -9,7 +9,7 @@ import (
 	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/apkdb"
 	"github.com/anchore/syft/syft/cataloger/bundler"
-	"github.com/anchore/syft/syft/cataloger/dpkg"
+	"github.com/anchore/syft/syft/cataloger/deb"
 	"github.com/anchore/syft/syft/cataloger/golang"
 	"github.com/anchore/syft/syft/cataloger/java"
 	"github.com/anchore/syft/syft/cataloger/javascript"
@@ -36,13 +36,13 @@ type Cataloger interface {
 // All returns a slice of all locally defined catalogers (defined in child packages).
 func All() []Cataloger {
 	return []Cataloger{
-		dpkg.New(),
+		deb.NewDpkgdbCataloger(),
-		bundler.New(),
+		bundler.NewGemfileLockCataloger(),
-		python.New(),
+		python.NewPythonCataloger(),
-		rpmdb.New(),
+		rpmdb.NewRpmdbCataloger(),
-		java.New(),
+		java.NewJavaCataloger(),
-		apkdb.New(),
+		apkdb.NewApkdbCataloger(),
-		golang.New(),
+		golang.NewGoModCataloger(),
-		javascript.New(),
+		javascript.NewJavascriptCataloger(),
 	}
 }

syft/cataloger/common/generic_cataloger.go

@@ -15,22 +15,29 @@ import (
 // GenericCataloger implements the Catalog interface and is responsible for dispatching the proper parser function for
 // a given path or glob pattern. This is intended to be reusable across many package cataloger types.
 type GenericCataloger struct {
-	globParsers   map[string]ParserFn
+	globParsers     map[string]ParserFn
-	pathParsers   map[string]ParserFn
+	pathParsers     map[string]ParserFn
-	selectedFiles []file.Reference
+	selectedFiles   []file.Reference
-	parsers       map[file.Reference]ParserFn
+	parsers         map[file.Reference]ParserFn
+	upstreamMatcher string
 }
 
 // NewGenericCataloger if provided path-to-parser-function and glob-to-parser-function lookups creates a GenericCataloger
-func NewGenericCataloger(pathParsers map[string]ParserFn, globParsers map[string]ParserFn) GenericCataloger {
+func NewGenericCataloger(pathParsers map[string]ParserFn, globParsers map[string]ParserFn, upstreamMatcher string) *GenericCataloger {
-	return GenericCataloger{
+	return &GenericCataloger{
-		globParsers:   globParsers,
+		globParsers:     globParsers,
-		pathParsers:   pathParsers,
+		pathParsers:     pathParsers,
-		selectedFiles: make([]file.Reference, 0),
+		selectedFiles:   make([]file.Reference, 0),
-		parsers:       make(map[file.Reference]ParserFn),
+		parsers:         make(map[file.Reference]ParserFn),
+		upstreamMatcher: upstreamMatcher,
 	}
 }
 
+// Name returns a string that uniquely describes the upstream cataloger that this Generic Cataloger represents.
+func (a *GenericCataloger) Name() string {
+	return a.upstreamMatcher
+}
+
 // register pairs a set of file references with a parser function for future cataloging (when the file contents are resolved)
 func (a *GenericCataloger) register(files []file.Reference, parser ParserFn) {
 	a.selectedFiles = append(a.selectedFiles, files...)
@@ -73,7 +80,7 @@ func (a *GenericCataloger) SelectFiles(resolver scope.FileResolver) []file.Refer
 }
 
 // Catalog takes a set of file contents and uses any configured parser functions to resolve and return discovered packages
-func (a *GenericCataloger) Catalog(contents map[file.Reference]string, upstreamMatcher string) ([]pkg.Package, error) {
+func (a *GenericCataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
 	defer a.clear()
 
 	packages := make([]pkg.Package, 0)
@@ -81,19 +88,19 @@ func (a *GenericCataloger) Catalog(contents map[file.Reference]string, upstreamM
 	for reference, parser := range a.parsers {
 		content, ok := contents[reference]
 		if !ok {
-			log.Errorf("cataloger '%s' missing file content: %+v", upstreamMatcher, reference)
+			log.Errorf("cataloger '%s' missing file content: %+v", a.upstreamMatcher, reference)
 			continue
 		}
 
 		entries, err := parser(string(reference.Path), strings.NewReader(content))
 		if err != nil {
 			// TODO: should we fail? or only log?
-			log.Errorf("cataloger '%s' failed to parse entries (reference=%+v): %+v", upstreamMatcher, reference, err)
+			log.Errorf("cataloger '%s' failed to parse entries (reference=%+v): %+v", a.upstreamMatcher, reference, err)
 			continue
 		}
 
 		for _, entry := range entries {
-			entry.FoundBy = upstreamMatcher
+			entry.FoundBy = a.upstreamMatcher
 			entry.Source = []file.Reference{reference}
 
 			packages = append(packages, entry)

syft/cataloger/common/generic_cataloger_test.go

@@ -60,9 +60,9 @@ func TestGenericCataloger(t *testing.T) {
 		"/another-path.txt": parser,
 		"/last/path.txt":    parser,
 	}
-
+	upstream := "some-other-cataloger"
 	resolver := newTestResolver()
-	cataloger := NewGenericCataloger(pathParsers, globParsers)
+	cataloger := NewGenericCataloger(pathParsers, globParsers, upstream)
 
 	selected := cataloger.SelectFiles(resolver)
 
@@ -79,7 +79,6 @@ func TestGenericCataloger(t *testing.T) {
 		selectionByPath[string(s.Path)] = s
 	}
 
-	upstream := "some-other-cataloger"
 	expectedPkgs := make(map[file.Reference]pkg.Package)
 	for path, ref := range selectionByPath {
 		expectedPkgs[ref] = pkg.Package{
@@ -89,7 +88,7 @@ func TestGenericCataloger(t *testing.T) {
 		}
 	}
 
-	actualPkgs, err := cataloger.Catalog(resolver.contents, upstream)
+	actualPkgs, err := cataloger.Catalog(resolver.contents)
 	if err != nil {
 		t.Fatalf("cataloger catalog action failed: %+v", err)
 	}

syft/cataloger/deb/cataloger.go

@@ -0,0 +1,17 @@
+/*
+Package dpkg provides a concrete Cataloger implementation for Debian package DB status files.
+*/
+package deb
+
+import (
+	"github.com/anchore/syft/syft/cataloger/common"
+)
+
+// NewDpkgdbCataloger returns a new Deb package cataloger object.
+func NewDpkgdbCataloger() *common.GenericCataloger {
+	globParsers := map[string]common.ParserFn{
+		"**/var/lib/dpkg/status": parseDpkgStatus,
+	}
+
+	return common.NewGenericCataloger(nil, globParsers, "dpkgdb-cataloger")
+}

syft/cataloger/deb/parse_dpkg_status.go

@@ -1,4 +1,4 @@
-package dpkg
+package deb
 
 import (
 	"bufio"

syft/cataloger/deb/parse_dpkg_status_test.go

@@ -1,4 +1,4 @@
-package dpkg
+package deb
 
 import (
 	"bufio"

syft/cataloger/dpkg/cataloger.go

@@ -1,42 +0,0 @@
-/*
-Package dpkg provides a concrete Cataloger implementation for Debian package DB status files.
-*/
-package dpkg
-
-import (
-	"github.com/anchore/stereoscope/pkg/file"
-	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
-)
-
-// Cataloger catalogs pkg.DebPkg Package Types defined in DPKG status files.
-type Cataloger struct {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Deb package cataloger object.
-func New() *Cataloger {
-	globParsers := map[string]common.ParserFn{
-		"**/var/lib/dpkg/status": parseDpkgStatus,
-	}
-
-	return &Cataloger{
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "dpkg-cataloger"
-}
-
-// SelectFiles returns a set of discovered DPKG status files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all DPKG status files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
-}

syft/cataloger/golang/cataloger.go

@@ -4,39 +4,14 @@ Package golang provides a concrete Cataloger implementation for go.mod files.
 package golang
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.GoModulePkg Package Types defined in go.mod files.
+// NewGoModCataloger returns a new Go module cataloger object.
-type Cataloger struct {
+func NewGoModCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Go module cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
 		"**/go.mod": parseGoMod,
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "go-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "go-cataloger"
-}
-
-// SelectFiles returns a set of discovered go.mod files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all go.mod files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/java/cataloger.go

@@ -4,40 +4,15 @@ Package java provides a concrete Cataloger implementation for Java archives (jar
 package java
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.JavaPkg and pkg.JenkinsPluginPkg Package Types defined in java archive files.
+// NewJavaCataloger returns a new Java archive cataloger object.
-type Cataloger struct {
+func NewJavaCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Java archive cataloger object.
-func New() *Cataloger {
 	globParsers := make(map[string]common.ParserFn)
 	for _, pattern := range archiveFormatGlobs {
 		globParsers[pattern] = parseJavaArchive
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "java-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "java-cataloger"
-}
-
-// SelectFiles returns a set of discovered Java archive files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all Java archive files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/javascript/cataloger.go

@@ -4,40 +4,15 @@ Package javascript provides a concrete Cataloger implementation for JavaScript e
 package javascript
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.YarnPkg and pkg.NpmPkg Package Types defined in package-lock.json and yarn.lock files.
+// NewJavascriptCataloger returns a new JavaScript cataloger object.
-type Cataloger struct {
+func NewJavascriptCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new JavaScript cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
 		"**/package-lock.json": parsePackageLock,
 		"**/yarn.lock":         parseYarnLock,
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "javascript-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "javascript-cataloger"
-}
-
-// SelectFiles returns a set of discovered Javascript ecosystem files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all Javascript ecosystem files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/python/cataloger.go

@@ -4,19 +4,11 @@ Package python provides a concrete Cataloger implementation for Python ecosystem
 package python
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.WheelPkg, pkg.EggPkg, and pkg.PythonRequirementsPkg Package Types defined in Python ecosystem files.
+// NewPythonCataloger returns a new Python cataloger object.
-type Cataloger struct {
+func NewPythonCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new Python cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
 		"**/*egg-info/PKG-INFO":  parseEggMetadata,
 		"**/*dist-info/METADATA": parseWheelMetadata,
@@ -25,22 +17,5 @@ func New() *Cataloger {
 		"**/setup.py":            parseSetup,
 	}
 
-	return &Cataloger{
+	return common.NewGenericCataloger(nil, globParsers, "python-cataloger")
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "python-cataloger"
-}
-
-// SelectFiles returns a set of discovered Python ecosystem files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all Python ecosystem files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }

syft/cataloger/rpmdb/cataloger.go

@@ -4,39 +4,13 @@ Package rpmdb provides a concrete Cataloger implementation for RPM "Package" DB
 package rpmdb
 
 import (
-	"github.com/anchore/stereoscope/pkg/file"
 	"github.com/anchore/syft/syft/cataloger/common"
-	"github.com/anchore/syft/syft/pkg"
-	"github.com/anchore/syft/syft/scope"
 )
 
-// Cataloger catalogs pkg.RpmPkg Package Types defined in RPM DB files.
+// NewRpmdbCataloger returns a new RPM DB cataloger object.
-type Cataloger struct {
+func NewRpmdbCataloger() *common.GenericCataloger {
-	cataloger common.GenericCataloger
-}
-
-// New returns a new RPM DB cataloger object.
-func New() *Cataloger {
 	globParsers := map[string]common.ParserFn{
 		"**/var/lib/rpm/Packages": parseRpmDB,
 	}
-
+	return common.NewGenericCataloger(nil, globParsers, "rpmdb-cataloger")
-	return &Cataloger{
-		cataloger: common.NewGenericCataloger(nil, globParsers),
-	}
-}
-
-// Name returns a string that uniquely describes this cataloger.
-func (a *Cataloger) Name() string {
-	return "rpmdb-cataloger"
-}
-
-// SelectFiles returns a set of discovered RPM DB files from the user content source.
-func (a *Cataloger) SelectFiles(resolver scope.FileResolver) []file.Reference {
-	return a.cataloger.SelectFiles(resolver)
-}
-
-// Catalog returns the Packages indexed from all RPM DB files discovered.
-func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
-	return a.cataloger.Catalog(contents, a.Name())
 }