Continue - Option to enable specific language or ecosystem cataloger

README.md

@@ -314,6 +314,29 @@ package:
     # same as -s ; SYFT_PACKAGE_CATALOGER_SCOPE env var
     scope: "squashed"
 
+  # Cataloger group select 
+  # When Empty default select for each scheme - Dir:Index, Image:installed
+  # Options: [index install all].
+  # cataloger-group: ""
+
+  # enable specific language or ecosystem cataloger
+  # default: select catalogers out of group
+  # catalogers:
+  # - "ruby-gemfile-cataloger"
+  # - "ruby-gemspec-cataloger"
+  # - "python-index-cataloger"
+  # - "python-package-cataloger"
+  # - "javascript-lock-cataloger"
+  # - "javascript-package-cataloger"
+  # - "php-composer-installed-cataloger"
+  # - "php-composer-lock-cataloger"
+  # - "dpkgdb-cataloger"
+  # - "rpmdb-cataloger"
+  # - "java-cataloger"
+  # - "apkdb-cataloger"
+  # - "go-module-binary-cataloger"
+  catalogers:
+
 # cataloging file classifications is exposed through the power-user subcommand
 file-classification:
   cataloger:

cmd/packages.go

@@ -137,18 +137,23 @@ func setPackageFlags(flags *pflag.FlagSet) {
 	)
 
 	flags.StringArrayP(
-		"exclude", "", nil,
-		"exclude paths from being scanned using a glob expression",
+		"exclude", "", nil, "exclude paths from being scanned using a glob expression",
+	)
+
+	flags.StringArrayP(
+		"cataloger", "C", nil, "enable specific language or ecosystem cataloger",
+	)
+
+	flags.StringP(
+		"cataloger-group", "", "", fmt.Sprintf("selection cataloger group, options=%v", cataloger.AllGroups),
 	)
 
 	flags.Bool(
-		"overwrite-existing-image", false,
-		"overwrite an existing image during the upload to Anchore Enterprise",
+		"overwrite-existing-image", false, "overwrite an existing image during the upload to Anchore Enterprise",
 	)
 
 	flags.Uint(
-		"import-timeout", 30,
-		"set a timeout duration (in seconds) for the upload to Anchore Enterprise",
+		"import-timeout", 30, "set a timeout duration (in seconds) for the upload to Anchore Enterprise",
 	)
 }
 
@@ -180,6 +185,14 @@ func bindExclusivePackagesConfigOptions(flags *pflag.FlagSet) error {
 		return err
 	}
 
+	if err := viper.BindPFlag("package.catalogers", flags.Lookup("cataloger")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("package.cataloger-group", flags.Lookup("cataloger-group")); err != nil {
+		return err
+	}
+
 	// Upload options //////////////////////////////////////////////////////////
 
 	if err := viper.BindPFlag("anchore.host", flags.Lookup("host")); err != nil {

internal/config/pkg.go

@@ -9,6 +9,8 @@ type pkg struct {
 	Cataloger               catalogerOptions `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
 	SearchUnindexedArchives bool             `yaml:"search-unindexed-archives" json:"search-unindexed-archives" mapstructure:"search-unindexed-archives"`
 	SearchIndexedArchives   bool             `yaml:"search-indexed-archives" json:"search-indexed-archives" mapstructure:"search-indexed-archives"`
+	Catalogers              []string         `yaml:"catalogers" json:"catalogers" mapstructure:"catalogers"`
+	CatalogerGroup          cataloger.Group  `yaml:"cataloger-group" json:"cataloger-group" mapstructure:"cataloger-group"`
 }
 
 func (cfg pkg) loadDefaultValues(v *viper.Viper) {
@@ -29,5 +31,7 @@ func (cfg pkg) ToConfig() cataloger.Config {
 			IncludeUnindexedArchives: cfg.SearchUnindexedArchives,
 			Scope:                    cfg.Cataloger.ScopeOpt,
 		},
+		Catalogers:     cfg.Catalogers,
+		CatalogerGroup: cfg.CatalogerGroup,
 	}
 }

syft/lib.go

@@ -48,23 +48,27 @@ func CatalogPackages(src *source.Source, cfg cataloger.Config) (*pkg.Catalog, []
 		log.Info("could not identify distro")
 	}
 
-	// conditionally use the correct set of loggers based on the input type (container image or directory)
-	var catalogers []cataloger.Cataloger
-	switch src.Metadata.Scheme {
-	case source.ImageScheme:
-		log.Info("cataloging image")
-		catalogers = cataloger.ImageCatalogers(cfg)
-	case source.FileScheme:
-		log.Info("cataloging file")
-		catalogers = cataloger.AllCatalogers(cfg)
-	case source.DirectoryScheme:
-		log.Info("cataloging directory")
-		catalogers = cataloger.DirectoryCatalogers(cfg)
-	default:
-		return nil, nil, nil, fmt.Errorf("unable to determine cataloger set from scheme=%+v", src.Metadata.Scheme)
+	// conditionally use the correct set of catalogers based on the scheme (container image or directory)
+	if cfg.CatalogerGroup == "" {
+		switch src.Metadata.Scheme {
+		case source.ImageScheme:
+			cfg.CatalogerGroup = cataloger.InstallationGroup
+		case source.FileScheme:
+			cfg.CatalogerGroup = cataloger.AllGroup
+		case source.DirectoryScheme:
+			cfg.CatalogerGroup = cataloger.IndexGroup
+		default:
+			return nil, nil, nil, fmt.Errorf("unable to determine cataloger set from scheme=%+v", src.Metadata.Scheme)
+		}
 	}
 
-	catalog, relationships, err := cataloger.Catalog(resolver, release, catalogers...)
+	groupCatalogers, err := cataloger.SelectGroup(cfg)
+	if err != nil {
+		return nil, nil, nil, err
+	}
+	enabledCatalogers := cataloger.FilterCatalogers(cfg, groupCatalogers)
+
+	catalog, relationships, err := cataloger.Catalog(resolver, release, enabledCatalogers...)
 	if err != nil {
 		return nil, nil, nil, err
 	}

syft/pkg/cataloger/cataloger.go

@@ -6,6 +6,9 @@ catalogers defined in child packages as well as the interface definition to impl
 package cataloger
 
 import (
+	"fmt"
+
+	"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/apkdb"
@@ -22,6 +25,20 @@ import (
 	"github.com/anchore/syft/syft/source"
 )
 
+type Group string
+
+const (
+	IndexGroup        Group = "index"
+	InstallationGroup Group = "install"
+	AllGroup          Group = "all"
+)
+
+var AllGroups = []Group{
+	IndexGroup,
+	InstallationGroup,
+	AllGroup,
+}
+
 // 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.
@@ -32,8 +49,8 @@ type Cataloger interface {
 	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 {
+// InstallationCatalogers returns a slice of locally implemented catalogers that are fit for detecting installations of packages.
+func InstallationCatalogers(cfg Config) []Cataloger {
 	return []Cataloger{
 		ruby.NewGemSpecCataloger(),
 		python.NewPythonPackageCataloger(),
@@ -47,8 +64,8 @@ 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 {
+// IndexCatalogers returns a slice of locally implemented catalogers that are fit for detecting packages from index files (and select installations)
+func IndexCatalogers(cfg Config) []Cataloger {
 	return []Cataloger{
 		ruby.NewGemFileLockCataloger(),
 		python.NewPythonIndexCataloger(),
@@ -85,3 +102,47 @@ func AllCatalogers(cfg Config) []Cataloger {
 		dart.NewPubspecLockCataloger(),
 	}
 }
+
+func SelectGroup(cfg Config) ([]Cataloger, error) {
+	switch cfg.CatalogerGroup {
+	case IndexGroup:
+		log.Info("cataloging index group")
+		return IndexCatalogers(cfg), nil
+	case InstallationGroup:
+		log.Info("cataloging installation group")
+		return InstallationCatalogers(cfg), nil
+	case AllGroup:
+		log.Info("cataloging all group")
+		return AllCatalogers(cfg), nil
+	default:
+		return nil, fmt.Errorf("unknown cataloger group, Group: %s", cfg.CatalogerGroup)
+	}
+}
+
+func FilterCatalogers(cfg Config, groupCatalogers []Cataloger) []Cataloger {
+	return filterCatalogers(groupCatalogers, cfg.Catalogers)
+}
+
+func filterCatalogers(catalogers []Cataloger, enabledCatalogers []string) []Cataloger {
+	// if enable-cataloger is not set, all applicable catalogers are enabled by default
+	if len(enabledCatalogers) == 0 {
+		return catalogers
+	}
+	var filteredCatalogers []Cataloger
+	for _, cataloger := range catalogers {
+		if contains(enabledCatalogers, cataloger.Name()) {
+			filteredCatalogers = append(filteredCatalogers, cataloger)
+		}
+	}
+	return filteredCatalogers
+}
+
+func contains(catalogers []string, str string) bool {
+	for _, cataloger := range catalogers {
+		if cataloger == str ||
+			fmt.Sprintf("%s-cataloger", cataloger) == str {
+			return true
+		}
+	}
+	return false
+}

syft/pkg/cataloger/config.go

@@ -5,7 +5,9 @@ import (
 )
 
 type Config struct {
-	Search SearchConfig
+	Search         SearchConfig
+	Catalogers     []string
+	CatalogerGroup Group
 }
 
 func DefaultConfig() Config {

test/integration/catalog_packages_test.go

@@ -22,7 +22,7 @@ func BenchmarkImagePackageCatalogers(b *testing.B) {
 	tarPath := imagetest.GetFixtureImageTarPath(b, fixtureImageName)
 
 	var pc *pkg.Catalog
-	for _, c := range cataloger.ImageCatalogers(cataloger.DefaultConfig()) {
+	for _, c := range cataloger.InstallationCatalogers(cataloger.DefaultConfig()) {
 		// in case of future alteration where state is persisted, assume no dependency is safe to reuse
 		userInput := "docker-archive:" + tarPath
 		sourceInput, err := source.ParseInput(userInput, "", false)