Remove internal string set (#2219)

* remove internal string set Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> * incorporate changes from #2227 Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> * beef up the pkg.License.Merg() doc string Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> --------- Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>
anchore · Oct 17, 2023 · 7018573 · 7018573
1 parent f3ad8cf
commit 7018573
Show file tree

Hide file tree

Showing 24 changed files with 333 additions and 150 deletions.
diff --git a/cmd/syft/cli/options/catalog.go b/cmd/syft/cli/options/catalog.go
@@ -7,10 +7,10 @@ import (
 
 	"github.com/iancoleman/strcase"
 	"github.com/mitchellh/go-homedir"
+	"github.com/scylladb/go-set/strset"
 
 	"github.com/anchore/clio"
 	"github.com/anchore/fangs"
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/pkg/cataloger"
 	golangCataloger "github.com/anchore/syft/syft/pkg/cataloger/golang"
@@ -147,8 +147,8 @@ func (cfg Catalog) ToCatalogerConfig() cataloger.Config {
 var validDefaultSourceValues = []string{"registry", "docker", "podman", ""}
 
 func checkDefaultSourceValues(source string) error {
-	validValues := internal.NewStringSet(validDefaultSourceValues...)
-	if !validValues.Contains(source) {
+	validValues := strset.New(validDefaultSourceValues...)
+	if !validValues.Has(source) {
 		validValuesString := strings.Join(validDefaultSourceValues, ", ")
 		return fmt.Errorf("%s is not a valid default source; please use one of the following: %s''", source, validValuesString)
 	}

diff --git a/internal/file/zip_file_manifest.go b/internal/file/zip_file_manifest.go
@@ -6,7 +6,8 @@ import (
 	"sort"
 	"strings"
 
-	"github.com/anchore/syft/internal"
+	"github.com/scylladb/go-set/strset"
+
 	"github.com/anchore/syft/internal/log"
 )
 
@@ -40,7 +41,7 @@ func (z ZipFileManifest) Add(entry string, info os.FileInfo) {
 
 // GlobMatch returns the path keys that match the given value(s).
 func (z ZipFileManifest) GlobMatch(patterns ...string) []string {
-	uniqueMatches := internal.NewStringSet()
+	uniqueMatches := strset.New()
 
 	for _, pattern := range patterns {
 		for entry := range z {
@@ -54,7 +55,7 @@ func (z ZipFileManifest) GlobMatch(patterns ...string) []string {
 		}
 	}
 
-	results := uniqueMatches.ToSlice()
+	results := uniqueMatches.List()
 	sort.Strings(results)
 
 	return results

diff --git a/internal/licenses/list.go b/internal/licenses/list.go
@@ -1,20 +1,21 @@
 package licenses
 
-import "github.com/anchore/syft/internal"
-
 // all of these taken from https://github.com/golang/pkgsite/blob/8996ff632abee854aef1b764ca0501f262f8f523/internal/licenses/licenses.go#L338
 // which unfortunately is not exported. But fortunately is under BSD-style license. Take note that this list has
 // been manually updated to include more license filenames (see https://github.com/anchore/syft/pull/2227).
 
-var (
-	FileNames = []string{
+func FileNames() []string {
+	return []string{
+		"AL2.0",
 		"COPYING",
 		"COPYING.md",
 		"COPYING.markdown",
 		"COPYING.txt",
+		"LGPL2.1",
 		"LICENCE",
 		"LICENCE.md",
 		"LICENCE.markdown",
+		"licence.txt",
 		"LICENCE.txt",
 		"LICENSE",
 		"LICENSE.md",
@@ -49,9 +50,5 @@ var (
 		"MIT_LICENCE",
 		"UNLICENSE",
 		"UNLICENCE",
-		"AL2.0",
-		"LGPL2.1",
 	}
-
-	FileNameSet = internal.NewStringSet(FileNames...)
-)
+}
diff --git a/syft/formats/common/cyclonedxhelpers/licenses.go b/syft/formats/common/cyclonedxhelpers/licenses.go
@@ -95,7 +95,7 @@ func separateLicenses(p pkg.Package) (spdx, other cyclonedx.Licenses, expression
 		// singular expression case
 		// only ID field here since we guarantee that the license is valid
 		if value, exists := spdxlicense.ID(l.SPDXExpression); exists {
-			if !l.URLs.Empty() {
+			if len(l.URLs) > 0 {
 				processLicenseURLs(l, value, &spdxc)
 				continue
 			}
@@ -123,7 +123,7 @@ func separateLicenses(p pkg.Package) (spdx, other cyclonedx.Licenses, expression
 
 		// license string that are not valid spdx expressions or ids
 		// we only use license Name here since we cannot guarantee that the license is a valid SPDX expression
-		if !l.URLs.Empty() {
+		if len(l.URLs) > 0 {
 			processLicenseURLs(l, "", &otherc)
 			continue
 		}
@@ -137,7 +137,7 @@ func separateLicenses(p pkg.Package) (spdx, other cyclonedx.Licenses, expression
 }
 
 func processLicenseURLs(l pkg.License, spdxID string, populate *cyclonedx.Licenses) {
-	for _, url := range l.URLs.ToSlice() {
+	for _, url := range l.URLs {
 		if spdxID == "" {
 			*populate = append(*populate, cyclonedx.LicenseChoice{
 				License: &cyclonedx.License{

diff --git a/syft/formats/common/cyclonedxhelpers/licenses_test.go b/syft/formats/common/cyclonedxhelpers/licenses_test.go
@@ -4,9 +4,9 @@ import (
 	"testing"
 
 	"github.com/CycloneDX/cyclonedx-go"
+	"github.com/google/go-cmp/cmp"
 	"github.com/stretchr/testify/assert"
 
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/syft/license"
 	"github.com/anchore/syft/syft/pkg"
 )
@@ -191,7 +191,9 @@ func Test_encodeLicense(t *testing.T) {
 	}
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
-			assert.Equal(t, test.expected, encodeLicenses(test.input))
+			if d := cmp.Diff(test.expected, encodeLicenses(test.input)); d != "" {
+				t.Errorf("unexpected license (-want +got):\n%s", d)
+			}
 		})
 	}
 }
@@ -223,7 +225,7 @@ func TestDecodeLicenses(t *testing.T) {
 					Value: "RandomLicense",
 					// CycloneDX specification doesn't give a field for determining the license type
 					Type: license.Declared,
-					URLs: internal.NewStringSet(),
+					URLs: []string{},
 				},
 			},
 		},
@@ -243,7 +245,7 @@ func TestDecodeLicenses(t *testing.T) {
 					Value:          "MIT",
 					SPDXExpression: "MIT",
 					Type:           license.Declared,
-					URLs:           internal.NewStringSet(),
+					URLs:           []string{},
 				},
 			},
 		},
@@ -262,7 +264,7 @@ func TestDecodeLicenses(t *testing.T) {
 					Value:          "MIT AND GPL-3.0-only WITH Classpath-exception-2.0",
 					SPDXExpression: "MIT AND GPL-3.0-only WITH Classpath-exception-2.0",
 					Type:           license.Declared,
-					URLs:           internal.NewStringSet(),
+					URLs:           []string{},
 				},
 			},
 		},

diff --git a/syft/formats/syftjson/decoder_test.go b/syft/formats/syftjson/decoder_test.go
@@ -12,7 +12,6 @@ import (
 	"github.com/stretchr/testify/require"
 
 	stereoscopeFile "github.com/anchore/stereoscope/pkg/file"
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/syft/cpe"
 	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/formats/internal/testutils"
@@ -118,7 +117,7 @@ func Test_encodeDecodeFileMetadata(t *testing.T) {
 			Value:          "MIT",
 			SPDXExpression: "MIT",
 			Type:           "MIT",
-			URLs:           internal.NewStringSet("https://example.org/license"),
+			URLs:           []string{"https://example.org/license"},
 			Locations:      file.LocationSet{},
 		}),
 		Language: "language",

diff --git a/syft/formats/syftjson/to_format_model.go b/syft/formats/syftjson/to_format_model.go
@@ -214,11 +214,18 @@ func toLicenseModel(pkgLicenses []pkg.License) (modelLicenses []model.License) {
 		if v := l.Locations.ToSlice(); v != nil {
 			locations = v
 		}
+
+		// format model must have allocated collections
+		urls := l.URLs
+		if urls == nil {
+			urls = []string{}
+		}
+
 		modelLicenses = append(modelLicenses, model.License{
 			Value:          l.Value,
 			SPDXExpression: l.SPDXExpression,
 			Type:           l.Type,
-			URLs:           l.URLs.ToSlice(),
+			URLs:           urls,
 			Locations:      locations,
 		})
 	}

diff --git a/syft/formats/syftjson/to_syft_model.go b/syft/formats/syftjson/to_syft_model.go
@@ -10,7 +10,6 @@ import (
 	"github.com/google/go-cmp/cmp"
 
 	stereoscopeFile "github.com/anchore/stereoscope/pkg/file"
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/cpe"
@@ -136,7 +135,7 @@ func toSyftLicenses(m []model.License) (p []pkg.License) {
 			Value:          l.Value,
 			SPDXExpression: l.SPDXExpression,
 			Type:           l.Type,
-			URLs:           internal.NewStringSet(l.URLs...),
+			URLs:           l.URLs,
 			Locations:      file.NewLocationSet(l.Locations...),
 		})
 	}

diff --git a/syft/pkg/catalog.go b/syft/pkg/catalog.go
@@ -4,8 +4,8 @@ import (
 	"sync"
 
 	"github.com/jinzhu/copier"
+	"github.com/scylladb/go-set/strset"
 
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/artifact"
 )
@@ -124,13 +124,13 @@ func (c *Collection) addTypeToIndex(p Package) {
 }
 
 func (c *Collection) addPathsToIndex(p Package) {
-	observedPaths := internal.NewStringSet()
+	observedPaths := strset.New()
 	for _, l := range p.Locations.ToSlice() {
-		if l.RealPath != "" && !observedPaths.Contains(l.RealPath) {
+		if l.RealPath != "" && !observedPaths.Has(l.RealPath) {
 			c.addPathToIndex(p.id, l.RealPath)
 			observedPaths.Add(l.RealPath)
 		}
-		if l.VirtualPath != "" && l.RealPath != l.VirtualPath && !observedPaths.Contains(l.VirtualPath) {
+		if l.VirtualPath != "" && l.RealPath != l.VirtualPath && !observedPaths.Has(l.VirtualPath) {
 			c.addPathToIndex(p.id, l.VirtualPath)
 			observedPaths.Add(l.VirtualPath)
 		}
@@ -173,13 +173,13 @@ func (c *Collection) deleteTypeFromIndex(p Package) {
 }
 
 func (c *Collection) deletePathsFromIndex(p Package) {
-	observedPaths := internal.NewStringSet()
+	observedPaths := strset.New()
 	for _, l := range p.Locations.ToSlice() {
-		if l.RealPath != "" && !observedPaths.Contains(l.RealPath) {
+		if l.RealPath != "" && !observedPaths.Has(l.RealPath) {
 			c.deletePathFromIndex(p.id, l.RealPath)
 			observedPaths.Add(l.RealPath)
 		}
-		if l.VirtualPath != "" && l.RealPath != l.VirtualPath && !observedPaths.Contains(l.VirtualPath) {
+		if l.VirtualPath != "" && l.RealPath != l.VirtualPath && !observedPaths.Has(l.VirtualPath) {
 			c.deletePathFromIndex(p.id, l.VirtualPath)
 			observedPaths.Add(l.VirtualPath)
 		}

diff --git a/syft/pkg/cataloger/common/cpe/generate.go b/syft/pkg/cataloger/common/cpe/generate.go
@@ -13,7 +13,6 @@ import (
 	"github.com/facebookincubator/nvdtools/wfn"
 	"github.com/scylladb/go-set/strset"
 
-	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/cpe"
 	"github.com/anchore/syft/syft/pkg"
@@ -118,13 +117,13 @@ func Generate(p pkg.Package) []cpe.CPE {
 		return nil
 	}
 
-	keys := internal.NewStringSet()
+	keys := strset.New()
 	cpes := make([]cpe.CPE, 0)
 	for _, product := range products {
 		for _, vendor := range vendors {
 			// prevent duplicate entries...
 			key := fmt.Sprintf("%s|%s|%s", product, vendor, p.Version)
-			if keys.Contains(key) {
+			if keys.Has(key) {
 				continue
 			}
 			keys.Add(key)

diff --git a/syft/pkg/cataloger/deb/parse_copyright.go b/syft/pkg/cataloger/deb/parse_copyright.go
@@ -7,6 +7,8 @@ import (
 	"sort"
 	"strings"
 
+	"github.com/scylladb/go-set/strset"
+
 	"github.com/anchore/syft/internal"
 )
 
@@ -18,7 +20,7 @@ var (
 )
 
 func parseLicensesFromCopyright(reader io.Reader) []string {
-	findings := internal.NewStringSet()
+	findings := strset.New()
 	scanner := bufio.NewScanner(reader)
 
 	for scanner.Scan() {
@@ -31,7 +33,7 @@ func parseLicensesFromCopyright(reader io.Reader) []string {
 		}
 	}
 
-	results := findings.ToSlice()
+	results := findings.List()
 
 	sort.Strings(results)