Add relationships for ALPM packages (arch linux) (#2851)

* add alpm relationships Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> * tweak reader linter rule to check for reader impl Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> * update JSON schema with alpm dependency information Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com> --------- Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>
2025-11-17 08:23:15 +01:00 · 2024-05-07 13:29:46 -04:00 · 2024-05-07 13:29:46 -04:00 · ada8f009d2
commit ada8f009d2
parent e7b6284039
18 changed files with 3078 additions and 266 deletions
--- a/internal/constants.go
+++ b/internal/constants.go
@ -3,5 +3,5 @@ package internal
 const (
 	// JSONSchemaVersion is the current schema version output by the JSON encoder
 	// This is roughly following the "SchemaVer" guidelines for versioning the JSON schema. Please see schema/json/README.md for details on how to increment.
-	JSONSchemaVersion = "16.0.7"
+	JSONSchemaVersion = "16.0.8"
 )
--- a/schema/json/schema-16.0.8.json
+++ b/schema/json/schema-16.0.8.json
--- a/schema/json/schema-latest.json
+++ b/schema/json/schema-latest.json
@ -1,6 +1,6 @@
 {
  "$schema": "https://json-schema.org/draft/2020-12/schema",
-  "$id": "anchore.io/schema/syft/json/16.0.7/document",
+  "$id": "anchore.io/schema/syft/json/16.0.8/document",
  "$ref": "#/$defs/Document",
  "$defs": {
    "AlpmDbEntry": {
@ -46,6 +46,18 @@
            "$ref": "#/$defs/AlpmFileRecord"
          },
          "type": "array"
+        },
+        "provides": {
+          "items": {
+            "type": "string"
+          },
+          "type": "array"
+        },
+        "depends": {
+          "items": {
+            "type": "string"
+          },
+          "type": "array"
        }
      },
      "type": "object",
--- a/syft/pkg/alpm.go
+++ b/syft/pkg/alpm.go
@ -27,6 +27,8 @@ type AlpmDBEntry struct {
 	Reason       int              `mapstructure:"reason" json:"reason"`
 	Files        []AlpmFileRecord `mapstructure:"files" json:"files"`
 	Backup       []AlpmFileRecord `mapstructure:"backup" json:"backup"`
+	Provides     []string         `mapstructure:"provides" json:"provides,omitempty"`
+	Depends      []string         `mapstructure:"depends" json:"depends,omitempty"`
 }

 type AlpmFileRecord struct {
--- a/syft/pkg/cataloger/arch/cataloger.go
+++ b/syft/pkg/cataloger/arch/cataloger.go
@ -4,12 +4,96 @@ Package arch provides a concrete Cataloger implementations for packages relating
 package arch

 import (
+	"context"
+	"strings"
+
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/pkg"
 	"github.com/anchore/syft/syft/pkg/cataloger/generic"
 )

+type cataloger struct {
+	*generic.Cataloger
+}
+
 // NewDBCataloger returns a new cataloger object initialized for arch linux pacman database flat-file stores.
 func NewDBCataloger() pkg.Cataloger {
-	return generic.NewCataloger("alpm-db-cataloger").
-		WithParserByGlobs(parseAlpmDB, pkg.AlpmDBGlob)
+	return cataloger{
+		Cataloger: generic.NewCataloger("alpm-db-cataloger").
+			WithParserByGlobs(parseAlpmDB, pkg.AlpmDBGlob),
+	}
+}
+
+func (c cataloger) Catalog(ctx context.Context, resolver file.Resolver) ([]pkg.Package, []artifact.Relationship, error) {
+	pkgs, rels, err := c.Cataloger.Catalog(ctx, resolver)
+	if err != nil {
+		return nil, nil, err
+	}
+
+	rels = append(rels, associateRelationships(pkgs)...)
+
+	return pkgs, rels, nil
+}
+
+// associateRelationships will create relationships between packages based on the "Depends" and "Provides"
+// fields for installed packages. If there is an installed package that has a dependency that is (somehow) not installed,
+// then that relationship (between the installed and uninstalled package) will NOT be created.
+func associateRelationships(pkgs []pkg.Package) (relationships []artifact.Relationship) {
+	// map["provides" + "package"] -> packages that provide that package
+	lookup := make(map[string][]pkg.Package)
+
+	// read providers and add lookup keys as needed
+	for _, p := range pkgs {
+		meta, ok := p.Metadata.(pkg.AlpmDBEntry)
+		if !ok {
+			log.Warnf("cataloger failed to extract alpm 'provides' metadata for package %+v", p.Name)
+			continue
+		}
+		// allow for lookup by package name
+		lookup[p.Name] = append(lookup[p.Name], p)
+
+		for _, provides := range meta.Provides {
+			// allow for lookup by exact specification
+			lookup[provides] = append(lookup[provides], p)
+
+			// allow for lookup by library name only
+			k := stripVersionSpecifier(provides)
+			lookup[k] = append(lookup[k], p)
+		}
+	}
+
+	// read "Depends" and match with provider keys
+	for _, p := range pkgs {
+		meta, ok := p.Metadata.(pkg.AlpmDBEntry)
+		if !ok {
+			log.Warnf("cataloger failed to extract alpm 'dependency' metadata for package %+v", p.Name)
+			continue
+		}
+
+		for _, dep := range meta.Depends {
+			for _, depPkg := range lookup[dep] {
+				relationships = append(relationships, artifact.Relationship{
+					From: depPkg,
+					To:   p,
+					Type: artifact.DependencyOfRelationship,
+				})
+			}
+		}
+	}
+	return relationships
+}
+
+func stripVersionSpecifier(s string) string {
+	// examples:
+	// gcc-libs                  -->  gcc-libs
+	// libtree-sitter.so=0-64    -->  libtree-sitter.so
+
+	items := strings.Split(s, "=")
+	if len(items) == 0 {
+		return s
+	}
+
+	return strings.TrimSpace(items[0])
 }
--- a/syft/pkg/cataloger/arch/cataloger_test.go
+++ b/syft/pkg/cataloger/arch/cataloger_test.go
@ -12,20 +12,117 @@ import (
 )

 func TestAlpmCataloger(t *testing.T) {
-	dbLocation := file.NewLocation("var/lib/pacman/local/gmp-6.2.1-2/desc")
-	expectedPkgs := []pkg.Package{
+	gmpDbLocation := file.NewLocation("var/lib/pacman/local/gmp-6.2.1-2/desc")
+	treeSitterDbLocation := file.NewLocation("var/lib/pacman/local/tree-sitter-0.22.6-1/desc")
+	emacsDbLocation := file.NewLocation("var/lib/pacman/local/emacs-29.3-3/desc")
+	fuzzyDbLocation := file.NewLocation("var/lib/pacman/local/fuzzy-1.2-3/desc")
+	madeupDbLocation := file.NewLocation("var/lib/pacman/local/madeup-20.30-4/desc")
+
+	treeSitterPkg := pkg.Package{
+		Name:    "tree-sitter",
+		Version: "0.22.6-1",
+		Type:    pkg.AlpmPkg,
+		FoundBy: "alpm-db-cataloger",
+		Licenses: pkg.NewLicenseSet(
+			pkg.NewLicenseFromLocations("MIT", treeSitterDbLocation),
+		),
+		Locations: file.NewLocationSet(treeSitterDbLocation),
+		Metadata: pkg.AlpmDBEntry{
+			BasePackage:  "tree-sitter",
+			Package:      "tree-sitter",
+			Version:      "0.22.6-1",
+			Description:  "Incremental parsing library",
+			Architecture: "x86_64",
+			Size:         223539,
+			Packager:     "Daniel M. Capella <polyzen@archlinux.org>",
+			URL:          "https://github.com/tree-sitter/tree-sitter",
+			Validation:   "pgp",
+			Reason:       1,
+			Files:        []pkg.AlpmFileRecord{},
+			Backup:       []pkg.AlpmFileRecord{},
+			Provides:     []string{"libtree-sitter.so=0-64"},
+		},
+	}
+
+	emacsPkg := pkg.Package{
+		Name:    "emacs",
+		Version: "29.3-3",
+		Type:    pkg.AlpmPkg,
+		FoundBy: "alpm-db-cataloger",
+		Licenses: pkg.NewLicenseSet(
+			pkg.NewLicenseFromLocations("GPL3", emacsDbLocation),
+		),
+		Locations: file.NewLocationSet(emacsDbLocation),
+		Metadata: pkg.AlpmDBEntry{
+			BasePackage:  "emacs",
+			Package:      "emacs",
+			Version:      "29.3-3",
+			Description:  "The extensible, customizable, self-documenting real-time display editor",
+			Architecture: "x86_64",
+			Size:         126427862,
+			Packager:     "Frederik Schwan <freswa@archlinux.org>",
+			URL:          "https://www.gnu.org/software/emacs/emacs.html",
+			Validation:   "pgp",
+			Files:        []pkg.AlpmFileRecord{},
+			Backup:       []pkg.AlpmFileRecord{},
+			Depends:      []string{"libtree-sitter.so=0-64"},
+		},
+	}
+
+	fuzzyPkg := pkg.Package{
+		Name:    "fuzzy",
+		Version: "1.2-3",
+		Type:    pkg.AlpmPkg,
+		FoundBy: "alpm-db-cataloger",
+		Locations: file.NewLocationSet(
+			fuzzyDbLocation,
+			file.NewLocation("var/lib/pacman/local/fuzzy-1.2-3/files"),
+		),
+		Metadata: pkg.AlpmDBEntry{
+			Package: "fuzzy",
+			Version: "1.2-3",
+			Files:   []pkg.AlpmFileRecord{},
+			Backup: []pkg.AlpmFileRecord{
 				{
+					Path: "/etc/fuzzy.conf",
+					Digests: []file.Digest{
+						{Algorithm: "md5", Value: "79fce043df7dfc676ae5ecb903762d8b"},
+					},
+				},
+			},
+			Depends: []string{"tree-sitter"},
+		},
+	}
+
+	madeupPkg := pkg.Package{
+		Name:      "madeup",
+		Version:   "20.30-4",
+		Type:      pkg.AlpmPkg,
+		FoundBy:   "alpm-db-cataloger",
+		Locations: file.NewLocationSet(madeupDbLocation),
+		Metadata: pkg.AlpmDBEntry{
+			Package: "madeup",
+			Version: "20.30-4",
+			Files:   []pkg.AlpmFileRecord{},
+			Backup:  []pkg.AlpmFileRecord{},
+			Depends: []string{"libtree-sitter.so"},
+		},
+	}
+
+	gmpPkg := pkg.Package{
 		Name:    "gmp",
 		Version: "6.2.1-2",
 		Type:    pkg.AlpmPkg,
 		FoundBy: "alpm-db-cataloger",
 		Licenses: pkg.NewLicenseSet(
-				pkg.NewLicenseFromLocations("LGPL3", dbLocation),
-				pkg.NewLicenseFromLocations("GPL", dbLocation),
+			pkg.NewLicenseFromLocations("LGPL3", gmpDbLocation),
+			pkg.NewLicenseFromLocations("GPL", gmpDbLocation),
+		),
+		Locations: file.NewLocationSet(
+			gmpDbLocation,
+			file.NewLocation("var/lib/pacman/local/gmp-6.2.1-2/files"),
+			file.NewLocation("var/lib/pacman/local/gmp-6.2.1-2/mtree"),
 		),
-			Locations: file.NewLocationSet(dbLocation),
-			CPEs:      nil,
-			PURL:      "",
 		Metadata: pkg.AlpmDBEntry{
 			BasePackage:  "gmp",
 			Package:      "gmp",
@ -37,6 +134,7 @@ func TestAlpmCataloger(t *testing.T) {
 			URL:          "https://gmplib.org/",
 			Validation:   "pgp",
 			Reason:       1,
+			Depends:      []string{"gcc-libs", "sh", "libtree-sitter.so=1-64"},
 			Files: []pkg.AlpmFileRecord{
 				{
 					Path:    "/usr",
@ -167,14 +265,36 @@ func TestAlpmCataloger(t *testing.T) {
 			},
 			Backup: []pkg.AlpmFileRecord{},
 		},
+	}
+
+	expectedPkgs := []pkg.Package{
+		treeSitterPkg,
+		emacsPkg,
+		fuzzyPkg,
+		madeupPkg,
+		gmpPkg,
+	}
+
+	expectedRelationships := []artifact.Relationship{
+		{ // exact spec lookup
+			From: treeSitterPkg,
+			To:   emacsPkg,
+			Type: artifact.DependencyOfRelationship,
+		},
+		{ // package name lookup
+			From: treeSitterPkg,
+			To:   fuzzyPkg,
+			Type: artifact.DependencyOfRelationship,
+		},
+		{ // library name lookup
+			From: treeSitterPkg,
+			To:   madeupPkg,
+			Type: artifact.DependencyOfRelationship,
 		},
 	}

-	// TODO: relationships are not under test yet
-	var expectedRelationships []artifact.Relationship
-
 	pkgtest.NewCatalogTester().
-		FromDirectory(t, "test-fixtures/gmp-fixture").
+		FromDirectory(t, "test-fixtures/installed").
 		WithCompareOptions(cmpopts.IgnoreFields(pkg.AlpmFileRecord{}, "Time")).
 		Expects(expectedPkgs, expectedRelationships).
 		TestCataloger(t, NewDBCataloger())
--- a/syft/pkg/cataloger/arch/package.go
+++ b/syft/pkg/cataloger/arch/package.go
@ -9,13 +9,16 @@ import (
 	"github.com/anchore/syft/syft/pkg"
 )

-func newPackage(m *parsedData, release *linux.Release, dbLocation file.Location) pkg.Package {
+func newPackage(m *parsedData, release *linux.Release, dbLocation file.Location, otherLocations ...file.Location) pkg.Package {
 	licenseCandidates := strings.Split(m.Licenses, "\n")

+	locs := file.NewLocationSet(dbLocation)
+	locs.Add(otherLocations...)
+
 	p := pkg.Package{
 		Name:      m.Package,
 		Version:   m.Version,
-		Locations: file.NewLocationSet(dbLocation),
+		Locations: locs,
 		Licenses:  pkg.NewLicenseSet(pkg.NewLicensesFromLocation(dbLocation.WithoutAnnotations(), licenseCandidates...)...),
 		Type:      pkg.AlpmPkg,
 		PURL:      packageURL(m, release),
--- a/syft/pkg/cataloger/arch/parse_alpm_db.go
+++ b/syft/pkg/cataloger/arch/parse_alpm_db.go
@ -6,6 +6,7 @@ import (
 	"context"
 	"fmt"
 	"io"
+	"path"
 	"path/filepath"
 	"strconv"
 	"strings"
@ -15,6 +16,7 @@ import (
 	"github.com/vbatts/go-mtree"

 	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/artifact"
 	"github.com/anchore/syft/syft/file"
 	"github.com/anchore/syft/syft/pkg"
@ -44,33 +46,26 @@ func parseAlpmDB(_ context.Context, resolver file.Resolver, env *generic.Environ
 		return nil, nil, err
 	}

-	base := filepath.Dir(reader.RealPath)
-	r, err := getFileReader(filepath.Join(base, "mtree"), resolver)
-	if err != nil {
-		return nil, nil, err
+	if data == nil {
+		return nil, nil, nil
 	}

-	pkgFiles, err := parseMtree(r)
-	if err != nil {
-		return nil, nil, err
-	}
+	base := path.Dir(reader.RealPath)

 	// replace the files found the pacman database with the files from the mtree These contain more metadata and
 	// thus more useful.
-	// TODO: probably want to use MTREE and PKGINFO here
-	data.Files = pkgFiles
+	files, fileLoc := fetchPkgFiles(base, resolver)
+	backups, backupLoc := fetchBackupFiles(base, resolver)

-	// We only really do this to get any backup database entries from the files database
-	files := filepath.Join(base, "files")
-	_, err = getFileReader(files, resolver)
-	if err != nil {
-		return nil, nil, err
+	var locs []file.Location
+	if fileLoc != nil {
+		locs = append(locs, fileLoc.WithAnnotation(pkg.EvidenceAnnotationKey, pkg.SupportingEvidenceAnnotation))
+		data.Files = files
 	}
-	filesMetadata, err := parseAlpmDBEntry(reader)
-	if err != nil {
-		return nil, nil, err
-	} else if filesMetadata != nil {
-		data.Backup = filesMetadata.Backup
+
+	if backupLoc != nil {
+		locs = append(locs, backupLoc.WithAnnotation(pkg.EvidenceAnnotationKey, pkg.SupportingEvidenceAnnotation))
+		data.Backup = backups
 	}

 	if data.Package == "" {
@ -82,10 +77,67 @@ func parseAlpmDB(_ context.Context, resolver file.Resolver, env *generic.Environ
 			data,
 			env.LinuxRelease,
 			reader.Location.WithAnnotation(pkg.EvidenceAnnotationKey, pkg.PrimaryEvidenceAnnotation),
+			locs...,
 		),
 	}, nil, nil
 }

+func fetchPkgFiles(base string, resolver file.Resolver) ([]pkg.AlpmFileRecord, *file.Location) {
+	// TODO: probably want to use MTREE and PKGINFO here
+	target := path.Join(base, "mtree")
+
+	loc, err := getLocation(target, resolver)
+	if err != nil {
+		log.WithFields("error", err, "path", target).Trace("failed to find mtree file")
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	if loc == nil {
+		return []pkg.AlpmFileRecord{}, nil
+	}
+
+	reader, err := resolver.FileContentsByLocation(*loc)
+	if err != nil {
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	defer internal.CloseAndLogError(reader, loc.RealPath)
+
+	pkgFiles, err := parseMtree(reader)
+	if err != nil {
+		log.WithFields("error", err, "path", target).Trace("failed to parse mtree file")
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	return pkgFiles, loc
+}
+
+func fetchBackupFiles(base string, resolver file.Resolver) ([]pkg.AlpmFileRecord, *file.Location) {
+	// We only really do this to get any backup database entries from the files database
+	target := filepath.Join(base, "files")
+
+	loc, err := getLocation(target, resolver)
+	if err != nil {
+		log.WithFields("error", err, "path", target).Trace("failed to find alpm files")
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	if loc == nil {
+		return []pkg.AlpmFileRecord{}, nil
+	}
+
+	reader, err := resolver.FileContentsByLocation(*loc)
+	if err != nil {
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	defer internal.CloseAndLogError(reader, loc.RealPath)
+
+	filesMetadata, err := parseAlpmDBEntry(reader)
+	if err != nil {
+		return []pkg.AlpmFileRecord{}, nil
+	}
+	if filesMetadata != nil {
+		return filesMetadata.Backup, loc
+	}
+	return []pkg.AlpmFileRecord{}, loc
+}
+
 func parseAlpmDBEntry(reader io.Reader) (*parsedData, error) {
 	scanner := newScanner(reader)
 	metadata, err := parseDatabase(scanner)
@ -119,7 +171,7 @@ func newScanner(reader io.Reader) *bufio.Scanner {
 	return scanner
 }

-func getFileReader(path string, resolver file.Resolver) (io.Reader, error) {
+func getLocation(path string, resolver file.Resolver) (*file.Location, error) {
 	locs, err := resolver.FilesByPath(path)
 	if err != nil {
 		return nil, err
@ -128,13 +180,11 @@ func getFileReader(path string, resolver file.Resolver) (io.Reader, error) {
 	if len(locs) == 0 {
 		return nil, fmt.Errorf("could not find file: %s", path)
 	}
-	// TODO: Should we maybe check if we found the file
-	dbContentReader, err := resolver.FileContentsByLocation(locs[0])
-	if err != nil {
-		return nil, err
+
+	if len(locs) > 1 {
+		log.WithFields("path", path).Trace("multiple files found for path, using first path")
 	}
-	defer internal.CloseAndLogError(dbContentReader, locs[0].RealPath)
-	return dbContentReader, nil
+	return &locs[0], nil
 }

 func parseDatabase(b *bufio.Scanner) (*parsedData, error) {
@ -157,9 +207,9 @@ func parseDatabase(b *bufio.Scanner) (*parsedData, error) {
 		case "files":
 			var files []map[string]string
 			for _, f := range strings.Split(value, "\n") {
-				path := fmt.Sprintf("/%s", f)
-				if ok := ignoredFiles[path]; !ok {
-					files = append(files, map[string]string{"path": path})
+				p := fmt.Sprintf("/%s", f)
+				if ok := ignoredFiles[p]; !ok {
+					files = append(files, map[string]string{"path": p})
 				}
 			}
 			pkgFields[key] = files
@ -167,10 +217,10 @@ func parseDatabase(b *bufio.Scanner) (*parsedData, error) {
 			var backup []map[string]interface{}
 			for _, f := range strings.Split(value, "\n") {
 				fields := strings.SplitN(f, "\t", 2)
-				path := fmt.Sprintf("/%s", fields[0])
-				if ok := ignoredFiles[path]; !ok {
+				p := fmt.Sprintf("/%s", fields[0])
+				if ok := ignoredFiles[p]; !ok {
 					backup = append(backup, map[string]interface{}{
-						"path": path,
+						"path": p,
 						"digests": []file.Digest{{
 							Algorithm: "md5",
 							Value:     fields[1],
@ -178,6 +228,8 @@ func parseDatabase(b *bufio.Scanner) (*parsedData, error) {
 				}
 			}
 			pkgFields[key] = backup
+		case "depends", "provides":
+			pkgFields[key] = processLibrarySpecs(value)
 		case "reason":
 			fallthrough
 		case "size":
@ -193,6 +245,19 @@ func parseDatabase(b *bufio.Scanner) (*parsedData, error) {
 	return parsePkgFiles(pkgFields)
 }

+func processLibrarySpecs(value string) []string {
+	lines := strings.Split(value, "\n")
+	librarySpecs := make([]string, 0)
+	for _, line := range lines {
+		line = strings.TrimSpace(line)
+		if line == "" {
+			continue
+		}
+		librarySpecs = append(librarySpecs, line)
+	}
+	return librarySpecs
+}
+
 func parsePkgFiles(pkgFields map[string]interface{}) (*parsedData, error) {
 	var entry parsedData
 	if err := mapstructure.Decode(pkgFields, &entry); err != nil {
@ -203,6 +268,10 @@ func parsePkgFiles(pkgFields map[string]interface{}) (*parsedData, error) {
 		entry.Backup = make([]pkg.AlpmFileRecord, 0)
 	}

+	if entry.Files == nil {
+		entry.Files = make([]pkg.AlpmFileRecord, 0)
+	}
+
 	if entry.Package == "" && len(entry.Files) == 0 && len(entry.Backup) == 0 {
 		return nil, nil
 	}
--- a/syft/pkg/cataloger/arch/parse_alpm_db_test.go
+++ b/syft/pkg/cataloger/arch/parse_alpm_db_test.go
@ -17,12 +17,13 @@ func TestDatabaseParser(t *testing.T) {
 	tests := []struct {
 		name     string
 		fixture  string
-		expected pkg.AlpmDBEntry
+		expected *parsedData
 	}{
 		{
-			name:    "test alpm database parsing",
+			name:    "simple desc parsing",
 			fixture: "test-fixtures/files",
-			expected: pkg.AlpmDBEntry{
+			expected: &parsedData{
+				AlpmDBEntry: pkg.AlpmDBEntry{
 					Backup: []pkg.AlpmFileRecord{
 						{
 							Path: "/etc/pacman.conf",
@ -88,6 +89,51 @@ func TestDatabaseParser(t *testing.T) {
 					},
 				},
 			},
+		},
+		{
+			name:    "with dependencies",
+			fixture: "test-fixtures/installed/var/lib/pacman/local/gmp-6.2.1-2/desc",
+			expected: &parsedData{
+				Licenses: "LGPL3\nGPL",
+				AlpmDBEntry: pkg.AlpmDBEntry{
+					BasePackage:  "gmp",
+					Package:      "gmp",
+					Version:      "6.2.1-2",
+					Description:  "A free library for arbitrary precision arithmetic",
+					Architecture: "x86_64",
+					Size:         1044438,
+					Packager:     "Antonio Rojas <arojas@archlinux.org>",
+					URL:          "https://gmplib.org/",
+					Validation:   "pgp",
+					Reason:       1,
+					Files:        []pkg.AlpmFileRecord{},
+					Backup:       []pkg.AlpmFileRecord{},
+					Depends:      []string{"gcc-libs", "sh", "libtree-sitter.so=1-64"},
+				},
+			},
+		},
+		{
+			name:    "with provides",
+			fixture: "test-fixtures/installed/var/lib/pacman/local/tree-sitter-0.22.6-1/desc",
+			expected: &parsedData{
+				Licenses: "MIT",
+				AlpmDBEntry: pkg.AlpmDBEntry{
+					BasePackage:  "tree-sitter",
+					Package:      "tree-sitter",
+					Version:      "0.22.6-1",
+					Description:  "Incremental parsing library",
+					Architecture: "x86_64",
+					Size:         223539,
+					Packager:     "Daniel M. Capella <polyzen@archlinux.org>",
+					URL:          "https://github.com/tree-sitter/tree-sitter",
+					Validation:   "pgp",
+					Reason:       1,
+					Files:        []pkg.AlpmFileRecord{},
+					Backup:       []pkg.AlpmFileRecord{},
+					Provides:     []string{"libtree-sitter.so=0-64"},
+				},
+			},
+		},
 	}

 	for _, test := range tests {
@ -101,13 +147,10 @@ func TestDatabaseParser(t *testing.T) {
 			entry, err := parseAlpmDBEntry(reader)
 			require.NoError(t, err)

-			if diff := cmp.Diff(entry.Files, test.expected.Files); diff != "" {
-				t.Errorf("Files mismatch (-want +got):\n%s", diff)
+			if diff := cmp.Diff(test.expected, entry); diff != "" {
+				t.Errorf("parsed data mismatch (-want +got):\n%s", diff)
 			}

-			if diff := cmp.Diff(entry.Backup, test.expected.Backup); diff != "" {
-				t.Errorf("Backup mismatch (-want +got):\n%s", diff)
-			}
 		})
 	}
 }
--- a/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/emacs-29.3-3/desc
+++ b/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/emacs-29.3-3/desc
@ -0,0 +1,38 @@
+%NAME%
+emacs
+
+%VERSION%
+29.3-3
+
+%BASE%
+emacs
+
+%DESC%
+The extensible, customizable, self-documenting real-time display editor
+
+%URL%
+https://www.gnu.org/software/emacs/emacs.html
+
+%ARCH%
+x86_64
+
+%BUILDDATE%
+1714249917
+
+%INSTALLDATE%
+1715026363
+
+%PACKAGER%
+Frederik Schwan <freswa@archlinux.org>
+
+%SIZE%
+126427862
+
+%LICENSE%
+GPL3
+
+%VALIDATION%
+pgp
+
+%DEPENDS%
+libtree-sitter.so=0-64
--- a/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/fuzzy-1.2-3/desc
+++ b/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/fuzzy-1.2-3/desc
@ -0,0 +1,8 @@
+%NAME%
+fuzzy
+
+%VERSION%
+1.2-3
+
+%DEPENDS%
+tree-sitter
--- a/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/fuzzy-1.2-3/files
+++ b/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/fuzzy-1.2-3/files
@ -0,0 +1,6 @@
+%FILES%
+etc/
+etc/fuzzy.conf
+
+%BACKUP%
+etc/fuzzy.conf	79fce043df7dfc676ae5ecb903762d8b
--- a/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/desc
+++ b/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/desc
@ -41,4 +41,4 @@ pgp
 %DEPENDS%
 gcc-libs
 sh
-
+libtree-sitter.so=1-64
--- a/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/files
+++ b/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/files
--- a/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/mtree
+++ b/syft/pkg/cataloger/arch/test-fixtures/gmp-fixture/var/lib/pacman/local/gmp-6.2.1-2/mtree
--- a/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/madeup-20.30-4/desc
+++ b/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/madeup-20.30-4/desc
@ -0,0 +1,8 @@
+%NAME%
+madeup
+
+%VERSION%
+20.30-4
+
+%DEPENDS%
+libtree-sitter.so
--- a/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/tree-sitter-0.22.6-1/desc
+++ b/syft/pkg/cataloger/arch/test-fixtures/installed/var/lib/pacman/local/tree-sitter-0.22.6-1/desc
@ -0,0 +1,41 @@
+%NAME%
+tree-sitter
+
+%VERSION%
+0.22.6-1
+
+%BASE%
+tree-sitter
+
+%DESC%
+Incremental parsing library
+
+%URL%
+https://github.com/tree-sitter/tree-sitter
+
+%ARCH%
+x86_64
+
+%BUILDDATE%
+1714945746
+
+%INSTALLDATE%
+1715026360
+
+%PACKAGER%
+Daniel M. Capella <polyzen@archlinux.org>
+
+%SIZE%
+223539
+
+%REASON%
+1
+
+%LICENSE%
+MIT
+
+%VALIDATION%
+pgp
+
+%PROVIDES%
+libtree-sitter.so=0-64
--- a/test/rules/rules.go
+++ b/test/rules/rules.go
@ -8,8 +8,8 @@ import "github.com/quasilyte/go-ruleguard/dsl"
 func resourceCleanup(m dsl.Matcher) {
 	m.Match(`$res, $err := $resolver.FileContentsByLocation($loc); if $*_ { $*_ }; $next`).
 		Where(m["res"].Type.Implements(`io.Closer`) &&
+			m["res"].Type.Implements(`io.Reader`) &&
 			m["err"].Type.Implements(`error`) &&
-			m["res"].Type.Implements(`io.Closer`) &&
 			!m["next"].Text.Matches(`defer internal.CloseAndLogError`)).
 		Report(`please call "defer internal.CloseAndLogError($res, $loc.RealPath)" right after checking the error returned from $resolver.FileContentsByLocation.`)
 }