fix: better clean up of file handles (#2823)

* always close ELF cataloger file handles

The elf-binary-package-cataloger does its own file IO to account for the
possibility of a logical ELF package being broken across multiple
physical files. However, this casued it to skip the normal invocation
pattern in the generic cataloger code that prevented file leaks. Ensure
this cataloger always closes its file handles.

Signed-off-by: Will Murphy <will.murphy@anchore.com>

* defer closing of generic cataloger file handles

Otherwise, a panicking cataloger could leak file handles.

Signed-off-by: Will Murphy <will.murphy@anchore.com>

* add unit test for file closed on panic parser

Signed-off-by: Will Murphy <will.murphy@anchore.com>

* make invoke parser a static function

Signed-off-by: Will Murphy <will.murphy@anchore.com>

* push error logging down into invoke parser

Signed-off-by: Will Murphy <will.murphy@anchore.com>

---------

Signed-off-by: Will Murphy <will.murphy@anchore.com>

syft/pkg/cataloger/generic/cataloger.go

@@ -3,6 +3,7 @@ package generic
 import (
 	"context"
 
+	"github.com/anchore/go-logger"
 	"github.com/anchore/syft/internal"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/syft/artifact"
@@ -122,17 +123,9 @@ func (c *Cataloger) Catalog(ctx context.Context, resolver file.Resolver) ([]pkg.
 
 		log.WithFields("path", location.RealPath).Trace("parsing file contents")
 
-		contentReader, err := resolver.FileContentsByLocation(location)
+		discoveredPackages, discoveredRelationships, err := invokeParser(ctx, resolver, location, logger, parser, &env)
 		if err != nil {
-			logger.WithFields("location", location.RealPath, "error", err).Warn("unable to fetch contents")
-			continue
-		}
-
-		discoveredPackages, discoveredRelationships, err := parser(ctx, resolver, &env, file.NewLocationReadCloser(location, contentReader))
-		internal.CloseAndLogError(contentReader, location.AccessPath)
-		if err != nil {
-			logger.WithFields("location", location.RealPath, "error", err).Warnf("cataloger failed")
-			continue
+			continue // logging is handled within invokeParser
 		}
 
 		for _, p := range discoveredPackages {
@@ -145,6 +138,23 @@ func (c *Cataloger) Catalog(ctx context.Context, resolver file.Resolver) ([]pkg.
 	return packages, relationships, nil
 }
 
+func invokeParser(ctx context.Context, resolver file.Resolver, location file.Location, logger logger.Logger, parser Parser, env *Environment) ([]pkg.Package, []artifact.Relationship, error) {
+	contentReader, err := resolver.FileContentsByLocation(location)
+	if err != nil {
+		logger.WithFields("location", location.RealPath, "error", err).Warn("unable to fetch contents")
+		return nil, nil, err
+	}
+	defer internal.CloseAndLogError(contentReader, location.AccessPath)
+
+	discoveredPackages, discoveredRelationships, err := parser(ctx, resolver, env, file.NewLocationReadCloser(location, contentReader))
+	if err != nil {
+		logger.WithFields("location", location.RealPath, "error", err).Warnf("cataloger failed")
+		return nil, nil, err
+	}
+
+	return discoveredPackages, discoveredRelationships, nil
+}
+
 // selectFiles takes a set of file trees and resolves and file references of interest for future cataloging
 func (c *Cataloger) selectFiles(resolver file.Resolver) []request {
 	var requests []request

syft/pkg/cataloger/generic/cataloger_test.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"fmt"
 	"io"
+	"strings"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
@@ -86,3 +87,103 @@ func Test_Cataloger(t *testing.T) {
 		}
 	}
 }
+
+type spyReturningFileResolver struct {
+	m *file.MockResolver
+	s *spyingIoReadCloser
+}
+
+type spyingIoReadCloser struct {
+	rc     io.ReadCloser
+	closed bool
+}
+
+func newSpyReturningFileResolver(s *spyingIoReadCloser, paths ...string) file.Resolver {
+	m := file.NewMockResolverForPaths(paths...)
+	return spyReturningFileResolver{
+		m: m,
+		s: s,
+	}
+}
+
+func (s *spyingIoReadCloser) Read(p []byte) (n int, err error) {
+	return s.Read(p)
+}
+
+func (s *spyingIoReadCloser) Close() error {
+	s.closed = true
+	return s.rc.Close()
+}
+
+var _ io.ReadCloser = (*spyingIoReadCloser)(nil)
+
+func (m spyReturningFileResolver) FileContentsByLocation(location file.Location) (io.ReadCloser, error) {
+	return m.s, nil
+}
+
+func (m spyReturningFileResolver) HasPath(path string) bool {
+	return m.m.HasPath(path)
+}
+
+func (m spyReturningFileResolver) FilesByPath(paths ...string) ([]file.Location, error) {
+	return m.m.FilesByPath(paths...)
+}
+
+func (m spyReturningFileResolver) FilesByGlob(patterns ...string) ([]file.Location, error) {
+	return m.m.FilesByGlob(patterns...)
+}
+
+func (m spyReturningFileResolver) FilesByMIMEType(types ...string) ([]file.Location, error) {
+	return m.m.FilesByMIMEType(types...)
+}
+
+func (m spyReturningFileResolver) RelativeFileByPath(f file.Location, path string) *file.Location {
+	return m.m.RelativeFileByPath(f, path)
+}
+
+func (m spyReturningFileResolver) AllLocations(ctx context.Context) <-chan file.Location {
+	return m.m.AllLocations(ctx)
+}
+
+func (m spyReturningFileResolver) FileMetadataByLocation(location file.Location) (file.Metadata, error) {
+	return m.m.FileMetadataByLocation(location)
+}
+
+var _ file.Resolver = (*spyReturningFileResolver)(nil)
+
+func TestClosesFileOnParserPanic(t *testing.T) {
+	rc := io.NopCloser(strings.NewReader("some string"))
+	spy := spyingIoReadCloser{
+		rc: rc,
+	}
+	resolver := newSpyReturningFileResolver(&spy, "test-fixtures/another-path.txt")
+	ctx := context.TODO()
+
+	processors := []processor{
+		func(resolver file.Resolver, env Environment) []request {
+			return []request{
+				{
+					Location: file.Location{
+						LocationData: file.LocationData{
+							Coordinates: file.Coordinates{},
+							AccessPath:  "/some/access/path",
+						},
+					},
+					Parser: func(context.Context, file.Resolver, *Environment, file.LocationReadCloser) ([]pkg.Package, []artifact.Relationship, error) {
+						panic("panic!")
+					},
+				},
+			}
+		},
+	}
+
+	c := Cataloger{
+		processor:         processors,
+		upstreamCataloger: "unit-test-cataloger",
+	}
+
+	assert.PanicsWithValue(t, "panic!", func() {
+		_, _, _ = c.Catalog(ctx, resolver)
+	})
+	require.True(t, spy.closed)
+}