fix: more robust go main version extraction (#2767)

Signed-off-by: Keith Zantow <kzantow@gmail.com>
Signed-off-by: Will Murphy <will.murphy@anchore.com>
Co-authored-by: Will Murphy <will.murphy@anchore.com>

syft/pkg/cataloger/golang/parse_go_binary.go

@@ -151,7 +151,10 @@ func (c *goBinaryCataloger) makeGoMainPackage(resolver file.Resolver, mod *exten
 	return main
 }
 
-var semverPattern = regexp.MustCompile(`\x00(?P<version>v?(\d+\.\d+\.\d+[-\w]*[+\w]*))\x00`)
+// this is checking for (.L)? because at least one binary seems to have \xA0L preceding the version string, but for some reason
+// this is unable to be matched by the regex here as \x00\xA0L;
+// the only thing that seems to work is to just look for version strings following both \x00 and \x00.L for now
+var semverPattern = regexp.MustCompile(`\x00(.L)?(?P<version>v?(\d+\.\d+\.\d+[-\w]*[+\w]*))\x00`)
 
 func (c *goBinaryCataloger) findMainModuleVersion(metadata *pkg.GolangBinaryBuildinfoEntry, gbs pkg.KeyValues, reader io.ReadSeekCloser) string {
 	vcsVersion, hasVersion := gbs.Get("vcs.revision")
@@ -179,16 +182,8 @@ func (c *goBinaryCataloger) findMainModuleVersion(metadata *pkg.GolangBinaryBuil
 		if err != nil {
 			log.WithFields("error", err).Trace("unable to seek to start of go binary reader")
 		} else {
-			contents, err := io.ReadAll(reader)
+			if v := extractVersionFromContents(reader); v != "" {
-			if err != nil {
+				return v
-				log.WithFields("error", err).Trace("unable to read from go binary reader")
-			} else {
-				matchMetadata := internal.MatchNamedCaptureGroups(semverPattern, string(contents))
-
-				version, ok := matchMetadata["version"]
-				if ok {
-					return version
-				}
 			}
 		}
 	}
@@ -210,6 +205,21 @@ func (c *goBinaryCataloger) findMainModuleVersion(metadata *pkg.GolangBinaryBuil
 	return ""
 }
 
+func extractVersionFromContents(reader io.Reader) string {
+	contents, err := io.ReadAll(reader)
+	if err != nil {
+		log.WithFields("error", err).Trace("unable to read from go binary reader")
+		return ""
+	}
+	matchMetadata := internal.MatchNamedCaptureGroups(semverPattern, string(contents))
+
+	version, ok := matchMetadata["version"]
+	if ok {
+		return version
+	}
+	return ""
+}
+
 func extractVersionFromLDFlags(ldflags string) (majorVersion string, fullVersion string) {
 	if ldflags == "" {
 		return "", ""

syft/pkg/cataloger/golang/parse_go_binary_test.go

@@ -2,6 +2,8 @@ package golang
 
 import (
 	"bufio"
+	"bytes"
+	"errors"
 	"io"
 	"os"
 	"os/exec"
@@ -1090,3 +1092,53 @@ func Test_extractVersionFromLDFlags(t *testing.T) {
 		})
 	}
 }
+
+func Test_extractVersionFromContents(t *testing.T) {
+	tests := []struct {
+		name     string
+		contents io.Reader
+		want     string
+	}{
+		{
+			name:     "empty string on error",
+			contents: &alwaysErrorReader{},
+			want:     "",
+		},
+		{
+			name:     "empty string on empty reader",
+			contents: bytes.NewReader([]byte{}),
+			want:     "",
+		},
+		{
+			name:     "null-byte delimited semver",
+			contents: strings.NewReader("\x001.2.3\x00"),
+			want:     "1.2.3",
+		},
+		{
+			name:     "null-byte delimited semver with v prefix",
+			contents: strings.NewReader("\x00v1.2.3\x00"),
+			want:     "v1.2.3",
+		},
+		{
+			// 01a0bfc8: 0e74 5a3b 0000 a04c 7631 2e39 2e35 0000  .tZ;...Lv1.9.5.. from nginx-ingress-controller
+			// at /nginx-ingress-controller in registry.k8s.io/ingress-nginx/controller:v1.9.5
+			// digest: sha256:b3aba22b1da80e7acfc52b115cae1d4c687172cbf2b742d5b502419c25ff340e
+			// TODO: eventually use something for managing snippets, similar to what's used with binary classifier tests
+			name:     "null byte, then random byte, then L then semver",
+			contents: strings.NewReader("\x0e\x74\x5a\x3b\x00\x00\xa0\x4cv1.9.5\x00\x00"),
+			want:     "v1.9.5",
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := extractVersionFromContents(tt.contents)
+			assert.Equal(t, tt.want, got)
+		})
+	}
+}
+
+type alwaysErrorReader struct{}
+
+func (alwaysErrorReader) Read(_ []byte) (int, error) {
+	return 0, errors.New("read from always error reader")
+}

syft/pkg/cataloger/php/parse_pecl_serialized.go

@@ -5,12 +5,13 @@ import (
 	"fmt"
 	"io"
 
+	"github.com/elliotchance/phpserialize"
+
 	"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"
-	"github.com/elliotchance/phpserialize"
 )
 
 // parsePeclSerialized is a parser function for PECL metadata contents, returning "Default" php packages discovered.