rename analyzers to catalogers; fix generic cataloger path dispatcher

imgbom/cataloger/dpkg/cataloger.go

@@ -0,0 +1,34 @@
+package dpkg
+
+import (
+	"github.com/anchore/imgbom/imgbom/cataloger/common"
+	"github.com/anchore/imgbom/imgbom/pkg"
+	"github.com/anchore/stereoscope/pkg/file"
+	"github.com/anchore/stereoscope/pkg/tree"
+)
+
+type Cataloger struct {
+	cataloger common.GenericCataloger
+}
+
+func NewCataloger() *Cataloger {
+	pathParsers := map[string]common.ParserFn{
+		"/var/lib/dpkg/status": parseDpkgStatus,
+	}
+
+	return &Cataloger{
+		cataloger: common.NewGenericCataloger(pathParsers, nil),
+	}
+}
+
+func (a *Cataloger) Name() string {
+	return "dpkg-cataloger"
+}
+
+func (a *Cataloger) SelectFiles(trees []tree.FileTreeReader) []file.Reference {
+	return a.cataloger.SelectFiles(trees)
+}
+
+func (a *Cataloger) Catalog(contents map[file.Reference]string) ([]pkg.Package, error) {
+	return a.cataloger.Catalog(contents, a.Name())
+}

imgbom/cataloger/dpkg/parse_dpkg_status.go

@@ -0,0 +1,99 @@
+package dpkg
+
+import (
+	"bufio"
+	"fmt"
+	"io"
+	"strings"
+
+	"github.com/anchore/imgbom/imgbom/pkg"
+	"github.com/mitchellh/mapstructure"
+)
+
+var errEndOfPackages = fmt.Errorf("no more packages to read")
+
+func parseDpkgStatus(reader io.Reader) ([]pkg.Package, error) {
+	buffedReader := bufio.NewReader(reader)
+	var packages = make([]pkg.Package, 0)
+
+	for {
+		entry, err := parseDpkgStatusEntry(buffedReader)
+		if err != nil {
+			if err == errEndOfPackages {
+				break
+			}
+			return nil, err
+		}
+		packages = append(packages, pkg.Package{
+			Name:     entry.Package,
+			Version:  entry.Version,
+			Type:     pkg.DebPkg,
+			Metadata: entry,
+		})
+	}
+
+	return packages, nil
+}
+
+func parseDpkgStatusEntry(reader *bufio.Reader) (entry pkg.DpkgMetadata, err error) {
+	dpkgFields := make(map[string]string)
+	var key string
+
+	for {
+		line, err := reader.ReadString('\n')
+		if err != nil {
+			if err == io.EOF {
+				return pkg.DpkgMetadata{}, errEndOfPackages
+			}
+			return pkg.DpkgMetadata{}, err
+		}
+
+		line = strings.TrimRight(line, "\n")
+
+		// empty line indicates end of entry
+		if len(line) == 0 {
+			// if the entry has not started, keep parsing lines
+			if len(dpkgFields) == 0 {
+				continue
+			}
+			break
+		}
+
+		switch {
+		case strings.HasPrefix(line, " "):
+			// a field-body continuation
+			if len(key) == 0 {
+				return pkg.DpkgMetadata{}, fmt.Errorf("no match for continuation: line: '%s'", line)
+			}
+
+			val, ok := dpkgFields[key]
+			if !ok {
+				return pkg.DpkgMetadata{}, fmt.Errorf("no previous key exists, expecting: %s", key)
+			}
+			// concatenate onto previous value
+			val = fmt.Sprintf("%s\n %s", val, strings.TrimSpace(line))
+			dpkgFields[key] = val
+		default:
+			// parse a new key
+			if i := strings.Index(line, ":"); i > 0 {
+				key = strings.TrimSpace(line[0:i])
+				val := strings.TrimSpace(line[i+1:])
+
+				if _, ok := dpkgFields[key]; ok {
+					return pkg.DpkgMetadata{}, fmt.Errorf("duplicate key discovered: %s", key)
+				}
+
+				dpkgFields[key] = val
+			} else {
+				return pkg.DpkgMetadata{}, fmt.Errorf("cannot parse field from line: '%s'", line)
+			}
+		}
+	}
+
+	err = mapstructure.Decode(dpkgFields, &entry)
+	if err != nil {
+		return pkg.DpkgMetadata{}, err
+	}
+
+	return entry, nil
+}

imgbom/cataloger/dpkg/parse_dpkg_status_test.go

@@ -0,0 +1,108 @@
+package dpkg
+
+import (
+	"bufio"
+	"os"
+	"testing"
+
+	"github.com/anchore/imgbom/imgbom/pkg"
+	"github.com/go-test/deep"
+)
+
+func compareEntries(t *testing.T, left, right pkg.DpkgMetadata) {
+	t.Helper()
+	if diff := deep.Equal(left, right); diff != nil {
+		t.Error(diff)
+	}
+}
+
+func TestSinglePackage(t *testing.T) {
+	tests := []struct {
+		name     string
+		expected pkg.DpkgMetadata
+	}{
+		{
+			name: "Test Single Package",
+			expected: pkg.DpkgMetadata{
+				Package: "apt",
+				Source:  "apt-dev",
+				Version: "1.8.2",
+			},
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			file, err := os.Open("test-fixtures/single")
+			if err != nil {
+				t.Fatal("Unable to read test_fixtures/single: ", err)
+			}
+			defer func() {
+				err := file.Close()
+				if err != nil {
+					t.Fatal("closing file failed:", err)
+				}
+			}()
+
+			reader := bufio.NewReader(file)
+
+			entry, err := parseDpkgStatusEntry(reader)
+			if err != nil {
+				t.Fatal("Unable to read file contents: ", err)
+			}
+
+			compareEntries(t, entry, test.expected)
+		})
+	}
+}
+
+func TestMultiplePackages(t *testing.T) {
+	tests := []struct {
+		name     string
+		expected []pkg.DpkgMetadata
+	}{
+		{
+			name: "Test Multiple Package",
+			expected: []pkg.DpkgMetadata{
+				{
+					Package: "tzdata",
+					Version: "2020a-0+deb10u1",
+					Source:  "tzdata-dev",
+				},
+				{
+					Package: "util-linux",
+					Version: "2.33.1-0.1",
+				},
+			},
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			file, err := os.Open("test-fixtures/multiple")
+			if err != nil {
+				t.Fatal("Unable to read: ", err)
+			}
+			defer func() {
+				err := file.Close()
+				if err != nil {
+					t.Fatal("closing file failed:", err)
+				}
+			}()
+
+			pkgs, err := parseDpkgStatus(file)
+			if err != nil {
+				t.Fatal("Unable to read file contents: ", err)
+			}
+
+			if len(pkgs) != 2 {
+				t.Fatalf("unexpected number of entries: %d", len(pkgs))
+			}
+
+			for idx, entry := range pkgs {
+				compareEntries(t, entry.Metadata.(pkg.DpkgMetadata), test.expected[idx])
+			}
+
+		})
+	}
+}

imgbom/cataloger/dpkg/test-fixtures/multiple

@@ -0,0 +1,52 @@
+Package: tzdata
+Status: install ok installed
+Priority: required
+Source: tzdata-dev
+Section: localization
+Installed-Size: 3036
+Maintainer: GNU Libc Maintainers <debian-glibc@lists.debian.org>
+Architecture: all
+Multi-Arch: foreign
+Version: 2020a-0+deb10u1
+Replaces: libc0.1, libc0.3, libc6, libc6.1
+Provides: tzdata-buster
+Depends: debconf (>= 0.5) | debconf-2.0
+Description: time zone and daylight-saving time data
+ This package contains data required for the implementation of
+ standard local time for many representative locations around the
+ globe. It is updated periodically to reflect changes made by
+ political bodies to time zone boundaries, UTC offsets, and
+ daylight-saving rules.
+Homepage: https://www.iana.org/time-zones
+
+Package: util-linux
+Essential: yes
+Status: install ok installed
+Priority: required
+Section: utils
+Installed-Size: 4327
+Maintainer: LaMont Jones <lamont@debian.org>
+Architecture: amd64
+Multi-Arch: foreign
+Version: 2.33.1-0.1
+Replaces: bash-completion (<< 1:2.8), initscripts (<< 2.88dsf-59.2~), login (<< 1:4.5-1.1~), mount (<< 2.29.2-3~), s390-tools (<< 2.2.0-1~), setpriv (<< 2.32.1-0.2~), sysvinit-utils (<< 2.88dsf-59.1~)
+Depends: fdisk, login (>= 1:4.5-1.1~)
+Pre-Depends: libaudit1 (>= 1:2.2.1), libblkid1 (>= 2.31.1), libc6 (>= 2.25), libcap-ng0 (>= 0.7.9), libmount1 (>= 2.25), libpam0g (>= 0.99.7.1), libselinux1 (>= 2.6-3~), libsmartcols1 (>= 2.33), libsystemd0, libtinfo6 (>= 6), libudev1 (>= 183), libuuid1 (>= 2.16), zlib1g (>= 1:1.1.4)
+Suggests: dosfstools, kbd | console-tools, util-linux-locales
+Breaks: bash-completion (<< 1:2.8), grml-debootstrap (<< 0.68), mount (<< 2.29.2-3~), s390-tools (<< 2.2.0-1~), setpriv (<< 2.32.1-0.2~), sysvinit-utils (<< 2.88dsf-59.4~)
+Conffiles:
+ /etc/default/hwclock 3916544450533eca69131f894db0ca12
+ /etc/init.d/hwclock.sh 1ca5c0743fa797ffa364db95bb8d8d8e
+ /etc/pam.d/runuser b8b44b045259525e0fae9e38fdb2aeeb
+ /etc/pam.d/runuser-l 2106ea05877e8913f34b2c77fa02be45
+ /etc/pam.d/su ce6dcfda3b190a27a455bb38a45ff34a
+ /etc/pam.d/su-l 756fef5687fecc0d986e5951427b0c4f
+Description: miscellaneous system utilities
+ This package contains a number of important utilities, most of which
+ are oriented towards maintenance of your system. Some of the more
+ important utilities included in this package allow you to view kernel
+ messages, create new filesystems, view block device information,
+ interface with real time clock, etc.
+
+
+

imgbom/cataloger/dpkg/test-fixtures/single

@@ -0,0 +1,35 @@
+Package: apt
+Status: install ok installed
+Priority: required
+Section: admin
+Installed-Size: 4064
+Maintainer: APT Development Team <deity@lists.debian.org>
+Architecture: amd64
+Version: 1.8.2
+Source: apt-dev
+Replaces: apt-transport-https (<< 1.5~alpha4~), apt-utils (<< 1.3~exp2~)
+Provides: apt-transport-https (= 1.8.2)
+Depends: adduser, gpgv | gpgv2 | gpgv1, debian-archive-keyring, libapt-pkg5.0 (>= 1.7.0~alpha3~), libc6 (>= 2.15), libgcc1 (>= 1:3.0), libgnutls30 (>= 3.6.6), libseccomp2 (>= 1.0.1), libstdc++6 (>= 5.2)
+Recommends: ca-certificates
+Suggests: apt-doc, aptitude | synaptic | wajig, dpkg-dev (>= 1.17.2), gnupg | gnupg2 | gnupg1, powermgmt-base
+Breaks: apt-transport-https (<< 1.5~alpha4~), apt-utils (<< 1.3~exp2~), aptitude (<< 0.8.10)
+Conffiles:
+ /etc/apt/apt.conf.d/01autoremove 76120d358bc9037bb6358e737b3050b5
+ /etc/cron.daily/apt-compat 49e9b2cfa17849700d4db735d04244f3
+ /etc/kernel/postinst.d/apt-auto-removal 4ad976a68f045517cf4696cec7b8aa3a
+ /etc/logrotate.d/apt 179f2ed4f85cbaca12fa3d69c2a4a1c3
+Description: commandline package manager
+ This package provides commandline tools for searching and
+ managing as well as querying information about packages
+ as a low-level access to all features of the libapt-pkg library.
+ .
+ These include:
+  * apt-get for retrieval of packages and information about them
+    from authenticated sources and for installation, upgrade and
+    removal of packages together with their dependencies
+  * apt-cache for querying available information about installed
+    as well as installable packages
+  * apt-cdrom to use removable media as a source for packages
+  * apt-config as an interface to the configuration settings
+  * apt-key as an interface to manage authentication keys
+