feat: add Debian archive (.deb) file cataloger (#3704)

* feat: add Debian archive (.deb) file cataloger

Add a cataloger that parses Debian package (.deb) archive files directly,
allowing Syft to discover packages from .deb files without requiring
them to be installed on the system. This implements issue #3315.

Key features:
- Parse .deb AR archives to extract package metadata
- Support for gzip, xz, and zstd compressed control files
- Extract package metadata from control files
- Process file information from md5sums files
- Mark configuration files from conffiles entries
- Handle trailing slashes in archive member names

Signed-off-by: Alan Pope <alan.pope@anchore.com>

* chore: run go mod tidy to fix failing workflow

Signed-off-by: Alan Pope <alan.pope@anchore.com>

* add license processing to dpkg archive cataloger + add tests

Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>

* update json schema with dpkg archive type

Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>

* update comments

Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>

---------

Signed-off-by: Alan Pope <alan.pope@anchore.com>
Signed-off-by: Alex Goodman <wagoodman@users.noreply.github.com>
Co-authored-by: Alex Goodman <wagoodman@users.noreply.github.com>
This commit is contained in:
Alan Pope
2025-03-19 20:03:21 +00:00
committed by GitHub
co-authored by Alex Goodman
parent be0959cabf
commit 5fa8e9c6e9
19 changed files with 3506 additions and 12 deletions
+20 -1
View File
@@ -24,7 +24,7 @@ const (
func newDpkgPackage(d pkg.DpkgDBEntry, dbLocation file.Location, resolver file.Resolver, release *linux.Release, evidence ...file.Location) pkg.Package {
// TODO: separate pr to license refactor, but explore extracting dpkg-specific license parsing into a separate function
licenses := make([]pkg.License, 0)
var licenses []pkg.License
locations := file.NewLocationSet(dbLocation.WithAnnotation(pkg.EvidenceAnnotationKey, pkg.PrimaryEvidenceAnnotation))
locations.Add(evidence...)
@@ -54,6 +54,25 @@ func newDpkgPackage(d pkg.DpkgDBEntry, dbLocation file.Location, resolver file.R
return p
}
func newDebArchivePackage(location file.Location, metadata pkg.DpkgArchiveEntry, licenseStrings []string) pkg.Package {
p := pkg.Package{
Name: metadata.Package,
Version: metadata.Version,
Licenses: pkg.NewLicenseSet(pkg.NewLicensesFromValues(licenseStrings...)...),
Type: pkg.DebPkg,
PURL: packageURL(
pkg.DpkgDBEntry(metadata),
// we don't know the distro information, but since this is a deb file then we can reasonably assume it is a debian-based distro
&linux.Release{IDLike: []string{"debian"}},
),
Metadata: metadata,
Locations: file.NewLocationSet(location.WithAnnotation(pkg.EvidenceAnnotationKey, pkg.PrimaryEvidenceAnnotation)),
}
p.SetID()
return p
}
// PackageURL returns the PURL for the specific Debian package (see https://github.com/package-url/purl-spec)
func packageURL(m pkg.DpkgDBEntry, distro *linux.Release) string {
if distro == nil {