[wip] git-root config search concept

Signed-off-by: Alex Goodman <alex.goodman@anchore.com>

.bouncer.yaml

@@ -3,4 +3,20 @@ permit:
   - MIT.*
   - Apache.*
   - MPL.*
-  - ISC
+  - ISC
+ignore-packages:
+  # packageurl-go is released under the MIT license located in the root of the repo at /mit.LICENSE
+  - github.com/anchore/packageurl-go
+
+  # from: https://github.com/spdx/tools-golang/blob/main/LICENSE.code
+  # The tools-golang source code is provided and may be used, at your option,
+  # under either:
+  # * Apache License, version 2.0 (Apache-2.0), OR
+  # * GNU General Public License, version 2.0 or later (GPL-2.0-or-later).
+  # (we choose Apache-2.0)
+  - github.com/spdx/tools-golang
+
+  # from: https://github.com/xi2/xz/blob/master/LICENSE
+  # All these files have been put into the public domain.
+  # You can do whatever you want with these files.
+  - github.com/xi2/xz

.circleci/config.yml

@@ -1,127 +0,0 @@
-version: 2.1
-
-jobs:
-  run-static-analysis:
-    parameters:
-      version:
-        type: string
-    docker:
-      - image: circleci/golang:<< parameters.version >>
-        environment:
-          GO111MODULE: "on"
-          # work around for recent circle CI breaking change
-          # Error: "Error response from daemon: client version 1.39 is too new. Maximum supported API version is 1.38"
-          DOCKER_API_VERSION: "1.38"
-    # 2CPU / 4GB RAM
-    resource_class: medium
-    steps:
-
-      - checkout
-
-      - restore_cache:
-          keys:
-            - go-<< parameters.version >>-{{ checksum "go.sum" }}-{{ checksum "Makefile" }}
-
-      - run: make ci-bootstrap
-
-      - save_cache:
-          key: go-<< parameters.version >>-{{ checksum "go.sum" }}-{{ checksum "Makefile" }}
-          paths:
-            - "/go/pkg/mod"
-            - ".tmp"
-
-      - run:
-          name: run static analysis
-          command: make static-analysis
-
-  run-tests:
-    parameters:
-      version:
-        type: string
-    docker:
-      - image: circleci/golang:<< parameters.version >>
-        environment:
-          GO111MODULE: "on"
-          # work around for recent circle CI breaking change
-          # Error: "Error response from daemon: client version 1.39 is too new. Maximum supported API version is 1.38"
-          DOCKER_API_VERSION: "1.38"
-    # 2CPU / 4GB RAM
-    resource_class: medium
-    steps:
-
-      - checkout
-
-      - restore_cache:
-          keys:
-            - go-<< parameters.version >>-{{ checksum "go.sum" }}-{{ checksum "Makefile" }}
-
-      - run: make ci-bootstrap
-
-      - save_cache:
-          key: go-<< parameters.version >>-{{ checksum "go.sum" }}-{{ checksum "Makefile" }}
-          paths:
-            - "/go/pkg/mod"
-            - ".tmp"
-
-      - setup_remote_docker:
-          version: 18.06.0-ce
-
-      - run:
-          name: enable docker client
-          command: |
-              # all of this to enable "circleci local execute ..." cli commands for /var/run/docker.sock
-              mkdir -p ${HOME}/.local/bin
-              cat \<< EOF > ${HOME}/.local/bin/docker
-              #!/bin/bash
-              set -xue
-              sudo -E ${HOME}/.local/bin/docker.bin \$@
-              EOF
-              sudo mv /usr/bin/docker ${HOME}/.local/bin/docker.bin
-              chmod 755 ${HOME}/.local/bin/docker
-
-      - run:
-          name: build cache key for java test-fixture blobs
-          command: make java-packages-fingerprint
-
-      - restore_cache:
-          keys:
-            - syft-unit-test-java-cache-{{ checksum "syft/cataloger/java/test-fixtures/java-builds/packages.fingerprint" }}
-
-      - run:
-          name: run unit tests
-          command: make unit
-
-      - save_cache:
-          key: syft-unit-test-java-cache-{{ checksum "syft/cataloger/java/test-fixtures/java-builds/packages.fingerprint" }}
-          paths:
-            - "syft/cataloger/java/test-fixtures/java-builds/packages"
-
-      - run:
-          name: build hash key for integration test-fixtures blobs
-          command: make integration-fingerprint
-
-      - restore_cache:
-          keys:
-            - syft-integration-test-tar-cache-{{ checksum "test/integration/test-fixtures/tar-cache.fingerprint" }}
-      - run:
-          name: run integration tests
-          command: make integration
-
-      - save_cache:
-          key: syft-integration-test-tar-cache-{{ checksum "test/integration/test-fixtures/tar-cache.fingerprint" }}
-          paths:
-            - "test/integration/test-fixtures/tar-cache"
-
-workflows:
-  # Note: changing this workflow name requires making the same update in the .github/workflows/release.yaml pipeline
-  "Static Analysis + Unit + Integration":
-    jobs:
-      - run-static-analysis:
-          name: "Static Analysis"
-          version: "1.14"
-      - run-tests:
-          name: "Unit & Integration Tests (go-1.13)"
-          version: "1.13"
-      - run-tests:
-          name: "Unit & Integration Tests (go-latest)"
-          version: "latest"

.github/ISSUE_TEMPLATE/bug_report.md

@@ -16,5 +16,5 @@ assignees: ''
 **Anything else we need to know?**:
 
 **Environment**:
-- syft version (use `syft version -v`):
+- Output of `syft version`:
 - OS (e.g: `cat /etc/os-release` or similar):

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,6 @@
+contact_links:
+
+  - name: Join the Slack community 💬
+    # link to our community Slack registration page
+    url: https://anchore.com/slack
+    about: 'Come chat with us! Ask for help, join our software development efforts, or just give us feedback!'

.github/scripts/go-mod-tidy-check.sh

@@ -0,0 +1,31 @@
+#!/usr/bin/env bash
+set -eu
+
+ORIGINAL_STATE_DIR=$(mktemp -d "TEMP-original-state-XXXXXXXXX")
+TIDY_STATE_DIR=$(mktemp -d "TEMP-tidy-state-XXXXXXXXX")
+
+trap "cp -v ${ORIGINAL_STATE_DIR}/* ./ && rm -fR ${ORIGINAL_STATE_DIR} ${TIDY_STATE_DIR}" EXIT
+
+echo "Capturing original state of files..."
+cp -v go.mod go.sum "${ORIGINAL_STATE_DIR}"
+
+echo "Capturing state of go.mod and go.sum after running go mod tidy..."
+go mod tidy
+cp -v go.mod go.sum "${TIDY_STATE_DIR}"
+echo ""
+
+set +e
+
+# Detect difference between the git HEAD state and the go mod tidy state
+DIFF_MOD=$(diff -u "${ORIGINAL_STATE_DIR}/go.mod" "${TIDY_STATE_DIR}/go.mod")
+DIFF_SUM=$(diff -u "${ORIGINAL_STATE_DIR}/go.sum" "${TIDY_STATE_DIR}/go.sum")
+
+if [[ -n "${DIFF_MOD}" || -n "${DIFF_SUM}" ]]; then
+    echo "go.mod diff:"
+    echo "${DIFF_MOD}"
+    echo "go.sum diff:"
+    echo "${DIFF_SUM}"
+    echo ""
+    printf "FAILED! go.mod and/or go.sum are NOT tidy; please run 'go mod tidy'.\n\n"
+    exit 1
+fi

.github/scripts/goreleaser-install.sh

@@ -0,0 +1,398 @@
+#!/bin/sh
+set -e
+# Code generated by godownloader on 2019-12-25T12:47:14Z. DO NOT EDIT.
+#
+
+usage() {
+	this=$1
+	cat <<EOF
+$this: download go binaries for goreleaser/goreleaser
+
+Usage: $this [-b] bindir [-d] [tag]
+  -b sets bindir or installation directory, Defaults to ./bin
+  -d turns on debug logging
+   [tag] is a tag from
+   https://github.com/goreleaser/goreleaser/releases
+   If tag is missing, then the latest will be used.
+
+ Generated by godownloader
+  https://github.com/goreleaser/godownloader
+
+EOF
+	exit 2
+}
+
+parse_args() {
+	#BINDIR is ./bin unless set be ENV
+	# over-ridden by flag below
+
+	BINDIR=${BINDIR:-./bin}
+	while getopts "b:dh?x" arg; do
+		case "$arg" in
+		b) BINDIR="$OPTARG" ;;
+		d) log_set_priority 10 ;;
+		h | \?) usage "$0" ;;
+		x) set -x ;;
+		esac
+	done
+	shift $((OPTIND - 1))
+	TAG=$1
+}
+# this function wraps all the destructive operations
+# if a curl|bash cuts off the end of the script due to
+# network, either nothing will happen or will syntax error
+# out preventing half-done work
+execute() {
+	tmpdir=$(mktemp -d)
+	log_debug "downloading files into ${tmpdir}"
+	http_download "${tmpdir}/${TARBALL}" "${TARBALL_URL}"
+	http_download "${tmpdir}/${CHECKSUM}" "${CHECKSUM_URL}"
+	hash_sha256_verify "${tmpdir}/${TARBALL}" "${tmpdir}/${CHECKSUM}"
+	srcdir="${tmpdir}"
+	(cd "${tmpdir}" && untar "${TARBALL}")
+	test ! -d "${BINDIR}" && install -d "${BINDIR}"
+	for binexe in $BINARIES; do
+		if [ "$OS" = "windows" ]; then
+			binexe="${binexe}.exe"
+		fi
+		install "${srcdir}/${binexe}" "${BINDIR}/"
+		log_info "installed ${BINDIR}/${binexe}"
+	done
+	rm -rf "${tmpdir}"
+}
+get_binaries() {
+	case "$PLATFORM" in
+	darwin/386) BINARIES="goreleaser" ;;
+	darwin/amd64) BINARIES="goreleaser" ;;
+	darwin/arm64) BINARIES="goreleaser" ;;
+	darwin/armv6) BINARIES="goreleaser" ;;
+	linux/386) BINARIES="goreleaser" ;;
+	linux/amd64) BINARIES="goreleaser" ;;
+	linux/arm64) BINARIES="goreleaser" ;;
+	linux/armv6) BINARIES="goreleaser" ;;
+	windows/386) BINARIES="goreleaser" ;;
+	windows/amd64) BINARIES="goreleaser" ;;
+	windows/arm64) BINARIES="goreleaser" ;;
+	windows/armv6) BINARIES="goreleaser" ;;
+	*)
+		log_crit "platform $PLATFORM is not supported.  Make sure this script is up-to-date and file request at https://github.com/${PREFIX}/issues/new"
+		exit 1
+		;;
+	esac
+}
+tag_to_version() {
+	if [ -z "${TAG}" ]; then
+		log_info "checking GitHub for latest tag"
+	else
+		log_info "checking GitHub for tag '${TAG}'"
+	fi
+	REALTAG=$(github_release "$OWNER/$REPO" "${TAG}") && true
+	if test -z "$REALTAG"; then
+		log_crit "unable to find '${TAG}' - use 'latest' or see https://github.com/${PREFIX}/releases for details"
+		exit 1
+	fi
+	# if version starts with 'v', remove it
+	TAG="$REALTAG"
+	VERSION=${TAG#v}
+}
+adjust_format() {
+	# change format (tar.gz or zip) based on OS
+	case ${OS} in
+	windows) FORMAT=zip ;;
+	esac
+	true
+}
+adjust_os() {
+	# adjust archive name based on OS
+	case ${OS} in
+	386) OS=i386 ;;
+	amd64) OS=x86_64 ;;
+	darwin) OS=Darwin ;;
+	linux) OS=Linux ;;
+	windows) OS=Windows ;;
+	esac
+	true
+}
+adjust_arch() {
+	# adjust archive name based on ARCH
+	case ${ARCH} in
+	386) ARCH=i386 ;;
+	amd64) ARCH=x86_64 ;;
+	darwin) ARCH=Darwin ;;
+	linux) ARCH=Linux ;;
+	windows) ARCH=Windows ;;
+	esac
+	true
+}
+
+cat /dev/null <<EOF
+------------------------------------------------------------------------
+https://github.com/client9/shlib - portable posix shell functions
+Public domain - http://unlicense.org
+https://github.com/client9/shlib/blob/master/LICENSE.md
+but credit (and pull requests) appreciated.
+------------------------------------------------------------------------
+EOF
+is_command() {
+	command -v "$1" >/dev/null
+}
+echoerr() {
+	echo "$@" 1>&2
+}
+log_prefix() {
+	echo "$0"
+}
+_logp=6
+log_set_priority() {
+	_logp="$1"
+}
+log_priority() {
+	if test -z "$1"; then
+		echo "$_logp"
+		return
+	fi
+	[ "$1" -le "$_logp" ]
+}
+log_tag() {
+	case $1 in
+	0) echo "emerg" ;;
+	1) echo "alert" ;;
+	2) echo "crit" ;;
+	3) echo "err" ;;
+	4) echo "warning" ;;
+	5) echo "notice" ;;
+	6) echo "info" ;;
+	7) echo "debug" ;;
+	*) echo "$1" ;;
+	esac
+}
+log_debug() {
+	log_priority 7 || return 0
+	echoerr "$(log_prefix)" "$(log_tag 7)" "$@"
+}
+log_info() {
+	log_priority 6 || return 0
+	echoerr "$(log_prefix)" "$(log_tag 6)" "$@"
+}
+log_err() {
+	log_priority 3 || return 0
+	echoerr "$(log_prefix)" "$(log_tag 3)" "$@"
+}
+log_crit() {
+	log_priority 2 || return 0
+	echoerr "$(log_prefix)" "$(log_tag 2)" "$@"
+}
+uname_os() {
+	os=$(uname -s | tr '[:upper:]' '[:lower:]')
+	case "$os" in
+	cygwin_nt*) os="windows" ;;
+	mingw*) os="windows" ;;
+	msys_nt*) os="windows" ;;
+	esac
+	echo "$os"
+}
+uname_arch() {
+	arch=$(uname -m)
+	case $arch in
+	x86_64) arch="amd64" ;;
+	x86) arch="386" ;;
+	i686) arch="386" ;;
+	i386) arch="386" ;;
+	aarch64) arch="arm64" ;;
+	armv5*) arch="armv5" ;;
+	armv6*) arch="armv6" ;;
+	armv7*) arch="armv7" ;;
+	esac
+	echo ${arch}
+}
+uname_os_check() {
+	os=$(uname_os)
+	case "$os" in
+	darwin) return 0 ;;
+	dragonfly) return 0 ;;
+	freebsd) return 0 ;;
+	linux) return 0 ;;
+	android) return 0 ;;
+	nacl) return 0 ;;
+	netbsd) return 0 ;;
+	openbsd) return 0 ;;
+	plan9) return 0 ;;
+	solaris) return 0 ;;
+	windows) return 0 ;;
+	esac
+	log_crit "uname_os_check '$(uname -s)' got converted to '$os' which is not a GOOS value. Please file bug at https://github.com/client9/shlib"
+	return 1
+}
+uname_arch_check() {
+	arch=$(uname_arch)
+	case "$arch" in
+	386) return 0 ;;
+	amd64) return 0 ;;
+	arm64) return 0 ;;
+	armv5) return 0 ;;
+	armv6) return 0 ;;
+	armv7) return 0 ;;
+	ppc64) return 0 ;;
+	ppc64le) return 0 ;;
+	mips) return 0 ;;
+	mipsle) return 0 ;;
+	mips64) return 0 ;;
+	mips64le) return 0 ;;
+	s390x) return 0 ;;
+	amd64p32) return 0 ;;
+	esac
+	log_crit "uname_arch_check '$(uname -m)' got converted to '$arch' which is not a GOARCH value.  Please file bug report at https://github.com/client9/shlib"
+	return 1
+}
+untar() {
+	tarball=$1
+	case "${tarball}" in
+	*.tar.gz | *.tgz) tar --no-same-owner -xzf "${tarball}" ;;
+	*.tar) tar --no-same-owner -xf "${tarball}" ;;
+	*.zip) unzip "${tarball}" ;;
+	*)
+		log_err "untar unknown archive format for ${tarball}"
+		return 1
+		;;
+	esac
+}
+http_download_curl() {
+	local_file=$1
+	source_url=$2
+	header=$3
+	if [ -z "$header" ]; then
+		code=$(curl -w '%{http_code}' -sL -o "$local_file" "$source_url")
+	else
+		code=$(curl -w '%{http_code}' -sL -H "$header" -o "$local_file" "$source_url")
+	fi
+	if [ "$code" != "200" ]; then
+		log_debug "http_download_curl received HTTP status $code"
+		return 1
+	fi
+	return 0
+}
+http_download_wget() {
+	local_file=$1
+	source_url=$2
+	header=$3
+	if [ -z "$header" ]; then
+		wget -q -O "$local_file" "$source_url"
+	else
+		wget -q --header "$header" -O "$local_file" "$source_url"
+	fi
+}
+http_download() {
+	log_debug "http_download $2"
+	if is_command curl; then
+		http_download_curl "$@"
+		return
+	elif is_command wget; then
+		http_download_wget "$@"
+		return
+	fi
+	log_crit "http_download unable to find wget or curl"
+	return 1
+}
+http_copy() {
+	tmp=$(mktemp)
+	http_download "${tmp}" "$1" "$2" || return 1
+	body=$(cat "$tmp")
+	rm -f "${tmp}"
+	echo "$body"
+}
+github_release() {
+	owner_repo=$1
+	version=$2
+	test -z "$version" && version="latest"
+	giturl="https://github.com/${owner_repo}/releases/${version}"
+	json=$(http_copy "$giturl" "Accept:application/json")
+	test -z "$json" && return 1
+	version=$(echo "$json" | tr -s '\n' ' ' | sed 's/.*"tag_name":"//' | sed 's/".*//')
+	test -z "$version" && return 1
+	echo "$version"
+}
+hash_sha256() {
+	TARGET=${1:-/dev/stdin}
+	if is_command gsha256sum; then
+		hash=$(gsha256sum "$TARGET") || return 1
+		echo "$hash" | cut -d ' ' -f 1
+	elif is_command sha256sum; then
+		hash=$(sha256sum "$TARGET") || return 1
+		echo "$hash" | cut -d ' ' -f 1
+	elif is_command shasum; then
+		hash=$(shasum -a 256 "$TARGET" 2>/dev/null) || return 1
+		echo "$hash" | cut -d ' ' -f 1
+	elif is_command openssl; then
+		hash=$(openssl -dst openssl dgst -sha256 "$TARGET") || return 1
+		echo "$hash" | cut -d ' ' -f a
+	else
+		log_crit "hash_sha256 unable to find command to compute sha-256 hash"
+		return 1
+	fi
+}
+hash_sha256_verify() {
+	TARGET=$1
+	checksums=$2
+	if [ -z "$checksums" ]; then
+		log_err "hash_sha256_verify checksum file not specified in arg2"
+		return 1
+	fi
+	BASENAME=${TARGET##*/}
+	want=$(grep "${BASENAME}" "${checksums}" 2>/dev/null | tr '\t' ' ' | cut -d ' ' -f 1)
+	if [ -z "$want" ]; then
+		log_err "hash_sha256_verify unable to find checksum for '${TARGET}' in '${checksums}'"
+		return 1
+	fi
+	got=$(hash_sha256 "$TARGET")
+	if [ "$want" != "$got" ]; then
+		log_err "hash_sha256_verify checksum for '$TARGET' did not verify ${want} vs $got"
+		return 1
+	fi
+}
+cat /dev/null <<EOF
+------------------------------------------------------------------------
+End of functions from https://github.com/client9/shlib
+------------------------------------------------------------------------
+EOF
+
+PROJECT_NAME="goreleaser"
+OWNER=goreleaser
+REPO="goreleaser"
+BINARY=goreleaser
+FORMAT=tar.gz
+OS=$(uname_os)
+ARCH=$(uname_arch)
+PREFIX="$OWNER/$REPO"
+
+# use in logging routines
+log_prefix() {
+	echo "$PREFIX"
+}
+PLATFORM="${OS}/${ARCH}"
+GITHUB_DOWNLOAD=https://github.com/${OWNER}/${REPO}/releases/download
+
+uname_os_check "$OS"
+uname_arch_check "$ARCH"
+
+parse_args "$@"
+
+get_binaries
+
+tag_to_version
+
+adjust_format
+
+adjust_os
+
+adjust_arch
+
+log_info "found version: ${VERSION} for ${TAG}/${OS}/${ARCH}"
+
+NAME=${PROJECT_NAME}_${OS}_${ARCH}
+TARBALL=${NAME}.${FORMAT}
+TARBALL_URL=${GITHUB_DOWNLOAD}/${TAG}/${TARBALL}
+CHECKSUM=checksums.txt
+CHECKSUM_URL=${GITHUB_DOWNLOAD}/${TAG}/${CHECKSUM}
+
+execute
+log_err "this script is deprecated, please do not use it anymore. check https://github.com/goreleaser/godownloader/issues/207"

.github/scripts/mac-prepare-for-signing.sh

@@ -0,0 +1,31 @@
+#!/usr/bin/env bash
+set -eu
+
+CI_HOME="/Users/runner"
+if [[ "${HOME}" != "${CI_HOME}" ]]; then
+  printf "WARNING! It looks like this isn't the CI environment. This script modifies the macOS Keychain setup in ways you probably wouldn't want for your own machine. It also requires an Apple Developer ID Certificate that you shouldn't have outside of the CI environment.\n\nExiting early to make sure nothing bad happens.\n"
+  exit 1
+fi
+
+# Install gon (see https://github.com/mitchellh/gon for details).
+brew tap mitchellh/gon
+brew install mitchellh/gon/gon
+
+# Write signing certificate to disk from environment variable.
+CERT_FILE="$HOME/developer_id_certificate.p12"
+echo -n "$APPLE_DEVELOPER_ID_CERT" | base64 --decode > "$CERT_FILE"
+
+# In order to have all keychain interactions avoid an interactive user prompt, we need to control the password for the keychain in question, which means we need to create a new keychain into which we'll import the signing certificate and from which we'll later access this certificate during code signing.
+EPHEMERAL_KEYCHAIN="ci-ephemeral-keychain"
+EPHEMERAL_KEYCHAIN_PASSWORD="$(openssl rand -base64 100)"
+security create-keychain -p "${EPHEMERAL_KEYCHAIN_PASSWORD}" "${EPHEMERAL_KEYCHAIN}"
+
+# Import signing certificate into the keychain. (This is a pre-requisite for gon, which is invoked via goreleaser.)
+EPHEMERAL_KEYCHAIN_FULL_PATH="$HOME/Library/Keychains/${EPHEMERAL_KEYCHAIN}-db"
+security import "${CERT_FILE}" -k "${EPHEMERAL_KEYCHAIN_FULL_PATH}" -P "${APPLE_DEVELOPER_ID_CERT_PASS}" -T "$(command -v codesign)"
+
+# Setting the partition list for this certificate's private key to include "apple-tool:" and "apple:" allows the codesign command to access this keychain item without an interactive user prompt. (codesign is invoked by gon.)
+security set-key-partition-list -S "apple-tool:,apple:" -s -k "${EPHEMERAL_KEYCHAIN_PASSWORD}" "${EPHEMERAL_KEYCHAIN_FULL_PATH}"
+
+# Make this new keychain the user's default keychain, so that codesign will be able to find this certificate when we specify it during signing.
+security default-keychain -d "user" -s "${EPHEMERAL_KEYCHAIN_FULL_PATH}"

.github/scripts/mac-sign-and-notarize.sh

@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+set -eu
+
+IS_SNAPSHOT="$1"                        # e.g. "true", "false"
+
+if [[ "${IS_SNAPSHOT}" == "true" ]]; then
+  # This is a snapshot build —— skipping signing and notarization...
+  exit 0
+fi
+
+GON_CONFIG="$2"                                       # e.g. "gon.hcl"
+NEW_NAME_WITHOUT_EXTENSION="$3"                       # e.g. "./dist/syft-0.1.0"
+ORIGINAL_NAME_WITHOUT_EXTENSION="./dist/output"       # This should match dmg and zip output_path in the gon config file, without the extension.
+
+gon "${GON_CONFIG}"
+
+# Rename outputs with specified desired name
+mv -v "${ORIGINAL_NAME_WITHOUT_EXTENSION}.dmg" "${NEW_NAME_WITHOUT_EXTENSION}.dmg"
+mv -v "${ORIGINAL_NAME_WITHOUT_EXTENSION}.zip" "${NEW_NAME_WITHOUT_EXTENSION}.zip"

.github/scripts/update-version-file.sh

@@ -5,7 +5,9 @@ BIN="syft"
 DISTDIR=$1
 VERSION=$2
 
-if [[ $VERSION == *-* ]] ; then
+# the source of truth as to whether we want to notify users of an update is if the release just created is NOT
+# flagged as a pre-release on github
+if [[ "$(curl -SsL https://api.github.com/repos/anchore/${BIN}/releases/tags/${VERSION} | jq .prerelease)" == "true" ]] ; then
    echo "skipping publishing a version file (this is a pre-release: ${VERSION})"
    exit 0
 fi
@@ -13,14 +15,9 @@ fi
 echo "creating and publishing version file"
 
 # create a version file for version-update checks
-echo "${VERSION}" | tee ${DISTDIR}/VERSION
+VERSION_FILE="${DISTDIR}/VERSION"
+echo "${VERSION}" | tee "${VERSION_FILE}"
 
 # upload the version file that supports the application version update check
-docker run --rm \
-    -i \
-    -e AWS_DEFAULT_REGION=us-west-2 \
-    -e AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID} \
-    -e AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY} \
-    -v $(pwd)/${DISTDIR}/:/distmount \
-    amazon/aws-cli \
-        s3 cp /distmount/VERSION s3://toolbox-data.anchore.io/${BIN}/releases/latest/VERSION
+export AWS_DEFAULT_REGION=us-west-2
+aws s3 cp "${VERSION_FILE}" s3://toolbox-data.anchore.io/${BIN}/releases/latest/VERSION

.github/workflows/acceptance-test.yaml

@@ -1,129 +0,0 @@
-name: 'Acceptance'
-on:
-  workflow_dispatch:
-  push:
-    # ... only act on pushes to main
-    branches:
-      - main
-    # ... do not act on release tags
-    tags-ignore:
-      - v*
-
-env:
-  GO_VERSION: "1.14.x"
-
-jobs:
-  Build-Snapshot-Artifacts:
-    runs-on: ubuntu-latest
-    steps:
-
-      # TODO: remove me after release
-      - name: Configure git for private modules
-        env:
-          TOKEN: ${{ secrets.ANCHORE_GIT_READ_TOKEN }}
-        run: git config --global url."https://anchore:${TOKEN}@github.com".insteadOf "https://github.com"
-
-      - uses: actions/setup-go@v2
-        with:
-          go-version: ${{ env.GO_VERSION }}
-
-      - uses: actions/checkout@v2
-
-      - name: Restore bootstrap cache
-        id: cache
-        uses: actions/cache@v2
-        with:
-          path: |
-            ~/go/pkg/mod
-            ${{ github.workspace }}/.tmp
-          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}-${{ hashFiles('Makefile') }}
-          restore-keys: |
-            ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}-
-            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
-
-      - name: Bootstrap dependencies
-        if: steps.cache.outputs.cache-hit != 'true'
-        run: make ci-bootstrap
-
-      - name: Import GPG key
-        id: import_gpg
-        uses: crazy-max/ghaction-import-gpg@v2
-        env:
-          GPG_PRIVATE_KEY: ${{ secrets.SIGNING_GPG_PRIVATE_KEY }}
-          PASSPHRASE: ${{ secrets.SIGNING_GPG_PASSPHRASE }}
-
-      - name: GPG signing info
-        run: |
-          echo "fingerprint: ${{ steps.import_gpg.outputs.fingerprint }}"
-          echo "keyid:       ${{ steps.import_gpg.outputs.keyid }}"
-          echo "name:        ${{ steps.import_gpg.outputs.name }}"
-          echo "email:       ${{ steps.import_gpg.outputs.email }}"
-
-      - name: Build snapshot artifacts
-        run: make snapshot
-        env:
-          GPG_PRIVATE_KEY: ${{ secrets.SIGNING_GPG_PRIVATE_KEY }}
-          PASSPHRASE: ${{ secrets.SIGNING_GPG_PASSPHRASE }}
-
-      - uses: actions/upload-artifact@v2
-        with:
-          name: artifacts
-          path: snapshot/**/*
-
-  # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
-  Acceptance-Linux:
-    needs: [ Build-Snapshot-Artifacts ]
-    runs-on: ubuntu-latest
-    steps:
-
-      - uses: actions/checkout@v2
-
-      - uses: actions/download-artifact@v2
-        with:
-          name: artifacts
-          path: snapshot
-
-      - name: Run Acceptance Tests (Linux)
-        run: make acceptance-linux
-
-  # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
-  Acceptance-Mac:
-    needs: [ Build-Snapshot-Artifacts ]
-    runs-on: macos-latest
-    steps:
-
-      - uses: actions/checkout@v2
-
-      - uses: actions/download-artifact@v2
-        with:
-          name: artifacts
-          path: snapshot
-
-      - name: Run Acceptance Tests (Mac)
-        run: make acceptance-mac
-
-  # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
-  Inline-Compare:
-    needs: [ Build-Snapshot-Artifacts ]
-    runs-on: ubuntu-latest
-    steps:
-
-      - uses: actions/checkout@v2
-
-      - name: Fingerprint inline-compare sources
-        run: make compare-fingerprint
-
-      - name: Restore inline reports cache
-        id: cache
-        uses: actions/cache@v2
-        with:
-          path: ${{ github.workspace }}/test/inline-compare/inline-reports
-          key: inline-reports-${{ hashFiles('**/inline-compare.fingerprint') }}
-
-      - uses: actions/download-artifact@v2
-        with:
-          name: artifacts
-          path: snapshot
-
-      - name: Compare Anchore inline-scan results against snapshot build output
-        run: make compare-snapshot

.github/workflows/codeql-analysis.yml

@@ -0,0 +1,62 @@
+# For most projects, this workflow file will not need changing; you simply need
+# to commit it to your repository.
+#
+# You may wish to alter this file to override the set of languages analyzed,
+# or to provide custom queries or build logic.
+name: "CodeQL Security Scan"
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    # The branches below must be a subset of the branches above
+    branches: [main]
+  schedule:
+    - cron: '0 0 * * 3'
+
+jobs:
+  analyze:
+    name: Analyze
+    runs-on: ubuntu-latest
+
+    strategy:
+      fail-fast: false
+      matrix:
+        # Override automatic language detection by changing the below list
+        # Supported options are ['csharp', 'cpp', 'go', 'java', 'javascript', 'python']
+        language: ['go', 'python']
+        # Learn more...
+        # https://docs.github.com/en/github/finding-security-vulnerabilities-and-errors-in-your-code/configuring-code-scanning#overriding-automatic-language-detection
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v2
+
+    # Initializes the CodeQL tools for scanning.
+    - name: Initialize CodeQL
+      uses: github/codeql-action/init@v1
+      with:
+        languages: ${{ matrix.language }}
+        # If you wish to specify custom queries, you can do so here or in a config file.
+        # By default, queries listed here will override any specified in a config file. 
+        # Prefix the list here with "+" to use these queries and those in the config file.
+        # queries: ./path/to/local/query, your-org/your-repo/queries@main
+
+    # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
+    # If this step fails, then you should remove it and run the build manually (see below)
+    - name: Autobuild
+      uses: github/codeql-action/autobuild@v1
+
+    # ℹ️ Command-line programs to run using the OS shell.
+    # 📚 https://git.io/JvXDl
+
+    # ✏️ If the Autobuild fails above, remove it and uncomment the following three lines
+    #    and modify them (or add more) to build your code if your project
+    #    uses a compiled language
+
+    #- run: |
+    #   make bootstrap
+    #   make release
+
+    - name: Perform CodeQL Analysis
+      uses: github/codeql-action/analyze@v1

.github/workflows/release.yaml

@@ -1,21 +1,21 @@
-name: 'Release'
+name: "Release"
 on:
   push:
     # take no actions on push to any branch...
     branches-ignore:
-      - '**'
+      - "**"
     # ... only act on release tags
     tags:
-      - 'v*'
+      - "v*"
 
 env:
-  GO_VERSION: "1.14.x"
+  GO_VERSION: "1.16.x"
 
 jobs:
-  wait-for-checks:
-    runs-on: ubuntu-latest
+  quality-gate:
+    environment: release
+    runs-on: ubuntu-latest # This OS choice is arbitrary. None of the steps in this job are specific to either Linux or macOS.
     steps:
-
       - uses: actions/checkout@v2
 
       # we don't want to release commits that have been pushed and tagged, but not necessarily merged onto main
@@ -25,13 +25,31 @@ jobs:
           git fetch origin main
           git merge-base --is-ancestor ${GITHUB_REF##*/} origin/main && echo "${GITHUB_REF##*/} is a commit on main!"
 
-      - name: Check static anaylysis, unit, and integration test results
+      - name: Check static analysis results
         uses: fountainhead/action-wait-for-check@v1.0.0
-        id: sa-unit-int
+        id: static-analysis
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
-          # This check name is defined as the circle-ci workflow name (in .circleci/config.yaml)
-          checkName: "Static Analysis + Unit + Integration"
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "Static analysis"
+          ref: ${{ github.event.pull_request.head.sha || github.sha }}
+
+      - name: Check unit test results
+        uses: fountainhead/action-wait-for-check@v1.0.0
+        id: unit
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "Unit tests"
+          ref: ${{ github.event.pull_request.head.sha || github.sha }}
+
+      - name: Check integration test results
+        uses: fountainhead/action-wait-for-check@v1.0.0
+        id: integration
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "Integration tests"
           ref: ${{ github.event.pull_request.head.sha || github.sha }}
 
       - name: Check acceptance test results (linux)
@@ -39,8 +57,8 @@ jobs:
         id: acceptance-linux
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
-          # This check name is defined as the github action job name (in .github/workflows/acceptance-test.yaml)
-          checkName: "Acceptance-Linux"
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "Acceptance tests (Linux)"
           ref: ${{ github.event.pull_request.head.sha || github.sha }}
 
       - name: Check acceptance test results (mac)
@@ -48,38 +66,35 @@ jobs:
         id: acceptance-mac
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
-          # This check name is defined as the github action job name (in .github/workflows/acceptance-test.yaml)
-          checkName: "Acceptance-Mac"
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "Acceptance tests (Mac)"
           ref: ${{ github.event.pull_request.head.sha || github.sha }}
 
-      - name: Check inline comparison test results
+      - name: Check cli test results (linux)
         uses: fountainhead/action-wait-for-check@v1.0.0
-        id: inline-compare
+        id: cli-linux
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
-          # This check name is defined as the github action job name (in .github/workflows/acceptance-test.yaml)
-          checkName: "Inline-Compare"
+          # This check name is defined as the github action job name (in .github/workflows/testing.yaml)
+          checkName: "CLI tests (Linux)"
           ref: ${{ github.event.pull_request.head.sha || github.sha }}
 
       - name: Quality gate
-        if: steps.sa-unit-int.outputs.conclusion != 'success' || steps.inline-compare.outputs.conclusion != 'success' || steps.acceptance-linux.outputs.conclusion != 'success' || steps.acceptance-mac.outputs.conclusion != 'success'
+        if: steps.static-analysis.outputs.conclusion != 'success' || steps.unit.outputs.conclusion != 'success' || steps.integration.outputs.conclusion != 'success' || steps.cli-linux.outputs.conclusion != 'success' || steps.acceptance-linux.outputs.conclusion != 'success' || steps.acceptance-mac.outputs.conclusion != 'success'
         run: |
-          echo "Static/Unit/Integration Status: ${{ steps.sa-unit-int.outputs.conclusion }}"
+          echo "Static Analysis Status: ${{ steps.static-analysis.conclusion }}"
+          echo "Unit Test Status: ${{ steps.unit.outputs.conclusion }}"
+          echo "Integration Test Status: ${{ steps.integration.outputs.conclusion }}"
           echo "Acceptance Test (Linux) Status: ${{ steps.acceptance-linux.outputs.conclusion }}"
           echo "Acceptance Test (Mac) Status: ${{ steps.acceptance-mac.outputs.conclusion }}"
-          echo "Inline Compare Status: ${{ steps.inline-compare.outputs.conclusion }}"
+          echo "CLI Test (Linux) Status: ${{ steps.cli-linux.outputs.conclusion }}"
           false
 
   release:
-    needs: [ wait-for-checks ]
-    runs-on: ubuntu-latest
+    needs: [quality-gate]
+    runs-on: macos-latest # Due to our code signing process, it's vital that we run our release steps on macOS.
     steps:
-
-      # TODO: remove me after release
-      - name: Configure git for private modules
-        env:
-          TOKEN: ${{ secrets.ANCHORE_GIT_READ_TOKEN }}
-        run: git config --global url."https://anchore:${TOKEN}@github.com".insteadOf "https://github.com"
+      - uses: docker-practice/actions-setup-docker@v1
 
       - uses: actions/setup-go@v2
         with:
@@ -89,21 +104,25 @@ jobs:
         with:
           fetch-depth: 0
 
-      - name: Restore bootstrap cache
-        id: cache
-        uses: actions/cache@v2
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
         with:
-          path: |
-            ~/go/pkg/mod
-            ${{ github.workspace }}/.tmp
-          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}-${{ hashFiles('Makefile') }}
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
           restore-keys: |
-            ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}-
             ${{ runner.os }}-go-${{ env.GO_VERSION }}-
 
-      - name: Bootstrap dependencies
-        if: steps.cache.outputs.cache-hit != 'true'
-        run: make ci-bootstrap
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
 
       - name: Import GPG key
         id: import_gpg
@@ -122,14 +141,33 @@ jobs:
       - name: Build & publish release artifacts
         run: make release
         env:
+          DOCKER_USERNAME: ${{ secrets.TOOLBOX_DOCKER_USER }}
+          DOCKER_PASSWORD: ${{ secrets.TOOLBOX_DOCKER_PASS }}
           GITHUB_TOKEN: ${{ secrets.ANCHORE_GIT_READ_TOKEN }}
           GPG_PRIVATE_KEY: ${{ secrets.SIGNING_GPG_PRIVATE_KEY }}
           PASSPHRASE: ${{ secrets.SIGNING_GPG_PASSPHRASE }}
           SIGNING_FINGERPRINT: ${{ steps.import_gpg.outputs.fingerprint }}
           AWS_ACCESS_KEY_ID: ${{ secrets.TOOLBOX_AWS_ACCESS_KEY_ID }}
           AWS_SECRET_ACCESS_KEY: ${{ secrets.TOOLBOX_AWS_SECRET_ACCESS_KEY }}
+          APPLE_DEVELOPER_ID_CERT: ${{ secrets.APPLE_DEVELOPER_ID_CERT }} # Used during macOS code signing.
+          APPLE_DEVELOPER_ID_CERT_PASS: ${{ secrets.APPLE_DEVELOPER_ID_CERT_PASS }} # Used during macOS code signing.
+          AC_USERNAME: ${{ secrets.ENG_CI_APPLE_ID }} # Used during macOS notarization.
+          AC_PASSWORD: ${{ secrets.ENG_CI_APPLE_ID_PASS }} # Used during macOS notarization.
+
+      - uses: anchore/sbom-action@v0
+        with:
+          artifact-name: sbom.spdx.json
+
+      - uses: 8398a7/action-slack@v3
+        with:
+          status: ${{ job.status }}
+          fields: repo,workflow,action,eventName
+          text: "A new Syft release is ready to be manually published: https://github.com/anchore/syft/releases"
+        env:
+          SLACK_WEBHOOK_URL: ${{ secrets.SLACK_TOOLBOX_WEBHOOK_URL }}
+        if: ${{ success() }}
 
       - uses: actions/upload-artifact@v2
         with:
           name: artifacts
-          path: dist/**/*
+          path: dist/**/*

.github/workflows/validations.yaml

@@ -0,0 +1,345 @@
+name: "Validations"
+on:
+  workflow_dispatch:
+  push:
+  pull_request:
+
+env:
+  GO_VERSION: "1.16.x"
+
+jobs:
+
+  Static-Analysis:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "Static analysis"
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
+
+      - name: Bootstrap CI environment dependencies
+        run: make ci-bootstrap
+
+      - name: Run static analysis
+        run: make static-analysis
+
+  Unit-Test:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "Unit tests"
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
+
+      - name: Bootstrap CI environment dependencies
+        run: make ci-bootstrap
+
+      - name: Build cache key for java test-fixture blobs (for unit tests)
+        run: make java-packages-fingerprint
+
+      - name: Restore Java test-fixture cache
+        id: unit-java-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: syft/pkg/cataloger/java/test-fixtures/java-builds/packages
+          key: ${{ runner.os }}-unit-java-cache-${{ hashFiles( 'syft/pkg/cataloger/java/test-fixtures/java-builds/packages.fingerprint' ) }}
+
+      - name: Run unit tests
+        run: make unit
+
+      - uses: actions/upload-artifact@v2
+        with:
+          name: unit-test-results
+          path: test/results/**/*
+
+  Integration-Test:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "Integration tests"
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
+
+      - name: Bootstrap CI environment dependencies
+        run: make ci-bootstrap
+
+      - name: Validate syft output against the CycloneDX schema
+        run: make validate-cyclonedx-schema
+
+      - name: Build key for tar cache
+        run: make integration-fingerprint
+
+      - name: Restore integration test cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/test/integration/test-fixtures/cache
+          key: ${{ runner.os }}-integration-test-cache-${{ hashFiles('test/integration/test-fixtures/cache.fingerprint') }}
+
+      - name: Run integration tests
+        run: make integration
+
+  Benchmark-Test:
+    name: "Benchmark tests"
+    runs-on: ubuntu-20.04
+    # note: we want benchmarks to run on pull_request events in order to publish results to a sticky comment, and
+    # we also want to run on push such that merges to main are recorded to the cache. For this reason we don't filter
+    # the job by event.
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
+
+      - name: Bootstrap CI environment dependencies
+        run: make ci-bootstrap
+
+      - name: Restore base benchmark result
+        uses: actions/cache@v2
+        with:
+          path: test/results/benchmark-main.txt
+          # use base sha for PR or new commit hash for main push in benchmark result key
+          key: ${{ runner.os }}-bench-${{ (github.event.pull_request.base.sha != github.event.after) && github.event.pull_request.base.sha || github.event.after }}
+
+      - name: Run benchmark tests
+        id: benchmark
+        run: |
+          REF_NAME=${GITHUB_REF##*/} make benchmark
+          OUTPUT=$(make show-benchstat)
+          OUTPUT="${OUTPUT//'%'/'%25'}"   # URL encode all '%' characters
+          OUTPUT="${OUTPUT//$'\n'/'%0A'}" # URL encode all '\n' characters
+          OUTPUT="${OUTPUT//$'\r'/'%0D'}" # URL encode all '\r' characters
+          echo "::set-output name=result::$OUTPUT"
+
+      - uses: actions/upload-artifact@v2
+        with:
+          name: benchmark-test-results
+          path: test/results/**/*
+
+      - name: Update PR benchmark results comment
+        uses: marocchino/sticky-pull-request-comment@v2
+        continue-on-error: true
+        with:
+          header: benchmark
+          message: |
+            ### Benchmark Test Results
+
+            <details>
+              <summary>Benchmark results from the latest changes vs base branch</summary>
+
+            ```
+            ${{ steps.benchmark.outputs.result }}
+            ```
+
+            </details>
+
+  Build-Snapshot-Artifacts:
+    name: "Build snapshot artifacts"
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v1
+
+      - name: Restore tool cache
+        id: tool-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/.tmp
+          key: ${{ runner.os }}-tool-${{ hashFiles('Makefile') }}
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap all project dependencies
+        if: steps.tool-cache.outputs.cache-hit != 'true' || steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap
+
+      - name: Build snapshot artifacts
+        run: make snapshot
+
+      - uses: actions/upload-artifact@v2
+        with:
+          name: artifacts
+          path: snapshot/**/*
+
+  Acceptance-Linux:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "Acceptance tests (Linux)"
+    needs: [Build-Snapshot-Artifacts]
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/checkout@v2
+
+      - uses: actions/download-artifact@v2
+        with:
+          name: artifacts
+          path: snapshot
+
+      - name: Run Acceptance Tests (Linux)
+        run: make acceptance-linux
+
+  Acceptance-Mac:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "Acceptance tests (Mac)"
+    needs: [Build-Snapshot-Artifacts]
+    runs-on: macos-latest
+    steps:
+      - uses: actions/checkout@v2
+
+      - uses: actions/download-artifact@v2
+        with:
+          name: artifacts
+          path: snapshot
+
+      - name: Restore docker image cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: image.tar
+          key: ${{ runner.os }}-${{ hashFiles('test/acceptance/mac.sh') }}
+
+      - name: Run Acceptance Tests (Mac)
+        run: make acceptance-mac
+
+  Cli-Linux:
+    # Note: changing this job name requires making the same update in the .github/workflows/release.yaml pipeline
+    name: "CLI tests (Linux)"
+    needs: [Build-Snapshot-Artifacts]
+    runs-on: ubuntu-20.04
+    steps:
+      - uses: actions/setup-go@v2
+        with:
+          go-version: ${{ env.GO_VERSION }}
+
+      - uses: actions/checkout@v2
+
+      - name: Restore go cache
+        id: go-cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ env.GO_VERSION }}-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-${{ env.GO_VERSION }}-
+
+      - name: (cache-miss) Bootstrap go dependencies
+        if: steps.go-cache.outputs.cache-hit != 'true'
+        run: make bootstrap-go
+
+      - name: Build key for tar cache
+        run: make cli-fingerprint
+
+      - name: Restore CLI test cache
+        uses: actions/cache@v2.1.3
+        with:
+          path: ${{ github.workspace }}/test/cli/test-fixtures/cache
+          key: ${{ runner.os }}-cli-test-cache-${{ hashFiles('test/cli/test-fixtures/cache.fingerprint') }}
+
+      - uses: actions/download-artifact@v2
+        with:
+          name: artifacts
+          path: snapshot
+
+      - name: Run CLI Tests (Linux)
+        run: make cli

.gitignore

@@ -1,7 +1,10 @@
+CHANGELOG.md
+/test/results
 /dist
 /snapshot
 .server/
 .vscode/
+.history/
 *.fingerprint
 *.tar
 *.jar
@@ -28,3 +31,8 @@ coverage.txt
 
 # Output of the go coverage tool, specifically when used with LiteIDE
 *.out
+
+# macOS Finder metadata
+.DS_STORE
+
+*.profile

.golangci.yaml

@@ -1,3 +1,9 @@
+# TODO: enable this when we have coverage on docstring comments
+#issues:
+#  # The list of ids of default excludes to include or disable.
+#  include:
+#    - EXC0002 # disable excluding of issues about comments from golint
+
 linters:
   # inverted configuration with `enable-all` and `disable` is not scalable during updates of golangci-lint
   disable-all: true
@@ -9,6 +15,7 @@ linters:
     - dogsled
     - dupl
     - errcheck
+    - exportloopref
     - funlen
     - gocognit
     - goconst
@@ -16,7 +23,6 @@ linters:
     - gocyclo
     - gofmt
     - goimports
-    - golint
     - goprintffuncname
     - gosec
     - gosimple
@@ -25,9 +31,8 @@ linters:
     - misspell
     - nakedret
     - nolintlint
-    - prealloc
+    - revive
     - rowserrcheck
-    - scopelint
     - staticcheck
     - structcheck
     - stylecheck
@@ -44,10 +49,13 @@ linters:
 #    - godot
 #    - godox
 #    - goerr113
+#    - golint      # deprecated
 #    - gomnd       # this is too aggressive
 #    - interfacer  # this is a good idea, but is no longer supported and is prone to false positives
 #    - lll         # without a way to specify per-line exception cases, this is not usable
 #    - maligned    # this is an excellent linter, but tricky to optimize and we are not sensitive to memory layout optimizations
 #    - nestif
+#    - prealloc    # following this rule isn't consistently a good idea, as it sometimes forces unnecessary allocations that result in less idiomatic code
+#    - scopelint   # deprecated
 #    - testpackage
-#    - wsl
+#    - wsl         # this doens't have an auto-fixer yet and is pretty noisy (https://github.com/bombsimon/wsl/issues/90)

.goreleaser.yaml

@@ -1,22 +1,38 @@
 release:
-  # If set to auto, will mark the release as not ready for production
-  # in case there is an indicator for this in the tag e.g. v1.0.0-rc1
-  # If set to true, will mark the release as not ready for production.
-  prerelease: false
+  # If set to auto, will mark the release as not ready for production in case there is an indicator for this in the
+  # tag e.g. v1.0.0-rc1 .If set to true, will mark the release as not ready for production.
+  prerelease: auto
 
-signs:
-  - artifacts: checksum
-    args: ["--output", "${signature}", "--detach-sign", "${artifact}"]
+  # If set to true, will not auto-publish the release. This is done to allow us to review the changelog before publishing.
+  draft: true
 
 builds:
   - binary: syft
+    id: syft
     env:
       - CGO_ENABLED=0
     goos:
-      # windows not supported yet (due to jotframe)
-      # - windows
       - linux
-      - darwin
+    goarch:
+      - amd64
+      - arm64
+    # Set the modified timestamp on the output binary to the git timestamp (to ensure a reproducible build)
+    mod_timestamp: '{{ .CommitTimestamp }}'
+    ldflags: |
+      -w
+      -s
+      -extldflags '-static'
+      -X github.com/anchore/syft/internal/version.version={{.Version}}
+      -X github.com/anchore/syft/internal/version.gitCommit={{.Commit}}
+      -X github.com/anchore/syft/internal/version.buildDate={{.Date}}
+      -X github.com/anchore/syft/internal/version.gitTreeState={{.Env.BUILD_GIT_TREE_STATE}}
+
+  - binary: syft
+    id: syft-win
+    env:
+      - CGO_ENABLED=0
+    goos:
+      - windows
     goarch:
       - amd64
     # Set the modified timestamp on the output binary to the git timestamp (to ensure a reproducible build)
@@ -30,6 +46,56 @@ builds:
       -X github.com/anchore/syft/internal/version.buildDate={{.Date}}
       -X github.com/anchore/syft/internal/version.gitTreeState={{.Env.BUILD_GIT_TREE_STATE}}
 
+    # For more info on this macOS build, see: https://github.com/mitchellh/gon#usage-with-goreleaser
+  - binary: syft
+    id: syft-macos
+    env:
+      - CGO_ENABLED=0
+    goos:
+      - darwin
+    goarch:
+      - amd64
+      - arm64
+    # Set the modified timestamp on the output binary to the git timestamp (to ensure a reproducible build)
+    mod_timestamp: '{{ .CommitTimestamp }}'
+    ldflags: |
+      -w
+      -s
+      -extldflags '-static'
+      -X github.com/anchore/syft/internal/version.version={{.Version}}
+      -X github.com/anchore/syft/internal/version.gitCommit={{.Commit}}
+      -X github.com/anchore/syft/internal/version.buildDate={{.Date}}
+      -X github.com/anchore/syft/internal/version.gitTreeState={{.Env.BUILD_GIT_TREE_STATE}}
+
+archives:
+  - format: tar.gz
+    builds:
+      - syft # i.e. Linux only
+  - format: zip # This is a hack for syft-macos! We don't actually intend to use _this_ ZIP file, we just need goreleaser to consider the ZIP file produced by gon (which will have the same file name) to be an artifact so we can use it downstream in publishing (e.g. to a homebrew tap)
+    id: syft-zip
+    builds:
+      - syft-macos
+      - syft-win
+
+signs:
+  - artifacts: checksum
+    cmd: sh
+    args:
+      - '-c'
+      # we should not include the zip artifact, as the artifact is mutated throughout the next macOS notarization step
+      # note: sed -i is not portable
+      - 'sed "/.*\.zip/d" ${artifact} > tmpfile && mv tmpfile ${artifact} && gpg --output ${signature} --detach-sign ${artifact}'
+  - id: syft-macos-signing
+    ids:
+      - syft-macos
+    cmd: ./.github/scripts/mac-sign-and-notarize.sh
+    signature: "syft_${VERSION}_darwin_amd64.dmg" # This is somewhat unintuitive. This gets the DMG file recognized as an artifact. In fact, both a DMG and a ZIP file are being produced by this signing step.
+    args:
+      - "{{ .IsSnapshot }}"
+      - "gon.hcl"
+      - "./dist/syft_{{ .Version }}_darwin_amd64"
+    artifacts: all
+
 nfpms:
   - license: "Apache 2.0"
     maintainer: "Anchore, Inc"
@@ -46,9 +112,49 @@ brews:
     homepage: *website
     description: *description
 
-archives:
-  - format: tar.gz
-    format_overrides:
-      - goos: windows
-        format: zip
+dockers:
+  - image_templates:
+      - "anchore/syft:latest"
+      - "anchore/syft:{{ .Tag }}-amd64"
+      - "anchore/syft:v{{ .Major }}-amd64"
+      - "anchore/syft:v{{ .Major }}.{{ .Minor }}-amd64"
+    dockerfile: Dockerfile
+    use: buildx
+    build_flag_templates:
+      - "--platform=linux/amd64"
+      - "--build-arg=BUILD_DATE={{.Date}}"
+      - "--build-arg=BUILD_VERSION={{.Version}}"
+      - "--build-arg=VCS_REF={{.FullCommit}}"
+      - "--build-arg=VCS_URL={{.GitURL}}"
 
+  - image_templates:
+      - "anchore/syft:{{ .Tag }}-arm64v8"
+      - "anchore/syft:v{{ .Major }}-arm64v8"
+      - "anchore/syft:v{{ .Major }}.{{ .Minor }}-arm64v8"
+    goarch: arm64
+    dockerfile: Dockerfile
+    use: buildx
+    build_flag_templates:
+      - "--platform=linux/arm64/v8"
+      - "--build-arg=BUILD_DATE={{.Date}}"
+      - "--build-arg=BUILD_VERSION={{.Version}}"
+      - "--build-arg=VCS_REF={{.FullCommit}}"
+      - "--build-arg=VCS_URL={{.GitURL}}"
+
+docker_manifests:
+  - name_template: anchore/syft:{{ .Tag }}
+    image_templates:
+      - anchore/syft:{{ .Tag }}-amd64
+      - anchore/syft:v{{ .Major }}-amd64
+      - anchore/syft:v{{ .Major }}.{{ .Minor }}-amd64
+      - anchore/syft:{{ .Tag }}-arm64v8
+      - anchore/syft:v{{ .Major }}-arm64v8
+      - anchore/syft:v{{ .Major }}.{{ .Minor }}-arm64v8
+  - name_template: anchore/syft:latest
+    image_templates:
+      - anchore/syft:{{ .Tag }}-amd64
+      - anchore/syft:v{{ .Major }}-amd64
+      - anchore/syft:v{{ .Major }}.{{ .Minor }}-amd64
+      - anchore/syft:{{ .Tag }}-arm64v8
+      - anchore/syft:v{{ .Major }}-arm64v8
+      - anchore/syft:v{{ .Major }}.{{ .Minor }}-arm64v8

CONTRIBUTING.md

@@ -1,12 +1,12 @@
-# Contributing to `syft`
+# Contributing to Syft
 
-If you are looking to contribute to this project and want to open a Github pull request ("PR"), there are a few guidelines of what we are looking for in patches. Make sure you go through this document and ensure that your code proposal is aligned.
+If you are looking to contribute to this project and want to open a GitHub pull request ("PR"), there are a few guidelines of what we are looking for in patches. Make sure you go through this document and ensure that your code proposal is aligned.
 
-## Sign your work
+## Sign off your work
 
 The `sign-off` is an added line at the end of the explanation for the commit, certifying that you wrote it or otherwise have the right to submit it as an open-source patch. By submitting a contribution, you agree to be bound by the terms of the DCO Version 1.1 and Apache License Version 2.0.
 
-Signing a commit certifies the below Developer's Certificate of Origin (DCO):
+Signing off a commit certifies the below Developer's Certificate of Origin (DCO):
 
 ```text
 Developer's Certificate of Origin 1.1
@@ -51,13 +51,13 @@ $ git config --global user.name "John Doe"
 $ git config --global user.email "john.doe@example.com"
 ```
 
-Creating a signed commit is then possible with `-s`:
+Creating a signed-off commit is then possible with `-s` or `--signoff`:
 
 ```text
 $ git commit -s -m "this is a commit message"
 ```
 
-To double check if the commit was signed, look at the log output:
+To double-check that the commit was signed-off, look at the log output:
 
 ```text
 $ git log -1
@@ -82,6 +82,11 @@ Date:   Mon Aug 1 11:27:13 2020 -0400
 
 [//]: # (TODO: PR title and description)
 
+## Sign your commits
+
+To ensure the authenticity and integrity of code contributions, **we require that all commits are signed**. Signing commits proves that your commits were truly created by you, as the holder of a private key.
+
+Configuring git to sign your commits is a straightforward process. To get set up, see [GitHub's documentation on signing commits](https://docs.github.com/en/authentication/managing-commit-signature-verification/signing-commits).
 
 ## Test your changes
 

Dockerfile

@@ -0,0 +1,28 @@
+FROM alpine:latest AS build
+
+RUN apk --no-cache add ca-certificates
+
+FROM scratch
+# needed for version check HTTPS request
+COPY --from=build /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certificates.crt
+
+# create the /tmp dir, which is needed for image content cache
+WORKDIR /tmp
+
+COPY syft /
+
+ARG BUILD_DATE
+ARG BUILD_VERSION
+ARG VCS_REF
+ARG VCS_URL
+
+LABEL org.label-schema.schema-version="1.0"
+LABEL org.label-schema.build-date=$BUILD_DATE
+LABEL org.label-schema.name="syft"
+LABEL org.label-schema.description="CLI tool and library for generating a Software Bill of Materials from container images and filesystems"
+LABEL org.label-schema.vcs-url=$VCS_URL
+LABEL org.label-schema.vcs-ref=$VCS_REF
+LABEL org.label-schema.vendor="Anchore, Inc."
+LABEL org.label-schema.version=$BUILD_VERSION
+
+ENTRYPOINT ["/syft"]

Makefile

@@ -1,9 +1,9 @@
 BIN = syft
 TEMPDIR = ./.tmp
-RESULTSDIR = $(TEMPDIR)/results
-COVER_REPORT = $(RESULTSDIR)/cover.report
-COVER_TOTAL = $(RESULTSDIR)/cover.total
-LINTCMD = $(TEMPDIR)/golangci-lint run --tests=false --config .golangci.yaml
+RESULTSDIR = test/results
+COVER_REPORT = $(RESULTSDIR)/unit-coverage-details.txt
+COVER_TOTAL = $(RESULTSDIR)/unit-coverage-summary.txt
+LINTCMD = $(TEMPDIR)/golangci-lint run --tests=false --timeout=2m --config .golangci.yaml
 ACC_TEST_IMAGE = centos:8.2.2004
 ACC_DIR = ./test/acceptance
 BOLD := $(shell tput -T linux bold)
@@ -15,13 +15,23 @@ RESET := $(shell tput -T linux sgr0)
 TITLE := $(BOLD)$(PURPLE)
 SUCCESS := $(BOLD)$(GREEN)
 # the quality gate lower threshold for unit test total % coverage (by function statements)
-COVERAGE_THRESHOLD := 72
+COVERAGE_THRESHOLD := 62
+# CI cache busting values; change these if you want CI to not use previous stored cache
+INTEGRATION_CACHE_BUSTER="88738d2f"
+CLI_CACHE_BUSTER="9a2c03cf"
+BOOTSTRAP_CACHE="c7afb99ad"
 
 ## Build variables
 DISTDIR=./dist
 SNAPSHOTDIR=./snapshot
 GITTREESTATE=$(if $(shell git status --porcelain),dirty,clean)
-SNAPSHOT_CMD=$(shell realpath $(shell pwd)/$(SNAPSHOTDIR)/syft_linux_amd64/syft)
+OS := $(shell uname)
+
+ifeq ($(OS),Darwin)
+	SNAPSHOT_CMD=$(shell realpath $(shell pwd)/$(SNAPSHOTDIR)/$(BIN)-macos_darwin_amd64/$(BIN))
+else
+	SNAPSHOT_CMD=$(shell realpath $(shell pwd)/$(SNAPSHOTDIR)/$(BIN)_linux_amd64/$(BIN))
+endif
 
 ifeq "$(strip $(VERSION))" ""
  override VERSION = $(shell git describe --always --tags --dirty)
@@ -49,6 +59,10 @@ ifndef SNAPSHOTDIR
 	$(error SNAPSHOTDIR is not set)
 endif
 
+ifndef REF_NAME
+	REF_NAME = $(VERSION)
+endif
+
 define title
     @printf '$(TITLE)$(1)$(RESET)\n'
 endef
@@ -60,32 +74,44 @@ all: clean static-analysis test ## Run all linux-based checks (linting, license
 	@printf '$(SUCCESS)All checks pass!$(RESET)\n'
 
 .PHONY: test
-test: unit integration acceptance-linux ## Run all tests (currently unit, integration, and linux acceptance tests)
+test: unit validate-cyclonedx-schema integration benchmark acceptance-linux cli ## Run all tests (currently unit, integration, linux acceptance, and cli tests)
 
 .PHONY: help
 help:
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "$(BOLD)$(CYAN)%-25s$(RESET)%s\n", $$1, $$2}'
 
 .PHONY: ci-bootstrap
-ci-bootstrap: bootstrap
-	sudo apt install -y bc
+ci-bootstrap:
+	DEBIAN_FRONTEND=noninteractive sudo apt update && sudo -E apt install -y bc jq libxml2-utils
+
+.PHONY:
+ci-bootstrap-mac:
+	github_changelog_generator --version || sudo gem install github_changelog_generator
+
+$(RESULTSDIR):
+	mkdir -p $(RESULTSDIR)
+
+$(TEMPDIR):
+	mkdir -p $(TEMPDIR)
+
+.PHONY: bootstrap-tools
+bootstrap-tools: $(TEMPDIR)
+	GO111MODULE=off GOBIN=$(shell realpath $(TEMPDIR)) go get -u golang.org/x/perf/cmd/benchstat
+	curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(TEMPDIR)/ v1.42.1
+	curl -sSfL https://raw.githubusercontent.com/wagoodman/go-bouncer/master/bouncer.sh | sh -s -- -b $(TEMPDIR)/ v0.2.0
+	curl -sSfL https://raw.githubusercontent.com/anchore/chronicle/main/install.sh | sh -s -- -b $(TEMPDIR)/ v0.3.0
+	.github/scripts/goreleaser-install.sh -b $(TEMPDIR)/ v0.177.0
+
+.PHONY: bootstrap-go
+bootstrap-go:
+	go mod download
 
 .PHONY: bootstrap
-bootstrap: ## Download and install all go dependencies (+ prep tooling in the ./tmp dir)
+bootstrap: $(RESULTSDIR) bootstrap-go bootstrap-tools ## Download and install all go dependencies (+ prep tooling in the ./tmp dir)
 	$(call title,Bootstrapping dependencies)
-	@pwd
-	# prep temp dirs
-	mkdir -p $(TEMPDIR)
-	mkdir -p $(RESULTSDIR)
-	# install go dependencies
-	go mod download
-	# install utilities
-	[ -f "$(TEMPDIR)/golangci" ] || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(TEMPDIR)/ v1.26.0
-	[ -f "$(TEMPDIR)/bouncer" ] || curl -sSfL https://raw.githubusercontent.com/wagoodman/go-bouncer/master/bouncer.sh | sh -s -- -b $(TEMPDIR)/ v0.2.0
-	[ -f "$(TEMPDIR)/goreleaser" ] || curl -sfL https://install.goreleaser.com/github.com/goreleaser/goreleaser.sh | sh -s -- -b $(TEMPDIR)/ v0.140.0
 
 .PHONY: static-analysis
-static-analysis: lint check-licenses
+static-analysis: lint check-go-mod-tidy check-licenses
 
 .PHONY: lint
 lint: ## Run gofmt + golangci lint checks
@@ -106,64 +132,68 @@ lint-fix: ## Auto-format all source code + run golangci lint fixers
 	$(call title,Running lint fixers)
 	gofmt -w -s .
 	$(LINTCMD) --fix
+	go mod tidy
 
 .PHONY: check-licenses
-check-licenses:
+check-licenses: ## Ensure transitive dependencies are compliant with the current license policy
 	$(TEMPDIR)/bouncer check
 
+check-go-mod-tidy:
+	@ .github/scripts/go-mod-tidy-check.sh && echo "go.mod and go.sum are tidy!"
+
+.PHONY: validate-cyclonedx-schema
+validate-cyclonedx-schema:
+	cd schema/cyclonedx && make
+
 .PHONY: unit
-unit: fixtures ## Run unit tests (with coverage)
+unit: $(RESULTSDIR) fixtures ## Run unit tests (with coverage)
 	$(call title,Running unit tests)
-	go test -coverprofile $(COVER_REPORT) ./...
+	go test  -coverprofile $(COVER_REPORT) $(shell go list ./... | grep -v anchore/syft/test)
 	@go tool cover -func $(COVER_REPORT) | grep total |  awk '{print substr($$3, 1, length($$3)-1)}' > $(COVER_TOTAL)
 	@echo "Coverage: $$(cat $(COVER_TOTAL))"
 	@if [ $$(echo "$$(cat $(COVER_TOTAL)) >= $(COVERAGE_THRESHOLD)" | bc -l) -ne 1 ]; then echo "$(RED)$(BOLD)Failed coverage quality gate (> $(COVERAGE_THRESHOLD)%)$(RESET)" && false; fi
 
+.PHONY: benchmark
+benchmark: $(RESULTSDIR) ## Run benchmark tests and compare against the baseline (if available)
+	$(call title,Running benchmark tests)
+	go test -p 1 -run=^Benchmark -bench=. -count=5 -benchmem ./... | tee $(RESULTSDIR)/benchmark-$(REF_NAME).txt
+	(test -s $(RESULTSDIR)/benchmark-main.txt && \
+		$(TEMPDIR)/benchstat $(RESULTSDIR)/benchmark-main.txt $(RESULTSDIR)/benchmark-$(REF_NAME).txt || \
+		$(TEMPDIR)/benchstat $(RESULTSDIR)/benchmark-$(REF_NAME).txt) \
+			| tee $(RESULTSDIR)/benchstat.txt
+
+.PHONY: show-benchstat
+show-benchstat:
+	@cat $(RESULTSDIR)/benchstat.txt
+
 .PHONY: integration
 integration: ## Run integration tests
 	$(call title,Running integration tests)
-	go test -v -tags=integration ./test/integration
+
+	go test -v ./test/integration
 
 # note: this is used by CI to determine if the integration test fixture cache (docker image tars) should be busted
 integration-fingerprint:
-	find test/integration/test-fixtures/image-* -type f -exec md5sum {} + | awk '{print $1}' | sort | md5sum | tee test/integration/test-fixtures/tar-cache.fingerprint
+	find test/integration/test-fixtures/image-* -type f -exec md5sum {} + | awk '{print $1}' | sort | md5sum | tee test/integration/test-fixtures/cache.fingerprint && echo "$(INTEGRATION_CACHE_BUSTER)" >> test/integration/test-fixtures/cache.fingerprint
 
 .PHONY: java-packages-fingerprint
 java-packages-fingerprint:
-	@cd syft/cataloger/java/test-fixtures/java-builds && \
+	@cd syft/pkg/cataloger/java/test-fixtures/java-builds && \
 	make packages.fingerprint
 
 .PHONY: fixtures
 fixtures:
 	$(call title,Generating test fixtures)
-	cd syft/cataloger/java/test-fixtures/java-builds && make
+	cd syft/pkg/cataloger/java/test-fixtures/java-builds && make
 
 .PHONY: generate-json-schema
-generate-json-schema: clean-json-schema-examples integration ## Generate a new json schema for the json presenter, derived from integration test cases
-	docker run \
-		-i \
-		--rm \
-		-v $(shell pwd)/json-schema:/work \
-		-w /work \
-		python:3.8 \
-			bash -x -c "\
-				pip install -r requirements.txt && \
-				python generate.py \
-			"
+generate-json-schema:  ## Generate a new json schema
+	cd schema/json && go run generate.go
 
-.PHONY: clear-test-cache
-clear-test-cache: ## Delete all test cache (built docker image tars)
-	find . -type f -wholename "**/test-fixtures/tar-cache/*.tar" -delete
-
-.PHONY: check-pipeline
-check-pipeline: ## Run local CircleCI pipeline locally (sanity check)
-	$(call title,Check pipeline)
-	# note: this is meant for local development & testing of the pipeline, NOT to be run in CI
-	mkdir -p $(TEMPDIR)
-	circleci config process .circleci/config.yml > .tmp/circleci.yml
-	circleci local execute -c .tmp/circleci.yml --job "Static Analysis"
-	circleci local execute -c .tmp/circleci.yml --job "Unit & Integration Tests (go-latest)"
-	@printf '$(SUCCESS)Pipeline checks pass!$(RESET)\n'
+.PHONY: generate-license-list
+generate-license-list: ## Generate an updated spdx license list
+	go generate ./internal/spdxlicense/...
+	gofmt -s -w ./internal/spdxlicense
 
 .PHONY: build
 build: $(SNAPSHOTDIR) ## Build release snapshot binaries and packages
@@ -175,11 +205,14 @@ $(SNAPSHOTDIR): ## Build snapshot release binaries and packages
 	cat .goreleaser.yaml >> $(TEMPDIR)/goreleaser.yaml
 
 	# build release snapshots
+	# DOCKER_CLI_EXPERIMENTAL needed to support multi architecture builds for goreleaser
 	BUILD_GIT_TREE_STATE=$(GITTREESTATE) \
-	$(TEMPDIR)/goreleaser release --skip-publish --rm-dist --snapshot --config $(TEMPDIR)/goreleaser.yaml
+	DOCKER_CLI_EXPERIMENTAL=enabled \
+	$(TEMPDIR)/goreleaser release --skip-publish --skip-sign --rm-dist --snapshot --config $(TEMPDIR)/goreleaser.yaml
 
+# note: we cannot clean the snapshot directory since the pipeline builds the snapshot separately
 .PHONY: acceptance-mac
-acceptance-mac: $(SNAPSHOTDIR) ## Run acceptance tests on build snapshot binaries and packages (Mac)
+acceptance-mac: $(RESULTSDIR) $(SNAPSHOTDIR) ## Run acceptance tests on build snapshot binaries and packages (Mac)
 	$(call title,Running acceptance test: Run on Mac)
 	$(ACC_DIR)/mac.sh \
 			$(SNAPSHOTDIR) \
@@ -187,25 +220,12 @@ acceptance-mac: $(SNAPSHOTDIR) ## Run acceptance tests on build snapshot binarie
 			$(ACC_TEST_IMAGE) \
 			$(RESULTSDIR)
 
+# note: we cannot clean the snapshot directory since the pipeline builds the snapshot separately
 .PHONY: acceptance-linux
 acceptance-linux: acceptance-test-deb-package-install acceptance-test-rpm-package-install ## Run acceptance tests on build snapshot binaries and packages (Linux)
 
-# note: this is used by CI to determine if the inline-scan report cache should be busted for the inline-compare tests
-.PHONY: compare-fingerprint
-compare-fingerprint:
-	find test/inline-compare/* -type f -exec md5sum {} + | grep -v '\-reports' | grep -v 'fingerprint' | awk '{print $1}' | sort | md5sum | tee test/inline-compare/inline-compare.fingerprint
-
-.PHONY: compare-snapshot
-compare-snapshot: $(SNAPSHOTDIR) ## Compare the reports of a run of a snapshot build of syft against inline-scan
-	chmod 755 $(SNAPSHOT_CMD)
-	@cd test/inline-compare && SYFT_CMD=$(SNAPSHOT_CMD) make
-
-.PHONY: compare
-compare:  ## Compare the reports of a run of a main-branch build of syft against inline-scan
-	@cd test/inline-compare && make
-
 .PHONY: acceptance-test-deb-package-install
-acceptance-test-deb-package-install: $(SNAPSHOTDIR)
+acceptance-test-deb-package-install: $(RESULTSDIR) $(SNAPSHOTDIR)
 	$(call title,Running acceptance test: DEB install)
 	$(ACC_DIR)/deb.sh \
 			$(SNAPSHOTDIR) \
@@ -214,7 +234,7 @@ acceptance-test-deb-package-install: $(SNAPSHOTDIR)
 			$(RESULTSDIR)
 
 .PHONY: acceptance-test-rpm-package-install
-acceptance-test-rpm-package-install: $(SNAPSHOTDIR)
+acceptance-test-rpm-package-install: $(RESULTSDIR) $(SNAPSHOTDIR)
 	$(call title,Running acceptance test: RPM install)
 	$(ACC_DIR)/rpm.sh \
 			$(SNAPSHOTDIR) \
@@ -222,16 +242,53 @@ acceptance-test-rpm-package-install: $(SNAPSHOTDIR)
 			$(ACC_TEST_IMAGE) \
 			$(RESULTSDIR)
 
+# note: this is used by CI to determine if the integration test fixture cache (docker image tars) should be busted
+cli-fingerprint:
+	find test/cli/test-fixtures/image-* -type f -exec md5sum {} + | awk '{print $1}' | sort | md5sum | tee test/cli/test-fixtures/cache.fingerprint && echo "$(CLI_CACHE_BUSTER)" >> test/cli/test-fixtures/cache.fingerprint
+
+.PHONY: cli
+cli: $(SNAPSHOTDIR) ## Run CLI tests
+	chmod 755 "$(SNAPSHOT_CMD)"
+	$(SNAPSHOT_CMD) version
+	SYFT_BINARY_LOCATION='$(SNAPSHOT_CMD)' \
+		go test -count=1 -v ./test/cli
+
+.PHONY: changelog
+changelog: clean-changelog CHANGELOG.md
+	@docker run -it --rm \
+		-v $(shell pwd)/CHANGELOG.md:/CHANGELOG.md \
+		rawkode/mdv \
+			-t 748.5989 \
+			/CHANGELOG.md
+
+CHANGELOG.md:
+	$(TEMPDIR)/chronicle -vv > CHANGELOG.md
+
 .PHONY: release
-release: clean-dist ## Build and publish final binaries and packages
+release: clean-dist CHANGELOG.md  ## Build and publish final binaries and packages. Intended to be run only on macOS.
 	$(call title,Publishing release artifacts)
+
+	# Prepare for macOS-specific signing process
+	.github/scripts/mac-prepare-for-signing.sh
+
+	# login to docker
+	# note: the previous step creates a new keychain, so it is important to reauth into docker.io
+	@echo $${DOCKER_PASSWORD} | docker login docker.io -u $${DOCKER_USERNAME}  --password-stdin
+
 	# create a config with the dist dir overridden
 	echo "dist: $(DISTDIR)" > $(TEMPDIR)/goreleaser.yaml
 	cat .goreleaser.yaml >> $(TEMPDIR)/goreleaser.yaml
 
-	# release
-	BUILD_GIT_TREE_STATE=$(GITTREESTATE) \
-	$(TEMPDIR)/goreleaser --rm-dist --config $(TEMPDIR)/goreleaser.yaml
+	# release (note the version transformation from v0.7.0 --> 0.7.0)
+	# DOCKER_CLI_EXPERIMENTAL needed to support multi architecture builds for goreleaser
+	bash -c "\
+		BUILD_GIT_TREE_STATE=$(GITTREESTATE) \
+		VERSION=$(VERSION:v%=%) \
+		DOCKER_CLI_EXPERIMENTAL=enabled \
+		$(TEMPDIR)/goreleaser \
+			--rm-dist \
+			--config $(TEMPDIR)/goreleaser.yaml \
+			--release-notes <(cat CHANGELOG.md)"
 
 	# verify checksum signatures
 	.github/scripts/verify-signature.sh "$(DISTDIR)"
@@ -239,8 +296,9 @@ release: clean-dist ## Build and publish final binaries and packages
 	# upload the version file that supports the application version update check (excluding pre-releases)
 	.github/scripts/update-version-file.sh "$(DISTDIR)" "$(VERSION)"
 
+
 .PHONY: clean
-clean: clean-dist clean-snapshot clean-json-schema-examples ## Remove previous builds and result reports
+clean: clean-dist clean-snapshot clean-test-image-cache ## Remove previous builds, result reports, and test cache
 	rm -rf $(RESULTSDIR)/*
 
 .PHONY: clean-snapshot
@@ -248,9 +306,32 @@ clean-snapshot:
 	rm -rf $(SNAPSHOTDIR) $(TEMPDIR)/goreleaser.yaml
 
 .PHONY: clean-dist
-clean-dist:
+clean-dist: clean-changelog
 	rm -rf $(DISTDIR) $(TEMPDIR)/goreleaser.yaml
 
-.PHONY: clean-json-schema-examples
-clean-json-schema-examples:
-	rm json-schema/examples/*
+.PHONY: clean-changelog
+clean-changelog:
+	rm -f CHANGELOG.md
+
+clean-test-image-cache: clean-test-image-tar-cache clean-test-image-docker-cache
+
+.PHONY: clear-test-image-tar-cache
+clean-test-image-tar-cache: ## Delete all test cache (built docker image tars)
+	find . -type f -wholename "**/test-fixtures/cache/stereoscope-fixture-*.tar" -delete
+
+.PHONY: clear-test-image-docker-cache
+clean-test-image-docker-cache: ## Purge all test docker images
+	docker images --format '{{.ID}} {{.Repository}}' | grep stereoscope-fixture- | awk '{print $$1}' | uniq | xargs docker rmi --force
+
+.PHONY: show-test-image-cache
+show-test-image-cache: ## Show all docker and image tar cache
+	$(call title,Docker daemon cache)
+	@docker images --format '{{.ID}} {{.Repository}}:{{.Tag}}' | grep stereoscope-fixture- | sort
+
+	$(call title,Tar cache)
+	@find . -type f -wholename "**/test-fixtures/cache/stereoscope-fixture-*.tar" | sort
+
+.PHONY: show-test-snapshots
+show-test-snapshots: ## Show all test snapshots
+	$(call title,Test snapshots)
+	@find . -type f -wholename "**/test-fixtures/snapshot/*" | sort

README.md

@@ -1,69 +1,199 @@
-# syft
+<p align="center">
+    <img src="https://user-images.githubusercontent.com/5199289/136844524-1527b09f-c5cb-4aa9-be54-5aa92a6086c1.png" width="271" alt="Cute pink owl syft logo">
+</p>
 
-[![CircleCI](https://circleci.com/gh/anchore/syft.svg?style=svg)](https://circleci.com/gh/anchore/syft)
+[![Validations](https://github.com/anchore/syft/actions/workflows/validations.yaml/badge.svg)](https://github.com/anchore/syft/actions/workflows/validations.yaml)
 [![Go Report Card](https://goreportcard.com/badge/github.com/anchore/syft)](https://goreportcard.com/report/github.com/anchore/syft)
 [![GitHub release](https://img.shields.io/github/release/anchore/syft.svg)](https://github.com/anchore/syft/releases/latest)
-[![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://github.com/anchore/syft/blob/master/LICENSE)
+[![GitHub go.mod Go version](https://img.shields.io/github/go-mod/go-version/anchore/syft.svg)](https://github.com/anchore/syft)
+[![License: Apache-2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://github.com/anchore/syft/blob/main/LICENSE)
+[![Slack Invite](https://img.shields.io/badge/Slack-Join-blue?logo=slack)](https://anchore.com/slack)
 
-A CLI tool and go library for generating a Software Bill of Materials (SBOM) from container images and filesystems.
+A CLI tool and Go library for generating a Software Bill of Materials (SBOM) from container images and filesystems. Exceptional for vulnerability detection when used with a scanner tool like [Grype](https://github.com/anchore/grype).
 
-[//]: # (TODO: add example gif here)
+### Join us at the virtual OSS meetup!
 
-**Features**
+When: December 1, 2021 @ 11am-noon PT (2-3pm ET)
+
+We’ve got security experts from VMware and Anchore who will share their tips, tricks, and lessons learned on securing containers using [Syft](https://github.com/anchore/syft), [Grype](https://github.com/anchore/grype), and [Cartographer](https://github.com/vmware-tanzu/cartographer).
+
+Register now: https://get.anchore.com/anchore-oss-meetup-dec-1-2021/
+
+---
+
+![syft-demo](https://user-images.githubusercontent.com/590471/90277200-2a253000-de33-11ea-893f-32c219eea11a.gif)
+
+## Features
 - Catalog container images and filesystems to discover packages and libraries.
 - Supports packages and libraries from various ecosystems (APK, DEB, RPM, Ruby Bundles, Python Wheel/Egg/requirements.txt, JavaScript NPM/Yarn, Java JAR/EAR/WAR, Jenkins plugins JPI/HPI, Go modules)
-- OS distribution detection (supports alpine, busybox, centos/redhat, debian/ubuntu flavored distributions)
+- Linux distribution identification (supports Alpine, BusyBox, CentOS/RedHat, Debian/Ubuntu flavored distributions)
+- Supports Docker and OCI image formats
+- Direct support for [Grype](https://github.com/anchore/grype), a fast and powerful vulnerability matcher.
+
+
+If you encounter an issue, please [let us know using the issue tracker](https://github.com/anchore/syft/issues).
+
+## Installation
+
+### Recommended
+```bash
+curl -sSfL https://raw.githubusercontent.com/anchore/syft/main/install.sh | sh -s -- -b /usr/local/bin
+```
+
+...or, you can specify a release version and destination directory for the installation:
+
+```
+curl -sSfL https://raw.githubusercontent.com/anchore/syft/main/install.sh | sh -s -- -b <DESTINATION_DIR> <RELEASE_VERSION>
+```
+
+### Homebrew
+```bash
+brew tap anchore/syft
+brew install syft
+```
+
+**Note**: Currently, Syft is built only for macOS and Linux.
 
 ## Getting started
 
-To generate an SBOM for an image:
+To generate an SBOM for a Docker or OCI image:
 ```
 syft <image>
 ```
 
+**Note**: This is equivalent to specifying the `packages` subcommand:
+```
+syft packages <image>
+```
+
 The above output includes only software that is visible in the container (i.e., the squashed representation of the image).
 To include software from all image layers in the SBOM, regardless of its presence in the final image, provide `--scope all-layers`:
 
 ```
-syft <image> --scope all-layers
+syft packages <image> --scope all-layers
 ```
 
+### Supported sources
+
 Syft can generate a SBOM from a variety of sources:
+
 ```
-# catalog a docker image tar (from the result of "docker image save ... -o image.tar" command)
-syft docker-archive://path/to/image.tar
+# catalog a container image archive (from the result of `docker image save ...`, `podman save ...`, or `skopeo copy` commands)
+syft packages path/to/image.tar
 
 # catalog a directory
-syft dir://path/to/dir
+syft packages path/to/dir
 ```
 
-By default Syft shows a summary table, however, more detailed `text` and `json` formats are also available.
+Sources can be explicitly provided with a scheme:
+
 ```
-syft <image> -o json
-syft <image> -o text
+docker:yourrepo/yourimage:tag          use images from the Docker daemon
+docker-archive:path/to/yourimage.tar   use a tarball from disk for archives created from "docker save"
+oci-archive:path/to/yourimage.tar      use a tarball from disk for OCI archives (from Skopeo or otherwise)
+oci-dir:path/to/yourimage              read directly from a path on disk for OCI layout directories (from Skopeo or otherwise)
+dir:path/to/yourproject                read directly from a path on disk (any directory)
+file:path/to/yourproject/file          read directly from a path on disk (any single file)
+registry:yourrepo/yourimage:tag        pull image directly from a registry (no container runtime required)
 ```
 
-## Installation
+### Output formats
 
-**Recommended**
-```bash
-# install the latest version to /usr/local/bin
-curl -sSfL https://raw.githubusercontent.com/anchore/syft/master/install.sh | sh -s -- -b /usr/local/bin
+The output format for Syft is configurable as well:
 
-# install a specific version into a specific dir
-curl -sSfL https://raw.githubusercontent.com/anchore/syft/master/install.sh | sh -s <RELEASE_VERSION> -b <SOME_BIN_PATH>
+```
+syft packages <image> -o <format>
 ```
 
-**macOS**
-```bash
-brew tap anchore/syft
-brew install syft
+Where the `formats` available are:
+- `json`: Use this to get as much information out of Syft as possible!
+- `text`: A row-oriented, human-and-machine-friendly output.
+- `cyclonedx`: A XML report conforming to the [CycloneDX 1.2 specification](https://cyclonedx.org/specification/overview/).
+- `spdx`: A tag-value formatted report conforming to the [SPDX 2.2 specification](https://spdx.github.io/spdx-spec/).
+- `spdx-json`: A JSON report conforming to the [SPDX 2.2 JSON Schema](https://github.com/spdx/spdx-spec/blob/v2.2/schemas/spdx-schema.json).
+- `table`: A columnar summary (default).
+
+## Private Registry Authentication
+
+### Local Docker Credentials
+When a container runtime is not present, Syft can still utilize credentials configured in common credential sources (such as `~/.docker/config.json`). 
+It will pull images from private registries using these credentials. The config file is where your credentials are stored when authenticating with private registries via some command like `docker login`. 
+For more information see the `go-containerregistry` [documentation](https://github.com/google/go-containerregistry/tree/main/pkg/authn).
+
+
+An example `config.json` looks something like this:
+```
+// config.json
+{
+	"auths": {
+		"registry.example.com": {
+			"username": "AzureDiamond",
+			"password": "hunter2"
+		}
+	}
+}
 ```
 
-You may experience a "macOS cannot verify app is free from malware" error upon running Syft because it is not yet signed and notarized. You can override this using `xattr`.
-```bash
-xattr -rd com.apple.quarantine syft
-```
+You can run the following command as an example. It details the mount/environment configuration a container needs to access a private registry:
+
+`docker run -v ./config.json:/config/config.json -e "DOCKER_CONFIG=/config" anchore/syft:latest  <private_image>`
+
+
+### Docker Credentials in Kubernetes
+The below section shows a simple workflow on how to mount this config file as a secret into a container on kubernetes.
+1. Create a secret. The value of `config.json` is important. It refers to the specification detailed [here](https://github.com/google/go-containerregistry/tree/main/pkg/authn#the-config-file). 
+Below this section is the `secret.yaml` file that the pod configuration will consume as a volume. 
+The key `config.json` is important. It will end up being the name of the file when mounted into the pod.
+    ```
+    # secret.yaml
+    
+    apiVersion: v1
+    kind: Secret
+    metadata:
+      name: registry-config
+      namespace: syft
+    data:
+      config.json: <base64 encoded config.json>
+    ```
+
+    `kubectl apply -f secret.yaml`
+
+
+2. Create your pod running syft. The env `DOCKER_CONFIG` is important because it advertises where to look for the credential file. 
+In the below example, setting `DOCKER_CONFIG=/config` informs syft that credentials can be found at `/config/config.json`. 
+This is why we used `config.json` as the key for our secret. When mounted into containers the secrets' key is used as the filename. 
+The `volumeMounts` section mounts our secret to `/config`. The `volumes` section names our volume and leverages the secret we created in step one.
+    ```
+    # pod.yaml
+    
+    apiVersion: v1
+    kind: Pod
+    spec:
+      containers:
+        - image: anchore/syft:latest
+          name: syft-private-registry-demo
+          env:
+            - name: DOCKER_CONFIG
+              value: /config
+          volumeMounts:
+          - mountPath: /config
+            name: registry-config
+            readOnly: true
+          args:
+            - <private_image>
+      volumes:
+      - name: registry-config
+        secret:
+          secretName: registry-config
+    ```
+
+    `kubectl apply -f pod.yaml`
+
+
+3. The user can now run `kubectl logs syft-private-registry-demo`. The logs should show the syft analysis for the `<private_image>` provided in the pod configuration.
+
+Using the above information, users should be able to configure private registry access without having to do so in the `grype` or `syft` configuration files.
+They will also not be dependent on a docker daemon, (or some other runtime software) for registry configuration and access.
 
 ## Configuration
 
@@ -71,31 +201,172 @@ Configuration search paths:
 
 - `.syft.yaml`
 - `.syft/config.yaml`
+- `<git repo root>/.syft.yaml`
+- `<git repo root>/.syft/config.yaml`
 - `~/.syft.yaml`
 - `<XDG_CONFIG_HOME>/syft/config.yaml`
 
 Configuration options (example values are the default):
 
 ```yaml
-# same as -o ; the output format of the SBOM report (options: table, text, json)
+# the output format of the SBOM report (options: table, text, json)
+# same as -o ; SYFT_OUTPUT env var
 output: "table"
 
-# same as -s ; the search space to look for packages (options: all-layers, squashed)
-scope: "squashed"
-
-# same as -q ; suppress all output (except for the SBOM report)
+# suppress all output (except for the SBOM report)
+# same as -q ; SYFT_QUIET env var
 quiet: false
 
+# same as --file; write output report to a file (default is to write to stdout)
+file: ""
+
+# enable/disable checking for application updates on startup
+# same as SYFT_CHECK_FOR_APP_UPDATE env var
+check-for-app-update: true
+
+# cataloging packages is exposed through the packages and power-user subcommands
+package:
+  cataloger:
+    # enable/disable cataloging of packages
+    # SYFT_PACKAGE_CATALOGER_ENABLED env var
+    enabled: true
+
+    # the search space to look for packages (options: all-layers, squashed)
+    # same as -s ; SYFT_PACKAGE_CATALOGER_SCOPE env var
+    scope: "squashed"
+
+# cataloging file classifications is exposed through the power-user subcommand
+file-classification:
+  cataloger:
+    # enable/disable cataloging of file classifications
+    # SYFT_FILE_CLASSIFICATION_CATALOGER_ENABLED env var
+    enabled: true
+
+    # the search space to look for file classifications (options: all-layers, squashed)
+    # SYFT_FILE_CLASSIFICATION_CATALOGER_SCOPE env var
+    scope: "squashed"
+
+# cataloging file contents is exposed through the power-user subcommand
+file-contents:
+  cataloger:
+    # enable/disable cataloging of secrets
+    # SYFT_FILE_CONTENTS_CATALOGER_ENABLED env var
+    enabled: true
+
+    # the search space to look for secrets (options: all-layers, squashed)
+    # SYFT_FILE_CONTENTS_CATALOGER_SCOPE env var
+    scope: "squashed"
+
+  # skip searching a file entirely if it is above the given size (default = 1MB; unit = bytes)
+  # SYFT_FILE_CONTENTS_SKIP_FILES_ABOVE_SIZE env var
+  skip-files-above-size: 1048576
+
+  # file globs for the cataloger to match on
+  # SYFT_FILE_CONTENTS_GLOBS env var
+  globs: []
+
+# cataloging file metadata is exposed through the power-user subcommand
+file-metadata:
+  cataloger:
+    # enable/disable cataloging of file metadata
+    # SYFT_FILE_METADATA_CATALOGER_ENABLED env var
+    enabled: true
+
+    # the search space to look for file metadata (options: all-layers, squashed)
+    # SYFT_FILE_METADATA_CATALOGER_SCOPE env var
+    scope: "squashed"
+
+  # the file digest algorithms to use when cataloging files (options: "sha256", "md5", "sha1")
+  # SYFT_FILE_METADATA_DIGESTS env var
+  digests: ["sha256"]
+
+# cataloging secrets is exposed through the power-user subcommand
+secrets:
+  cataloger:
+    # enable/disable cataloging of secrets
+    # SYFT_SECRETS_CATALOGER_ENABLED env var
+    enabled: true
+
+    # the search space to look for secrets (options: all-layers, squashed)
+    # SYFT_SECRETS_CATALOGER_SCOPE env var
+    scope: "all-layers"
+
+  # show extracted secret values in the final JSON report
+  # SYFT_SECRETS_REVEAL_VALUES env var
+  reveal-values: false
+
+  # skip searching a file entirely if it is above the given size (default = 1MB; unit = bytes)
+  # SYFT_SECRETS_SKIP_FILES_ABOVE_SIZE env var
+  skip-files-above-size: 1048576
+
+  # name-regex pairs to consider when searching files for secrets. Note: the regex must match single line patterns
+  # but may also have OPTIONAL multiline capture groups. Regexes with a named capture group of "value" will
+  # use the entire regex to match, but the secret value will be assumed to be entirely contained within the
+  # "value" named capture group.
+  additional-patterns: {}
+
+  # names to exclude from the secrets search, valid values are: "aws-access-key", "aws-secret-key", "pem-private-key",
+  # "docker-config-auth", and "generic-api-key". Note: this does not consider any names introduced in the
+  # "secrets.additional-patterns" config option.
+  # SYFT_SECRETS_EXCLUDE_PATTERN_NAMES env var
+  exclude-pattern-names: []
+
+# options when pulling directly from a registry via the "registry:" scheme
+registry:
+  # skip TLS verification when communicating with the registry
+  # SYFT_REGISTRY_INSECURE_SKIP_TLS_VERIFY env var
+  insecure-skip-tls-verify: false
+  # use http instead of https when connecting to the registry
+  # SYFT_REGISTRY_INSECURE_USE_HTTP env var
+  insecure-use-http: false
+
+  # credentials for specific registries
+  auth:
+    - # the URL to the registry (e.g. "docker.io", "localhost:5000", etc.)
+      # SYFT_REGISTRY_AUTH_AUTHORITY env var
+      authority: ""
+      # SYFT_REGISTRY_AUTH_USERNAME env var
+      username: ""
+      # SYFT_REGISTRY_AUTH_PASSWORD env var
+      password: ""
+      # note: token and username/password are mutually exclusive
+      # SYFT_REGISTRY_AUTH_TOKEN env var
+      token: ""
+    - ... # note, more credentials can be provided via config file only
+
 log:
   # use structured logging
+  # same as SYFT_LOG_STRUCTURED env var
   structured: false
 
   # the log level; note: detailed logging suppress the ETUI
+  # same as SYFT_LOG_LEVEL env var
   level: "error"
 
   # location to write the log file (default is not to have a log file)
+  # same as SYFT_LOG_FILE env var
   file: ""
 
-# enable/disable checking for application updates on startup
-check-for-app-update: true
+# uploading package SBOM is exposed through the packages subcommand
+anchore:
+  # (feature-preview) the Anchore Enterprise Host or URL to upload results to (supported on Enterprise 3.0+)
+  # same as -H ; SYFT_ANCHORE_HOST env var
+  host: ""
+
+  # (feature-preview) the path after the host to the Anchore External API (supported on Enterprise 3.0+)
+  # same as SYFT_ANCHORE_PATH env var
+  path: ""
+
+  # (feature-preview) the username to authenticate against Anchore Enterprise (supported on Enterprise 3.0+)
+  # same as -u ; SYFT_ANCHORE_USERNAME env var
+  username: ""
+
+  # (feature-preview) the password to authenticate against Anchore Enterprise (supported on Enterprise 3.0+)
+  # same as -p ; SYFT_ANCHORE_PASSWORD env var
+  password: ""
+
+  # (feature-preview) path to dockerfile to be uploaded with the syft results to Anchore Enterprise (supported on Enterprise 3.0+)
+  # same as -d ; SYFT_ANCHORE_DOCKERFILE env var
+  dockerfile: ""
+
 ```

RELEASE.md

@@ -0,0 +1,104 @@
+# Release Process
+
+A good release process has the following qualities:
+
+1. There is a way to plan what should be in a release
+1. There is a way to see what is actually in a release
+1. Allow for different kinds of releases (major breaking vs backwards compatible
+   enhancements vs patch updates)
+1. Specify a repeatable way to build and publish software artifacts
+
+## Planning a release
+
+To indicate a set of features to be released together add each issue to an in-repository
+Milestone named with major-minor version to be released (e.g. `v0.1`). It is OK for other
+features to be in the release that were not originally planned, and these issues and PRs
+do not need to be added to the Milestone in question. Only the set of features that, when
+completed, would allow the release to be considered complete. A Milestone is only used to:
+
+- Plan what is desired to be in a release
+- Track progress to indicate when we may be ready to cut a new release
+
+Not all releases need to be planned. For instance, patch releases for fixes should be
+released when they are ready and when releasing would not interfere with another current
+release (where some partial or breaking features have already been merged). Beta releases
+and release candidates should not be independently planned from the non-beta release. That
+is, the features for a `v0.1-beta.1` release should be planned under the `v0.1` Milestone,
+not under a separate `v0.1-beta.1` Milestone.
+
+Unless necessary, feature releases should be small and frequent, which may obviate the
+need for regular release planning under a Milestone.
+
+## What is in a release
+
+Milestones are specifically for planning a release, not necessarily tracking all changes
+that a release may bring (and more importantly, not all releases are necessarily planned
+either).
+
+This is one of the (many) reasons for a Changelog. A good Changelog lists changes grouped
+by the type of change (new, enhancement, deprecation, breaking, bug fix, security fix), in
+chronological order (within groups), linking the PR where the change was made in the
+Changelog line. Furthermore, there should be a place to see all released versions, the
+release date for each release, the semantic version of the release, and the set of changes
+for each release.
+
+This project auto-generates the Changelog contents for each current release and posts the
+generated contents to the GitHub Release page. Leveraging the GitHub Releases feature
+allows GitHub to manage the Changelog on each release outside of the git repository while
+still being hosted with the released assets.
+
+The Changelog is generated from the metadata from in-repository issues and PRs, using
+labels to guide what kind of change each item is (e.g. breaking, new feature, bug fix,
+etx). Only issues/PRs with select labels are included in the Changelog, and only if the
+issue/PR was created after the last release. Additional labels are used to exclude items
+from the Changelog.
+
+The above suggestions imply that we should:
+
+- Ensure there is a sufficient title for each PR and issue title to be included in the
+  Changelog
+- The appropriate label is applied to PRs and/or issues to drive specific change type
+  sections (deprecated, breaking, security, bug, etc)
+
+With this approach as we cultivate good organization of PRs and issues we automatically
+get an equally good Changelog.
+
+## Major, minor, and patch releases
+
+The latest version of the tool is the only supported version, which implies that multiple
+parallel release branches will not be a regular process (if ever). Multiple releases can
+be planned in parallel, however, only one can be actively developed at a time. That is, if
+PRs attached to a release Milestone have been merged into the main branch, that release is
+now the "next" release. This implies that the source of truth for release lies with the
+git log and Changelog, not with the release Milestones (which are purely for planning and
+tracking).
+
+Semantic versioning should be used to indicate breaking changes, new features, and fixes.
+The exception to this is `< 1.0`, where the major version is not bumped for breaking changes,
+instead the minor version indicates both new features and breaking changes.
+
+## Cutting a release
+
+Ideally releasing should be done often with small increments when possible. Unless a
+breaking change is blocking the release, or no fixes/features have been merged, a good
+target release cadence is between every 2 or 4 weeks.
+
+This release process itself should be as automated as possible, and have only a few steps:
+
+1. Tag the main branch with a full semantic-version, prefixed with a `v`. If there is a
+   milestone with a partial version, the full version should be used for the git tag (e.g.
+   with a Milestone of `v0.1` the tag should be `v0.1.0`). You can determine the changes going
+   into a release by running `make changelog-unreleased`. Use this change list to determine the 
+   release increment. After determining the release increment (major, minor, patch), create the tag.
+   Given the above example the command to create the tag would be `git tag v0.1.0`.
+
+1. Push the tag. Given the above example the command to push the tag would be `git push origin v0.1.0`.
+   
+1. A release admin must approve the release on the GitHub Actions release pipeline run page.
+   Once approved, the release pipeline will generate all assets and draft a GitHub Release.
+
+1. Navigate to the GitHub Release draft page to review the final changelog and publish the
+   release. Once published, a release-follow-up pipeline will publish derivative artifacts
+   (docker image to DockerHub, brew formula to the external homebrew git repo, etc).
+
+1. If there is a release Milestone, close it.

cmd/check_for_application_update.go

@@ -0,0 +1,30 @@
+package cmd
+
+import (
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/bus"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/version"
+	"github.com/anchore/syft/syft/event"
+	"github.com/wagoodman/go-partybus"
+)
+
+func checkForApplicationUpdate() {
+	if appConfig.CheckForAppUpdate {
+		isAvailable, newVersion, err := version.IsUpdateAvailable()
+		if err != nil {
+			// this should never stop the application
+			log.Errorf(err.Error())
+		}
+		if isAvailable {
+			log.Infof("new version of %s is available: %s (current version is %s)", internal.ApplicationName, newVersion, version.FromBuild().Version)
+
+			bus.Publish(partybus.Event{
+				Type:  event.AppUpdateAvailable,
+				Value: newVersion,
+			})
+		} else {
+			log.Debugf("no new %s update available", internal.ApplicationName)
+		}
+	}
+}

cmd/cmd.go

@@ -6,29 +6,25 @@ import (
 
 	"github.com/spf13/cobra"
 
-	"github.com/anchore/syft/syft/presenter"
-	"github.com/anchore/syft/syft/scope"
-
 	"github.com/anchore/stereoscope"
 	"github.com/anchore/syft/internal/config"
-	"github.com/anchore/syft/internal/format"
 	"github.com/anchore/syft/internal/log"
 	"github.com/anchore/syft/internal/logger"
 	"github.com/anchore/syft/syft"
+	"github.com/gookit/color"
 	"github.com/spf13/viper"
 	"github.com/wagoodman/go-partybus"
-	"gopkg.in/yaml.v2"
 )
 
-var appConfig *config.Application
-var eventBus *partybus.Bus
-var eventSubscription *partybus.Subscription
-var cliOpts = config.CliOnlyOptions{}
+var (
+	appConfig         *config.Application
+	eventBus          *partybus.Bus
+	eventSubscription *partybus.Subscription
+)
 
 func init() {
-	setGlobalCliOptions()
-
 	cobra.OnInitialize(
+		initCmdAliasBindings,
 		initAppConfig,
 		initLogging,
 		logAppConfig,
@@ -38,54 +34,52 @@ func init() {
 
 func Execute() {
 	if err := rootCmd.Execute(); err != nil {
-		fmt.Fprintln(os.Stderr, err.Error())
+		fmt.Fprintln(os.Stderr, color.Red.Sprint(err.Error()))
 		os.Exit(1)
 	}
 }
 
-func setGlobalCliOptions() {
-	rootCmd.PersistentFlags().StringVarP(&cliOpts.ConfigPath, "config", "c", "", "application config file")
-
-	// scan options
-	flag := "scope"
-	rootCmd.Flags().StringP(
-		"scope", "s", scope.SquashedScope.String(),
-		fmt.Sprintf("selection of layers to catalog, options=%v", scope.Options))
-	if err := viper.BindPFlag(flag, rootCmd.Flags().Lookup(flag)); err != nil {
-		fmt.Printf("unable to bind flag '%s': %+v", flag, err)
-		os.Exit(1)
+// we must setup the config-cli bindings first before the application configuration is parsed. However, this cannot
+// be done without determining what the primary command that the config options should be bound to since there are
+// shared concerns (the root-packages alias).
+func initCmdAliasBindings() {
+	activeCmd, _, err := rootCmd.Find(os.Args[1:])
+	if err != nil {
+		panic(err)
 	}
 
-	// output & formatting options
-	flag = "output"
-	rootCmd.Flags().StringP(
-		flag, "o", presenter.TablePresenter.String(),
-		fmt.Sprintf("report output formatter, options=%v", presenter.Options),
-	)
-	if err := viper.BindPFlag(flag, rootCmd.Flags().Lookup(flag)); err != nil {
-		fmt.Printf("unable to bind flag '%s': %+v", flag, err)
-		os.Exit(1)
+	// enable all cataloger by default if power-user command is run
+	if activeCmd == powerUserCmd {
+		config.PowerUserCatalogerEnabledDefault()
 	}
 
-	flag = "quiet"
-	rootCmd.Flags().BoolP(
-		flag, "q", false,
-		"suppress all logging output",
-	)
-	if err := viper.BindPFlag(flag, rootCmd.Flags().Lookup(flag)); err != nil {
-		fmt.Printf("unable to bind flag '%s': %+v", flag, err)
-		os.Exit(1)
+	switch activeCmd {
+	case packagesCmd, rootCmd:
+		// note: we need to lazily bind config options since they are shared between both the root command
+		// and the packages command. Otherwise there will be global viper state that is in contention.
+		// See for more details: https://github.com/spf13/viper/issues/233 . Additionally, the bindings must occur BEFORE
+		// reading the application configuration, which implies that it must be an initializer (or rewrite the command
+		// initialization structure against typical patterns used with cobra, which is somewhat extreme for a
+		// temporary alias)
+		if err = bindPackagesConfigOptions(activeCmd.Flags()); err != nil {
+			panic(err)
+		}
+	default:
+		// even though the root command or packages command is NOT being run, we still need default bindings
+		// such that application config parsing passes.
+		if err = bindPackagesConfigOptions(packagesCmd.Flags()); err != nil {
+			panic(err)
+		}
 	}
-
-	rootCmd.Flags().CountVarP(&cliOpts.Verbosity, "verbose", "v", "increase verbosity (-v = info, -vv = debug)")
 }
 
 func initAppConfig() {
-	cfg, err := config.LoadConfigFromFile(viper.GetViper(), &cliOpts)
+	cfg, err := config.LoadApplicationConfig(viper.GetViper(), persistentOpts)
 	if err != nil {
 		fmt.Printf("failed to load application config: \n\t%+v\n", err)
 		os.Exit(1)
 	}
+
 	appConfig = cfg
 }
 
@@ -101,18 +95,12 @@ func initLogging() {
 	logWrapper := logger.NewLogrusLogger(cfg)
 	syft.SetLogger(logWrapper)
 	stereoscope.SetLogger(&logger.LogrusNestedLogger{
-		Logger: logWrapper.Logger.WithField("from-lib", "steroscope"),
+		Logger: logWrapper.Logger.WithField("from-lib", "stereoscope"),
 	})
 }
 
 func logAppConfig() {
-	appCfgStr, err := yaml.Marshal(&appConfig)
-
-	if err != nil {
-		log.Debugf("Could not display application config: %+v", err)
-	} else {
-		log.Debugf("Application config:\n%+v", format.Magenta.Format(string(appCfgStr)))
-	}
+	log.Debugf("application config:\n%+v", color.Magenta.Sprint(appConfig.String()))
 }
 
 func initEventBus() {

cmd/completion.go

@@ -0,0 +1,113 @@
+package cmd
+
+import (
+	"context"
+	"os"
+	"strings"
+
+	"github.com/docker/docker/api/types"
+	"github.com/docker/docker/api/types/filters"
+	"github.com/docker/docker/client"
+
+	"github.com/spf13/cobra"
+)
+
+// completionCmd represents the completion command
+var completionCmd = &cobra.Command{
+	Use:   "completion [bash|zsh|fish]",
+	Short: "Generate a shell completion for Syft (listing local docker images)",
+	Long: `To load completions (docker image list):
+
+Bash:
+
+$ source <(syft completion bash)
+
+# To load completions for each session, execute once:
+Linux:
+  $ syft completion bash > /etc/bash_completion.d/syft
+MacOS:
+  $ syft completion bash > /usr/local/etc/bash_completion.d/syft
+
+Zsh:
+
+# If shell completion is not already enabled in your environment you will need
+# to enable it.  You can execute the following once:
+
+$ echo "autoload -U compinit; compinit" >> ~/.zshrc
+
+# To load completions for each session, execute once:
+$ syft completion zsh > "${fpath[1]}/_syft"
+
+# You will need to start a new shell for this setup to take effect.
+
+Fish:
+
+$ syft completion fish | source
+
+# To load completions for each session, execute once:
+$ syft completion fish > ~/.config/fish/completions/syft.fish
+`,
+	DisableFlagsInUseLine: true,
+	ValidArgs:             []string{"bash", "zsh", "fish"},
+	Args:                  cobra.ExactValidArgs(1),
+	Run: func(cmd *cobra.Command, args []string) {
+		var err error
+		switch args[0] {
+		case "bash":
+			err = cmd.Root().GenBashCompletion(os.Stdout)
+		case "zsh":
+			err = cmd.Root().GenZshCompletion(os.Stdout)
+		case "fish":
+			err = cmd.Root().GenFishCompletion(os.Stdout, true)
+		}
+		if err != nil {
+			panic(err)
+		}
+	},
+}
+
+func init() {
+	rootCmd.AddCommand(completionCmd)
+}
+
+func dockerImageValidArgsFunction(cmd *cobra.Command, args []string, toComplete string) ([]string, cobra.ShellCompDirective) {
+	// Since we use ValidArgsFunction, Cobra will call this AFTER having parsed all flags and arguments provided
+	dockerImageRepoTags, err := listLocalDockerImages(toComplete)
+	if err != nil {
+		// Indicates that an error occurred and completions should be ignored
+		return []string{"completion failed"}, cobra.ShellCompDirectiveError
+	}
+	if len(dockerImageRepoTags) == 0 {
+		return []string{"no docker images found"}, cobra.ShellCompDirectiveError
+	}
+	// ShellCompDirectiveDefault indicates that the shell will perform its default behavior after completions have
+	// been provided (without implying other possible directives)
+	return dockerImageRepoTags, cobra.ShellCompDirectiveDefault
+}
+
+func listLocalDockerImages(prefix string) ([]string, error) {
+	var repoTags = make([]string, 0)
+	ctx := context.Background()
+	cli, err := client.NewClientWithOpts(client.FromEnv, client.WithAPIVersionNegotiation())
+	if err != nil {
+		return repoTags, err
+	}
+
+	// Only want to return tagged images
+	imageListArgs := filters.NewArgs()
+	imageListArgs.Add("dangling", "false")
+	images, err := cli.ImageList(ctx, types.ImageListOptions{All: false, Filters: imageListArgs})
+	if err != nil {
+		return repoTags, err
+	}
+
+	for _, image := range images {
+		// image may have multiple tags
+		for _, tag := range image.RepoTags {
+			if strings.HasPrefix(tag, prefix) {
+				repoTags = append(repoTags, tag)
+			}
+		}
+	}
+	return repoTags, nil
+}

cmd/event_loop.go

@@ -0,0 +1,98 @@
+package cmd
+
+import (
+	"errors"
+	"fmt"
+	"os"
+
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/ui"
+	"github.com/hashicorp/go-multierror"
+	"github.com/wagoodman/go-partybus"
+)
+
+// eventLoop listens to worker errors (from execution path), worker events (from a partybus subscription), and
+// signal interrupts. Is responsible for handling each event relative to a given UI an to coordinate eventing until
+// an eventual graceful exit.
+// nolint:funlen
+func eventLoop(workerErrs <-chan error, signals <-chan os.Signal, subscription *partybus.Subscription, cleanupFn func(), uxs ...ui.UI) error {
+	defer cleanupFn()
+	events := subscription.Events()
+	var err error
+	var ux ui.UI
+
+	if ux, err = setupUI(subscription.Unsubscribe, uxs...); err != nil {
+		return err
+	}
+
+	var retErr error
+	var forceTeardown bool
+
+	for {
+		if workerErrs == nil && events == nil {
+			break
+		}
+		select {
+		case err, isOpen := <-workerErrs:
+			if !isOpen {
+				workerErrs = nil
+				continue
+			}
+			if err != nil {
+				// capture the error from the worker and unsubscribe to complete a graceful shutdown
+				retErr = multierror.Append(retErr, err)
+				_ = subscription.Unsubscribe()
+				// the worker has exited, we may have been mid-handling events for the UI which should now be
+				// ignored, in which case forcing a teardown of the UI irregardless of the state is required.
+				forceTeardown = true
+			}
+		case e, isOpen := <-events:
+			if !isOpen {
+				events = nil
+				continue
+			}
+
+			if err := ux.Handle(e); err != nil {
+				if errors.Is(err, partybus.ErrUnsubscribe) {
+					events = nil
+				} else {
+					retErr = multierror.Append(retErr, err)
+					// TODO: should we unsubscribe? should we try to halt execution? or continue?
+				}
+			}
+		case <-signals:
+			// ignore further results from any event source and exit ASAP, but ensure that all cache is cleaned up.
+			// we ignore further errors since cleaning up the tmp directories will affect running catalogers that are
+			// reading/writing from/to their nested temp dirs. This is acceptable since we are bailing without result.
+
+			// TODO: potential future improvement would be to pass context into workers with a cancel function that is
+			// to the event loop. In this way we can have a more controlled shutdown even at the most nested levels
+			// of processing.
+			events = nil
+			workerErrs = nil
+			forceTeardown = true
+		}
+	}
+
+	if err := ux.Teardown(forceTeardown); err != nil {
+		retErr = multierror.Append(retErr, err)
+	}
+
+	return retErr
+}
+
+// setupUI takes one or more UIs that responds to events and takes a event bus unsubscribe function for use
+// during teardown. With the given UIs, the first UI which the ui.Setup() function does not return an error
+// will be utilized in execution. Providing a set of UIs allows for the caller to provide graceful fallbacks
+// when there are environmental problem (e.g. unable to setup a TUI with the current TTY).
+func setupUI(unsubscribe func() error, uis ...ui.UI) (ui.UI, error) {
+	for _, ux := range uis {
+		if err := ux.Setup(unsubscribe); err != nil {
+			log.Warnf("unable to setup given UI, falling back to alternative UI: %+v", err)
+			continue
+		}
+
+		return ux, nil
+	}
+	return nil, fmt.Errorf("unable to setup any UI")
+}

cmd/event_loop_test.go

@@ -0,0 +1,455 @@
+package cmd
+
+import (
+	"fmt"
+	"os"
+	"syscall"
+	"testing"
+	"time"
+
+	"github.com/anchore/syft/internal/ui"
+	"github.com/anchore/syft/syft/event"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/mock"
+	"github.com/wagoodman/go-partybus"
+)
+
+var _ ui.UI = (*uiMock)(nil)
+
+type uiMock struct {
+	t           *testing.T
+	finalEvent  partybus.Event
+	unsubscribe func() error
+	mock.Mock
+}
+
+func (u *uiMock) Setup(unsubscribe func() error) error {
+	u.t.Logf("UI Setup called")
+	u.unsubscribe = unsubscribe
+	return u.Called(unsubscribe).Error(0)
+}
+
+func (u *uiMock) Handle(event partybus.Event) error {
+	u.t.Logf("UI Handle called: %+v", event.Type)
+	if event == u.finalEvent {
+		assert.NoError(u.t, u.unsubscribe())
+	}
+	return u.Called(event).Error(0)
+}
+
+func (u *uiMock) Teardown(_ bool) error {
+	u.t.Logf("UI Teardown called")
+	return u.Called().Error(0)
+}
+
+func Test_eventLoop_gracefulExit(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		finalEvent := partybus.Event{
+			Type: event.PresenterReady,
+		}
+
+		worker := func() <-chan error {
+			ret := make(chan error)
+			go func() {
+				t.Log("worker running")
+				// send an empty item (which is ignored) ensuring we've entered the select statement,
+				// then close (a partial shutdown).
+				ret <- nil
+				t.Log("worker sent nothing")
+				close(ret)
+				t.Log("worker closed")
+				// do the other half of the shutdown
+				testBus.Publish(finalEvent)
+				t.Log("worker published final event")
+			}()
+			return ret
+		}
+
+		signaler := func() <-chan os.Signal {
+			return nil
+		}
+
+		ux := &uiMock{
+			t:          t,
+			finalEvent: finalEvent,
+		}
+
+		// ensure the mock sees at least the final event
+		ux.On("Handle", finalEvent).Return(nil)
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(nil)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		assert.NoError(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func Test_eventLoop_workerError(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		workerErr := fmt.Errorf("worker error")
+
+		worker := func() <-chan error {
+			ret := make(chan error)
+			go func() {
+				t.Log("worker running")
+				// send an empty item (which is ignored) ensuring we've entered the select statement,
+				// then close (a partial shutdown).
+				ret <- nil
+				t.Log("worker sent nothing")
+				ret <- workerErr
+				t.Log("worker sent error")
+				close(ret)
+				t.Log("worker closed")
+				// note: NO final event is fired
+			}()
+			return ret
+		}
+
+		signaler := func() <-chan os.Signal {
+			return nil
+		}
+
+		ux := &uiMock{
+			t: t,
+		}
+
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(nil)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		// ensure we see an error returned
+		assert.ErrorIs(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+			workerErr,
+			"should have seen a worker error, but did not",
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func Test_eventLoop_unsubscribeError(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		finalEvent := partybus.Event{
+			Type: event.PresenterReady,
+		}
+
+		worker := func() <-chan error {
+			ret := make(chan error)
+			go func() {
+				t.Log("worker running")
+				// send an empty item (which is ignored) ensuring we've entered the select statement,
+				// then close (a partial shutdown).
+				ret <- nil
+				t.Log("worker sent nothing")
+				close(ret)
+				t.Log("worker closed")
+				// do the other half of the shutdown
+				testBus.Publish(finalEvent)
+				t.Log("worker published final event")
+			}()
+			return ret
+		}
+
+		signaler := func() <-chan os.Signal {
+			return nil
+		}
+
+		ux := &uiMock{
+			t:          t,
+			finalEvent: finalEvent,
+		}
+
+		// ensure the mock sees at least the final event... note the unsubscribe error here
+		ux.On("Handle", finalEvent).Return(partybus.ErrUnsubscribe)
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(nil)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		// unsubscribe errors should be handled and ignored, not propagated. We are additionally asserting that
+		// this case is handled as a controlled shutdown (this test should not timeout)
+		assert.NoError(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func Test_eventLoop_handlerError(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		finalEvent := partybus.Event{
+			Type:  event.PresenterReady,
+			Error: fmt.Errorf("unable to create presenter"),
+		}
+
+		worker := func() <-chan error {
+			ret := make(chan error)
+			go func() {
+				t.Log("worker running")
+				// send an empty item (which is ignored) ensuring we've entered the select statement,
+				// then close (a partial shutdown).
+				ret <- nil
+				t.Log("worker sent nothing")
+				close(ret)
+				t.Log("worker closed")
+				// do the other half of the shutdown
+				testBus.Publish(finalEvent)
+				t.Log("worker published final event")
+			}()
+			return ret
+		}
+
+		signaler := func() <-chan os.Signal {
+			return nil
+		}
+
+		ux := &uiMock{
+			t:          t,
+			finalEvent: finalEvent,
+		}
+
+		// ensure the mock sees at least the final event... note the event error is propagated
+		ux.On("Handle", finalEvent).Return(finalEvent.Error)
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(nil)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		// handle errors SHOULD propagate the event loop. We are additionally asserting that this case is
+		// handled as a controlled shutdown (this test should not timeout)
+		assert.ErrorIs(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+			finalEvent.Error,
+			"should have seen a event error, but did not",
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func Test_eventLoop_signalsStopExecution(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		worker := func() <-chan error {
+			// the worker will never return work and the event loop will always be waiting...
+			return make(chan error)
+		}
+
+		signaler := func() <-chan os.Signal {
+			ret := make(chan os.Signal)
+			go func() {
+				ret <- syscall.SIGINT
+				// note: we do NOT close the channel to ensure the event loop does not depend on that behavior to exit
+			}()
+			return ret
+		}
+
+		ux := &uiMock{
+			t: t,
+		}
+
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(nil)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		assert.NoError(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func Test_eventLoop_uiTeardownError(t *testing.T) {
+	test := func(t *testing.T) {
+
+		testBus := partybus.NewBus()
+		subscription := testBus.Subscribe()
+		t.Cleanup(testBus.Close)
+
+		finalEvent := partybus.Event{
+			Type: event.PresenterReady,
+		}
+
+		worker := func() <-chan error {
+			ret := make(chan error)
+			go func() {
+				t.Log("worker running")
+				// send an empty item (which is ignored) ensuring we've entered the select statement,
+				// then close (a partial shutdown).
+				ret <- nil
+				t.Log("worker sent nothing")
+				close(ret)
+				t.Log("worker closed")
+				// do the other half of the shutdown
+				testBus.Publish(finalEvent)
+				t.Log("worker published final event")
+			}()
+			return ret
+		}
+
+		signaler := func() <-chan os.Signal {
+			return nil
+		}
+
+		ux := &uiMock{
+			t:          t,
+			finalEvent: finalEvent,
+		}
+
+		teardownError := fmt.Errorf("sorry, dave, the UI doesn't want to be torn down")
+
+		// ensure the mock sees at least the final event... note the event error is propagated
+		ux.On("Handle", finalEvent).Return(nil)
+		// ensure the mock sees basic setup/teardown events
+		ux.On("Setup", mock.AnythingOfType("func() error")).Return(nil)
+		ux.On("Teardown").Return(teardownError)
+
+		var cleanupCalled bool
+		cleanupFn := func() {
+			t.Log("cleanup called")
+			cleanupCalled = true
+		}
+
+		// ensure we see an error returned
+		assert.ErrorIs(t,
+			eventLoop(
+				worker(),
+				signaler(),
+				subscription,
+				cleanupFn,
+				ux,
+			),
+			teardownError,
+			"should have seen a UI teardown error, but did not",
+		)
+
+		assert.True(t, cleanupCalled, "cleanup function not called")
+		ux.AssertExpectations(t)
+	}
+
+	// if there is a bug, then there is a risk of the event loop never returning
+	testWithTimeout(t, 5*time.Second, test)
+}
+
+func testWithTimeout(t *testing.T, timeout time.Duration, test func(*testing.T)) {
+	done := make(chan bool)
+	go func() {
+		test(t)
+		done <- true
+	}()
+
+	select {
+	case <-time.After(timeout):
+		t.Fatal("test timed out")
+	case <-done:
+	}
+}

cmd/packages.go

@@ -0,0 +1,354 @@
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"io/ioutil"
+	"os"
+
+	"github.com/anchore/stereoscope"
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/anchore"
+	"github.com/anchore/syft/internal/bus"
+	"github.com/anchore/syft/internal/formats"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/ui"
+	"github.com/anchore/syft/internal/version"
+	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/event"
+	"github.com/anchore/syft/syft/format"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+	"github.com/pkg/profile"
+	"github.com/spf13/cobra"
+	"github.com/spf13/pflag"
+	"github.com/spf13/viper"
+	"github.com/wagoodman/go-partybus"
+)
+
+const (
+	packagesExample = `  {{.appName}} {{.command}} alpine:latest                a summary of discovered packages
+  {{.appName}} {{.command}} alpine:latest -o json        show all possible cataloging details
+  {{.appName}} {{.command}} alpine:latest -o cyclonedx   show a CycloneDX formatted SBOM
+  {{.appName}} {{.command}} alpine:latest -o spdx        show a SPDX 2.2 tag-value formatted SBOM
+  {{.appName}} {{.command}} alpine:latest -o spdx-json   show a SPDX 2.2 JSON formatted SBOM
+  {{.appName}} {{.command}} alpine:latest -vv            show verbose debug information
+
+  Supports the following image sources:
+    {{.appName}} {{.command}} yourrepo/yourimage:tag     defaults to using images from a Docker daemon. If Docker is not present, the image is pulled directly from the registry.
+    {{.appName}} {{.command}} path/to/a/file/or/dir      a Docker tar, OCI tar, OCI directory, or generic filesystem directory
+
+  You can also explicitly specify the scheme to use:
+    {{.appName}} {{.command}} docker:yourrepo/yourimage:tag          explicitly use the Docker daemon
+    {{.appName}} {{.command}} docker-archive:path/to/yourimage.tar   use a tarball from disk for archives created from "docker save"
+    {{.appName}} {{.command}} oci-archive:path/to/yourimage.tar      use a tarball from disk for OCI archives (from Skopeo or otherwise)
+    {{.appName}} {{.command}} oci-dir:path/to/yourimage              read directly from a path on disk for OCI layout directories (from Skopeo or otherwise)
+    {{.appName}} {{.command}} dir:path/to/yourproject                read directly from a path on disk (any directory)
+    {{.appName}} {{.command}} file:path/to/yourproject/file          read directly from a path on disk (any single file)
+    {{.appName}} {{.command}} registry:yourrepo/yourimage:tag        pull image directly from a registry (no container runtime required)
+`
+)
+
+var (
+	packagesPresenterOpt format.Option
+	packagesCmd          = &cobra.Command{
+		Use:   "packages [SOURCE]",
+		Short: "Generate a package SBOM",
+		Long:  "Generate a packaged-based Software Bill Of Materials (SBOM) from container images and filesystems",
+		Example: internal.Tprintf(packagesExample, map[string]interface{}{
+			"appName": internal.ApplicationName,
+			"command": "packages",
+		}),
+		Args:          validateInputArgs,
+		SilenceUsage:  true,
+		SilenceErrors: true,
+		PreRunE: func(cmd *cobra.Command, args []string) error {
+			// set the presenter
+			presenterOption := format.ParseOption(appConfig.Output)
+			if presenterOption == format.UnknownFormatOption {
+				return fmt.Errorf("bad --output value '%s'", appConfig.Output)
+			}
+			packagesPresenterOpt = presenterOption
+
+			if appConfig.Dev.ProfileCPU && appConfig.Dev.ProfileMem {
+				return fmt.Errorf("cannot profile CPU and memory simultaneously")
+			}
+			return nil
+		},
+		RunE: func(cmd *cobra.Command, args []string) error {
+			if appConfig.Dev.ProfileCPU {
+				defer profile.Start(profile.CPUProfile).Stop()
+			} else if appConfig.Dev.ProfileMem {
+				defer profile.Start(profile.MemProfile).Stop()
+			}
+
+			return packagesExec(cmd, args)
+		},
+		ValidArgsFunction: dockerImageValidArgsFunction,
+	}
+)
+
+func init() {
+	setPackageFlags(packagesCmd.Flags())
+
+	rootCmd.AddCommand(packagesCmd)
+}
+
+func setPackageFlags(flags *pflag.FlagSet) {
+	// Formatting & Input options //////////////////////////////////////////////
+
+	flags.StringP(
+		"scope", "s", source.SquashedScope.String(),
+		fmt.Sprintf("selection of layers to catalog, options=%v", source.AllScopes))
+
+	flags.StringP(
+		"output", "o", string(format.TableOption),
+		fmt.Sprintf("report output formatter, options=%v", format.AllOptions),
+	)
+
+	flags.StringP(
+		"file", "", "",
+		"file to write the report output to (default is STDOUT)",
+	)
+
+	// Upload options //////////////////////////////////////////////////////////
+	flags.StringP(
+		"host", "H", "",
+		"the hostname or URL of the Anchore Enterprise instance to upload to",
+	)
+
+	flags.StringP(
+		"username", "u", "",
+		"the username to authenticate against Anchore Enterprise",
+	)
+
+	flags.StringP(
+		"password", "p", "",
+		"the password to authenticate against Anchore Enterprise",
+	)
+
+	flags.StringP(
+		"dockerfile", "d", "",
+		"include dockerfile for upload to Anchore Enterprise",
+	)
+
+	flags.Bool(
+		"overwrite-existing-image", false,
+		"overwrite an existing image during the upload to Anchore Enterprise",
+	)
+
+	flags.Uint(
+		"import-timeout", 30,
+		"set a timeout duration (in seconds) for the upload to Anchore Enterprise",
+	)
+}
+
+func bindPackagesConfigOptions(flags *pflag.FlagSet) error {
+	// Formatting & Input options //////////////////////////////////////////////
+
+	if err := viper.BindPFlag("package.cataloger.scope", flags.Lookup("scope")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("output", flags.Lookup("output")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("file", flags.Lookup("file")); err != nil {
+		return err
+	}
+
+	// Upload options //////////////////////////////////////////////////////////
+
+	if err := viper.BindPFlag("anchore.host", flags.Lookup("host")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("anchore.username", flags.Lookup("username")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("anchore.password", flags.Lookup("password")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("anchore.dockerfile", flags.Lookup("dockerfile")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("anchore.overwrite-existing-image", flags.Lookup("overwrite-existing-image")); err != nil {
+		return err
+	}
+
+	if err := viper.BindPFlag("anchore.import-timeout", flags.Lookup("import-timeout")); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func validateInputArgs(cmd *cobra.Command, args []string) error {
+	if len(args) == 0 {
+		// in the case that no arguments are given we want to show the help text and return with a non-0 return code.
+		if err := cmd.Help(); err != nil {
+			return fmt.Errorf("unable to display help: %w", err)
+		}
+		return fmt.Errorf("an image/directory argument is required")
+	}
+
+	return cobra.MaximumNArgs(1)(cmd, args)
+}
+
+func packagesExec(_ *cobra.Command, args []string) error {
+	// could be an image or a directory, with or without a scheme
+	userInput := args[0]
+
+	reporter, closer, err := reportWriter()
+	defer func() {
+		if err := closer(); err != nil {
+			log.Warnf("unable to write to report destination: %+v", err)
+		}
+	}()
+
+	if err != nil {
+		return err
+	}
+
+	return eventLoop(
+		packagesExecWorker(userInput),
+		setupSignals(),
+		eventSubscription,
+		stereoscope.Cleanup,
+		ui.Select(isVerbose(), appConfig.Quiet, reporter)...,
+	)
+}
+
+func isVerbose() (result bool) {
+	isPipedInput, err := internal.IsPipedInput()
+	if err != nil {
+		// since we can't tell if there was piped input we assume that there could be to disable the ETUI
+		log.Warnf("unable to determine if there is piped input: %+v", err)
+		return true
+	}
+	// verbosity should consider if there is piped input (in which case we should not show the ETUI)
+	return appConfig.CliOptions.Verbosity > 0 || isPipedInput
+}
+
+func packagesExecWorker(userInput string) <-chan error {
+	errs := make(chan error)
+	go func() {
+		defer close(errs)
+
+		tasks, err := tasks()
+		if err != nil {
+			errs <- err
+			return
+		}
+
+		f := formats.ByOption(packagesPresenterOpt)
+		if f == nil {
+			errs <- fmt.Errorf("unknown format: %s", packagesPresenterOpt)
+			return
+		}
+
+		checkForApplicationUpdate()
+
+		src, cleanup, err := source.New(userInput, appConfig.Registry.ToOptions())
+		if err != nil {
+			errs <- fmt.Errorf("failed to determine image source: %w", err)
+			return
+		}
+		if cleanup != nil {
+			defer cleanup()
+		}
+
+		s := sbom.SBOM{
+			Source: src.Metadata,
+			Descriptor: sbom.Descriptor{
+				Name:          internal.ApplicationName,
+				Version:       version.FromBuild().Version,
+				Configuration: appConfig,
+			},
+		}
+
+		var relationships []<-chan artifact.Relationship
+		for _, task := range tasks {
+			c := make(chan artifact.Relationship)
+			relationships = append(relationships, c)
+
+			go runTask(task, &s.Artifacts, src, c, errs)
+		}
+		s.Relationships = append(s.Relationships, mergeRelationships(relationships...)...)
+
+		if appConfig.Anchore.Host != "" {
+			if err := runPackageSbomUpload(src, s); err != nil {
+				errs <- err
+				return
+			}
+		}
+
+		bus.Publish(partybus.Event{
+			Type:  event.PresenterReady,
+			Value: f.Presenter(s),
+		})
+	}()
+	return errs
+}
+
+func mergeRelationships(cs ...<-chan artifact.Relationship) (relationships []artifact.Relationship) {
+	for _, c := range cs {
+		for n := range c {
+			relationships = append(relationships, n)
+		}
+	}
+
+	return relationships
+}
+
+func runPackageSbomUpload(src *source.Source, s sbom.SBOM) error {
+	log.Infof("uploading results to %s", appConfig.Anchore.Host)
+
+	if src.Metadata.Scheme != source.ImageScheme {
+		return fmt.Errorf("unable to upload results: only images are supported")
+	}
+
+	var dockerfileContents []byte
+	if appConfig.Anchore.Dockerfile != "" {
+		if _, err := os.Stat(appConfig.Anchore.Dockerfile); os.IsNotExist(err) {
+			return fmt.Errorf("unable dockerfile=%q does not exist: %w", appConfig.Anchore.Dockerfile, err)
+		}
+
+		fh, err := os.Open(appConfig.Anchore.Dockerfile)
+		if err != nil {
+			return fmt.Errorf("unable to open dockerfile=%q: %w", appConfig.Anchore.Dockerfile, err)
+		}
+
+		dockerfileContents, err = ioutil.ReadAll(fh)
+		if err != nil {
+			return fmt.Errorf("unable to read dockerfile=%q: %w", appConfig.Anchore.Dockerfile, err)
+		}
+	}
+
+	c, err := anchore.NewClient(anchore.Configuration{
+		BaseURL:  appConfig.Anchore.Host,
+		Username: appConfig.Anchore.Username,
+		Password: appConfig.Anchore.Password,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to create anchore client: %w", err)
+	}
+
+	importCfg := anchore.ImportConfig{
+		ImageMetadata:           src.Image.Metadata,
+		SBOM:                    s,
+		Dockerfile:              dockerfileContents,
+		OverwriteExistingUpload: appConfig.Anchore.OverwriteExistingImage,
+		Timeout:                 appConfig.Anchore.ImportTimeout,
+	}
+
+	if err := c.Import(context.Background(), importCfg); err != nil {
+		return fmt.Errorf("failed to upload results to host=%s: %+v", appConfig.Anchore.Host, err)
+	}
+
+	return nil
+}

cmd/power_user.go

@@ -0,0 +1,151 @@
+package cmd
+
+import (
+	"fmt"
+	"os"
+
+	"github.com/anchore/stereoscope"
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/bus"
+	"github.com/anchore/syft/internal/formats/syftjson"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/internal/ui"
+	"github.com/anchore/syft/internal/version"
+	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/event"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+	"github.com/gookit/color"
+	"github.com/pkg/profile"
+	"github.com/spf13/cobra"
+	"github.com/wagoodman/go-partybus"
+)
+
+const powerUserExample = `  {{.appName}} {{.command}} <image>
+
+  DEPRECATED - THIS COMMAND WILL BE REMOVED in v1.0.0
+
+  Only image sources are supported (e.g. docker: , docker-archive: , oci: , etc.), the directory source (dir:) is not supported.
+
+  All behavior is controlled via application configuration and environment variables (see https://github.com/anchore/syft#configuration)
+`
+
+var powerUserOpts = struct {
+	configPath string
+}{}
+
+var powerUserCmd = &cobra.Command{
+	Use:   "power-user [IMAGE]",
+	Short: "Run bulk operations on container images",
+	Example: internal.Tprintf(powerUserExample, map[string]interface{}{
+		"appName": internal.ApplicationName,
+		"command": "power-user",
+	}),
+	Args:          validateInputArgs,
+	Hidden:        true,
+	SilenceUsage:  true,
+	SilenceErrors: true,
+	PreRunE: func(cmd *cobra.Command, args []string) error {
+		if appConfig.Dev.ProfileCPU && appConfig.Dev.ProfileMem {
+			return fmt.Errorf("cannot profile CPU and memory simultaneously")
+		}
+		return nil
+	},
+	RunE: func(cmd *cobra.Command, args []string) error {
+		if appConfig.Dev.ProfileCPU {
+			defer profile.Start(profile.CPUProfile).Stop()
+		} else if appConfig.Dev.ProfileMem {
+			defer profile.Start(profile.MemProfile).Stop()
+		}
+
+		return powerUserExec(cmd, args)
+	},
+	ValidArgsFunction: dockerImageValidArgsFunction,
+}
+
+func init() {
+	powerUserCmd.Flags().StringVarP(&powerUserOpts.configPath, "config", "c", "", "config file path with all power-user options")
+
+	rootCmd.AddCommand(powerUserCmd)
+}
+
+func powerUserExec(_ *cobra.Command, args []string) error {
+	// could be an image or a directory, with or without a scheme
+	userInput := args[0]
+
+	reporter, closer, err := reportWriter()
+	defer func() {
+		if err := closer(); err != nil {
+			log.Warnf("unable to write to report destination: %+v", err)
+		}
+
+		// inform user at end of run that command will be removed
+		deprecated := color.Style{color.Red, color.OpBold}.Sprint("DEPRECATED: This command will be removed in v1.0.0")
+		fmt.Fprintln(os.Stderr, deprecated)
+	}()
+
+	if err != nil {
+		return err
+	}
+
+	return eventLoop(
+		powerUserExecWorker(userInput),
+		setupSignals(),
+		eventSubscription,
+		stereoscope.Cleanup,
+		ui.Select(isVerbose(), appConfig.Quiet, reporter)...,
+	)
+}
+func powerUserExecWorker(userInput string) <-chan error {
+	errs := make(chan error)
+	go func() {
+		defer close(errs)
+
+		appConfig.Secrets.Cataloger.Enabled = true
+		appConfig.FileMetadata.Cataloger.Enabled = true
+		appConfig.FileContents.Cataloger.Enabled = true
+		appConfig.FileClassification.Cataloger.Enabled = true
+		tasks, err := tasks()
+		if err != nil {
+			errs <- err
+			return
+		}
+
+		checkForApplicationUpdate()
+
+		src, cleanup, err := source.New(userInput, appConfig.Registry.ToOptions())
+		if err != nil {
+			errs <- err
+			return
+		}
+		if cleanup != nil {
+			defer cleanup()
+		}
+
+		s := sbom.SBOM{
+			Source: src.Metadata,
+			Descriptor: sbom.Descriptor{
+				Name:          internal.ApplicationName,
+				Version:       version.FromBuild().Version,
+				Configuration: appConfig,
+			},
+		}
+
+		var relationships []<-chan artifact.Relationship
+		for _, task := range tasks {
+			c := make(chan artifact.Relationship)
+			relationships = append(relationships, c)
+
+			go runTask(task, &s.Artifacts, src, c, errs)
+		}
+
+		s.Relationships = append(s.Relationships, mergeRelationships(relationships...)...)
+
+		bus.Publish(partybus.Event{
+			Type:  event.PresenterReady,
+			Value: syftjson.Format().Presenter(s),
+		})
+	}()
+
+	return errs
+}

cmd/report_writer.go

@@ -0,0 +1,33 @@
+package cmd
+
+import (
+	"fmt"
+	"io"
+	"os"
+	"strings"
+
+	"github.com/anchore/syft/internal/log"
+)
+
+func reportWriter() (io.Writer, func() error, error) {
+	nop := func() error { return nil }
+	path := strings.TrimSpace(appConfig.File)
+
+	switch len(path) {
+	case 0:
+		return os.Stdout, nop, nil
+
+	default:
+		reportFile, err := os.OpenFile(path, os.O_RDWR|os.O_CREATE|os.O_TRUNC, 0644)
+
+		if err != nil {
+			return nil, nop, fmt.Errorf("unable to create report file: %w", err)
+		}
+
+		return reportFile, func() error {
+			log.Infof("report written to file=%q", path)
+
+			return reportFile.Close()
+		}, nil
+	}
+}

cmd/root.go

@@ -4,79 +4,43 @@ import (
 	"fmt"
 	"os"
 
-	"github.com/anchore/syft/internal"
-	"github.com/anchore/syft/internal/bus"
-	"github.com/anchore/syft/internal/log"
-	"github.com/anchore/syft/internal/ui"
-	"github.com/anchore/syft/internal/version"
-	"github.com/anchore/syft/syft"
-	"github.com/anchore/syft/syft/event"
-	"github.com/anchore/syft/syft/presenter"
+	"github.com/anchore/syft/internal/config"
 	"github.com/spf13/cobra"
-	"github.com/wagoodman/go-partybus"
+	"github.com/spf13/viper"
 )
 
+var persistentOpts = config.CliOnlyOptions{}
+
+// rootCmd is currently an alias for the packages command
 var rootCmd = &cobra.Command{
-	Use:   fmt.Sprintf("%s [SOURCE]", internal.ApplicationName),
-	Short: "A tool for generating a Software Bill Of Materials (SBOM) from container images and filesystems",
-	Long: internal.Tprintf(`
-Supports the following image sources:
-    {{.appName}} yourrepo/yourimage:tag             defaults to using images from a docker daemon
-    {{.appName}} docker://yourrepo/yourimage:tag    explicitly use the docker daemon
-    {{.appName}} tar://path/to/yourimage.tar        use a tarball from disk
-    {{.appName}} dir://path/to/yourproject          read directly from a path in disk
-`, map[string]interface{}{
-		"appName": internal.ApplicationName,
-	}),
-	Args: cobra.ExactArgs(1),
-	Run: func(cmd *cobra.Command, args []string) {
-		err := doRunCmd(cmd, args)
-		if err != nil {
-			log.Errorf(err.Error())
-			os.Exit(1)
-		}
-	},
+	Short:             packagesCmd.Short,
+	Long:              packagesCmd.Long,
+	Args:              packagesCmd.Args,
+	Example:           packagesCmd.Example,
+	SilenceUsage:      true,
+	SilenceErrors:     true,
+	PreRunE:           packagesCmd.PreRunE,
+	RunE:              packagesCmd.RunE,
+	ValidArgsFunction: packagesCmd.ValidArgsFunction,
 }
 
-func startWorker(userInput string) <-chan error {
-	errs := make(chan error)
-	go func() {
-		defer close(errs)
+func init() {
+	// set universal flags
+	rootCmd.PersistentFlags().StringVarP(&persistentOpts.ConfigPath, "config", "c", "", "application config file")
 
-		if appConfig.CheckForAppUpdate {
-			isAvailable, newVersion, err := version.IsUpdateAvailable()
-			if err != nil {
-				log.Errorf(err.Error())
-			}
-			if isAvailable {
-				log.Infof("new version of %s is available: %s", internal.ApplicationName, newVersion)
+	flag := "quiet"
+	rootCmd.PersistentFlags().BoolP(
+		flag, "q", false,
+		"suppress all logging output",
+	)
 
-				bus.Publish(partybus.Event{
-					Type:  event.AppUpdateAvailable,
-					Value: newVersion,
-				})
-			} else {
-				log.Debugf("no new %s update available", internal.ApplicationName)
-			}
-		}
+	if err := viper.BindPFlag(flag, rootCmd.PersistentFlags().Lookup(flag)); err != nil {
+		fmt.Printf("unable to bind flag '%s': %+v", flag, err)
+		os.Exit(1)
+	}
 
-		catalog, scope, _, err := syft.Catalog(userInput, appConfig.ScopeOpt)
-		if err != nil {
-			errs <- fmt.Errorf("failed to catalog input: %+v", err)
-			return
-		}
+	rootCmd.PersistentFlags().CountVarP(&persistentOpts.Verbosity, "verbose", "v", "increase verbosity (-v = info, -vv = debug)")
 
-		bus.Publish(partybus.Event{
-			Type:  event.CatalogerFinished,
-			Value: presenter.GetPresenter(appConfig.PresenterOpt, *scope, catalog),
-		})
-	}()
-	return errs
-}
-
-func doRunCmd(_ *cobra.Command, args []string) error {
-	userInput := args[0]
-	errs := startWorker(userInput)
-	ux := ui.Select(appConfig.CliOptions.Verbosity > 0, appConfig.Quiet)
-	return ux(errs, eventSubscription)
+	// set common options that are not universal (package subcommand-alias specific)
+	setPackageFlags(rootCmd.Flags())
 }

cmd/signals.go

@@ -0,0 +1,20 @@
+package cmd
+
+import (
+	"os"
+	"os/signal"
+	"syscall"
+)
+
+func setupSignals() <-chan os.Signal {
+	c := make(chan os.Signal, 1) // Note: A buffered channel is recommended for this; see https://golang.org/pkg/os/signal/#Notify
+
+	interruptions := []os.Signal{
+		syscall.SIGINT,
+		syscall.SIGTERM,
+	}
+
+	signal.Notify(c, interruptions...)
+
+	return c
+}

cmd/tasks.go

@@ -0,0 +1,231 @@
+package cmd
+
+import (
+	"crypto"
+	"fmt"
+
+	"github.com/anchore/syft/syft"
+	"github.com/anchore/syft/syft/artifact"
+	"github.com/anchore/syft/syft/file"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+)
+
+type task func(*sbom.Artifacts, *source.Source) ([]artifact.Relationship, error)
+
+func tasks() ([]task, error) {
+	var tasks []task
+
+	generators := []func() (task, error){
+		generateCatalogPackagesTask,
+		generateCatalogFileMetadataTask,
+		generateCatalogFileDigestsTask,
+		generateCatalogSecretsTask,
+		generateCatalogFileClassificationsTask,
+		generateCatalogContentsTask,
+	}
+
+	for _, generator := range generators {
+		task, err := generator()
+		if err != nil {
+			return nil, err
+		}
+
+		if task != nil {
+			tasks = append(tasks, task)
+		}
+	}
+
+	return tasks, nil
+}
+
+func generateCatalogPackagesTask() (task, error) {
+	if !appConfig.Package.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		packageCatalog, relationships, theDistro, err := syft.CatalogPackages(src, appConfig.Package.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		results.PackageCatalog = packageCatalog
+		results.Distro = theDistro
+
+		return relationships, nil
+	}
+
+	return task, nil
+}
+
+func generateCatalogFileMetadataTask() (task, error) {
+	if !appConfig.FileMetadata.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	metadataCataloger := file.NewMetadataCataloger()
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		resolver, err := src.FileResolver(appConfig.FileMetadata.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		result, err := metadataCataloger.Catalog(resolver)
+		if err != nil {
+			return nil, err
+		}
+		results.FileMetadata = result
+		return nil, nil
+	}
+
+	return task, nil
+}
+
+func generateCatalogFileDigestsTask() (task, error) {
+	if !appConfig.FileMetadata.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	supportedHashAlgorithms := make(map[string]crypto.Hash)
+	for _, h := range []crypto.Hash{
+		crypto.MD5,
+		crypto.SHA1,
+		crypto.SHA256,
+	} {
+		supportedHashAlgorithms[file.DigestAlgorithmName(h)] = h
+	}
+
+	var hashes []crypto.Hash
+	for _, hashStr := range appConfig.FileMetadata.Digests {
+		name := file.CleanDigestAlgorithmName(hashStr)
+		hashObj, ok := supportedHashAlgorithms[name]
+		if !ok {
+			return nil, fmt.Errorf("unsupported hash algorithm: %s", hashStr)
+		}
+		hashes = append(hashes, hashObj)
+	}
+
+	digestsCataloger, err := file.NewDigestsCataloger(hashes)
+	if err != nil {
+		return nil, err
+	}
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		resolver, err := src.FileResolver(appConfig.FileMetadata.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		result, err := digestsCataloger.Catalog(resolver)
+		if err != nil {
+			return nil, err
+		}
+		results.FileDigests = result
+		return nil, nil
+	}
+
+	return task, nil
+}
+
+func generateCatalogSecretsTask() (task, error) {
+	if !appConfig.Secrets.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	patterns, err := file.GenerateSearchPatterns(file.DefaultSecretsPatterns, appConfig.Secrets.AdditionalPatterns, appConfig.Secrets.ExcludePatternNames)
+	if err != nil {
+		return nil, err
+	}
+
+	secretsCataloger, err := file.NewSecretsCataloger(patterns, appConfig.Secrets.RevealValues, appConfig.Secrets.SkipFilesAboveSize)
+	if err != nil {
+		return nil, err
+	}
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		resolver, err := src.FileResolver(appConfig.Secrets.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		result, err := secretsCataloger.Catalog(resolver)
+		if err != nil {
+			return nil, err
+		}
+		results.Secrets = result
+		return nil, nil
+	}
+
+	return task, nil
+}
+
+func generateCatalogFileClassificationsTask() (task, error) {
+	if !appConfig.FileClassification.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	// TODO: in the future we could expose out the classifiers via configuration
+	classifierCataloger, err := file.NewClassificationCataloger(file.DefaultClassifiers)
+	if err != nil {
+		return nil, err
+	}
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		resolver, err := src.FileResolver(appConfig.FileClassification.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		result, err := classifierCataloger.Catalog(resolver)
+		if err != nil {
+			return nil, err
+		}
+		results.FileClassifications = result
+		return nil, nil
+	}
+
+	return task, nil
+}
+
+func generateCatalogContentsTask() (task, error) {
+	if !appConfig.FileContents.Cataloger.Enabled {
+		return nil, nil
+	}
+
+	contentsCataloger, err := file.NewContentsCataloger(appConfig.FileContents.Globs, appConfig.FileContents.SkipFilesAboveSize)
+	if err != nil {
+		return nil, err
+	}
+
+	task := func(results *sbom.Artifacts, src *source.Source) ([]artifact.Relationship, error) {
+		resolver, err := src.FileResolver(appConfig.FileContents.Cataloger.ScopeOpt)
+		if err != nil {
+			return nil, err
+		}
+
+		result, err := contentsCataloger.Catalog(resolver)
+		if err != nil {
+			return nil, err
+		}
+		results.FileContents = result
+		return nil, nil
+	}
+
+	return task, nil
+}
+
+func runTask(t task, a *sbom.Artifacts, src *source.Source, c chan<- artifact.Relationship, errs chan<- error) {
+	defer close(c)
+
+	relationships, err := t(a, src)
+	if err != nil {
+		errs <- err
+		return
+	}
+
+	for _, relationship := range relationships {
+		c <- relationship
+	}
+}

cmd/version.go

@@ -1,14 +1,17 @@
 package cmd
 
 import (
+	"encoding/json"
 	"fmt"
+	"os"
 
 	"github.com/anchore/syft/internal"
+
 	"github.com/anchore/syft/internal/version"
 	"github.com/spf13/cobra"
 )
 
-var showVerboseVersionInfo bool
+var outputFormat string
 
 var versionCmd = &cobra.Command{
 	Use:   "version",
@@ -17,14 +20,15 @@ var versionCmd = &cobra.Command{
 }
 
 func init() {
-	versionCmd.Flags().BoolVarP(&showVerboseVersionInfo, "verbose", "v", false, "show additional version information")
-
+	versionCmd.Flags().StringVarP(&outputFormat, "output", "o", "text", "format to show version information (available=[text, json])")
 	rootCmd.AddCommand(versionCmd)
 }
 
 func printVersion(_ *cobra.Command, _ []string) {
 	versionInfo := version.FromBuild()
-	if showVerboseVersionInfo {
+
+	switch outputFormat {
+	case "text":
 		fmt.Println("Application:  ", internal.ApplicationName)
 		fmt.Println("Version:      ", versionInfo.Version)
 		fmt.Println("BuildDate:    ", versionInfo.BuildDate)
@@ -33,7 +37,24 @@ func printVersion(_ *cobra.Command, _ []string) {
 		fmt.Println("Platform:     ", versionInfo.Platform)
 		fmt.Println("GoVersion:    ", versionInfo.GoVersion)
 		fmt.Println("Compiler:     ", versionInfo.Compiler)
-	} else {
-		fmt.Printf("%s %s\n", internal.ApplicationName, versionInfo.Version)
+
+	case "json":
+		enc := json.NewEncoder(os.Stdout)
+		enc.SetEscapeHTML(false)
+		enc.SetIndent("", " ")
+		err := enc.Encode(&struct {
+			version.Version
+			Application string `json:"application"`
+		}{
+			Version:     versionInfo,
+			Application: internal.ApplicationName,
+		})
+		if err != nil {
+			fmt.Printf("failed to show version information: %+v\n", err)
+			os.Exit(1)
+		}
+	default:
+		fmt.Printf("unsupported output format: %s\n", outputFormat)
+		os.Exit(1)
 	}
 }

go.mod

@@ -1,39 +1,58 @@
 module github.com/anchore/syft
 
-go 1.14
+go 1.16
 
 require (
+	github.com/acarl005/stripansi v0.0.0-20180116102854-5a71ef0e047d
 	github.com/adrg/xdg v0.2.1
-	github.com/anchore/go-testutils v0.0.0-20200624184116-66aa578126db
+	github.com/alecthomas/jsonschema v0.0.0-20210301060011-54c507b6f074
+	github.com/anchore/client-go v0.0.0-20210222170800-9c70f9b80bcf
+	github.com/anchore/go-presenter v0.0.0-20211102174526-0dbf20f6c7fa
+	github.com/anchore/go-rpmdb v0.0.0-20210914181456-a9c52348da63
+	github.com/anchore/go-testutils v0.0.0-20200925183923-d5f45b0d3c04
 	github.com/anchore/go-version v1.2.2-0.20200701162849-18adb9c92b9b
-	github.com/anchore/stereoscope v0.0.0-20200803190343-146f38f8cc19
-	github.com/bmatcuk/doublestar v1.3.1
-	github.com/containerd/continuity v0.0.0-20200710164510-efbc4488d8fe // indirect
+	github.com/anchore/packageurl-go v0.0.0-20210922164639-b3fa992ebd29
+	github.com/anchore/stereoscope v0.0.0-20211116152349-7e4e1b56a15d
+	// we are hinting brotli to latest due to warning when installing archiver v3:
+	// go: warning: github.com/andybalholm/brotli@v1.0.1: retracted by module author: occasional panics and data corruption
+	github.com/andybalholm/brotli v1.0.4 // indirect
+	github.com/antihax/optional v1.0.0
+	github.com/bmatcuk/doublestar/v2 v2.0.4
+	github.com/docker/docker v20.10.10+incompatible
 	github.com/dustin/go-humanize v1.0.0
-	github.com/go-test/deep v1.0.6
-	github.com/google/go-containerregistry v0.1.1 // indirect
+	github.com/facebookincubator/nvdtools v0.1.4
+	github.com/go-test/deep v1.0.7
+	github.com/google/go-cmp v0.5.6
+	github.com/google/uuid v1.2.0
 	github.com/gookit/color v1.2.7
-	github.com/gopherjs/gopherjs v0.0.0-20190910122728-9d188e94fb99 // indirect
 	github.com/hashicorp/go-multierror v1.1.0
 	github.com/hashicorp/go-version v1.2.0
+	github.com/jinzhu/copier v0.3.2
+	github.com/mgutz/ansi v0.0.0-20200706080929-d51e80ef957d // indirect
+	github.com/mholt/archiver/v3 v3.5.1
 	github.com/mitchellh/go-homedir v1.1.0
-	github.com/mitchellh/mapstructure v1.3.1
+	github.com/mitchellh/hashstructure/v2 v2.0.2
+	github.com/mitchellh/mapstructure v1.4.1
 	github.com/olekukonko/tablewriter v0.0.4
-	github.com/pelletier/go-toml v1.8.0
-	github.com/rogpeppe/go-internal v1.5.2
+	github.com/pelletier/go-toml v1.9.3
+	github.com/pkg/profile v1.5.0
+	github.com/scylladb/go-set v1.0.2
 	github.com/sergi/go-diff v1.1.0
-	github.com/sirupsen/logrus v1.6.0
-	github.com/spf13/cobra v1.0.0
-	github.com/spf13/viper v1.7.0
-	github.com/wagoodman/go-partybus v0.0.0-20200526224238-eb215533f07d
+	github.com/sirupsen/logrus v1.8.1
+	github.com/spdx/tools-golang v0.1.0
+	github.com/spf13/afero v1.6.0
+	github.com/spf13/cobra v1.2.1
+	github.com/spf13/pflag v1.0.5
+	github.com/spf13/viper v1.8.1
+	github.com/stretchr/testify v1.7.0
+	github.com/vifraa/gopom v0.1.0
+	github.com/wagoodman/go-partybus v0.0.0-20210627031916-db1f5573bbc5
 	github.com/wagoodman/go-progress v0.0.0-20200731105512-1020f39e6240
-	github.com/wagoodman/go-rpmdb v0.0.0-20200810111121-8136676cb95c
-	github.com/wagoodman/jotframe v0.0.0-20200730190914-3517092dd163
+	github.com/wagoodman/jotframe v0.0.0-20211129225309-56b0d0a4aebb
 	github.com/x-cray/logrus-prefixed-formatter v0.5.2
 	github.com/xeipuuv/gojsonschema v1.2.0
-	golang.org/x/crypto v0.0.0-20200604202706-70a84ac30bf9
-	golang.org/x/sys v0.0.0-20200610111108-226ff32320da // indirect
-	google.golang.org/genproto v0.0.0-20200615140333-fd031eab31e7 // indirect
-	gopkg.in/ini.v1 v1.57.0 // indirect
-	gopkg.in/yaml.v2 v2.3.0
+	golang.org/x/mod v0.4.2
+	golang.org/x/net v0.0.0-20211111160137-58aab5ef257a
+	golang.org/x/term v0.0.0-20210615171337-6886f2dfbf5b
+	gopkg.in/yaml.v2 v2.4.0
 )

gon.hcl

@@ -0,0 +1,15 @@
+source = ["./dist/syft-macos_darwin_amd64/syft"] # The 'dist' directory path should ideally reference an env var, where the source of truth is the Makefile. I wasn't able to figure out how to solve this.
+bundle_id = "com.anchore.toolbox.syft"
+
+sign {
+  application_identity = "Developer ID Application: ANCHORE, INC. (9MJHKYX5AT)"
+}
+
+dmg {
+  output_path = "./dist/output.dmg"
+  volume_name = "Syft"
+}
+
+zip {
+  output_path = "./dist/output.zip"
+}

install.sh

@@ -1,6 +1,6 @@
 #!/bin/sh
 set -e
-# Code generated by godownloader on 2020-08-10T20:55:46Z. DO NOT EDIT.
+# Code generated by godownloader on 2020-08-10T20:55:46Z.
 #
 
 usage() {
@@ -45,11 +45,16 @@ parse_args() {
 execute() {
   tmpdir=$(mktemp -d)
   log_debug "downloading files into ${tmpdir}"
-  http_download "${tmpdir}/${TARBALL}" "${TARBALL_URL}"
+  http_download "${tmpdir}/${ARCHIVE}" "${ARCHIVE_URL}"
   http_download "${tmpdir}/${CHECKSUM}" "${CHECKSUM_URL}"
-  hash_sha256_verify "${tmpdir}/${TARBALL}" "${tmpdir}/${CHECKSUM}"
+
+  # macOS has its own secure verification mechanism, and checksums.txt is not used.
+  if [ "$OS" != "darwin" ]; then
+    hash_sha256_verify "${tmpdir}/${ARCHIVE}" "${tmpdir}/${CHECKSUM}"
+  fi
+
   srcdir="${tmpdir}"
-  (cd "${tmpdir}" && untar "${TARBALL}")
+  (cd "${tmpdir}" && unpack "${ARCHIVE}")
   test ! -d "${BINDIR}" && install -d "${BINDIR}"
   for binexe in $BINARIES; do
     if [ "$OS" = "windows" ]; then
@@ -89,6 +94,7 @@ tag_to_version() {
 adjust_format() {
   # change format (tar.gz or zip) based on OS
   case ${OS} in
+    darwin) FORMAT=dmg ;;
     windows) FORMAT=zip ;;
   esac
   true
@@ -221,18 +227,26 @@ uname_arch_check() {
   log_crit "uname_arch_check '$(uname -m)' got converted to '$arch' which is not a GOARCH value.  Please file bug report at https://github.com/client9/shlib"
   return 1
 }
-untar() {
-  tarball=$1
-  case "${tarball}" in
-    *.tar.gz | *.tgz) tar --no-same-owner -xzf "${tarball}" ;;
-    *.tar) tar --no-same-owner -xf "${tarball}" ;;
-    *.zip) unzip "${tarball}" ;;
+unpack() {
+  archive=$1
+  case "${archive}" in
+    *.tar.gz | *.tgz) tar --no-same-owner -xzf "${archive}" ;;
+    *.tar) tar --no-same-owner -xf "${archive}" ;;
+    *.zip) unzip "${archive}" ;;
+    *.dmg) extract_from_dmg "${archive}" ;;
     *)
-      log_err "untar unknown archive format for ${tarball}"
+      log_err "unpack unknown archive format for ${archive}"
       return 1
       ;;
   esac
 }
+extract_from_dmg() {
+  dmg_file=$1
+  mount_point="/Volumes/tmp-dmg"
+  hdiutil attach -quiet -nobrowse -mountpoint "${mount_point}" "${dmg_file}"
+  cp -fR "${mount_point}/" ./
+  hdiutil detach -quiet -force "${mount_point}"
+}
 http_download_curl() {
   local_file=$1
   source_url=$2
@@ -366,8 +380,8 @@ adjust_arch
 log_info "found version: ${VERSION} for ${TAG}/${OS}/${ARCH}"
 
 NAME=${PROJECT_NAME}_${VERSION}_${OS}_${ARCH}
-TARBALL=${NAME}.${FORMAT}
-TARBALL_URL=${GITHUB_DOWNLOAD}/${TAG}/${TARBALL}
+ARCHIVE=${NAME}.${FORMAT}
+ARCHIVE_URL=${GITHUB_DOWNLOAD}/${TAG}/${ARCHIVE}
 CHECKSUM=${PROJECT_NAME}_${VERSION}_checksums.txt
 CHECKSUM_URL=${GITHUB_DOWNLOAD}/${TAG}/${CHECKSUM}
 

internal/anchore/client.go

@@ -0,0 +1,115 @@
+package anchore
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"path"
+	"strings"
+	"unicode"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/syft/internal"
+	"github.com/anchore/syft/internal/version"
+)
+
+type Configuration struct {
+	BaseURL   string
+	Username  string
+	Password  string
+	UserAgent string
+}
+
+type Client struct {
+	config Configuration
+	client *external.APIClient
+}
+
+func NewClient(cfg Configuration) (*Client, error) {
+	if cfg.UserAgent == "" {
+		versionInfo := version.FromBuild()
+		// format: product / product-version comment
+		cfg.UserAgent = fmt.Sprintf("%s / %s %s", internal.ApplicationName, versionInfo.Version, versionInfo.Platform)
+	}
+
+	baseURL, err := prepareBaseURLForClient(cfg.BaseURL)
+	if err != nil {
+		return nil, fmt.Errorf("unable to create client: %w", err)
+	}
+
+	return &Client{
+		config: cfg,
+		client: external.NewAPIClient(&external.Configuration{
+			BasePath:  baseURL,
+			UserAgent: cfg.UserAgent,
+		}),
+	}, nil
+}
+
+func (c *Client) newRequestContext(parentContext context.Context) context.Context {
+	if parentContext == nil {
+		parentContext = context.Background()
+	}
+	return context.WithValue(
+		parentContext,
+		external.ContextBasicAuth,
+		external.BasicAuth{
+			UserName: c.config.Username,
+			Password: c.config.Password,
+		},
+	)
+}
+
+var ErrInvalidBaseURLInput = errors.New("invalid base URL input")
+
+func prepareBaseURLForClient(baseURL string) (string, error) {
+	if err := checkBaseURLInput(baseURL); err != nil {
+		return "", err
+	}
+
+	scheme, urlWithoutScheme := splitSchemeFromURL(baseURL)
+
+	if scheme == "" {
+		scheme = "http"
+	}
+
+	urlWithoutScheme = path.Clean(urlWithoutScheme)
+
+	const requiredSuffix = "v1"
+	if path.Base(urlWithoutScheme) != requiredSuffix {
+		urlWithoutScheme = path.Join(urlWithoutScheme, requiredSuffix)
+	}
+
+	preparedBaseURL := scheme + "://" + urlWithoutScheme
+	return preparedBaseURL, nil
+}
+
+func checkBaseURLInput(url string) error {
+	if url == "" {
+		return ErrInvalidBaseURLInput
+	}
+
+	firstCharacter := rune(url[0])
+	if !(unicode.IsLetter(firstCharacter)) {
+		return ErrInvalidBaseURLInput
+	}
+
+	return nil
+}
+
+func splitSchemeFromURL(url string) (scheme, urlWithoutScheme string) {
+	if hasScheme(url) {
+		urlParts := strings.SplitN(url, "://", 2)
+		scheme = urlParts[0]
+		urlWithoutScheme = urlParts[1]
+		return
+	}
+
+	return "", url
+}
+
+func hasScheme(url string) bool {
+	parts := strings.Split(url, "://")
+
+	return len(parts) > 1
+}

internal/anchore/client_test.go

@@ -0,0 +1,210 @@
+package anchore
+
+import "testing"
+
+func TestHasScheme(t *testing.T) {
+	cases := []struct {
+		url      string
+		expected bool
+	}{
+		{
+			url:      "http://localhost",
+			expected: true,
+		},
+		{
+			url:      "https://anchore.com:8443",
+			expected: true,
+		},
+		{
+			url:      "google.com",
+			expected: false,
+		},
+		{
+			url:      "",
+			expected: false,
+		},
+	}
+
+	for _, testCase := range cases {
+		t.Run(testCase.url, func(t *testing.T) {
+			result := hasScheme(testCase.url)
+
+			if testCase.expected != result {
+				t.Errorf("expected %t but got %t", testCase.expected, result)
+			}
+		})
+	}
+}
+
+func TestPrepareBaseURLForClient(t *testing.T) {
+	cases := []struct {
+		inputURL    string
+		expectedURL string
+		expectedErr error
+	}{
+		{
+			inputURL:    "",
+			expectedURL: "",
+			expectedErr: ErrInvalidBaseURLInput,
+		},
+		{
+			inputURL:    "localhost",
+			expectedURL: "http://localhost/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "https://localhost",
+			expectedURL: "https://localhost/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "https://localhost/",
+			expectedURL: "https://localhost/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "https://localhost/v1/",
+			expectedURL: "https://localhost/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "https://localhost/v1//",
+			expectedURL: "https://localhost/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "http://something.com/platform/v1/services/anchore",
+			expectedURL: "http://something.com/platform/v1/services/anchore/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "my-host:8228",
+			expectedURL: "http://my-host:8228/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "v1/v1",
+			expectedURL: "http://v1/v1",
+			expectedErr: nil,
+		},
+		{
+			inputURL:    "/v1",
+			expectedURL: "",
+			expectedErr: ErrInvalidBaseURLInput,
+		},
+		{
+			inputURL:    "/imports/images",
+			expectedURL: "",
+			expectedErr: ErrInvalidBaseURLInput,
+		},
+	}
+
+	for _, testCase := range cases {
+		t.Run(testCase.inputURL, func(t *testing.T) {
+			resultURL, err := prepareBaseURLForClient(testCase.inputURL)
+			if err != testCase.expectedErr {
+				t.Errorf("expected err to be '%v' but got '%v'", testCase.expectedErr, err)
+			}
+
+			if resultURL != testCase.expectedURL {
+				t.Errorf("expected URL to be '%v' but got '%v'", testCase.expectedURL, resultURL)
+			}
+		})
+	}
+}
+
+func TestCheckBaseURLInput(t *testing.T) {
+	cases := []struct {
+		input    string
+		expected error
+	}{
+		{
+			input:    "",
+			expected: ErrInvalidBaseURLInput,
+		},
+		{
+			input:    "x",
+			expected: nil,
+		},
+		{
+			input:    "localhost:8000",
+			expected: nil,
+		},
+		{
+			input:    ":80",
+			expected: ErrInvalidBaseURLInput,
+		},
+		{
+			input:    "/v1",
+			expected: ErrInvalidBaseURLInput,
+		},
+	}
+
+	for _, testCase := range cases {
+		t.Run(testCase.input, func(t *testing.T) {
+			resultErr := checkBaseURLInput(testCase.input)
+
+			if testCase.expected != resultErr {
+				t.Errorf("expected err to be '%v' but got '%v'", testCase.expected, resultErr)
+			}
+		})
+	}
+}
+
+func TestSplitSchemeFromURL(t *testing.T) {
+	cases := []struct {
+		input                    string
+		expectedScheme           string
+		expectedURLWithoutScheme string
+	}{
+		{
+			input:                    "",
+			expectedScheme:           "",
+			expectedURLWithoutScheme: "",
+		},
+		{
+			input:                    "localhost",
+			expectedScheme:           "",
+			expectedURLWithoutScheme: "localhost",
+		},
+		{
+			input:                    "https://anchore.com/path",
+			expectedScheme:           "https",
+			expectedURLWithoutScheme: "anchore.com/path",
+		},
+		{
+			input:                    "tcp://host:1234",
+			expectedScheme:           "tcp",
+			expectedURLWithoutScheme: "host:1234",
+		},
+		{
+			input:                    "/hello",
+			expectedScheme:           "",
+			expectedURLWithoutScheme: "/hello",
+		},
+		{
+			input:                    "://host",
+			expectedScheme:           "",
+			expectedURLWithoutScheme: "host",
+		},
+		{
+			input:                    "http//localhost",
+			expectedScheme:           "",
+			expectedURLWithoutScheme: "http//localhost",
+		},
+	}
+
+	for _, testCase := range cases {
+		t.Run(testCase.input, func(t *testing.T) {
+			resultScheme, resultURLWithoutScheme := splitSchemeFromURL(testCase.input)
+
+			if testCase.expectedScheme != resultScheme {
+				t.Errorf("expected scheme to be '%s' but got '%s'", testCase.expectedScheme, resultScheme)
+			}
+
+			if testCase.expectedURLWithoutScheme != resultURLWithoutScheme {
+				t.Errorf("expected urlWithoutScheme to be '%s' but got '%s'", testCase.expectedURLWithoutScheme, resultURLWithoutScheme)
+			}
+		})
+	}
+}

internal/anchore/import.go

@@ -0,0 +1,140 @@
+package anchore
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"time"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/stereoscope/pkg/image"
+	"github.com/anchore/syft/internal/bus"
+	"github.com/anchore/syft/syft/event"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/antihax/optional"
+	"github.com/wagoodman/go-partybus"
+	"github.com/wagoodman/go-progress"
+)
+
+type ImportConfig struct {
+	ImageMetadata           image.Metadata
+	SBOM                    sbom.SBOM
+	Dockerfile              []byte
+	OverwriteExistingUpload bool
+	Timeout                 uint
+}
+
+func importProgress(source string) (*progress.Stage, *progress.Manual) {
+	stage := &progress.Stage{}
+	prog := &progress.Manual{
+		// this is the number of stages to expect; start + individual endpoints + stop
+		Total: 6,
+	}
+	bus.Publish(partybus.Event{
+		Type:   event.ImportStarted,
+		Source: source,
+		Value: progress.StagedProgressable(&struct {
+			progress.Stager
+			progress.Progressable
+		}{
+			Stager:       progress.Stager(stage),
+			Progressable: prog,
+		}),
+	})
+
+	return stage, prog
+}
+
+// nolint:funlen
+func (c *Client) Import(ctx context.Context, cfg ImportConfig) error {
+	stage, prog := importProgress(c.config.BaseURL)
+
+	timeout := time.Duration(cfg.Timeout) * time.Second
+	ctxWithTimeout, cancel := context.WithTimeout(ctx, timeout)
+	defer cancel()
+
+	authedCtx := c.newRequestContext(ctxWithTimeout)
+
+	stage.Current = "starting session"
+	startOperation, _, err := c.client.ImportsApi.CreateOperation(authedCtx)
+	if err != nil {
+		var detail = "no details given"
+		var openAPIErr external.GenericOpenAPIError
+		if errors.As(err, &openAPIErr) {
+			detail = string(openAPIErr.Body())
+		}
+		return fmt.Errorf("unable to start import session: %w: %s", err, detail)
+	}
+	prog.N++
+	sessionID := startOperation.Uuid
+
+	packageDigest, err := importPackageSBOM(authedCtx, c.client.ImportsApi, sessionID, cfg.SBOM, stage)
+	if err != nil {
+		return fmt.Errorf("failed to import Package SBOM: %w", err)
+	}
+	prog.N++
+
+	manifestDigest, err := importManifest(authedCtx, c.client.ImportsApi, sessionID, cfg.SBOM.Source.ImageMetadata.RawManifest, stage)
+	if err != nil {
+		return fmt.Errorf("failed to import Manifest: %w", err)
+	}
+	prog.N++
+
+	configDigest, err := importConfig(authedCtx, c.client.ImportsApi, sessionID, cfg.SBOM.Source.ImageMetadata.RawConfig, stage)
+	if err != nil {
+		return fmt.Errorf("failed to import Config: %w", err)
+	}
+	prog.N++
+
+	dockerfileDigest, err := importDockerfile(authedCtx, c.client.ImportsApi, sessionID, cfg.Dockerfile, stage)
+	if err != nil {
+		return fmt.Errorf("failed to import Dockerfile: %w", err)
+	}
+	prog.N++
+
+	stage.Current = "finalizing"
+	imageModel := addImageModel(cfg.ImageMetadata, packageDigest, manifestDigest, dockerfileDigest, configDigest, sessionID)
+	opts := external.AddImageOpts{
+		Force: optional.NewBool(cfg.OverwriteExistingUpload),
+	}
+
+	_, _, err = c.client.ImagesApi.AddImage(authedCtx, imageModel, &opts)
+	if err != nil {
+		var detail = "no details given"
+		var openAPIErr external.GenericOpenAPIError
+		if errors.As(err, &openAPIErr) {
+			detail = string(openAPIErr.Body())
+		}
+		return fmt.Errorf("unable to complete import session=%q: %w: %s", sessionID, err, detail)
+	}
+	prog.N++
+
+	stage.Current = ""
+	prog.SetCompleted()
+
+	return nil
+}
+
+func addImageModel(imageMetadata image.Metadata, packageDigest, manifestDigest, dockerfileDigest, configDigest, sessionID string) external.ImageAnalysisRequest {
+	var tags = make([]string, len(imageMetadata.Tags))
+	for i, t := range imageMetadata.Tags {
+		tags[i] = t.String()
+	}
+
+	return external.ImageAnalysisRequest{
+		Source: external.ImageSource{
+			Import: &external.ImageImportManifest{
+				Contents: external.ImportContentDigests{
+					Packages:    packageDigest,
+					Manifest:    manifestDigest,
+					Dockerfile:  dockerfileDigest,
+					ImageConfig: configDigest,
+				},
+				Tags:          tags,
+				Digest:        imageMetadata.ManifestDigest,
+				LocalImageId:  imageMetadata.ID,
+				OperationUuid: sessionID,
+			},
+		},
+	}
+}

internal/anchore/import_config.go

@@ -0,0 +1,50 @@
+// nolint:dupl
+package anchore
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"net/http"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/syft/internal/log"
+)
+
+type configImportAPI interface {
+	ImportImageConfig(ctx context.Context, sessionID string, contents interface{}) (external.ImageImportContentResponse, *http.Response, error)
+}
+
+func importConfig(ctx context.Context, api configImportAPI, sessionID string, config []byte, stage *progress.Stage) (string, error) {
+	if len(config) > 0 {
+		log.Debug("importing image config")
+		stage.Current = "image config"
+
+		// API requires an object, but we do not verify the shape of this object locally
+		var sender map[string]interface{}
+		if err := json.Unmarshal(config, &sender); err != nil {
+			return "", err
+		}
+
+		response, httpResponse, err := api.ImportImageConfig(ctx, sessionID, sender)
+		if err != nil {
+			var openAPIErr external.GenericOpenAPIError
+			if errors.As(err, &openAPIErr) {
+				log.Errorf("api response: %+v", string(openAPIErr.Body()))
+			}
+			return "", fmt.Errorf("unable to import Config: %w", err)
+		}
+
+		defer httpResponse.Body.Close()
+
+		if httpResponse.StatusCode != 200 {
+			return "", fmt.Errorf("unable to import Config: %s", httpResponse.Status)
+		}
+
+		return response.Digest, nil
+	}
+	return "", nil
+}

internal/anchore/import_config_test.go

@@ -0,0 +1,129 @@
+package anchore
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"strings"
+	"testing"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/docker/docker/pkg/ioutils"
+	"github.com/go-test/deep"
+)
+
+type mockConfigImportAPI struct {
+	sessionID      string
+	model          interface{}
+	httpResponse   *http.Response
+	err            error
+	ctx            context.Context
+	responseDigest string
+	wasCalled      bool
+}
+
+func (m *mockConfigImportAPI) ImportImageConfig(ctx context.Context, sessionID string, contents interface{}) (external.ImageImportContentResponse, *http.Response, error) {
+	m.wasCalled = true
+	m.model = contents
+	m.sessionID = sessionID
+	m.ctx = ctx
+	if m.httpResponse == nil {
+		m.httpResponse = &http.Response{}
+	}
+	m.httpResponse.Body = ioutils.NewReadCloserWrapper(strings.NewReader(""), func() error { return nil })
+	return external.ImageImportContentResponse{Digest: m.responseDigest}, m.httpResponse, m.err
+}
+
+func TestConfigImport(t *testing.T) {
+
+	sessionID := "my-session"
+
+	tests := []struct {
+		name            string
+		manifestJSONStr string
+		api             *mockConfigImportAPI
+		expectsError    bool
+		expectsCall     bool
+	}{
+
+		{
+			name:            "Go case: import works",
+			manifestJSONStr: `{ "key": "the-manifest-contents!" }`,
+			api: &mockConfigImportAPI{
+				httpResponse:   &http.Response{StatusCode: 200},
+				responseDigest: "digest!",
+			},
+			expectsCall: true,
+		},
+		{
+			name:            "No manifest provided",
+			manifestJSONStr: "",
+			api:             &mockConfigImportAPI{},
+			expectsCall:     false,
+		},
+		{
+			name:            "API returns an error",
+			manifestJSONStr: `{ "key": "the-manifest-contents!" }`,
+			api: &mockConfigImportAPI{
+				err: fmt.Errorf("api error, something went wrong"),
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+		{
+			name:            "API HTTP-level error",
+			manifestJSONStr: `{ "key": "the-manifest-contents!" }`,
+			api: &mockConfigImportAPI{
+				httpResponse: &http.Response{StatusCode: 404},
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+
+			digest, err := importConfig(context.TODO(), test.api, sessionID, []byte(test.manifestJSONStr), &progress.Stage{})
+
+			// validate error handling
+			if err != nil && !test.expectsError {
+				t.Fatalf("did not expect an error, but got: %+v", err)
+			} else if err == nil && test.expectsError {
+				t.Fatalf("did expect an error, but got none")
+			}
+
+			if !test.api.wasCalled && test.expectsCall {
+				t.Fatalf("was not called!")
+			} else if test.api.wasCalled && !test.expectsCall {
+				t.Fatalf("should not have been called")
+			}
+
+			if !test.expectsCall {
+				return
+			}
+
+			if digest != test.api.responseDigest {
+				t.Errorf("unexpected content digest: %q != %q", digest, test.api.responseDigest)
+			}
+
+			// validating that the mock got the right parameters
+			if test.api.sessionID != sessionID {
+				t.Errorf("different session ID: %s != %s", test.api.sessionID, sessionID)
+			}
+
+			var expected map[string]interface{}
+			if err := json.Unmarshal([]byte(test.manifestJSONStr), &expected); err != nil {
+				t.Fatalf("could not unmarshal expected results")
+			}
+
+			for _, d := range deep.Equal(test.api.model, expected) {
+				t.Errorf("model difference: %s", d)
+			}
+
+		})
+	}
+}

internal/anchore/import_dockerfile.go

@@ -0,0 +1,43 @@
+package anchore
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net/http"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/syft/internal/log"
+
+	"github.com/anchore/client-go/pkg/external"
+)
+
+type dockerfileImportAPI interface {
+	ImportImageDockerfile(ctx context.Context, sessionID string, contents string) (external.ImageImportContentResponse, *http.Response, error)
+}
+
+func importDockerfile(ctx context.Context, api dockerfileImportAPI, sessionID string, dockerfile []byte, stage *progress.Stage) (string, error) {
+	if len(dockerfile) > 0 {
+		log.Debug("importing dockerfile")
+		stage.Current = "dockerfile"
+
+		response, httpResponse, err := api.ImportImageDockerfile(ctx, sessionID, string(dockerfile))
+		if err != nil {
+			var openAPIErr external.GenericOpenAPIError
+			if errors.As(err, &openAPIErr) {
+				log.Errorf("api response: %+v", string(openAPIErr.Body()))
+			}
+			return "", fmt.Errorf("unable to import Dockerfile: %w", err)
+		}
+
+		defer httpResponse.Body.Close()
+
+		if httpResponse.StatusCode != 200 {
+			return "", fmt.Errorf("unable to import Dockerfile: %s", httpResponse.Status)
+		}
+
+		return response.Digest, nil
+	}
+	return "", nil
+}

internal/anchore/import_dockerfile_test.go

@@ -0,0 +1,124 @@
+package anchore
+
+import (
+	"context"
+	"fmt"
+	"net/http"
+	"strings"
+	"testing"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/docker/docker/pkg/ioutils"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/go-test/deep"
+)
+
+type mockDockerfileImportAPI struct {
+	sessionID      string
+	model          string
+	httpResponse   *http.Response
+	err            error
+	ctx            context.Context
+	responseDigest string
+	wasCalled      bool
+}
+
+func (m *mockDockerfileImportAPI) ImportImageDockerfile(ctx context.Context, sessionID string, contents string) (external.ImageImportContentResponse, *http.Response, error) {
+	m.wasCalled = true
+	m.model = contents
+	m.sessionID = sessionID
+	m.ctx = ctx
+	if m.httpResponse == nil {
+		m.httpResponse = &http.Response{}
+	}
+	m.httpResponse.Body = ioutils.NewReadCloserWrapper(strings.NewReader(""), func() error { return nil })
+	return external.ImageImportContentResponse{Digest: m.responseDigest}, m.httpResponse, m.err
+}
+
+func TestDockerfileImport(t *testing.T) {
+
+	sessionID := "my-session"
+
+	tests := []struct {
+		name         string
+		dockerfile   string
+		api          *mockDockerfileImportAPI
+		expectsError bool
+		expectsCall  bool
+	}{
+
+		{
+			name:       "Go case: import works",
+			dockerfile: "the-manifest-contents!",
+			api: &mockDockerfileImportAPI{
+				httpResponse:   &http.Response{StatusCode: 200},
+				responseDigest: "digest!",
+			},
+			expectsCall: true,
+		},
+		{
+			name:        "No manifest provided",
+			dockerfile:  "",
+			api:         &mockDockerfileImportAPI{},
+			expectsCall: false,
+		},
+		{
+			name:       "API returns an error",
+			dockerfile: "the-manifest-contents!",
+			api: &mockDockerfileImportAPI{
+				err: fmt.Errorf("api error, something went wrong"),
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+		{
+			name:       "API HTTP-level error",
+			dockerfile: "the-manifest-contents!",
+			api: &mockDockerfileImportAPI{
+				httpResponse: &http.Response{StatusCode: 404},
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+
+			digest, err := importDockerfile(context.TODO(), test.api, sessionID, []byte(test.dockerfile), &progress.Stage{})
+
+			// validate error handling
+			if err != nil && !test.expectsError {
+				t.Fatalf("did not expect an error, but got: %+v", err)
+			} else if err == nil && test.expectsError {
+				t.Fatalf("did expect an error, but got none")
+			}
+
+			if !test.api.wasCalled && test.expectsCall {
+				t.Fatalf("was not called!")
+			} else if test.api.wasCalled && !test.expectsCall {
+				t.Fatalf("should not have been called")
+			}
+
+			if !test.expectsCall {
+				return
+			}
+
+			if digest != test.api.responseDigest {
+				t.Errorf("unexpected content digest: %q != %q", digest, test.api.responseDigest)
+			}
+
+			// validating that the mock got the right parameters
+			if test.api.sessionID != sessionID {
+				t.Errorf("different session ID: %s != %s", test.api.sessionID, sessionID)
+			}
+
+			for _, d := range deep.Equal(test.api.model, test.dockerfile) {
+				t.Errorf("model difference: %s", d)
+			}
+
+		})
+	}
+}

internal/anchore/import_manifest.go

@@ -0,0 +1,50 @@
+// nolint: dupl
+package anchore
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"net/http"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/syft/internal/log"
+)
+
+type manifestImportAPI interface {
+	ImportImageManifest(ctx context.Context, sessionID string, contents interface{}) (external.ImageImportContentResponse, *http.Response, error)
+}
+
+func importManifest(ctx context.Context, api manifestImportAPI, sessionID string, manifest []byte, stage *progress.Stage) (string, error) {
+	if len(manifest) > 0 {
+		log.Debug("importing image manifest")
+		stage.Current = "image manifest"
+
+		// API requires an object, but we do not verify the shape of this object locally
+		var sender map[string]interface{}
+		if err := json.Unmarshal(manifest, &sender); err != nil {
+			return "", err
+		}
+
+		response, httpResponse, err := api.ImportImageManifest(ctx, sessionID, sender)
+		if err != nil {
+			var openAPIErr external.GenericOpenAPIError
+			if errors.As(err, &openAPIErr) {
+				log.Errorf("api response: %+v", string(openAPIErr.Body()))
+			}
+			return "", fmt.Errorf("unable to import Manifest: %w", err)
+		}
+
+		defer httpResponse.Body.Close()
+
+		if httpResponse.StatusCode != 200 {
+			return "", fmt.Errorf("unable to import Manifest: %s", httpResponse.Status)
+		}
+
+		return response.Digest, nil
+	}
+	return "", nil
+}

internal/anchore/import_manifest_test.go

@@ -0,0 +1,129 @@
+package anchore
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"strings"
+	"testing"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/docker/docker/pkg/ioutils"
+	"github.com/go-test/deep"
+)
+
+type mockManifestImportAPI struct {
+	sessionID      string
+	model          interface{}
+	httpResponse   *http.Response
+	err            error
+	ctx            context.Context
+	responseDigest string
+	wasCalled      bool
+}
+
+func (m *mockManifestImportAPI) ImportImageManifest(ctx context.Context, sessionID string, contents interface{}) (external.ImageImportContentResponse, *http.Response, error) {
+	m.wasCalled = true
+	m.model = contents
+	m.sessionID = sessionID
+	m.ctx = ctx
+	if m.httpResponse == nil {
+		m.httpResponse = &http.Response{}
+	}
+	m.httpResponse.Body = ioutils.NewReadCloserWrapper(strings.NewReader(""), func() error { return nil })
+	return external.ImageImportContentResponse{Digest: m.responseDigest}, m.httpResponse, m.err
+}
+
+func TestManifestImport(t *testing.T) {
+
+	sessionID := "my-session"
+
+	tests := []struct {
+		name         string
+		manifest     string
+		api          *mockManifestImportAPI
+		expectsError bool
+		expectsCall  bool
+	}{
+
+		{
+			name:     "Go case: import works",
+			manifest: `{ "key": "the-config-contents!" }`,
+			api: &mockManifestImportAPI{
+				httpResponse:   &http.Response{StatusCode: 200},
+				responseDigest: "digest!",
+			},
+			expectsCall: true,
+		},
+		{
+			name:        "No manifest provided",
+			manifest:    "",
+			api:         &mockManifestImportAPI{},
+			expectsCall: false,
+		},
+		{
+			name:     "API returns an error",
+			manifest: `{ "key": "the-config-contents!" }`,
+			api: &mockManifestImportAPI{
+				err: fmt.Errorf("api error, something went wrong"),
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+		{
+			name:     "API HTTP-level error",
+			manifest: `{ "key": "the-config-contents!" }`,
+			api: &mockManifestImportAPI{
+				httpResponse: &http.Response{StatusCode: 404},
+			},
+			expectsError: true,
+			expectsCall:  true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+
+			digest, err := importManifest(context.TODO(), test.api, sessionID, []byte(test.manifest), &progress.Stage{})
+
+			// validate error handling
+			if err != nil && !test.expectsError {
+				t.Fatalf("did not expect an error, but got: %+v", err)
+			} else if err == nil && test.expectsError {
+				t.Fatalf("did expect an error, but got none")
+			}
+
+			if !test.api.wasCalled && test.expectsCall {
+				t.Fatalf("was not called!")
+			} else if test.api.wasCalled && !test.expectsCall {
+				t.Fatalf("should not have been called")
+			}
+
+			if !test.expectsCall {
+				return
+			}
+
+			if digest != test.api.responseDigest {
+				t.Errorf("unexpected content digest: %q != %q", digest, test.api.responseDigest)
+			}
+
+			// validating that the mock got the right parameters
+			if test.api.sessionID != sessionID {
+				t.Errorf("different session ID: %s != %s", test.api.sessionID, sessionID)
+			}
+
+			var expected map[string]interface{}
+			if err := json.Unmarshal([]byte(test.manifest), &expected); err != nil {
+				t.Fatalf("could not unmarshal expected results")
+			}
+
+			for _, d := range deep.Equal(test.api.model, expected) {
+				t.Errorf("model difference: %s", d)
+			}
+
+		})
+	}
+}

internal/anchore/import_package_sbom.go

@@ -0,0 +1,67 @@
+package anchore
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"net/http"
+
+	"github.com/anchore/syft/syft/sbom"
+
+	"github.com/anchore/syft/internal/formats/syftjson"
+
+	"github.com/wagoodman/go-progress"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/syft/internal/log"
+)
+
+type packageSBOMImportAPI interface {
+	ImportImagePackages(context.Context, string, external.ImagePackageManifest) (external.ImageImportContentResponse, *http.Response, error)
+}
+
+func packageSbomModel(s sbom.SBOM) (*external.ImagePackageManifest, error) {
+	var buf bytes.Buffer
+
+	err := syftjson.Format().Presenter(s).Present(&buf)
+	if err != nil {
+		return nil, fmt.Errorf("unable to serialize results: %w", err)
+	}
+
+	// the model is 1:1 the JSON output of today. As the schema changes, this will need to be converted into individual mappings.
+	var model external.ImagePackageManifest
+	if err = json.Unmarshal(buf.Bytes(), &model); err != nil {
+		return nil, fmt.Errorf("unable to convert JSON presenter output to import model: %w", err)
+	}
+
+	return &model, nil
+}
+
+func importPackageSBOM(ctx context.Context, api packageSBOMImportAPI, sessionID string, s sbom.SBOM, stage *progress.Stage) (string, error) {
+	log.Debug("importing package SBOM")
+	stage.Current = "package SBOM"
+
+	model, err := packageSbomModel(s)
+	if err != nil {
+		return "", fmt.Errorf("unable to create PackageSBOM model: %w", err)
+	}
+
+	response, httpResponse, err := api.ImportImagePackages(ctx, sessionID, *model)
+	if err != nil {
+		var openAPIErr external.GenericOpenAPIError
+		if errors.As(err, &openAPIErr) {
+			log.Errorf("api response: %+v", string(openAPIErr.Body()))
+		}
+		return "", fmt.Errorf("unable to import PackageSBOM: %w", err)
+	}
+
+	defer httpResponse.Body.Close()
+
+	if httpResponse.StatusCode != 200 {
+		return "", fmt.Errorf("unable to import PackageSBOM: %s", httpResponse.Status)
+	}
+
+	return response.Digest, nil
+}

internal/anchore/import_package_sbom_test.go

@@ -0,0 +1,283 @@
+package anchore
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"strings"
+	"testing"
+
+	"github.com/anchore/syft/syft/sbom"
+
+	"github.com/anchore/client-go/pkg/external"
+	"github.com/anchore/syft/internal/formats/syftjson"
+	syftjsonModel "github.com/anchore/syft/internal/formats/syftjson/model"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+	"github.com/docker/docker/pkg/ioutils"
+	"github.com/go-test/deep"
+	"github.com/wagoodman/go-progress"
+)
+
+func must(c pkg.CPE, e error) pkg.CPE {
+	if e != nil {
+		panic(e)
+	}
+	return c
+}
+
+// this test is tailored towards the assumption that the import doc shape and the syft json shape are the same.
+// TODO: replace this as the document shapes diverge.
+func TestPackageSbomToModel(t *testing.T) {
+
+	m := source.Metadata{
+		Scheme: source.ImageScheme,
+		ImageMetadata: source.ImageMetadata{
+			UserInput: "user-in",
+			Layers: []source.LayerMetadata{
+				{
+					MediaType: "layer-metadata-type!",
+					Digest:    "layer-digest",
+					Size:      20,
+				},
+			},
+			Size:           10,
+			ManifestDigest: "sha256:digest!",
+			MediaType:      "mediatype!",
+			Tags:           nil,
+		},
+	}
+
+	d, _ := distro.NewDistro(distro.CentOS, "8.0", "")
+
+	p := pkg.Package{
+		Name:    "name",
+		Version: "version",
+		FoundBy: "foundBy",
+		Locations: []source.Location{
+			{
+				Coordinates: source.Coordinates{
+					RealPath:     "path",
+					FileSystemID: "layerID",
+				},
+			},
+		},
+		Licenses: []string{"license"},
+		Language: pkg.Python,
+		Type:     pkg.PythonPkg,
+		CPEs: []pkg.CPE{
+			must(pkg.NewCPE("cpe:2.3:*:some:package:1:*:*:*:*:*:*:*")),
+		},
+		PURL: "purl",
+	}
+
+	c := pkg.NewCatalog(p)
+
+	sbomResult := sbom.SBOM{
+		Artifacts: sbom.Artifacts{
+			PackageCatalog: c,
+			Distro:         &d,
+		},
+		Source: m,
+	}
+
+	model, err := packageSbomModel(sbomResult)
+	if err != nil {
+		t.Fatalf("unable to generate model from source material: %+v", err)
+	}
+
+	var modelJSON []byte
+
+	modelJSON, err = json.Marshal(&model)
+	if err != nil {
+		t.Fatalf("unable to marshal model: %+v", err)
+	}
+
+	s := sbom.SBOM{
+		Artifacts: sbom.Artifacts{
+			PackageCatalog: c,
+			Distro:         &d,
+		},
+		Source: m,
+	}
+
+	var buf bytes.Buffer
+	pres := syftjson.Format().Presenter(s)
+	if err := pres.Present(&buf); err != nil {
+		t.Fatalf("unable to get expected json: %+v", err)
+	}
+
+	// unmarshal expected result
+	var expectedDoc syftjsonModel.Document
+	if err := json.Unmarshal(buf.Bytes(), &expectedDoc); err != nil {
+		t.Fatalf("unable to parse json doc: %+v", err)
+	}
+
+	// unmarshal actual result
+	var actualDoc syftjsonModel.Document
+	if err := json.Unmarshal(modelJSON, &actualDoc); err != nil {
+		t.Fatalf("unable to parse json doc: %+v", err)
+	}
+
+	for _, d := range deep.Equal(actualDoc, expectedDoc) {
+		if strings.HasSuffix(d, "<nil slice> != []") {
+			// do not consider nil vs empty collection semantics as a "difference"
+			continue
+		}
+		t.Errorf("diff: %+v", d)
+	}
+}
+
+type mockPackageSBOMImportAPI struct {
+	sessionID      string
+	model          external.ImagePackageManifest
+	httpResponse   *http.Response
+	err            error
+	ctx            context.Context
+	responseDigest string
+}
+
+func (m *mockPackageSBOMImportAPI) ImportImagePackages(ctx context.Context, sessionID string, model external.ImagePackageManifest) (external.ImageImportContentResponse, *http.Response, error) {
+	m.model = model
+	m.sessionID = sessionID
+	m.ctx = ctx
+	if m.httpResponse == nil {
+		m.httpResponse = &http.Response{}
+	}
+	m.httpResponse.Body = ioutils.NewReadCloserWrapper(strings.NewReader(""), func() error { return nil })
+	return external.ImageImportContentResponse{Digest: m.responseDigest}, m.httpResponse, m.err
+}
+
+func TestPackageSbomImport(t *testing.T) {
+
+	catalog := pkg.NewCatalog(pkg.Package{
+		Name:    "name",
+		Version: "version",
+		FoundBy: "foundBy",
+		Locations: []source.Location{
+			{
+				Coordinates: source.Coordinates{
+					RealPath:     "path",
+					FileSystemID: "layerID",
+				},
+			},
+		},
+		Licenses: []string{"license"},
+		Language: pkg.Python,
+		Type:     pkg.PythonPkg,
+		CPEs: []pkg.CPE{
+			must(pkg.NewCPE("cpe:2.3:*:some:package:1:*:*:*:*:*:*:*")),
+		},
+		PURL:         "purl",
+		MetadataType: pkg.PythonPackageMetadataType,
+		Metadata: pkg.PythonPackageMetadata{
+			Name:        "p-name",
+			Version:     "p-version",
+			License:     "p-license",
+			Author:      "p-author",
+			AuthorEmail: "p-email",
+			Platform:    "p-platform",
+			Files: []pkg.PythonFileRecord{
+				{
+					Path: "p-path",
+					Digest: &pkg.PythonFileDigest{
+						Algorithm: "p-alg",
+						Value:     "p-digest",
+					},
+					Size: "p-size",
+				},
+			},
+			SitePackagesRootPath: "p-site-packages-root",
+			TopLevelPackages:     []string{"top-level"},
+		},
+	})
+
+	m := source.Metadata{
+		Scheme: source.ImageScheme,
+		ImageMetadata: source.ImageMetadata{
+			UserInput:      "user-in",
+			Layers:         nil,
+			Size:           10,
+			ManifestDigest: "sha256:digest!",
+			MediaType:      "mediatype!",
+			Tags:           nil,
+		},
+	}
+
+	d, _ := distro.NewDistro(distro.CentOS, "8.0", "")
+
+	sbomResult := sbom.SBOM{
+		Artifacts: sbom.Artifacts{
+			PackageCatalog: catalog,
+			Distro:         &d,
+		},
+		Source: m,
+	}
+
+	theModel, err := packageSbomModel(sbomResult)
+	if err != nil {
+		t.Fatalf("could not get sbom model: %+v", err)
+	}
+
+	sessionID := "my-session"
+
+	tests := []struct {
+		name         string
+		api          *mockPackageSBOMImportAPI
+		expectsError bool
+	}{
+
+		{
+			name: "Go case: import works",
+			api: &mockPackageSBOMImportAPI{
+				httpResponse:   &http.Response{StatusCode: 200},
+				responseDigest: "digest!",
+			},
+		},
+		{
+			name: "API returns an error",
+			api: &mockPackageSBOMImportAPI{
+				err: fmt.Errorf("API error, something went wrong."),
+			},
+			expectsError: true,
+		},
+		{
+			name: "API HTTP-level error",
+			api: &mockPackageSBOMImportAPI{
+				httpResponse: &http.Response{StatusCode: 404},
+			},
+			expectsError: true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+
+			digest, err := importPackageSBOM(context.TODO(), test.api, sessionID, sbomResult, &progress.Stage{})
+
+			// validate error handling
+			if err != nil && !test.expectsError {
+				t.Fatalf("did not expect an error, but got: %+v", err)
+			} else if err == nil && test.expectsError {
+				t.Fatalf("did expect an error, but got none")
+			}
+
+			if digest != test.api.responseDigest {
+				t.Errorf("unexpected content digest: %q != %q", digest, test.api.responseDigest)
+			}
+
+			// validating that the mock got the right parameters (api.ImportImagePackages)
+			if test.api.sessionID != sessionID {
+				t.Errorf("different session ID: %s != %s", test.api.sessionID, sessionID)
+			}
+
+			for _, d := range deep.Equal(&test.api.model, theModel) {
+				t.Errorf("model difference: %s", d)
+			}
+
+		})
+	}
+}

internal/bus/bus.go

@@ -1,3 +1,16 @@
+/*
+Package bus provides access to a singleton instance of an event bus (provided by the calling application). The event bus
+is intended to allow for the syft library to publish events which library consumers can subscribe to. These events
+can provide static information, but also have an object as a payload for which the consumer can poll for updates.
+This is akin to a logger, except instead of only allowing strings to be logged, rich objects that can be interacted with.
+
+Note that the singleton instance is only allowed to publish events and not subscribe to them --this is intentional.
+Internal library interactions should continue to use traditional in-execution-path approaches for data sharing
+(e.g. function returns and channels) and not depend on bus subscriptions for critical interactions (e.g. one part of the
+lib publishes an event and another part of the lib subscribes and reacts to that event). The bus is provided only as a
+means for consumers to observe events emitted from the library (such as to provide a rich UI) and not to allow
+consumers to augment or otherwise change execution.
+*/
 package bus
 
 import "github.com/wagoodman/go-partybus"
@@ -5,6 +18,8 @@ import "github.com/wagoodman/go-partybus"
 var publisher partybus.Publisher
 var active bool
 
+// SetPublisher sets the singleton event bus publisher. This is optional; if no bus is provided, the library will
+// behave no differently than if a bus had been provided.
 func SetPublisher(p partybus.Publisher) {
 	publisher = p
 	if p != nil {
@@ -12,6 +27,7 @@ func SetPublisher(p partybus.Publisher) {
 	}
 }
 
+// Publish an event onto the bus. If there is no bus set by the calling application, this does nothing.
 func Publish(event partybus.Event) {
 	if active {
 		publisher.Publish(event)

internal/config/anchore.go

@@ -0,0 +1,21 @@
+package config
+
+import "github.com/spf13/viper"
+
+type anchore struct {
+	// upload options
+	Host string `yaml:"host" json:"host" mapstructure:"host"` // -H , hostname of the engine/enterprise instance to upload to (setting this value enables upload)
+	Path string `yaml:"path" json:"path" mapstructure:"path"` // override the engine/enterprise API upload path
+	// IMPORTANT: do not show the username in any YAML/JSON output (sensitive information)
+	Username string `yaml:"-" json:"-" mapstructure:"username"` // -u , username to authenticate upload
+	// IMPORTANT: do not show the password in any YAML/JSON output (sensitive information)
+	Password               string `yaml:"-" json:"-" mapstructure:"password"`                                                               // -p , password to authenticate upload
+	Dockerfile             string `yaml:"dockerfile" json:"dockerfile" mapstructure:"dockerfile"`                                           // -d , dockerfile to attach for upload
+	OverwriteExistingImage bool   `yaml:"overwrite-existing-image" json:"overwrite-existing-image" mapstructure:"overwrite-existing-image"` // --overwrite-existing-image , if any of the SBOM components have already been uploaded this flag will ensure they are overwritten with the current upload
+	ImportTimeout          uint   `yaml:"import-timeout" json:"import-timeout" mapstructure:"import-timeout"`                               // --import-timeout
+	// , customize the number of seconds within which the SBOM import must be completed or canceled
+}
+
+func (cfg anchore) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("anchore.path", "")
+}

internal/config/application.go

@@ -0,0 +1,288 @@
+package config
+
+import (
+	"errors"
+	"fmt"
+	"os"
+	"os/exec"
+	"path"
+	"path/filepath"
+	"reflect"
+	"strings"
+
+	"github.com/adrg/xdg"
+	"github.com/anchore/syft/internal"
+	"github.com/mitchellh/go-homedir"
+	"github.com/sirupsen/logrus"
+	"github.com/spf13/viper"
+	"gopkg.in/yaml.v2"
+)
+
+var ErrApplicationConfigNotFound = fmt.Errorf("application config not found")
+
+var catalogerEnabledDefault = false
+
+type defaultValueLoader interface {
+	loadDefaultValues(*viper.Viper)
+}
+
+type parser interface {
+	parseConfigValues() error
+}
+
+// Application is the main syft application configuration.
+type Application struct {
+	ConfigPath         string             `yaml:",omitempty" json:"configPath"`                                                         // the location where the application config was read from (either from -c or discovered while loading)
+	Output             string             `yaml:"output" json:"output" mapstructure:"output"`                                           // -o, the Presenter hint string to use for report formatting
+	File               string             `yaml:"file" json:"file" mapstructure:"file"`                                                 // --file, the file to write report output to
+	Quiet              bool               `yaml:"quiet" json:"quiet" mapstructure:"quiet"`                                              // -q, indicates to not show any status output to stderr (ETUI or logging UI)
+	CheckForAppUpdate  bool               `yaml:"check-for-app-update" json:"check-for-app-update" mapstructure:"check-for-app-update"` // whether to check for an application update on start up or not
+	Anchore            anchore            `yaml:"anchore" json:"anchore" mapstructure:"anchore"`                                        // options for interacting with Anchore Engine/Enterprise
+	CliOptions         CliOnlyOptions     `yaml:"-" json:"-"`                                                                           // all options only available through the CLI (not via env vars or config)
+	Dev                development        `yaml:"dev" json:"dev" mapstructure:"dev"`
+	Log                logging            `yaml:"log" json:"log" mapstructure:"log"` // all logging-related options
+	Package            packages           `yaml:"package" json:"package" mapstructure:"package"`
+	FileMetadata       FileMetadata       `yaml:"file-metadata" json:"file-metadata" mapstructure:"file-metadata"`
+	FileClassification fileClassification `yaml:"file-classification" json:"file-classification" mapstructure:"file-classification"`
+	FileContents       fileContents       `yaml:"file-contents" json:"file-contents" mapstructure:"file-contents"`
+	Secrets            secrets            `yaml:"secrets" json:"secrets" mapstructure:"secrets"`
+	Registry           registry           `yaml:"registry" json:"registry" mapstructure:"registry"`
+}
+
+// PowerUserCatalogerEnabledDefault switches all catalogers to be enabled when running power-user command
+func PowerUserCatalogerEnabledDefault() {
+	catalogerEnabledDefault = true
+}
+
+func newApplicationConfig(v *viper.Viper, cliOpts CliOnlyOptions) *Application {
+	config := &Application{
+		CliOptions: cliOpts,
+	}
+	config.loadDefaultValues(v)
+	return config
+}
+
+// LoadApplicationConfig populates the given viper object with application configuration discovered on disk
+func LoadApplicationConfig(v *viper.Viper, cliOpts CliOnlyOptions) (*Application, error) {
+	// the user may not have a config, and this is OK, we can use the default config + default cobra cli values instead
+	config := newApplicationConfig(v, cliOpts)
+
+	if err := readConfig(v, cliOpts.ConfigPath); err != nil && !errors.Is(err, ErrApplicationConfigNotFound) {
+		return nil, err
+	}
+
+	if err := v.Unmarshal(config); err != nil {
+		return nil, fmt.Errorf("unable to parse config: %w", err)
+	}
+	config.ConfigPath = v.ConfigFileUsed()
+
+	if err := config.parseConfigValues(); err != nil {
+		return nil, fmt.Errorf("invalid application config: %w", err)
+	}
+
+	return config, nil
+}
+
+// init loads the default configuration values into the viper instance (before the config values are read and parsed).
+func (cfg Application) loadDefaultValues(v *viper.Viper) {
+	// set the default values for primitive fields in this struct
+	v.SetDefault("check-for-app-update", true)
+
+	// for each field in the configuration struct, see if the field implements the defaultValueLoader interface and invoke it if it does
+	value := reflect.ValueOf(cfg)
+	for i := 0; i < value.NumField(); i++ {
+		// note: the defaultValueLoader method receiver is NOT a pointer receiver.
+		if loadable, ok := value.Field(i).Interface().(defaultValueLoader); ok {
+			// the field implements defaultValueLoader, call it
+			loadable.loadDefaultValues(v)
+		}
+	}
+}
+
+func (cfg *Application) parseConfigValues() error {
+	// parse application config options
+	for _, optionFn := range []func() error{
+		cfg.parseUploadOptions,
+		cfg.parseLogLevelOption,
+	} {
+		if err := optionFn(); err != nil {
+			return err
+		}
+	}
+
+	// parse nested config options
+	// for each field in the configuration struct, see if the field implements the parser interface
+	// note: the app config is a pointer, so we need to grab the elements explicitly (to traverse the address)
+	value := reflect.ValueOf(cfg).Elem()
+	for i := 0; i < value.NumField(); i++ {
+		// note: since the interface method of parser is a pointer receiver we need to get the value of the field as a pointer.
+		if parsable, ok := value.Field(i).Addr().Interface().(parser); ok {
+			// the field implements parser, call it
+			if err := parsable.parseConfigValues(); err != nil {
+				return err
+			}
+		}
+	}
+	return nil
+}
+
+func (cfg *Application) parseUploadOptions() error {
+	if cfg.Anchore.Host == "" && cfg.Anchore.Dockerfile != "" {
+		return fmt.Errorf("cannot provide dockerfile option without enabling upload")
+	}
+	return nil
+}
+
+func (cfg *Application) parseLogLevelOption() error {
+	switch {
+	case cfg.Quiet:
+		// TODO: this is bad: quiet option trumps all other logging options (such as to a file on disk)
+		// we should be able to quiet the console logging and leave file logging alone...
+		// ... this will be an enhancement for later
+		cfg.Log.LevelOpt = logrus.PanicLevel
+	case cfg.Log.Level != "":
+		if cfg.CliOptions.Verbosity > 0 {
+			return fmt.Errorf("cannot explicitly set log level (cfg file or env var) and use -v flag together")
+		}
+
+		lvl, err := logrus.ParseLevel(strings.ToLower(cfg.Log.Level))
+		if err != nil {
+			return fmt.Errorf("bad log level configured (%q): %w", cfg.Log.Level, err)
+		}
+
+		cfg.Log.LevelOpt = lvl
+		if cfg.Log.LevelOpt >= logrus.InfoLevel {
+			cfg.CliOptions.Verbosity = 1
+		}
+	default:
+
+		switch v := cfg.CliOptions.Verbosity; {
+		case v == 1:
+			cfg.Log.LevelOpt = logrus.InfoLevel
+		case v >= 2:
+			cfg.Log.LevelOpt = logrus.DebugLevel
+		default:
+			cfg.Log.LevelOpt = logrus.ErrorLevel
+		}
+	}
+
+	if cfg.Log.Level == "" {
+		cfg.Log.Level = cfg.Log.LevelOpt.String()
+	}
+
+	return nil
+}
+
+func (cfg Application) String() string {
+	// yaml is pretty human friendly (at least when compared to json)
+	appCfgStr, err := yaml.Marshal(&cfg)
+
+	if err != nil {
+		return err.Error()
+	}
+
+	return string(appCfgStr)
+}
+
+// readConfig attempts to read the given config path from disk or discover an alternate store location
+// nolint:funlen
+func readConfig(v *viper.Viper, configPath string) error {
+	var err error
+	v.AutomaticEnv()
+	v.SetEnvPrefix(internal.ApplicationName)
+	// allow for nested options to be specified via environment variables
+	// e.g. pod.context = APPNAME_POD_CONTEXT
+	v.SetEnvKeyReplacer(strings.NewReplacer(".", "_", "-", "_"))
+
+	// use explicitly the given user config
+	if configPath != "" {
+		v.SetConfigFile(configPath)
+		if err := v.ReadInConfig(); err != nil {
+			return fmt.Errorf("unable to read application config=%q : %w", configPath, err)
+		}
+		// don't fall through to other options if the config path was explicitly provided
+		return nil
+	}
+
+	// start searching for valid configs in order...
+
+	// 1. look for .<appname>.yaml (in the current directory)
+	v.AddConfigPath(".")
+	v.SetConfigName("." + internal.ApplicationName)
+	if exists, err := readViperConfig(v); exists || err != nil {
+		return err
+	}
+
+	// 2. look for .<appname>/config.yaml (in the current directory)
+	v.AddConfigPath("." + internal.ApplicationName)
+	v.SetConfigName("config")
+	if exists, err := readViperConfig(v); exists || err != nil {
+		return err
+	}
+
+	if root := gitRepoRoot(); root != "" {
+		// 3. look for <git repo root>/.<appname>.yaml, where the repo root is defined by "git rev-parse --show-toplevel"
+		v.AddConfigPath(root)
+		v.SetConfigName("." + internal.ApplicationName)
+		if exists, err := readViperConfig(v); exists {
+			if err != nil {
+				return err
+			}
+			return os.Chdir(root)
+		}
+
+		// 4. look for <git repo root>/.<appname>/config.yaml, where the repo root is defined by "git rev-parse --show-toplevel"
+		v.AddConfigPath(filepath.Join(root, "."+internal.ApplicationName))
+		v.SetConfigName("config")
+		if exists, err := readViperConfig(v); exists {
+			if err != nil {
+				return err
+			}
+			return os.Chdir(root)
+		}
+	}
+
+	// 5. look for ~/.<appname>.yaml
+	home, err := homedir.Dir()
+	if err == nil {
+		v.AddConfigPath(home)
+		v.SetConfigName("." + internal.ApplicationName)
+		if exists, err := readViperConfig(v); exists || err != nil {
+			return err
+		}
+	}
+
+	// 6. look for <appname>/config.yaml in xdg locations (starting with xdg home config dir, then moving upwards)
+	v.AddConfigPath(path.Join(xdg.ConfigHome, internal.ApplicationName))
+	for _, dir := range xdg.ConfigDirs {
+		v.AddConfigPath(path.Join(dir, internal.ApplicationName))
+	}
+	v.SetConfigName("config")
+	if exists, err := readViperConfig(v); exists || err != nil {
+		return err
+	}
+
+	return ErrApplicationConfigNotFound
+}
+
+func readViperConfig(v *viper.Viper) (bool, error) {
+	err := v.ReadInConfig()
+	if err == nil {
+		return true, nil
+	} else if errors.As(err, &viper.ConfigFileNotFoundError{}) {
+		return false, nil
+	}
+	return true, fmt.Errorf("unable to parse config=%q: %w", v.ConfigFileUsed(), err)
+}
+
+func gitRepoRoot() string {
+	root, err := exec.Command("git", "rev-parse", "--show-toplevel").Output()
+	if err != nil {
+		return ""
+	}
+	absRepoRoot, err := filepath.Abs(strings.TrimSpace(string(root)))
+	if err != nil {
+		return ""
+	}
+	return absRepoRoot
+}

internal/config/cataloger_options.go

@@ -0,0 +1,23 @@
+package config
+
+import (
+	"fmt"
+
+	"github.com/anchore/syft/syft/source"
+)
+
+type catalogerOptions struct {
+	Enabled  bool         `yaml:"enabled" json:"enabled" mapstructure:"enabled"`
+	Scope    string       `yaml:"scope" json:"scope" mapstructure:"scope"`
+	ScopeOpt source.Scope `yaml:"-" json:"-"`
+}
+
+func (cfg *catalogerOptions) parseConfigValues() error {
+	scopeOption := source.ParseScope(cfg.Scope)
+	if scopeOption == source.UnknownScope {
+		return fmt.Errorf("bad scope value %q", cfg.Scope)
+	}
+	cfg.ScopeOpt = scopeOption
+
+	return nil
+}

internal/config/cli_only_options.go

@@ -0,0 +1,7 @@
+package config
+
+// CliOnlyOptions are options that are in the application config in memory, but are only exposed via CLI switches (not from unmarshaling a config file)
+type CliOnlyOptions struct {
+	ConfigPath string // -c. where the read config is on disk
+	Verbosity  int    // -v or -vv , controlling which UI (ETUI vs logging) and what the log level should be
+}

internal/config/config.go

@@ -1,176 +0,0 @@
-package config
-
-import (
-	"fmt"
-	"path"
-	"strings"
-
-	"github.com/adrg/xdg"
-	"github.com/anchore/syft/internal"
-	"github.com/anchore/syft/syft/presenter"
-	"github.com/anchore/syft/syft/scope"
-	"github.com/mitchellh/go-homedir"
-	"github.com/sirupsen/logrus"
-	"github.com/spf13/viper"
-)
-
-type CliOnlyOptions struct {
-	ConfigPath string
-	Verbosity  int
-}
-
-type Application struct {
-	ConfigPath        string
-	PresenterOpt      presenter.Option
-	Output            string `mapstructure:"output"`
-	ScopeOpt          scope.Option
-	Scope             string  `mapstructure:"scope"`
-	Quiet             bool    `mapstructure:"quiet"`
-	Log               Logging `mapstructure:"log"`
-	CliOptions        CliOnlyOptions
-	CheckForAppUpdate bool `mapstructure:"check-for-app-update"`
-}
-
-type Logging struct {
-	Structured   bool `mapstructure:"structured"`
-	LevelOpt     logrus.Level
-	Level        string `mapstructure:"level"`
-	FileLocation string `mapstructure:"file"`
-}
-
-func setNonCliDefaultValues(v *viper.Viper) {
-	v.SetDefault("log.level", "")
-	v.SetDefault("log.file", "")
-	v.SetDefault("log.structured", false)
-	v.SetDefault("check-for-app-update", true)
-}
-
-func LoadConfigFromFile(v *viper.Viper, cliOpts *CliOnlyOptions) (*Application, error) {
-	// the user may not have a config, and this is OK, we can use the default config + default cobra cli values instead
-	setNonCliDefaultValues(v)
-	if cliOpts != nil {
-		_ = readConfig(v, cliOpts.ConfigPath)
-	} else {
-		_ = readConfig(v, "")
-	}
-
-	config := &Application{
-		CliOptions: *cliOpts,
-	}
-	err := v.Unmarshal(config)
-	if err != nil {
-		return nil, fmt.Errorf("unable to parse config: %w", err)
-	}
-	config.ConfigPath = v.ConfigFileUsed()
-
-	err = config.Build()
-	if err != nil {
-		return nil, fmt.Errorf("invalid config: %w", err)
-	}
-
-	return config, nil
-}
-
-func (cfg *Application) Build() error {
-	// set the presenter
-	presenterOption := presenter.ParseOption(cfg.Output)
-	if presenterOption == presenter.UnknownPresenter {
-		return fmt.Errorf("bad --output value '%s'", cfg.Output)
-	}
-	cfg.PresenterOpt = presenterOption
-
-	// set the scope
-	scopeOption := scope.ParseOption(cfg.Scope)
-	if scopeOption == scope.UnknownScope {
-		return fmt.Errorf("bad --scope value '%s'", cfg.Scope)
-	}
-	cfg.ScopeOpt = scopeOption
-
-	if cfg.Quiet {
-		// TODO: this is bad: quiet option trumps all other logging options
-		// we should be able to quiet the console logging and leave file logging alone...
-		// ... this will be an enhancement for later
-		cfg.Log.LevelOpt = logrus.PanicLevel
-	} else {
-		if cfg.Log.Level != "" {
-			if cfg.CliOptions.Verbosity > 0 {
-				return fmt.Errorf("cannot explicitly set log level (cfg file or env var) and use -v flag together")
-			}
-
-			lvl, err := logrus.ParseLevel(strings.ToLower(cfg.Log.Level))
-			if err != nil {
-				return fmt.Errorf("bad log level configured (%q): %w", cfg.Log.Level, err)
-			}
-			// set the log level explicitly
-			cfg.Log.LevelOpt = lvl
-		} else {
-			// set the log level implicitly
-			switch v := cfg.CliOptions.Verbosity; {
-			case v == 1:
-				cfg.Log.LevelOpt = logrus.InfoLevel
-			case v >= 2:
-				cfg.Log.LevelOpt = logrus.DebugLevel
-			default:
-				cfg.Log.LevelOpt = logrus.ErrorLevel
-			}
-		}
-	}
-
-	return nil
-}
-
-func readConfig(v *viper.Viper, configPath string) error {
-	v.AutomaticEnv()
-	v.SetEnvPrefix(internal.ApplicationName)
-	// allow for nested options to be specified via environment variables
-	// e.g. pod.context = APPNAME_POD_CONTEXT
-	v.SetEnvKeyReplacer(strings.NewReplacer(".", "_", "-", "_"))
-
-	// use explicitly the given user config
-	if configPath != "" {
-		v.SetConfigFile(configPath)
-		if err := v.ReadInConfig(); err == nil {
-			return nil
-		}
-		// don't fall through to other options if this fails
-		return fmt.Errorf("unable to read config: %v", configPath)
-	}
-
-	// start searching for valid configs in order...
-
-	// 1. look for .<appname>.yaml (in the current directory)
-	v.AddConfigPath(".")
-	v.SetConfigName(internal.ApplicationName)
-	if err := v.ReadInConfig(); err == nil {
-		return nil
-	}
-
-	// 2. look for .<appname>/config.yaml (in the current directory)
-	v.AddConfigPath("." + internal.ApplicationName)
-	v.SetConfigName("config")
-	if err := v.ReadInConfig(); err == nil {
-		return nil
-	}
-
-	// 3. look for ~/.<appname>.yaml
-	home, err := homedir.Dir()
-	if err == nil {
-		v.AddConfigPath(home)
-		v.SetConfigName("." + internal.ApplicationName)
-		if err := v.ReadInConfig(); err == nil {
-			return nil
-		}
-	}
-
-	// 4. look for <appname>/config.yaml in xdg locations (starting with xdg home config dir, then moving upwards)
-	v.AddConfigPath(path.Join(xdg.ConfigHome, internal.ApplicationName))
-	for _, dir := range xdg.ConfigDirs {
-		v.AddConfigPath(path.Join(dir, internal.ApplicationName))
-	}
-	v.SetConfigName("config")
-	if err := v.ReadInConfig(); err == nil {
-		return nil
-	}
-
-	return fmt.Errorf("application config not found")
-}

internal/config/development.go

@@ -0,0 +1,13 @@
+package config
+
+import "github.com/spf13/viper"
+
+type development struct {
+	ProfileCPU bool `yaml:"profile-cpu" json:"profile-cpu" mapstructure:"profile-cpu"`
+	ProfileMem bool `yaml:"profile-mem" json:"profile-mem" mapstructure:"profile-mem"`
+}
+
+func (cfg development) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("dev.profile-cpu", false)
+	v.SetDefault("dev.profile-mem", false)
+}

internal/config/file_classification.go

@@ -0,0 +1,19 @@
+package config
+
+import (
+	"github.com/anchore/syft/syft/source"
+	"github.com/spf13/viper"
+)
+
+type fileClassification struct {
+	Cataloger catalogerOptions `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
+}
+
+func (cfg fileClassification) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("file-classification.cataloger.enabled", catalogerEnabledDefault)
+	v.SetDefault("file-classification.cataloger.scope", source.SquashedScope)
+}
+
+func (cfg *fileClassification) parseConfigValues() error {
+	return cfg.Cataloger.parseConfigValues()
+}

internal/config/file_contents.go

@@ -0,0 +1,24 @@
+package config
+
+import (
+	"github.com/anchore/syft/internal/file"
+	"github.com/anchore/syft/syft/source"
+	"github.com/spf13/viper"
+)
+
+type fileContents struct {
+	Cataloger          catalogerOptions `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
+	SkipFilesAboveSize int64            `yaml:"skip-files-above-size" json:"skip-files-above-size" mapstructure:"skip-files-above-size"`
+	Globs              []string         `yaml:"globs" json:"globs" mapstructure:"globs"`
+}
+
+func (cfg fileContents) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("file-contents.cataloger.enabled", catalogerEnabledDefault)
+	v.SetDefault("file-contents.cataloger.scope", source.SquashedScope)
+	v.SetDefault("file-contents.skip-files-above-size", 1*file.MB)
+	v.SetDefault("file-contents.globs", []string{})
+}
+
+func (cfg *fileContents) parseConfigValues() error {
+	return cfg.Cataloger.parseConfigValues()
+}

internal/config/file_metadata.go

@@ -0,0 +1,21 @@
+package config
+
+import (
+	"github.com/anchore/syft/syft/source"
+	"github.com/spf13/viper"
+)
+
+type FileMetadata struct {
+	Cataloger catalogerOptions `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
+	Digests   []string         `yaml:"digests" json:"digests" mapstructure:"digests"`
+}
+
+func (cfg FileMetadata) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("file-metadata.cataloger.enabled", catalogerEnabledDefault)
+	v.SetDefault("file-metadata.cataloger.scope", source.SquashedScope)
+	v.SetDefault("file-metadata.digests", []string{"sha256"})
+}
+
+func (cfg *FileMetadata) parseConfigValues() error {
+	return cfg.Cataloger.parseConfigValues()
+}

internal/config/logging.go

@@ -0,0 +1,18 @@
+package config
+
+import (
+	"github.com/sirupsen/logrus"
+	"github.com/spf13/viper"
+)
+
+// logging contains all logging-related configuration options available to the user via the application config.
+type logging struct {
+	Structured   bool         `yaml:"structured" json:"structured" mapstructure:"structured"` // show all log entries as JSON formatted strings
+	LevelOpt     logrus.Level `yaml:"-" json:"-"`                                             // the native log level object used by the logger
+	Level        string       `yaml:"level" json:"level" mapstructure:"level"`                // the log level string hint
+	FileLocation string       `yaml:"file" json:"file-location" mapstructure:"file"`          // the file path to write logs to
+}
+
+func (cfg logging) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("log.structured", false)
+}

internal/config/packages.go

@@ -0,0 +1,15 @@
+package config
+
+import "github.com/spf13/viper"
+
+type packages struct {
+	Cataloger catalogerOptions `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
+}
+
+func (cfg packages) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("package.cataloger.enabled", true)
+}
+
+func (cfg *packages) parseConfigValues() error {
+	return cfg.Cataloger.parseConfigValues()
+}

internal/config/registry.go

@@ -0,0 +1,75 @@
+package config
+
+import (
+	"os"
+
+	"github.com/anchore/stereoscope/pkg/image"
+
+	"github.com/spf13/viper"
+)
+
+type RegistryCredentials struct {
+	Authority string `yaml:"authority" json:"authority" mapstructure:"authority"`
+	// IMPORTANT: do not show the username in any YAML/JSON output (sensitive information)
+	Username string `yaml:"-" json:"-" mapstructure:"username"`
+	// IMPORTANT: do not show the password in any YAML/JSON output (sensitive information)
+	Password string `yaml:"-" json:"-" mapstructure:"password"`
+	// IMPORTANT: do not show the token in any YAML/JSON output (sensitive information)
+	Token string `yaml:"-" json:"-" mapstructure:"token"`
+}
+
+type registry struct {
+	InsecureSkipTLSVerify bool                  `yaml:"insecure-skip-tls-verify" json:"insecure-skip-tls-verify" mapstructure:"insecure-skip-tls-verify"`
+	InsecureUseHTTP       bool                  `yaml:"insecure-use-http" json:"insecure-use-http" mapstructure:"insecure-use-http"`
+	Auth                  []RegistryCredentials `yaml:"auth" json:"auth" mapstructure:"auth"`
+}
+
+func (cfg registry) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("registry.insecure-skip-tls-verify", false)
+	v.SetDefault("registry.insecure-use-http", false)
+	v.SetDefault("registry.auth", []RegistryCredentials{})
+}
+
+// nolint: unparam
+func (cfg *registry) parseConfigValues() error {
+	// there may be additional credentials provided by env var that should be appended to the set of credentials
+	authority, username, password, token :=
+		os.Getenv("SYFT_REGISTRY_AUTH_AUTHORITY"),
+		os.Getenv("SYFT_REGISTRY_AUTH_USERNAME"),
+		os.Getenv("SYFT_REGISTRY_AUTH_PASSWORD"),
+		os.Getenv("SYFT_REGISTRY_AUTH_TOKEN")
+
+	if hasNonEmptyCredentials(username, password, token) {
+		// note: we prepend the credentials such that the environment variables take precedence over on-disk configuration.
+		cfg.Auth = append([]RegistryCredentials{
+			{
+				Authority: authority,
+				Username:  username,
+				Password:  password,
+				Token:     token,
+			},
+		}, cfg.Auth...)
+	}
+	return nil
+}
+
+func hasNonEmptyCredentials(username, password, token string) bool {
+	return password != "" && username != "" || token != ""
+}
+
+func (cfg *registry) ToOptions() *image.RegistryOptions {
+	var auth = make([]image.RegistryCredentials, len(cfg.Auth))
+	for i, a := range cfg.Auth {
+		auth[i] = image.RegistryCredentials{
+			Authority: a.Authority,
+			Username:  a.Username,
+			Password:  a.Password,
+			Token:     a.Token,
+		}
+	}
+	return &image.RegistryOptions{
+		InsecureSkipTLSVerify: cfg.InsecureSkipTLSVerify,
+		InsecureUseHTTP:       cfg.InsecureUseHTTP,
+		Credentials:           auth,
+	}
+}

internal/config/registry_test.go

@@ -0,0 +1,111 @@
+package config
+
+import (
+	"fmt"
+	"github.com/anchore/stereoscope/pkg/image"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestHasNonEmptyCredentials(t *testing.T) {
+	tests := []struct {
+		username, password, token string
+		expected                  bool
+	}{
+
+		{
+			"", "", "",
+			false,
+		},
+		{
+			"user", "", "",
+			false,
+		},
+		{
+			"", "pass", "",
+			false,
+		},
+		{
+			"", "pass", "tok",
+			true,
+		},
+		{
+			"user", "", "tok",
+			true,
+		},
+		{
+			"", "", "tok",
+			true,
+		},
+		{
+			"user", "pass", "tok",
+			true,
+		},
+
+		{
+			"user", "pass", "",
+			true,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(fmt.Sprintf("%+v", test), func(t *testing.T) {
+			assert.Equal(t, test.expected, hasNonEmptyCredentials(test.username, test.password, test.token))
+		})
+	}
+}
+
+func Test_registry_ToOptions(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    registry
+		expected image.RegistryOptions
+	}{
+		{
+			name:  "no registry options",
+			input: registry{},
+			expected: image.RegistryOptions{
+				Credentials: []image.RegistryCredentials{},
+			},
+		},
+		{
+			name: "set InsecureSkipTLSVerify",
+			input: registry{
+				InsecureSkipTLSVerify: true,
+			},
+			expected: image.RegistryOptions{
+				InsecureSkipTLSVerify: true,
+				Credentials:           []image.RegistryCredentials{},
+			},
+		},
+		{
+			name: "set InsecureUseHTTP",
+			input: registry{
+				InsecureUseHTTP: true,
+			},
+			expected: image.RegistryOptions{
+				InsecureUseHTTP: true,
+				Credentials:     []image.RegistryCredentials{},
+			},
+		},
+		{
+			name: "set all bool options",
+			input: registry{
+				InsecureSkipTLSVerify: true,
+				InsecureUseHTTP:       true,
+			},
+			expected: image.RegistryOptions{
+				InsecureSkipTLSVerify: true,
+				InsecureUseHTTP:       true,
+				Credentials:           []image.RegistryCredentials{},
+			},
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, &test.expected, test.input.ToOptions())
+		})
+	}
+}

internal/config/secrets.go

@@ -0,0 +1,28 @@
+package config
+
+import (
+	"github.com/anchore/syft/internal/file"
+	"github.com/anchore/syft/syft/source"
+	"github.com/spf13/viper"
+)
+
+type secrets struct {
+	Cataloger           catalogerOptions  `yaml:"cataloger" json:"cataloger" mapstructure:"cataloger"`
+	AdditionalPatterns  map[string]string `yaml:"additional-patterns" json:"additional-patterns" mapstructure:"additional-patterns"`
+	ExcludePatternNames []string          `yaml:"exclude-pattern-names" json:"exclude-pattern-names" mapstructure:"exclude-pattern-names"`
+	RevealValues        bool              `yaml:"reveal-values" json:"reveal-values" mapstructure:"reveal-values"`
+	SkipFilesAboveSize  int64             `yaml:"skip-files-above-size" json:"skip-files-above-size" mapstructure:"skip-files-above-size"`
+}
+
+func (cfg secrets) loadDefaultValues(v *viper.Viper) {
+	v.SetDefault("secrets.cataloger.enabled", catalogerEnabledDefault)
+	v.SetDefault("secrets.cataloger.scope", source.AllLayersScope)
+	v.SetDefault("secrets.reveal-values", false)
+	v.SetDefault("secrets.skip-files-above-size", 1*file.MB)
+	v.SetDefault("secrets.additional-patterns", map[string]string{})
+	v.SetDefault("secrets.exclude-pattern-names", []string{})
+}
+
+func (cfg *secrets) parseConfigValues() error {
+	return cfg.Cataloger.parseConfigValues()
+}

internal/constants.go

@@ -1,4 +1,10 @@
 package internal
 
-// note: do not change this
-const ApplicationName = "syft"
+const (
+	// ApplicationName is the non-capitalized name of the application (do not change this)
+	ApplicationName = "syft"
+
+	// JSONSchemaVersion is the current schema version output by the JSON presenter
+	// 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 = "2.0.0"
+)

internal/docs.go

@@ -0,0 +1,4 @@
+/*
+Package internal contains miscellaneous functions and objects useful within syft but should not be used externally.
+*/
+package internal

internal/err_helper.go

@@ -0,0 +1,38 @@
+package internal
+
+import (
+	"fmt"
+	"io"
+	"os"
+
+	"github.com/anchore/syft/internal/log"
+)
+
+// CloseAndLogError closes the given io.Closer and reports any errors found as a warning in the log
+func CloseAndLogError(closer io.Closer, location string) {
+	if err := closer.Close(); err != nil {
+		log.Warnf("unable to close file for location=%q: %+v", location, err)
+	}
+}
+
+type ErrPath struct {
+	Path string
+	Err  error
+}
+
+func (e ErrPath) Error() string {
+	return fmt.Sprintf("unable to observe contents of %+v: %v", e.Path, e.Err)
+}
+
+func IsErrPath(err error) bool {
+	_, ok := err.(ErrPath)
+	return ok
+}
+
+func IsErrPathPermission(err error) bool {
+	pathErr, ok := err.(ErrPath)
+	if ok {
+		return os.IsPermission(pathErr.Err)
+	}
+	return ok
+}

internal/file/glob_match.go

@@ -1,5 +1,6 @@
 package file
 
+// GlobMatch evaluates the given glob pattern against the given "name" string, indicating if there is a match or not.
 // Source: https://research.swtch.com/glob.go
 func GlobMatch(pattern, name string) bool {
 	px := 0

internal/file/opener.go

@@ -0,0 +1,16 @@
+package file
+
+import (
+	"io"
+	"os"
+)
+
+// Opener is an object that stores a path to later be opened as a file.
+type Opener struct {
+	path string
+}
+
+// Open the stored path as a io.ReadCloser.
+func (o Opener) Open() (io.ReadCloser, error) {
+	return os.Open(o.path)
+}

internal/file/test-fixtures/generate-zip-fixture-from-source-dir.sh

@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+set -eux
+
+# $1 —— absolute path to destination file, should end with .zip, ideally
+# $2 —— absolute path to directory from which to add entries to the archive
+
+pushd "$2" && find . -print | zip "$1" -@ && popd

internal/file/test-fixtures/generate-zip-fixture.sh

@@ -1,4 +0,0 @@
-#!/usr/bin/env bash
-set -eux
-
-zip -r "$1" zip-source

internal/file/zip_file_helpers_test.go

@@ -0,0 +1,115 @@
+package file
+
+import (
+	"io/ioutil"
+	"os"
+	"os/exec"
+	"path"
+	"path/filepath"
+	"syscall"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+var expectedZipArchiveEntries = []string{
+	"some-dir" + string(os.PathSeparator),
+	filepath.Join("some-dir", "a-file.txt"),
+	"b-file.txt",
+	"nested.zip",
+}
+
+// createZipArchive creates a new ZIP archive file at destinationArchivePath based on the directory found at
+// sourceDirPath.
+func createZipArchive(t testing.TB, sourceDirPath, destinationArchivePath string) {
+	t.Helper()
+
+	cwd, err := os.Getwd()
+	if err != nil {
+		t.Fatalf("unable to get cwd: %+v", err)
+	}
+
+	cmd := exec.Command("./generate-zip-fixture-from-source-dir.sh", destinationArchivePath, path.Base(sourceDirPath))
+	cmd.Dir = filepath.Join(cwd, "test-fixtures")
+
+	if err := cmd.Start(); err != nil {
+		t.Fatalf("unable to start generate zip fixture script: %+v", err)
+	}
+
+	if err := cmd.Wait(); err != nil {
+		if exiterr, ok := err.(*exec.ExitError); ok {
+			// The program has exited with an exit code != 0
+
+			// This works on both Unix and Windows. Although package
+			// syscall is generally platform dependent, WaitStatus is
+			// defined for both Unix and Windows and in both cases has
+			// an ExitStatus() method with the same signature.
+			if status, ok := exiterr.Sys().(syscall.WaitStatus); ok {
+				if status.ExitStatus() != 0 {
+					t.Fatalf("failed to generate fixture: rc=%d", status.ExitStatus())
+				}
+			}
+		} else {
+			t.Fatalf("unable to get generate fixture script result: %+v", err)
+		}
+	}
+
+}
+
+func assertNoError(t testing.TB, fn func() error) func() {
+	return func() {
+		assert.NoError(t, fn())
+	}
+}
+
+// setupZipFileTest encapsulates common test setup work for zip file tests. It returns a cleanup function,
+// which should be called (typically deferred) by the caller, the path of the created zip archive, and an error,
+// which should trigger a fatal test failure in the consuming test. The returned cleanup function will never be nil
+// (even if there's an error), and it should always be called.
+func setupZipFileTest(t testing.TB, sourceDirPath string) string {
+	t.Helper()
+
+	archivePrefix, err := ioutil.TempFile("", "syft-ziputil-archive-TEST-")
+	if err != nil {
+		t.Fatalf("unable to create tempfile: %+v", err)
+	}
+
+	t.Cleanup(
+		assertNoError(t,
+			func() error {
+				return os.Remove(archivePrefix.Name())
+			},
+		),
+	)
+
+	destinationArchiveFilePath := archivePrefix.Name() + ".zip"
+	t.Logf("archive path: %s", destinationArchiveFilePath)
+	createZipArchive(t, sourceDirPath, destinationArchiveFilePath)
+
+	t.Cleanup(
+		assertNoError(t,
+			func() error {
+				return os.Remove(destinationArchiveFilePath)
+			},
+		),
+	)
+
+	cwd, err := os.Getwd()
+	if err != nil {
+		t.Fatalf("unable to get cwd: %+v", err)
+	}
+
+	t.Logf("running from: %s", cwd)
+
+	return destinationArchiveFilePath
+}
+
+// TODO: Consider moving any non-git asset generation to a task (e.g. make) that's run ahead of running go tests.
+func ensureNestedZipExists(t *testing.T, sourceDirPath string) error {
+	t.Helper()
+
+	nestedArchiveFilePath := path.Join(sourceDirPath, "nested.zip")
+	createZipArchive(t, sourceDirPath, nestedArchiveFilePath)
+
+	return nil
+}

internal/file/zip_file_manifest.go

@@ -1,32 +1,54 @@
 package file
 
 import (
-	"archive/zip"
 	"fmt"
 	"os"
 	"sort"
+	"strings"
 
 	"github.com/anchore/syft/internal"
-
 	"github.com/anchore/syft/internal/log"
 )
 
+// ZipFileManifest is a collection of paths and their file metadata.
 type ZipFileManifest map[string]os.FileInfo
 
-func newZipManifest() ZipFileManifest {
-	return make(ZipFileManifest)
+// NewZipFileManifest creates and returns a new ZipFileManifest populated with path and metadata from the given zip archive path.
+func NewZipFileManifest(archivePath string) (ZipFileManifest, error) {
+	zipReader, err := OpenZip(archivePath)
+	manifest := make(ZipFileManifest)
+	if err != nil {
+		return manifest, fmt.Errorf("unable to open zip archive (%s): %w", archivePath, err)
+	}
+	defer func() {
+		err = zipReader.Close()
+		if err != nil {
+			log.Warnf("unable to close zip archive (%s): %+v", archivePath, err)
+		}
+	}()
+
+	for _, file := range zipReader.Reader.File {
+		manifest.Add(file.Name, file.FileInfo())
+	}
+	return manifest, nil
 }
 
+// Add a new path and it's file metadata to the collection.
 func (z ZipFileManifest) Add(entry string, info os.FileInfo) {
 	z[entry] = info
 }
 
+// GlobMatch returns the path keys that match the given value(s).
 func (z ZipFileManifest) GlobMatch(patterns ...string) []string {
 	uniqueMatches := internal.NewStringSet()
 
 	for _, pattern := range patterns {
 		for entry := range z {
-			if GlobMatch(pattern, entry) {
+			// We want to match globs as if entries begin with a leading slash (akin to an absolute path)
+			// so that glob logic is consistent inside and outside of ZIP archives
+			normalizedEntry := normalizeZipEntryName(entry)
+
+			if GlobMatch(pattern, normalizedEntry) {
 				uniqueMatches.Add(entry)
 			}
 		}
@@ -38,21 +60,11 @@ func (z ZipFileManifest) GlobMatch(patterns ...string) []string {
 	return results
 }
 
-func NewZipFileManifest(archivePath string) (ZipFileManifest, error) {
-	zipReader, err := zip.OpenReader(archivePath)
-	manifest := newZipManifest()
-	if err != nil {
-		return manifest, fmt.Errorf("unable to open zip archive (%s): %w", archivePath, err)
+// normalizeZipEntryName takes the given path entry and ensures it is prefixed with "/".
+func normalizeZipEntryName(entry string) string {
+	if !strings.HasPrefix(entry, "/") {
+		return "/" + entry
 	}
-	defer func() {
-		err = zipReader.Close()
-		if err != nil {
-			log.Errorf("unable to close zip archive (%s): %w", archivePath, err)
-		}
-	}()
 
-	for _, file := range zipReader.Reader.File {
-		manifest.Add(file.Name, file.FileInfo())
-	}
-	return manifest, nil
+	return entry
 }

internal/file/zip_file_manifest_test.go

@@ -0,0 +1,100 @@
+package file
+
+import (
+	"encoding/json"
+	"os"
+	"path"
+	"testing"
+)
+
+func TestNewZipFileManifest(t *testing.T) {
+	cwd, err := os.Getwd()
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	sourceDirPath := path.Join(cwd, "test-fixtures", "zip-source")
+	err = ensureNestedZipExists(t, sourceDirPath)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	archiveFilePath := setupZipFileTest(t, sourceDirPath)
+
+	actual, err := NewZipFileManifest(archiveFilePath)
+	if err != nil {
+		t.Fatalf("unable to extract from unzip archive: %+v", err)
+	}
+
+	if len(expectedZipArchiveEntries) != len(actual) {
+		t.Fatalf("mismatched manifest: %d != %d", len(actual), len(expectedZipArchiveEntries))
+	}
+
+	for _, e := range expectedZipArchiveEntries {
+		_, ok := actual[e]
+		if !ok {
+			t.Errorf("missing path: %s", e)
+		}
+	}
+
+	if t.Failed() {
+		b, err := json.MarshalIndent(actual, "", "  ")
+		if err != nil {
+			t.Fatalf("can't show results: %+v", err)
+		}
+
+		t.Errorf("full result: %s", string(b))
+	}
+}
+
+func TestZipFileManifest_GlobMatch(t *testing.T) {
+	cwd, err := os.Getwd()
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	sourceDirPath := path.Join(cwd, "test-fixtures", "zip-source")
+	err = ensureNestedZipExists(t, sourceDirPath)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	archiveFilePath := setupZipFileTest(t, sourceDirPath)
+
+	z, err := NewZipFileManifest(archiveFilePath)
+	if err != nil {
+		t.Fatalf("unable to extract from unzip archive: %+v", err)
+	}
+
+	cases := []struct {
+		glob     string
+		expected string
+	}{
+		{
+			"/b*",
+			"b-file.txt",
+		},
+		{
+			"*/a-file.txt",
+			"some-dir/a-file.txt",
+		},
+		{
+			"**/*.zip",
+			"nested.zip",
+		},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.glob, func(t *testing.T) {
+			glob := tc.glob
+
+			results := z.GlobMatch(glob)
+
+			if len(results) == 1 && results[0] == tc.expected {
+				return
+			}
+
+			t.Errorf("unexpected results for glob '%s': %+v", glob, results)
+		})
+	}
+}

internal/file/zip_file_traversal.go

@@ -15,6 +15,7 @@ import (
 )
 
 const (
+	// represents the order of bytes
 	_  = iota
 	KB = 1 << (10 * iota)
 	MB
@@ -23,6 +24,15 @@ const (
 
 const perFileReadLimit = 2 * GB
 
+type errZipSlipDetected struct {
+	Prefix   string
+	JoinArgs []string
+}
+
+func (e *errZipSlipDetected) Error() string {
+	return fmt.Sprintf("paths are not allowed to resolve outside of the root prefix (%q). Destination: %q", e.Prefix, e.JoinArgs)
+}
+
 type zipTraversalRequest map[string]struct{}
 
 func newZipTraverseRequest(paths ...string) zipTraversalRequest {
@@ -33,17 +43,18 @@ func newZipTraverseRequest(paths ...string) zipTraversalRequest {
 	return results
 }
 
+// TraverseFilesInZip enumerates all paths stored within a zip archive using the visitor pattern.
 func TraverseFilesInZip(archivePath string, visitor func(*zip.File) error, paths ...string) error {
 	request := newZipTraverseRequest(paths...)
 
-	zipReader, err := zip.OpenReader(archivePath)
+	zipReader, err := OpenZip(archivePath)
 	if err != nil {
 		return fmt.Errorf("unable to open zip archive (%s): %w", archivePath, err)
 	}
 	defer func() {
 		err = zipReader.Close()
 		if err != nil {
-			log.Errorf("unable to close zip archive (%s): %w", archivePath, err)
+			log.Errorf("unable to close zip archive (%s): %+v", archivePath, err)
 		}
 	}()
 
@@ -63,8 +74,9 @@ func TraverseFilesInZip(archivePath string, visitor func(*zip.File) error, paths
 	return nil
 }
 
-func ExtractFromZipToUniqueTempFile(archivePath, dir string, paths ...string) (map[string]io.Reader, error) {
-	results := make(map[string]io.Reader)
+// ExtractFromZipToUniqueTempFile extracts select paths for the given archive to a temporary directory, returning file openers for each file extracted.
+func ExtractFromZipToUniqueTempFile(archivePath, dir string, paths ...string) (map[string]Opener, error) {
+	results := make(map[string]Opener)
 
 	// don't allow for full traversal, only select traversal from given paths
 	if len(paths) == 0 {
@@ -78,11 +90,21 @@ func ExtractFromZipToUniqueTempFile(archivePath, dir string, paths ...string) (m
 		if err != nil {
 			return fmt.Errorf("unable to create temp file: %w", err)
 		}
+		// we shouldn't try and keep the tempfile open as the returned result may have several files, which takes up
+		// resources (leading to "too many open files"). Instead we'll return a file opener to the caller which
+		// provides a ReadCloser. It is up to the caller to handle closing the file explicitly.
+		defer tempFile.Close()
 
 		zippedFile, err := file.Open()
 		if err != nil {
 			return fmt.Errorf("unable to read file=%q from zip=%q: %w", file.Name, archivePath, err)
 		}
+		defer func() {
+			err := zippedFile.Close()
+			if err != nil {
+				log.Errorf("unable to close source file=%q from zip=%q: %+v", file.Name, archivePath, err)
+			}
+		}()
 
 		if file.FileInfo().IsDir() {
 			return fmt.Errorf("unable to extract directories, only files: %s", file.Name)
@@ -103,18 +125,15 @@ func ExtractFromZipToUniqueTempFile(archivePath, dir string, paths ...string) (m
 			return fmt.Errorf("unable to reset file pointer (%s): %w", tempFile.Name(), err)
 		}
 
-		results[file.Name] = tempFile
+		results[file.Name] = Opener{path: tempFile.Name()}
 
-		err = zippedFile.Close()
-		if err != nil {
-			return fmt.Errorf("unable to close source file=%q from zip=%q: %w", file.Name, archivePath, err)
-		}
 		return nil
 	}
 
 	return results, TraverseFilesInZip(archivePath, visitor, paths...)
 }
 
+// ContentsFromZip extracts select paths for the given archive and returns a set of string contents for each path.
 func ContentsFromZip(archivePath string, paths ...string) (map[string]string, error) {
 	results := make(map[string]string)
 
@@ -156,19 +175,15 @@ func ContentsFromZip(archivePath string, paths ...string) (map[string]string, er
 	return results, TraverseFilesInZip(archivePath, visitor, paths...)
 }
 
+// UnzipToDir extracts a zip archive to a target directory.
 func UnzipToDir(archivePath, targetDir string) error {
 	visitor := func(file *zip.File) error {
-		// the zip-slip attack protection is still being erroneously detected
-		// nolint:gosec
-		expandedFilePath := filepath.Clean(filepath.Join(targetDir, file.Name))
-
-		// protect against zip slip attacks (traversing unintended parent paths from maliciously crafted relative-path entries)
-		if !strings.HasPrefix(expandedFilePath, filepath.Clean(targetDir)+string(os.PathSeparator)) {
-			return fmt.Errorf("potential zip slip attack: %q", expandedFilePath)
+		joinedPath, err := safeJoin(targetDir, file.Name)
+		if err != nil {
+			return err
 		}
 
-		err := extractSingleFile(file, expandedFilePath, archivePath)
-		if err != nil {
+		if err = extractSingleFile(file, joinedPath, archivePath); err != nil {
 			return err
 		}
 		return nil
@@ -177,6 +192,20 @@ func UnzipToDir(archivePath, targetDir string) error {
 	return TraverseFilesInZip(archivePath, visitor)
 }
 
+// safeJoin ensures that any destinations do not resolve to a path above the prefix path.
+func safeJoin(prefix string, dest ...string) (string, error) {
+	joinResult := filepath.Join(append([]string{prefix}, dest...)...)
+	cleanJoinResult := filepath.Clean(joinResult)
+	if !strings.HasPrefix(cleanJoinResult, filepath.Clean(prefix)) {
+		return "", &errZipSlipDetected{
+			Prefix:   prefix,
+			JoinArgs: dest,
+		}
+	}
+	// why not return the clean path? the called may not be expected it from what should only be a join operation.
+	return joinResult, nil
+}
+
 func extractSingleFile(file *zip.File, expandedFilePath, archivePath string) error {
 	zippedFile, err := file.Open()
 	if err != nil {

internal/file/zip_file_traversal_test.go

@@ -3,50 +3,20 @@ package file
 import (
 	"crypto/sha256"
 	"encoding/json"
+	"errors"
+	"fmt"
 	"io"
 	"io/ioutil"
 	"os"
-	"os/exec"
+	"path"
 	"path/filepath"
 	"strings"
-	"syscall"
 	"testing"
 
 	"github.com/go-test/deep"
+	"github.com/stretchr/testify/assert"
 )
 
-func generateFixture(t *testing.T, archivePath string) {
-	cwd, err := os.Getwd()
-	if err != nil {
-		t.Errorf("unable to get cwd: %+v", err)
-	}
-
-	cmd := exec.Command("./generate-zip-fixture.sh", archivePath)
-	cmd.Dir = filepath.Join(cwd, "test-fixtures")
-
-	if err := cmd.Start(); err != nil {
-		t.Fatalf("unable to start generate zip fixture script: %+v", err)
-	}
-
-	if err := cmd.Wait(); err != nil {
-		if exiterr, ok := err.(*exec.ExitError); ok {
-			// The program has exited with an exit code != 0
-
-			// This works on both Unix and Windows. Although package
-			// syscall is generally platform dependent, WaitStatus is
-			// defined for both Unix and Windows and in both cases has
-			// an ExitStatus() method with the same signature.
-			if status, ok := exiterr.Sys().(syscall.WaitStatus); ok {
-				if status.ExitStatus() != 0 {
-					t.Fatalf("failed to generate fixture: rc=%d", status.ExitStatus())
-				}
-			}
-		} else {
-			t.Fatalf("unable to get generate fixture script result: %+v", err)
-		}
-	}
-}
-
 func equal(r1, r2 io.Reader) (bool, error) {
 	w1 := sha256.New()
 	w2 := sha256.New()
@@ -67,54 +37,48 @@ func equal(r1, r2 io.Reader) (bool, error) {
 }
 
 func TestUnzipToDir(t *testing.T) {
-	archivePrefix, err := ioutil.TempFile("", "syft-ziputil-archive-TEST-")
+	cwd, err := os.Getwd()
 	if err != nil {
-		t.Fatalf("unable to create tempfile: %+v", err)
+		t.Fatal(err)
 	}
-	defer os.Remove(archivePrefix.Name())
-	// the zip utility will add ".zip" to the end of the given name
-	archivePath := archivePrefix.Name() + ".zip"
-	defer os.Remove(archivePath)
-	t.Logf("archive path: %s", archivePath)
 
-	generateFixture(t, archivePrefix.Name())
+	goldenRootDir := filepath.Join(cwd, "test-fixtures")
+	sourceDirPath := path.Join(goldenRootDir, "zip-source")
+	archiveFilePath := setupZipFileTest(t, sourceDirPath)
 
-	contentsDir, err := ioutil.TempDir("", "syft-ziputil-contents-TEST-")
+	unzipDestinationDir, err := ioutil.TempDir("", "syft-ziputil-contents-TEST-")
+	t.Cleanup(assertNoError(t, func() error {
+		return os.RemoveAll(unzipDestinationDir)
+	}))
 	if err != nil {
 		t.Fatalf("unable to create tempdir: %+v", err)
 	}
-	defer os.RemoveAll(contentsDir)
 
-	t.Logf("content path: %s", contentsDir)
+	t.Logf("content path: %s", unzipDestinationDir)
 
-	cwd, err := os.Getwd()
-	if err != nil {
-		t.Errorf("unable to get cwd: %+v", err)
-	}
-
-	t.Logf("running from: %s", cwd)
-
-	// note: zip utility already includes "zip-source" as a parent dir for all contained files
-	goldenRootDir := filepath.Join(cwd, "test-fixtures")
-	expectedPaths := 4
+	expectedPaths := len(expectedZipArchiveEntries)
 	observedPaths := 0
 
-	err = UnzipToDir(archivePath, contentsDir)
+	err = UnzipToDir(archiveFilePath, unzipDestinationDir)
 	if err != nil {
 		t.Fatalf("unable to unzip archive: %+v", err)
 	}
 
 	// compare the source dir tree and the unzipped tree
-	err = filepath.Walk(filepath.Join(contentsDir, "zip-source"),
+	err = filepath.Walk(unzipDestinationDir,
 		func(path string, info os.FileInfo, err error) error {
-			t.Logf("unzipped path: %s", path)
-			observedPaths++
+			// We don't unzip the root archive dir, since there's no archive entry for it
+			if path != unzipDestinationDir {
+				t.Logf("unzipped path: %s", path)
+				observedPaths++
+			}
+
 			if err != nil {
 				t.Fatalf("this should not happen")
 				return err
 			}
 
-			goldenPath := filepath.Join(goldenRootDir, strings.TrimPrefix(path, contentsDir))
+			goldenPath := filepath.Join(sourceDirPath, strings.TrimPrefix(path, unzipDestinationDir))
 
 			if info.IsDir() {
 				i, err := os.Stat(goldenPath)
@@ -156,44 +120,124 @@ func TestUnzipToDir(t *testing.T) {
 	}
 
 	if observedPaths != expectedPaths {
-		t.Errorf("missed test paths: %d!=%d", observedPaths, expectedPaths)
+		t.Errorf("missed test paths: %d != %d", observedPaths, expectedPaths)
 	}
-
 }
 
-func TestExtractFilesFromZipFile(t *testing.T) {
-	archivePrefix, err := ioutil.TempFile("", "syft-ziputil-archive-TEST-")
+func TestContentsFromZip(t *testing.T) {
+	tests := []struct {
+		name        string
+		archivePrep func(tb testing.TB) string
+	}{
+		{
+			name:        "standard, non-nested zip",
+			archivePrep: prepZipSourceFixture,
+		},
+		{
+			name:        "zip with prepended bytes",
+			archivePrep: prependZipSourceFixtureWithString(t, "junk at the beginning of the file..."),
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			archivePath := test.archivePrep(t)
+			expected := zipSourceFixtureExpectedContents()
+
+			var paths []string
+			for p := range expected {
+				paths = append(paths, p)
+			}
+
+			actual, err := ContentsFromZip(archivePath, paths...)
+			if err != nil {
+				t.Fatalf("unable to extract from unzip archive: %+v", err)
+			}
+
+			assertZipSourceFixtureContents(t, actual, expected)
+		})
+	}
+}
+
+func prependZipSourceFixtureWithString(tb testing.TB, value string) func(tb testing.TB) string {
+	if len(value) == 0 {
+		tb.Fatalf("no bytes given to prefix")
+	}
+	return func(t testing.TB) string {
+		archivePath := prepZipSourceFixture(t)
+
+		// create a temp file
+		tmpFile, err := ioutil.TempFile("", "syft-ziputil-prependZipSourceFixtureWithString-")
+		if err != nil {
+			t.Fatalf("unable to create tempfile: %+v", err)
+		}
+		defer tmpFile.Close()
+
+		// write value to the temp file
+		if _, err := tmpFile.WriteString(value); err != nil {
+			t.Fatalf("unable to write to tempfile: %+v", err)
+		}
+
+		// open the original archive
+		sourceFile, err := os.Open(archivePath)
+		if err != nil {
+			t.Fatalf("unable to read source file: %+v", err)
+		}
+
+		// copy all contents from the archive to the temp file
+		if _, err := io.Copy(tmpFile, sourceFile); err != nil {
+			t.Fatalf("unable to copy source to dest: %+v", err)
+		}
+
+		sourceFile.Close()
+
+		// remove the original archive and replace it with the temp file
+		if err := os.Remove(archivePath); err != nil {
+			t.Fatalf("unable to remove original source archive (%q): %+v", archivePath, err)
+		}
+
+		if err := os.Rename(tmpFile.Name(), archivePath); err != nil {
+			t.Fatalf("unable to move new archive to old path (%q): %+v", tmpFile.Name(), err)
+		}
+
+		return archivePath
+	}
+}
+
+func prepZipSourceFixture(t testing.TB) string {
+	t.Helper()
+	archivePrefix, err := ioutil.TempFile("", "syft-ziputil-prepZipSourceFixture-")
 	if err != nil {
 		t.Fatalf("unable to create tempfile: %+v", err)
 	}
-	defer os.Remove(archivePrefix.Name())
+
+	t.Cleanup(func() {
+		assert.NoError(t, os.Remove(archivePrefix.Name()))
+	})
+
 	// the zip utility will add ".zip" to the end of the given name
 	archivePath := archivePrefix.Name() + ".zip"
-	defer os.Remove(archivePath)
+
+	t.Cleanup(func() {
+		assert.NoError(t, os.Remove(archivePath))
+	})
+
 	t.Logf("archive path: %s", archivePath)
 
-	generateFixture(t, archivePrefix.Name())
+	createZipArchive(t, "zip-source", archivePrefix.Name())
 
-	cwd, err := os.Getwd()
-	if err != nil {
-		t.Errorf("unable to get cwd: %+v", err)
-	}
+	return archivePath
+}
 
-	t.Logf("running from: %s", cwd)
-
-	aFilePath := filepath.Join("zip-source", "some-dir", "a-file.txt")
-	bFilePath := filepath.Join("zip-source", "b-file.txt")
-
-	expected := map[string]string{
-		aFilePath: "A file! nice!",
-		bFilePath: "B file...",
-	}
-
-	actual, err := ContentsFromZip(archivePath, aFilePath, bFilePath)
-	if err != nil {
-		t.Fatalf("unable to extract from unzip archive: %+v", err)
+func zipSourceFixtureExpectedContents() map[string]string {
+	return map[string]string{
+		filepath.Join("some-dir", "a-file.txt"): "A file! nice!",
+		filepath.Join("b-file.txt"):             "B file...",
 	}
+}
 
+func assertZipSourceFixtureContents(t testing.TB, actual map[string]string, expected map[string]string) {
+	t.Helper()
 	diffs := deep.Equal(actual, expected)
 	if len(diffs) > 0 {
 		for _, d := range diffs {
@@ -207,60 +251,80 @@ func TestExtractFilesFromZipFile(t *testing.T) {
 
 		t.Errorf("full result: %s", string(b))
 	}
-
 }
 
-func TestZipFileManifest(t *testing.T) {
-	archivePrefix, err := ioutil.TempFile("", "syft-ziputil-archive-TEST-")
-	if err != nil {
-		t.Fatalf("unable to create tempfile: %+v", err)
+// looks like there isn't a helper for this yet? https://github.com/stretchr/testify/issues/497
+func assertErrorAs(expectedErr interface{}) assert.ErrorAssertionFunc {
+	return func(t assert.TestingT, actualErr error, i ...interface{}) bool {
+		return errors.As(actualErr, &expectedErr)
+	}
+}
+
+func TestSafeJoin(t *testing.T) {
+	tests := []struct {
+		prefix       string
+		args         []string
+		expected     string
+		errAssertion assert.ErrorAssertionFunc
+	}{
+		// go cases...
+		{
+			prefix: "/a/place",
+			args: []string{
+				"somewhere/else",
+			},
+			expected:     "/a/place/somewhere/else",
+			errAssertion: assert.NoError,
+		},
+		{
+			prefix: "/a/place",
+			args: []string{
+				"somewhere/../else",
+			},
+			expected:     "/a/place/else",
+			errAssertion: assert.NoError,
+		},
+		{
+			prefix: "/a/../place",
+			args: []string{
+				"somewhere/else",
+			},
+			expected:     "/place/somewhere/else",
+			errAssertion: assert.NoError,
+		},
+		// zip slip examples....
+		{
+			prefix: "/a/place",
+			args: []string{
+				"../../../etc/passwd",
+			},
+			expected:     "",
+			errAssertion: assertErrorAs(&errZipSlipDetected{}),
+		},
+		{
+			prefix: "/a/place",
+			args: []string{
+				"../",
+				"../",
+			},
+			expected:     "",
+			errAssertion: assertErrorAs(&errZipSlipDetected{}),
+		},
+		{
+			prefix: "/a/place",
+			args: []string{
+				"../",
+			},
+			expected:     "",
+			errAssertion: assertErrorAs(&errZipSlipDetected{}),
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(fmt.Sprintf("%+v:%+v", test.prefix, test.args), func(t *testing.T) {
+			actual, err := safeJoin(test.prefix, test.args...)
+			test.errAssertion(t, err)
+			assert.Equal(t, test.expected, actual)
+		})
 	}
-	defer os.Remove(archivePrefix.Name())
-	// the zip utility will add ".zip" to the end of the given name
-	archivePath := archivePrefix.Name() + ".zip"
-	defer os.Remove(archivePath)
-	t.Logf("archive path: %s", archivePath)
-
-	generateFixture(t, archivePrefix.Name())
-
-	cwd, err := os.Getwd()
-	if err != nil {
-		t.Errorf("unable to get cwd: %+v", err)
-	}
-
-	t.Logf("running from: %s", cwd)
-
-	expected := []string{
-		filepath.Join("zip-source") + string(os.PathSeparator),
-		filepath.Join("zip-source", "some-dir") + string(os.PathSeparator),
-		filepath.Join("zip-source", "some-dir", "a-file.txt"),
-		filepath.Join("zip-source", "b-file.txt"),
-	}
-
-	actual, err := NewZipFileManifest(archivePath)
-	if err != nil {
-		t.Fatalf("unable to extract from unzip archive: %+v", err)
-	}
-
-	if len(expected) != len(actual) {
-		t.Fatalf("mismatched manifest: %d != %d", len(actual), len(expected))
-	}
-
-	for _, e := range expected {
-		_, ok := actual[e]
-		if !ok {
-			t.Errorf("missing path: %s", e)
-		}
-	}
-
-	if t.Failed() {
-
-		b, err := json.MarshalIndent(actual, "", "  ")
-		if err != nil {
-			t.Fatalf("can't show results: %+v", err)
-		}
-
-		t.Errorf("full result: %s", string(b))
-	}
-
 }

internal/file/zip_read_closer.go

@@ -0,0 +1,149 @@
+package file
+
+import (
+	"archive/zip"
+	"encoding/binary"
+	"fmt"
+	"io"
+	"os"
+)
+
+// directoryEndLen, readByf, directoryEnd, and findSignatureInBlock were copied from the golang stdlib, specifically:
+// - https://github.com/golang/go/blob/go1.16.4/src/archive/zip/struct.go
+// - https://github.com/golang/go/blob/go1.16.4/src/archive/zip/reader.go
+// findArchiveStartOffset is derived from the same stdlib utils, specifically the readDirectoryEnd function.
+
+const directoryEndLen = 22
+
+// ZipReadCloser is a drop-in replacement for zip.ReadCloser (from zip.OpenReader) that additionally considers zips
+// that have bytes prefixed to the front of the archive (common with self-extracting jars).
+type ZipReadCloser struct {
+	*zip.Reader
+	io.Closer
+}
+
+// OpenZip provides a ZipReadCloser for the given filepath.
+func OpenZip(filepath string) (*ZipReadCloser, error) {
+	f, err := os.Open(filepath)
+	if err != nil {
+		return nil, err
+	}
+	fi, err := f.Stat()
+	if err != nil {
+		f.Close()
+		return nil, err
+	}
+
+	// some archives may have bytes prepended to the front of the archive, such as with self executing JARs. We first
+	// need to find the start of the archive and keep track of this offset.
+	offset, err := findArchiveStartOffset(f, fi.Size())
+	if err != nil {
+		return nil, fmt.Errorf("cannot find beginning of zip archive=%q : %w", filepath, err)
+	}
+
+	if _, err := f.Seek(0, io.SeekStart); err != nil {
+		return nil, fmt.Errorf("unable to seek to beginning of archive: %w", err)
+	}
+
+	size := fi.Size() - int64(offset)
+
+	r, err := zip.NewReader(io.NewSectionReader(f, int64(offset), size), size)
+	if err != nil {
+		return nil, fmt.Errorf("unable to open ZipReadCloser @ %q: %w", filepath, err)
+	}
+
+	return &ZipReadCloser{
+		Reader: r,
+		Closer: f,
+	}, nil
+}
+
+type readBuf []byte
+
+func (b *readBuf) uint16() uint16 {
+	v := binary.LittleEndian.Uint16(*b)
+	*b = (*b)[2:]
+	return v
+}
+
+func (b *readBuf) uint32() uint32 {
+	v := binary.LittleEndian.Uint32(*b)
+	*b = (*b)[4:]
+	return v
+}
+
+type directoryEnd struct {
+	diskNbr            uint32 // unused
+	dirDiskNbr         uint32 // unused
+	dirRecordsThisDisk uint64 // unused
+	directoryRecords   uint64
+	directorySize      uint64
+	directoryOffset    uint64 // relative to file
+}
+
+// note: this is derived from readDirectoryEnd within the archive/zip package
+func findArchiveStartOffset(r io.ReaderAt, size int64) (startOfArchive uint64, err error) {
+	// look for directoryEndSignature in the last 1k, then in the last 65k
+	var buf []byte
+	var directoryEndOffset int64
+	for i, bLen := range []int64{1024, 65 * 1024} {
+		if bLen > size {
+			bLen = size
+		}
+		buf = make([]byte, int(bLen))
+		if _, err := r.ReadAt(buf, size-bLen); err != nil && err != io.EOF {
+			return 0, err
+		}
+		if p := findSignatureInBlock(buf); p >= 0 {
+			buf = buf[p:]
+			directoryEndOffset = size - bLen + int64(p)
+			break
+		}
+		if i == 1 || bLen == size {
+			return 0, zip.ErrFormat
+		}
+	}
+
+	if buf == nil {
+		// we were unable to find the directoryEndSignature block
+		return 0, zip.ErrFormat
+	}
+
+	// read header into struct
+	b := readBuf(buf[4:]) // skip signature
+	d := &directoryEnd{
+		diskNbr:            uint32(b.uint16()),
+		dirDiskNbr:         uint32(b.uint16()),
+		dirRecordsThisDisk: uint64(b.uint16()),
+		directoryRecords:   uint64(b.uint16()),
+		directorySize:      uint64(b.uint32()),
+		directoryOffset:    uint64(b.uint32()),
+	}
+	// Calculate where the zip data actually begins
+	startOfArchive = uint64(directoryEndOffset) - d.directorySize - d.directoryOffset
+
+	// These values mean that the file can be a zip64 file
+	if d.directoryRecords == 0xffff || d.directorySize == 0xffff || d.directoryOffset == 0xffffffff {
+		startOfArchive = 0 // Prefixed data not supported
+	}
+
+	// Make sure directoryOffset points to somewhere in our file.
+	if o := int64(d.directoryOffset); o < 0 || o >= size {
+		return 0, zip.ErrFormat
+	}
+	return startOfArchive, nil
+}
+
+func findSignatureInBlock(b []byte) int {
+	for i := len(b) - directoryEndLen; i >= 0; i-- {
+		// defined from directoryEndSignature
+		if b[i] == 'P' && b[i+1] == 'K' && b[i+2] == 0x05 && b[i+3] == 0x06 {
+			// n is length of comment
+			n := int(b[i+directoryEndLen-2]) | int(b[i+directoryEndLen-1])<<8
+			if n+directoryEndLen+i <= len(b) {
+				return i
+			}
+		}
+	}
+	return -1
+}

internal/file/zip_read_closer_test.go

@@ -0,0 +1,47 @@
+package file
+
+import (
+	"os"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestFindArchiveStartOffset(t *testing.T) {
+	tests := []struct {
+		name        string
+		archivePrep func(tb testing.TB) string
+		expected    uint64
+	}{
+		{
+			name:        "standard, non-nested zip",
+			archivePrep: prepZipSourceFixture,
+			expected:    0,
+		},
+		{
+			name:        "zip with prepended bytes",
+			archivePrep: prependZipSourceFixtureWithString(t, "junk at the beginning of the file..."),
+			expected:    36,
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			archivePath := test.archivePrep(t)
+			f, err := os.Open(archivePath)
+			if err != nil {
+				t.Fatalf("could not open archive %q: %+v", archivePath, err)
+			}
+			fi, err := os.Stat(f.Name())
+			if err != nil {
+				t.Fatalf("unable to stat archive: %+v", err)
+			}
+
+			actual, err := findArchiveStartOffset(f, fi.Size())
+			if err != nil {
+				t.Fatalf("unable to find offset: %+v", err)
+			}
+			assert.Equal(t, test.expected, actual)
+		})
+	}
+}

internal/format/color.go

@@ -1,21 +0,0 @@
-package format
-
-import "fmt"
-
-const (
-	DefaultColor Color = iota + 30
-	Red
-	Green
-	Yellow
-	Blue
-	Magenta
-	Cyan
-	White
-)
-
-type Color uint8
-
-// TODO: not cross platform (windows...)
-func (c Color) Format(s string) string {
-	return fmt.Sprintf("\x1b[%dm%s\x1b[0m", c, s)
-}

internal/formats/common/spdxhelpers/description.go

@@ -0,0 +1,19 @@
+package spdxhelpers
+
+import "github.com/anchore/syft/syft/pkg"
+
+func Description(p pkg.Package) string {
+	if hasMetadata(p) {
+		switch metadata := p.Metadata.(type) {
+		case pkg.ApkMetadata:
+			return metadata.Description
+		case pkg.NpmPackageJSONMetadata:
+			return metadata.Description
+		}
+	}
+	return ""
+}
+
+func hasMetadata(p pkg.Package) bool {
+	return p.Metadata != nil
+}

internal/formats/common/spdxhelpers/description_test.go

@@ -0,0 +1,56 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_Description(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected string
+	}{
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name:     "no metadata",
+			input:    pkg.Package{},
+			expected: "",
+		},
+		{
+			name: "from apk",
+			input: pkg.Package{
+				Metadata: pkg.ApkMetadata{
+					Description: "a description!",
+				},
+			},
+			expected: "a description!",
+		},
+		{
+			name: "from npm",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Description: "a description!",
+				},
+			},
+			expected: "a description!",
+		},
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name: "empty",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Homepage: "",
+				},
+			},
+			expected: "",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, Description(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/download_location.go

@@ -0,0 +1,23 @@
+package spdxhelpers
+
+import "github.com/anchore/syft/syft/pkg"
+
+func DownloadLocation(p pkg.Package) string {
+	// 3.7: Package Download Location
+	// Cardinality: mandatory, one
+	// NONE if there is no download location whatsoever.
+	// NOASSERTION if:
+	//   (i) the SPDX file creator has attempted to but cannot reach a reasonable objective determination;
+	//   (ii) the SPDX file creator has made no attempt to determine this field; or
+	//   (iii) the SPDX file creator has intentionally provided no information (no meaning should be implied by doing so).
+
+	if hasMetadata(p) {
+		switch metadata := p.Metadata.(type) {
+		case pkg.ApkMetadata:
+			return NoneIfEmpty(metadata.URL)
+		case pkg.NpmPackageJSONMetadata:
+			return NoneIfEmpty(metadata.URL)
+		}
+	}
+	return "NOASSERTION"
+}

internal/formats/common/spdxhelpers/download_location_test.go

@@ -0,0 +1,54 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_DownloadLocation(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected string
+	}{
+		{
+			name:     "no metadata",
+			input:    pkg.Package{},
+			expected: "NOASSERTION",
+		},
+		{
+			name: "from apk",
+			input: pkg.Package{
+				Metadata: pkg.ApkMetadata{
+					URL: "http://a-place.gov",
+				},
+			},
+			expected: "http://a-place.gov",
+		},
+		{
+			name: "from npm",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					URL: "http://a-place.gov",
+				},
+			},
+			expected: "http://a-place.gov",
+		},
+		{
+			name: "empty",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					URL: "",
+				},
+			},
+			expected: "NONE",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, DownloadLocation(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/external_refs.go

@@ -0,0 +1,51 @@
+package spdxhelpers
+
+import (
+	"github.com/anchore/syft/internal/formats/spdx22json/model"
+	"github.com/anchore/syft/internal/log"
+	"github.com/anchore/syft/syft/pkg"
+)
+
+func ExternalRefs(p pkg.Package) (externalRefs []model.ExternalRef) {
+	externalRefs = make([]model.ExternalRef, 0)
+
+	for _, c := range p.CPEs {
+		externalRefs = append(externalRefs, model.ExternalRef{
+			ReferenceCategory: model.SecurityReferenceCategory,
+			ReferenceLocator:  c.BindToFmtString(),
+			ReferenceType:     model.Cpe23ExternalRefType,
+		})
+	}
+
+	if p.PURL != "" {
+		externalRefs = append(externalRefs, model.ExternalRef{
+			ReferenceCategory: model.PackageManagerReferenceCategory,
+			ReferenceLocator:  p.PURL,
+			ReferenceType:     model.PurlExternalRefType,
+		})
+	}
+	return externalRefs
+}
+
+func ExtractPURL(refs []model.ExternalRef) string {
+	for _, r := range refs {
+		if r.ReferenceType == model.PurlExternalRefType {
+			return r.ReferenceLocator
+		}
+	}
+	return ""
+}
+
+func ExtractCPEs(refs []model.ExternalRef) (cpes []pkg.CPE) {
+	for _, r := range refs {
+		if r.ReferenceType == model.Cpe23ExternalRefType {
+			cpe, err := pkg.NewCPE(r.ReferenceLocator)
+			if err != nil {
+				log.Warnf("unable to extract SPDX CPE=%q: %+v", r.ReferenceLocator, err)
+				continue
+			}
+			cpes = append(cpes, cpe)
+		}
+	}
+	return cpes
+}

internal/formats/common/spdxhelpers/external_refs_test.go

@@ -0,0 +1,45 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/internal/formats/spdx22json/model"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_ExternalRefs(t *testing.T) {
+	testCPE := pkg.MustCPE("cpe:2.3:a:name:name:3.2:*:*:*:*:*:*:*")
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected []model.ExternalRef
+	}{
+		{
+			name: "cpe + purl",
+			input: pkg.Package{
+				CPEs: []pkg.CPE{
+					testCPE,
+				},
+				PURL: "a-purl",
+			},
+			expected: []model.ExternalRef{
+				{
+					ReferenceCategory: model.SecurityReferenceCategory,
+					ReferenceLocator:  testCPE.BindToFmtString(),
+					ReferenceType:     model.Cpe23ExternalRefType,
+				},
+				{
+					ReferenceCategory: model.PackageManagerReferenceCategory,
+					ReferenceLocator:  "a-purl",
+					ReferenceType:     model.PurlExternalRefType,
+				},
+			},
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.ElementsMatch(t, test.expected, ExternalRefs(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/homepage.go

@@ -0,0 +1,15 @@
+package spdxhelpers
+
+import "github.com/anchore/syft/syft/pkg"
+
+func Homepage(p pkg.Package) string {
+	if hasMetadata(p) {
+		switch metadata := p.Metadata.(type) {
+		case pkg.GemMetadata:
+			return metadata.Homepage
+		case pkg.NpmPackageJSONMetadata:
+			return metadata.Homepage
+		}
+	}
+	return ""
+}

internal/formats/common/spdxhelpers/homepage_test.go

@@ -0,0 +1,56 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_Homepage(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected string
+	}{
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name:     "no metadata",
+			input:    pkg.Package{},
+			expected: "",
+		},
+		{
+			name: "from gem",
+			input: pkg.Package{
+				Metadata: pkg.GemMetadata{
+					Homepage: "http://a-place.gov",
+				},
+			},
+			expected: "http://a-place.gov",
+		},
+		{
+			name: "from npm",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Homepage: "http://a-place.gov",
+				},
+			},
+			expected: "http://a-place.gov",
+		},
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name: "empty",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Homepage: "",
+				},
+			},
+			expected: "",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, Homepage(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/license.go

@@ -0,0 +1,37 @@
+package spdxhelpers
+
+import (
+	"strings"
+
+	"github.com/anchore/syft/internal/spdxlicense"
+	"github.com/anchore/syft/syft/pkg"
+)
+
+func License(p pkg.Package) string {
+	// source: https://spdx.github.io/spdx-spec/3-package-information/#313-concluded-license
+	// The options to populate this field are limited to:
+	// A valid SPDX License Expression as defined in Appendix IV;
+	// NONE, if the SPDX file creator concludes there is no license available for this package; or
+	// NOASSERTION if:
+	//   (i) the SPDX file creator has attempted to but cannot reach a reasonable objective determination;
+	//   (ii) the SPDX file creator has made no attempt to determine this field; or
+	//   (iii) the SPDX file creator has intentionally provided no information (no meaning should be implied by doing so).
+
+	if len(p.Licenses) == 0 {
+		return "NONE"
+	}
+
+	// take all licenses and assume an AND expression; for information about license expressions see https://spdx.github.io/spdx-spec/appendix-IV-SPDX-license-expressions/
+	var parsedLicenses []string
+	for _, l := range p.Licenses {
+		if value, exists := spdxlicense.ID(l); exists {
+			parsedLicenses = append(parsedLicenses, value)
+		}
+	}
+
+	if len(parsedLicenses) == 0 {
+		return "NOASSERTION"
+	}
+
+	return strings.Join(parsedLicenses, " AND ")
+}

internal/formats/common/spdxhelpers/license_test.go

@@ -0,0 +1,73 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_License(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected string
+	}{
+		{
+			name:     "no licenses",
+			input:    pkg.Package{},
+			expected: "NONE",
+		},
+		{
+			name: "no SPDX licenses",
+			input: pkg.Package{
+				Licenses: []string{
+					"made-up",
+				},
+			},
+			expected: "NOASSERTION",
+		},
+		{
+			name: "with SPDX license",
+			input: pkg.Package{
+				Licenses: []string{
+					"MIT",
+				},
+			},
+			expected: "MIT",
+		},
+		{
+			name: "with SPDX license expression",
+			input: pkg.Package{
+				Licenses: []string{
+					"MIT",
+					"GPL-3.0",
+				},
+			},
+			expected: "MIT AND GPL-3.0",
+		},
+		{
+			name: "cap insensitive",
+			input: pkg.Package{
+				Licenses: []string{
+					"gpl-3.0",
+				},
+			},
+			expected: "GPL-3.0",
+		},
+		{
+			name: "debian to spdx conversion",
+			input: pkg.Package{
+				Licenses: []string{
+					"GPL-2",
+				},
+			},
+			expected: "GPL-2.0",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, License(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/none_if_empty.go

@@ -0,0 +1,12 @@
+package spdxhelpers
+
+import (
+	"strings"
+)
+
+func NoneIfEmpty(value string) string {
+	if strings.TrimSpace(value) == "" {
+		return "NONE"
+	}
+	return value
+}

internal/formats/common/spdxhelpers/none_if_empty_test.go

@@ -0,0 +1,41 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_noneIfEmpty(t *testing.T) {
+	tests := []struct {
+		name     string
+		value    string
+		expected string
+	}{
+		{
+			name:     "non-zero value",
+			value:    "something",
+			expected: "something",
+		},
+		{
+			name:     "empty",
+			value:    "",
+			expected: "NONE",
+		},
+		{
+			name:     "space",
+			value:    " ",
+			expected: "NONE",
+		},
+		{
+			name:     "tab",
+			value:    "\t",
+			expected: "NONE",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, NoneIfEmpty(test.value))
+		})
+	}
+}

internal/formats/common/spdxhelpers/originator_test.go

@@ -0,0 +1,114 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_Originator(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected string
+	}{
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name:     "no metadata",
+			input:    pkg.Package{},
+			expected: "",
+		},
+		{
+			name: "from gem",
+			input: pkg.Package{
+				Metadata: pkg.GemMetadata{
+					Authors: []string{
+						"auth1",
+						"auth2",
+					},
+				},
+			},
+			expected: "auth1",
+		},
+		{
+			name: "from npm",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Author: "auth",
+				},
+			},
+			expected: "auth",
+		},
+		{
+			name: "from apk",
+			input: pkg.Package{
+				Metadata: pkg.ApkMetadata{
+					Maintainer: "auth",
+				},
+			},
+			expected: "auth",
+		},
+		{
+			name: "from python - just name",
+			input: pkg.Package{
+				Metadata: pkg.PythonPackageMetadata{
+					Author: "auth",
+				},
+			},
+			expected: "auth",
+		},
+		{
+			name: "from python - just email",
+			input: pkg.Package{
+				Metadata: pkg.PythonPackageMetadata{
+					AuthorEmail: "auth@auth.gov",
+				},
+			},
+			expected: "auth@auth.gov",
+		},
+		{
+			name: "from python - both name and email",
+			input: pkg.Package{
+				Metadata: pkg.PythonPackageMetadata{
+					Author:      "auth",
+					AuthorEmail: "auth@auth.gov",
+				},
+			},
+			expected: "auth <auth@auth.gov>",
+		},
+		{
+			name: "from rpm",
+			input: pkg.Package{
+				Metadata: pkg.RpmdbMetadata{
+					Vendor: "auth",
+				},
+			},
+			expected: "auth",
+		},
+		{
+			name: "from dpkg",
+			input: pkg.Package{
+				Metadata: pkg.DpkgMetadata{
+					Maintainer: "auth",
+				},
+			},
+			expected: "auth",
+		},
+		{
+			// note: since this is an optional field, no value is preferred over NONE or NOASSERTION
+			name: "empty",
+			input: pkg.Package{
+				Metadata: pkg.NpmPackageJSONMetadata{
+					Author: "",
+				},
+			},
+			expected: "",
+		},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			assert.Equal(t, test.expected, Originator(test.input))
+		})
+	}
+}

internal/formats/common/spdxhelpers/origintor.go

@@ -0,0 +1,37 @@
+package spdxhelpers
+
+import (
+	"fmt"
+
+	"github.com/anchore/syft/syft/pkg"
+)
+
+func Originator(p pkg.Package) string {
+	if hasMetadata(p) {
+		switch metadata := p.Metadata.(type) {
+		case pkg.ApkMetadata:
+			return metadata.Maintainer
+		case pkg.NpmPackageJSONMetadata:
+			return metadata.Author
+		case pkg.PythonPackageMetadata:
+			author := metadata.Author
+			if author == "" {
+				return metadata.AuthorEmail
+			}
+			if metadata.AuthorEmail != "" {
+				author += fmt.Sprintf(" <%s>", metadata.AuthorEmail)
+			}
+			return author
+		case pkg.GemMetadata:
+			if len(metadata.Authors) > 0 {
+				return metadata.Authors[0]
+			}
+			return ""
+		case pkg.RpmdbMetadata:
+			return metadata.Vendor
+		case pkg.DpkgMetadata:
+			return metadata.Maintainer
+		}
+	}
+	return ""
+}

internal/formats/common/spdxhelpers/source_info.go

@@ -0,0 +1,41 @@
+package spdxhelpers
+
+import (
+	"strings"
+
+	"github.com/anchore/syft/syft/pkg"
+)
+
+func SourceInfo(p pkg.Package) string {
+	answer := ""
+	switch p.Type {
+	case pkg.RpmPkg:
+		answer = "acquired package info from RPM DB"
+	case pkg.ApkPkg:
+		answer = "acquired package info from APK DB"
+	case pkg.DebPkg:
+		answer = "acquired package info from DPKG DB"
+	case pkg.NpmPkg:
+		answer = "acquired package info from installed node module manifest file"
+	case pkg.PythonPkg:
+		answer = "acquired package info from installed python package manifest file"
+	case pkg.JavaPkg, pkg.JenkinsPluginPkg:
+		answer = "acquired package info from installed java archive"
+	case pkg.GemPkg:
+		answer = "acquired package info from installed gem metadata file"
+	case pkg.GoModulePkg:
+		answer = "acquired package info from go module information"
+	case pkg.RustPkg:
+		answer = "acquired package info from rust cargo manifest"
+	case pkg.PhpComposerPkg:
+		answer = "acquired package info from PHP composer manifest"
+	default:
+		answer = "acquired package info from the following paths"
+	}
+	var paths []string
+	for _, l := range p.Locations {
+		paths = append(paths, l.RealPath)
+	}
+
+	return answer + ": " + strings.Join(paths, ", ")
+}

internal/formats/common/spdxhelpers/source_info_test.go

@@ -0,0 +1,143 @@
+package spdxhelpers
+
+import (
+	"testing"
+
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/source"
+	"github.com/stretchr/testify/assert"
+)
+
+func Test_SourceInfo(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    pkg.Package
+		expected []string
+	}{
+		{
+			name: "locations are captured",
+			input: pkg.Package{
+				// note: no type given
+				Locations: []source.Location{
+					source.NewVirtualLocation("/a-place", "/b-place"),
+					source.NewVirtualLocation("/c-place", "/d-place"),
+				},
+			},
+			expected: []string{
+				"from the following paths",
+				"/a-place",
+				"/c-place",
+			},
+		},
+		{
+			// note: no specific support for this
+			input: pkg.Package{
+				Type: pkg.KbPkg,
+			},
+			expected: []string{
+				"from the following paths",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.RpmPkg,
+			},
+			expected: []string{
+				"from RPM DB",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.ApkPkg,
+			},
+			expected: []string{
+				"from APK DB",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.DebPkg,
+			},
+			expected: []string{
+				"from DPKG DB",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.NpmPkg,
+			},
+			expected: []string{
+				"from installed node module manifest file",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.PythonPkg,
+			},
+			expected: []string{
+				"from installed python package manifest file",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.JavaPkg,
+			},
+			expected: []string{
+				"from installed java archive",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.JenkinsPluginPkg,
+			},
+			expected: []string{
+				"from installed java archive",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.GemPkg,
+			},
+			expected: []string{
+				"from installed gem metadata file",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.GoModulePkg,
+			},
+			expected: []string{
+				"from go module information",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.RustPkg,
+			},
+			expected: []string{
+				"from rust cargo manifest",
+			},
+		},
+		{
+			input: pkg.Package{
+				Type: pkg.PhpComposerPkg,
+			},
+			expected: []string{
+				"from PHP composer manifest",
+			},
+		},
+	}
+	var pkgTypes []pkg.Type
+	for _, test := range tests {
+		t.Run(test.name+" "+string(test.input.Type), func(t *testing.T) {
+			if test.input.Type != "" {
+				pkgTypes = append(pkgTypes, test.input.Type)
+			}
+			actual := SourceInfo(test.input)
+			for _, expected := range test.expected {
+				assert.Contains(t, actual, expected)
+			}
+		})
+	}
+	assert.ElementsMatch(t, pkg.AllPkgs, pkgTypes, "missing one or more package types to test against (maybe a package type was added?)")
+}

internal/formats/common/testutils/utils.go

@@ -0,0 +1,260 @@
+package testutils
+
+import (
+	"bytes"
+	"testing"
+
+	"github.com/anchore/go-presenter"
+	"github.com/anchore/go-testutils"
+	"github.com/anchore/stereoscope/pkg/filetree"
+	"github.com/anchore/stereoscope/pkg/image"
+	"github.com/anchore/stereoscope/pkg/imagetest"
+	"github.com/anchore/syft/syft/distro"
+	"github.com/anchore/syft/syft/pkg"
+	"github.com/anchore/syft/syft/sbom"
+	"github.com/anchore/syft/syft/source"
+	"github.com/sergi/go-diff/diffmatchpatch"
+	"github.com/stretchr/testify/assert"
+)
+
+type redactor func(s []byte) []byte
+
+type imageCfg struct {
+	fromSnapshot bool
+}
+
+type ImageOption func(cfg *imageCfg)
+
+func FromSnapshot() ImageOption {
+	return func(cfg *imageCfg) {
+		cfg.fromSnapshot = true
+	}
+}
+
+func AssertPresenterAgainstGoldenImageSnapshot(t *testing.T, pres presenter.Presenter, testImage string, updateSnapshot bool, redactors ...redactor) {
+	var buffer bytes.Buffer
+
+	// grab the latest image contents and persist
+	if updateSnapshot {
+		imagetest.UpdateGoldenFixtureImage(t, testImage)
+	}
+
+	err := pres.Present(&buffer)
+	assert.NoError(t, err)
+	actual := buffer.Bytes()
+
+	// replace the expected snapshot contents with the current presenter contents
+	if updateSnapshot {
+		testutils.UpdateGoldenFileContents(t, actual)
+	}
+
+	var expected = testutils.GetGoldenFileContents(t)
+
+	// remove dynamic values, which should be tested independently
+	for _, r := range redactors {
+		actual = r(actual)
+		expected = r(expected)
+	}
+
+	// assert that the golden file snapshot matches the actual contents
+	if !bytes.Equal(expected, actual) {
+		dmp := diffmatchpatch.New()
+		diffs := dmp.DiffMain(string(expected), string(actual), true)
+		t.Errorf("mismatched output:\n%s", dmp.DiffPrettyText(diffs))
+	}
+}
+
+func AssertPresenterAgainstGoldenSnapshot(t *testing.T, pres presenter.Presenter, updateSnapshot bool, redactors ...redactor) {
+	var buffer bytes.Buffer
+
+	err := pres.Present(&buffer)
+	assert.NoError(t, err)
+	actual := buffer.Bytes()
+
+	// replace the expected snapshot contents with the current presenter contents
+	if updateSnapshot {
+		testutils.UpdateGoldenFileContents(t, actual)
+	}
+
+	var expected = testutils.GetGoldenFileContents(t)
+
+	// remove dynamic values, which should be tested independently
+	for _, r := range redactors {
+		actual = r(actual)
+		expected = r(expected)
+	}
+
+	if !bytes.Equal(expected, actual) {
+		dmp := diffmatchpatch.New()
+		diffs := dmp.DiffMain(string(expected), string(actual), true)
+		t.Errorf("mismatched output:\n%s", dmp.DiffPrettyText(diffs))
+	}
+}
+
+func ImageInput(t testing.TB, testImage string, options ...ImageOption) sbom.SBOM {
+	t.Helper()
+	catalog := pkg.NewCatalog()
+	var cfg imageCfg
+	var img *image.Image
+	for _, opt := range options {
+		opt(&cfg)
+	}
+
+	switch cfg.fromSnapshot {
+	case true:
+		img = imagetest.GetGoldenFixtureImage(t, testImage)
+	default:
+		img = imagetest.GetFixtureImage(t, "docker-archive", testImage)
+	}
+
+	populateImageCatalog(catalog, img)
+
+	// this is a hard coded value that is not given by the fixture helper and must be provided manually
+	img.Metadata.ManifestDigest = "sha256:2731251dc34951c0e50fcc643b4c5f74922dad1a5d98f302b504cf46cd5d9368"
+
+	src, err := source.NewFromImage(img, "user-image-input")
+	assert.NoError(t, err)
+
+	dist, err := distro.NewDistro(distro.Debian, "1.2.3", "like!")
+	assert.NoError(t, err)
+
+	return sbom.SBOM{
+		Artifacts: sbom.Artifacts{
+			PackageCatalog: catalog,
+			Distro:         &dist,
+		},
+		Source: src.Metadata,
+		Descriptor: sbom.Descriptor{
+			Name:    "syft",
+			Version: "v0.42.0-bogus",
+			// the application configuration should be persisted here, however, we do not want to import
+			// the application configuration in this package (it's reserved only for ingestion by the cmd package)
+			Configuration: map[string]string{
+				"config-key": "config-value",
+			},
+		},
+	}
+}
+
+func populateImageCatalog(catalog *pkg.Catalog, img *image.Image) {
+	_, ref1, _ := img.SquashedTree().File("/somefile-1.txt", filetree.FollowBasenameLinks)
+	_, ref2, _ := img.SquashedTree().File("/somefile-2.txt", filetree.FollowBasenameLinks)
+
+	// populate catalog with test data
+	catalog.Add(pkg.Package{
+		Name:    "package-1",
+		Version: "1.0.1",
+		Locations: []source.Location{
+			source.NewLocationFromImage(string(ref1.RealPath), *ref1, img),
+		},
+		Type:         pkg.PythonPkg,
+		FoundBy:      "the-cataloger-1",
+		Language:     pkg.Python,
+		MetadataType: pkg.PythonPackageMetadataType,
+		Licenses:     []string{"MIT"},
+		Metadata: pkg.PythonPackageMetadata{
+			Name:    "package-1",
+			Version: "1.0.1",
+		},
+		PURL: "a-purl-1",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:1:*:*:*:*:*:*:*"),
+		},
+	})
+	catalog.Add(pkg.Package{
+		Name:    "package-2",
+		Version: "2.0.1",
+		Locations: []source.Location{
+			source.NewLocationFromImage(string(ref2.RealPath), *ref2, img),
+		},
+		Type:         pkg.DebPkg,
+		FoundBy:      "the-cataloger-2",
+		MetadataType: pkg.DpkgMetadataType,
+		Metadata: pkg.DpkgMetadata{
+			Package: "package-2",
+			Version: "2.0.1",
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+}
+
+func DirectoryInput(t testing.TB) sbom.SBOM {
+	catalog := newDirectoryCatalog()
+
+	dist, err := distro.NewDistro(distro.Debian, "1.2.3", "like!")
+	assert.NoError(t, err)
+
+	src, err := source.NewFromDirectory("/some/path")
+	assert.NoError(t, err)
+
+	return sbom.SBOM{
+		Artifacts: sbom.Artifacts{
+			PackageCatalog: catalog,
+			Distro:         &dist,
+		},
+		Source: src.Metadata,
+		Descriptor: sbom.Descriptor{
+			Name:    "syft",
+			Version: "v0.42.0-bogus",
+			// the application configuration should be persisted here, however, we do not want to import
+			// the application configuration in this package (it's reserved only for ingestion by the cmd package)
+			Configuration: map[string]string{
+				"config-key": "config-value",
+			},
+		},
+	}
+}
+
+func newDirectoryCatalog() *pkg.Catalog {
+	catalog := pkg.NewCatalog()
+
+	// populate catalog with test data
+	catalog.Add(pkg.Package{
+		Name:    "package-1",
+		Version: "1.0.1",
+		Type:    pkg.PythonPkg,
+		FoundBy: "the-cataloger-1",
+		Locations: []source.Location{
+			source.NewLocation("/some/path/pkg1"),
+		},
+		Language:     pkg.Python,
+		MetadataType: pkg.PythonPackageMetadataType,
+		Licenses:     []string{"MIT"},
+		Metadata: pkg.PythonPackageMetadata{
+			Name:    "package-1",
+			Version: "1.0.1",
+			Files: []pkg.PythonFileRecord{
+				{
+					Path: "/some/path/pkg1/dependencies/foo",
+				},
+			},
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+	catalog.Add(pkg.Package{
+		Name:    "package-2",
+		Version: "2.0.1",
+		Type:    pkg.DebPkg,
+		FoundBy: "the-cataloger-2",
+		Locations: []source.Location{
+			source.NewLocation("/some/path/pkg1"),
+		},
+		MetadataType: pkg.DpkgMetadataType,
+		Metadata: pkg.DpkgMetadata{
+			Package: "package-2",
+			Version: "2.0.1",
+		},
+		PURL: "a-purl-2",
+		CPEs: []pkg.CPE{
+			pkg.MustCPE("cpe:2.3:*:some:package:2:*:*:*:*:*:*:*"),
+		},
+	})
+
+	return catalog
+}

internal/formats/cyclonedx12xml/encoder.go

@@ -0,0 +1,26 @@
+package cyclonedx12xml
+
+import (
+	"encoding/xml"
+	"io"
+
+	"github.com/anchore/syft/syft/sbom"
+)
+
+func encoder(output io.Writer, s sbom.SBOM) error {
+	enc := xml.NewEncoder(output)
+	enc.Indent("", "  ")
+
+	_, err := output.Write([]byte(xml.Header))
+	if err != nil {
+		return err
+	}
+
+	err = enc.Encode(toFormatModel(s))
+	if err != nil {
+		return err
+	}
+
+	_, err = output.Write([]byte("\n"))
+	return err
+}

internal/formats/cyclonedx12xml/encoder_test.go

@@ -0,0 +1,39 @@
+package cyclonedx12xml
+
+import (
+	"flag"
+	"regexp"
+	"testing"
+
+	"github.com/anchore/syft/internal/formats/common/testutils"
+)
+
+var updateCycloneDx = flag.Bool("update-cyclonedx", false, "update the *.golden files for cyclone-dx presenters")
+
+func TestCycloneDxDirectoryPresenter(t *testing.T) {
+	testutils.AssertPresenterAgainstGoldenSnapshot(t,
+		Format().Presenter(testutils.DirectoryInput(t)),
+		*updateCycloneDx,
+		cycloneDxRedactor,
+	)
+}
+
+func TestCycloneDxImagePresenter(t *testing.T) {
+	testImage := "image-simple"
+	testutils.AssertPresenterAgainstGoldenImageSnapshot(t,
+		Format().Presenter(testutils.ImageInput(t, testImage)),
+		testImage,
+		*updateCycloneDx,
+		cycloneDxRedactor,
+	)
+}
+
+func cycloneDxRedactor(s []byte) []byte {
+	serialPattern := regexp.MustCompile(`serialNumber="[a-zA-Z0-9\-:]+"`)
+	rfc3339Pattern := regexp.MustCompile(`([0-9]+)-(0[1-9]|1[012])-(0[1-9]|[12][0-9]|3[01])[Tt]([01][0-9]|2[0-3]):([0-5][0-9]):([0-5][0-9]|60)(\.[0-9]+)?(([Zz])|([\+|\-]([01][0-9]|2[0-3]):[0-5][0-9]))`)
+
+	for _, pattern := range []*regexp.Regexp{serialPattern, rfc3339Pattern} {
+		s = pattern.ReplaceAll(s, []byte("redacted"))
+	}
+	return s
+}

internal/formats/cyclonedx12xml/format.go

@@ -0,0 +1,12 @@
+package cyclonedx12xml
+
+import "github.com/anchore/syft/syft/format"
+
+func Format() format.Format {
+	return format.NewFormat(
+		format.CycloneDxOption,
+		encoder,
+		nil,
+		nil,
+	)
+}

internal/formats/cyclonedx12xml/model/bom_descriptor.go

@@ -0,0 +1,31 @@
+package model
+
+import (
+	"encoding/xml"
+)
+
+// Source: https://cyclonedx.org/ext/bom-descriptor/
+
+// BomDescriptor represents all metadata surrounding the BOM report (such as when the BOM was made, with which tool, and the item being cataloged).
+type BomDescriptor struct {
+	XMLName   xml.Name                `xml:"metadata"`
+	Timestamp string                  `xml:"timestamp,omitempty"` // The date and time (timestamp) when the document was created
+	Tools     []BomDescriptorTool     `xml:"tools>tool"`          // The tool used to create the BOM.
+	Component *BomDescriptorComponent `xml:"component"`           // The component that the BOM describes.
+}
+
+// BomDescriptorTool represents the tool that created the BOM report.
+type BomDescriptorTool struct {
+	XMLName xml.Name `xml:"tool"`
+	Vendor  string   `xml:"vendor,omitempty"`  // The vendor of the tool used to create the BOM.
+	Name    string   `xml:"name,omitempty"`    // The name of the tool used to create the BOM.
+	Version string   `xml:"version,omitempty"` // The version of the tool used to create the BOM.
+	// TODO: hashes, author, manufacture, supplier
+	// TODO: add user-defined fields for the remaining build/version parameters
+}
+
+// BomDescriptorComponent represents the software/package being cataloged.
+type BomDescriptorComponent struct {
+	XMLName xml.Name `xml:"component"`
+	Component
+}

internal/formats/cyclonedx12xml/model/component.go

@@ -0,0 +1,20 @@
+package model
+
+import "encoding/xml"
+
+// Component represents a single element in the CycloneDX BOM
+type Component struct {
+	XMLName     xml.Name   `xml:"component"`
+	Type        string     `xml:"type,attr"`             // Required; Describes if the component is a library, framework, application, container, operating system, firmware, hardware device, or file
+	Supplier    string     `xml:"supplier,omitempty"`    // The organization that supplied the component. The supplier may often be the manufacture, but may also be a distributor or repackager.
+	Author      string     `xml:"author,omitempty"`      // The person(s) or organization(s) that authored the component
+	Publisher   string     `xml:"publisher,omitempty"`   // The person(s) or organization(s) that published the component
+	Group       string     `xml:"group,omitempty"`       // The high-level classification that a project self-describes as. This will often be a shortened, single name of the company or project that produced the component, or the source package or domain name.
+	Name        string     `xml:"name"`                  // Required; The name of the component as defined by the project
+	Version     string     `xml:"version"`               // Required; The version of the component as defined by the project
+	Description string     `xml:"description,omitempty"` // A description of the component
+	Licenses    *[]License `xml:"licenses>license"`      // A node describing zero or more license names, SPDX license IDs or expressions
+	PackageURL  string     `xml:"purl,omitempty"`        // Specifies the package-url (PackageURL). The purl, if specified, must be valid and conform to the specification defined at: https://github.com/package-url/purl-spec
+	// TODO: source, hashes, copyright, cpe, purl, swid, modified, pedigree, externalReferences
+	// TODO: add user-defined parameters for syft-specific values (image layer index, cataloger, location path, etc.)
+}