1
0
mirror of https://git.yoctoproject.org/poky synced 2026-05-08 05:09:24 +00:00

classes/spdx-common: Move common SPDX to new class

Moves SPDX code that can be shared between different SPDX versions into
a common class

(From OE-Core rev: 769a390adc9fc0b52978abe0f19f885967af0117)

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This commit is contained in:
Joshua Watt
2024-06-10 15:41:48 -06:00
committed by Richard Purdie
parent a43f15565b
commit 92b202c246
2 changed files with 265 additions and 248 deletions
+9 -248
View File
@@ -4,65 +4,13 @@
# SPDX-License-Identifier: GPL-2.0-only
#
DEPLOY_DIR_SPDX ??= "${DEPLOY_DIR}/spdx"
inherit spdx-common
# The product name that the CVE database uses. Defaults to BPN, but may need to
# be overriden per recipe (for example tiff.bb sets CVE_PRODUCT=libtiff).
CVE_PRODUCT ??= "${BPN}"
CVE_VERSION ??= "${PV}"
SPDXDIR ??= "${WORKDIR}/spdx"
SPDXDEPLOY = "${SPDXDIR}/deploy"
SPDXWORK = "${SPDXDIR}/work"
SPDXIMAGEWORK = "${SPDXDIR}/image-work"
SPDXSDKWORK = "${SPDXDIR}/sdk-work"
SPDXDEPS = "${SPDXDIR}/deps.json"
SPDX_TOOL_NAME ??= "oe-spdx-creator"
SPDX_TOOL_VERSION ??= "1.0"
SPDXRUNTIMEDEPLOY = "${SPDXDIR}/runtime-deploy"
SPDX_INCLUDE_SOURCES ??= "0"
SPDX_ARCHIVE_SOURCES ??= "0"
SPDX_ARCHIVE_PACKAGED ??= "0"
SPDX_UUID_NAMESPACE ??= "sbom.openembedded.org"
SPDX_NAMESPACE_PREFIX ??= "http://spdx.org/spdxdocs"
SPDX_PRETTY ??= "0"
SPDX_LICENSES ??= "${COREBASE}/meta/files/spdx-licenses.json"
SPDX_CUSTOM_ANNOTATION_VARS ??= ""
SPDX_ORG ??= "OpenEmbedded ()"
SPDX_SUPPLIER ??= "Organization: ${SPDX_ORG}"
SPDX_SUPPLIER[doc] = "The SPDX PackageSupplier field for SPDX packages created from \
this recipe. For SPDX documents create using this class during the build, this \
is the contact information for the person or organization who is doing the \
build."
def extract_licenses(filename):
import re
lic_regex = re.compile(rb'^\W*SPDX-License-Identifier:\s*([ \w\d.()+-]+?)(?:\s+\W*)?$', re.MULTILINE)
try:
with open(filename, 'rb') as f:
size = min(15000, os.stat(filename).st_size)
txt = f.read(size)
licenses = re.findall(lic_regex, txt)
if licenses:
ascii_licenses = [lic.decode('ascii') for lic in licenses]
return ascii_licenses
except Exception as e:
bb.warn(f"Exception reading {filename}: {e}")
return None
def get_doc_namespace(d, doc):
def get_namespace(d, name):
import uuid
namespace_uuid = uuid.uuid5(uuid.NAMESPACE_DNS, d.getVar("SPDX_UUID_NAMESPACE"))
return "%s/%s-%s" % (d.getVar("SPDX_NAMESPACE_PREFIX"), doc.name, str(uuid.uuid5(namespace_uuid, doc.name)))
return "%s/%s-%s" % (d.getVar("SPDX_NAMESPACE_PREFIX"), name, str(uuid.uuid5(namespace_uuid, name)))
def create_annotation(d, comment):
from datetime import datetime, timezone
@@ -80,26 +28,6 @@ def recipe_spdx_is_native(d, recipe):
a.annotator == "Tool: %s - %s" % (d.getVar("SPDX_TOOL_NAME"), d.getVar("SPDX_TOOL_VERSION")) and
a.comment == "isNative" for a in recipe.annotations)
def is_work_shared_spdx(d):
return bb.data.inherits_class('kernel', d) or ('work-shared' in d.getVar('WORKDIR'))
def get_json_indent(d):
if d.getVar("SPDX_PRETTY") == "1":
return 2
return None
python() {
import json
if d.getVar("SPDX_LICENSE_DATA"):
return
with open(d.getVar("SPDX_LICENSES"), "r") as f:
data = json.load(f)
# Transform the license array to a dictionary
data["licenses"] = {l["licenseId"]: l for l in data["licenses"]}
d.setVar("SPDX_LICENSE_DATA", data)
}
def convert_license_to_spdx(lic, document, d, existing={}):
from pathlib import Path
import oe.spdx
@@ -172,34 +100,6 @@ def convert_license_to_spdx(lic, document, d, existing={}):
return ' '.join(convert(l) for l in lic_split)
def process_sources(d):
pn = d.getVar('PN')
assume_provided = (d.getVar("ASSUME_PROVIDED") or "").split()
if pn in assume_provided:
for p in d.getVar("PROVIDES").split():
if p != pn:
pn = p
break
# glibc-locale: do_fetch, do_unpack and do_patch tasks have been deleted,
# so avoid archiving source here.
if pn.startswith('glibc-locale'):
return False
if d.getVar('PN') == "libtool-cross":
return False
if d.getVar('PN') == "libgcc-initial":
return False
if d.getVar('PN') == "shadow-sysroot":
return False
# We just archive gcc-source for all the gcc related recipes
if d.getVar('BPN') in ['gcc', 'libgcc']:
bb.debug(1, 'spdx: There is bug in scan of %s is, do nothing' % pn)
return False
return True
def add_package_files(d, doc, spdx_pkg, topdir, get_spdxid, get_types, *, archive=None, ignore_dirs=[], ignore_top_level_dirs=[]):
from pathlib import Path
import oe.spdx
@@ -348,14 +248,12 @@ def collect_dep_recipes(d, doc, spdx_recipe):
import oe.spdx
deploy_dir_spdx = Path(d.getVar("DEPLOY_DIR_SPDX"))
spdx_deps_file = Path(d.getVar("SPDXDEPS"))
package_archs = d.getVar("SSTATE_ARCHS").split()
package_archs.reverse()
dep_recipes = []
with spdx_deps_file.open("r") as f:
deps = json.load(f)
deps = get_spdx_deps(d)
for dep_pn, dep_hashfn, in_taskhash in deps:
# If this dependency is not calculated in the taskhash skip it.
@@ -468,51 +366,6 @@ def add_download_packages(d, doc, recipe):
# but this should be sufficient for now
doc.add_relationship(package, "BUILD_DEPENDENCY_OF", recipe)
def collect_direct_deps(d, dep_task):
current_task = "do_" + d.getVar("BB_CURRENTTASK")
pn = d.getVar("PN")
taskdepdata = d.getVar("BB_TASKDEPDATA", False)
for this_dep in taskdepdata.values():
if this_dep[0] == pn and this_dep[1] == current_task:
break
else:
bb.fatal(f"Unable to find this {pn}:{current_task} in taskdepdata")
deps = set()
for dep_name in this_dep[3]:
dep_data = taskdepdata[dep_name]
if dep_data[1] == dep_task and dep_data[0] != pn:
deps.add((dep_data[0], dep_data[7], dep_name in this_dep[8]))
return sorted(deps)
collect_direct_deps[vardepsexclude] += "BB_TASKDEPDATA"
collect_direct_deps[vardeps] += "DEPENDS"
python do_collect_spdx_deps() {
# This task calculates the build time dependencies of the recipe, and is
# required because while a task can deptask on itself, those dependencies
# do not show up in BB_TASKDEPDATA. To work around that, this task does the
# deptask on do_create_spdx and writes out the dependencies it finds, then
# do_create_spdx reads in the found dependencies when writing the actual
# SPDX document
import json
from pathlib import Path
spdx_deps_file = Path(d.getVar("SPDXDEPS"))
deps = collect_direct_deps(d, "do_create_spdx")
with spdx_deps_file.open("w") as f:
json.dump(deps, f)
}
# NOTE: depending on do_unpack is a hack that is necessary to get it's dependencies for archive the source
addtask do_collect_spdx_deps after do_unpack
do_collect_spdx_deps[depends] += "${PATCHDEPENDENCY}"
do_collect_spdx_deps[deptask] = "do_create_spdx"
do_collect_spdx_deps[dirs] = "${SPDXDIR}"
python do_create_spdx() {
from datetime import datetime, timezone
@@ -551,7 +404,7 @@ python do_create_spdx() {
doc = oe.spdx.SPDXDocument()
doc.name = "recipe-" + d.getVar("PN")
doc.documentNamespace = get_doc_namespace(d, doc)
doc.documentNamespace = get_namespace(d, doc.name)
doc.creationInfo.created = creation_time
doc.creationInfo.comment = "This document was created by analyzing recipe files during the build."
doc.creationInfo.licenseListVersion = d.getVar("SPDX_LICENSE_DATA")["licenseListVersion"]
@@ -655,7 +508,7 @@ python do_create_spdx() {
package_doc = oe.spdx.SPDXDocument()
pkg_name = d.getVar("PKG:%s" % package) or package
package_doc.name = pkg_name
package_doc.documentNamespace = get_doc_namespace(d, package_doc)
package_doc.documentNamespace = get_namespace(d, package_doc.name)
package_doc.creationInfo.created = creation_time
package_doc.creationInfo.comment = "This document was created by analyzing packages created during the build."
package_doc.creationInfo.licenseListVersion = d.getVar("SPDX_LICENSE_DATA")["licenseListVersion"]
@@ -716,44 +569,6 @@ do_create_spdx[dirs] = "${SPDXWORK}"
do_create_spdx[cleandirs] = "${SPDXDEPLOY} ${SPDXWORK}"
do_create_spdx[depends] += "${PATCHDEPENDENCY}"
def collect_package_providers(d):
from pathlib import Path
import oe.sbom
import oe.spdx
import json
deploy_dir_spdx = Path(d.getVar("DEPLOY_DIR_SPDX"))
providers = {}
deps = collect_direct_deps(d, "do_create_spdx")
deps.append((d.getVar("PN"), d.getVar("BB_HASHFILENAME"), True))
for dep_pn, dep_hashfn, _ in deps:
localdata = d
recipe_data = oe.packagedata.read_pkgdata(dep_pn, localdata)
if not recipe_data:
localdata = bb.data.createCopy(d)
localdata.setVar("PKGDATA_DIR", "${PKGDATA_DIR_SDK}")
recipe_data = oe.packagedata.read_pkgdata(dep_pn, localdata)
for pkg in recipe_data.get("PACKAGES", "").split():
pkg_data = oe.packagedata.read_subpkgdata_dict(pkg, localdata)
rprovides = set(n for n, _ in bb.utils.explode_dep_versions2(pkg_data.get("RPROVIDES", "")).items())
rprovides.add(pkg)
if "PKG" in pkg_data:
pkg = pkg_data["PKG"]
rprovides.add(pkg)
for r in rprovides:
providers[r] = (pkg, dep_hashfn)
return providers
collect_package_providers[vardepsexclude] += "BB_TASKDEPDATA"
python do_create_runtime_spdx() {
from datetime import datetime, timezone
import oe.sbom
@@ -800,7 +615,7 @@ python do_create_runtime_spdx() {
runtime_doc = oe.spdx.SPDXDocument()
runtime_doc.name = "runtime-" + pkg_name
runtime_doc.documentNamespace = get_doc_namespace(localdata, runtime_doc)
runtime_doc.documentNamespace = get_namespace(localdata, runtime_doc.name)
runtime_doc.creationInfo.created = creation_time
runtime_doc.creationInfo.comment = "This document was created by analyzing package runtime dependencies."
runtime_doc.creationInfo.licenseListVersion = d.getVar("SPDX_LICENSE_DATA")["licenseListVersion"]
@@ -891,60 +706,6 @@ do_create_runtime_spdx[dirs] = "${SPDXRUNTIMEDEPLOY}"
do_create_runtime_spdx[cleandirs] = "${SPDXRUNTIMEDEPLOY}"
do_create_runtime_spdx[rdeptask] = "do_create_spdx"
def spdx_get_src(d):
"""
save patched source of the recipe in SPDX_WORKDIR.
"""
import shutil
spdx_workdir = d.getVar('SPDXWORK')
spdx_sysroot_native = d.getVar('STAGING_DIR_NATIVE')
pn = d.getVar('PN')
workdir = d.getVar("WORKDIR")
try:
# The kernel class functions require it to be on work-shared, so we dont change WORKDIR
if not is_work_shared_spdx(d):
# Change the WORKDIR to make do_unpack do_patch run in another dir.
d.setVar('WORKDIR', spdx_workdir)
# Restore the original path to recipe's native sysroot (it's relative to WORKDIR).
d.setVar('STAGING_DIR_NATIVE', spdx_sysroot_native)
# The changed 'WORKDIR' also caused 'B' changed, create dir 'B' for the
# possibly requiring of the following tasks (such as some recipes's
# do_patch required 'B' existed).
bb.utils.mkdirhier(d.getVar('B'))
bb.build.exec_func('do_unpack', d)
# Copy source of kernel to spdx_workdir
if is_work_shared_spdx(d):
share_src = d.getVar('WORKDIR')
d.setVar('WORKDIR', spdx_workdir)
d.setVar('STAGING_DIR_NATIVE', spdx_sysroot_native)
src_dir = spdx_workdir + "/" + d.getVar('PN')+ "-" + d.getVar('PV') + "-" + d.getVar('PR')
bb.utils.mkdirhier(src_dir)
if bb.data.inherits_class('kernel',d):
share_src = d.getVar('STAGING_KERNEL_DIR')
cmd_copy_share = "cp -rf " + share_src + "/* " + src_dir + "/"
cmd_copy_shared_res = os.popen(cmd_copy_share).read()
bb.note("cmd_copy_shared_result = " + cmd_copy_shared_res)
git_path = src_dir + "/.git"
if os.path.exists(git_path):
shutils.rmtree(git_path)
# Make sure gcc and kernel sources are patched only once
if not (d.getVar('SRC_URI') == "" or is_work_shared_spdx(d)):
bb.build.exec_func('do_patch', d)
# Some userland has no source.
if not os.path.exists( spdx_workdir ):
bb.utils.mkdirhier(spdx_workdir)
finally:
d.setVar("WORKDIR", workdir)
spdx_get_src[vardepsexclude] += "STAGING_KERNEL_DIR"
do_rootfs[recrdeptask] += "do_create_spdx do_create_runtime_spdx"
do_rootfs[cleandirs] += "${SPDXIMAGEWORK}"
@@ -1019,7 +780,7 @@ def combine_spdx(d, rootfs_name, rootfs_deploydir, rootfs_spdxid, packages, spdx
doc = oe.spdx.SPDXDocument()
doc.name = rootfs_name
doc.documentNamespace = get_doc_namespace(d, doc)
doc.documentNamespace = get_namespace(d, doc.name)
doc.creationInfo.created = creation_time
doc.creationInfo.comment = "This document was created by analyzing the source of the Yocto recipe during the build."
doc.creationInfo.licenseListVersion = d.getVar("SPDX_LICENSE_DATA")["licenseListVersion"]
+256
View File
@@ -0,0 +1,256 @@
#
# Copyright OpenEmbedded Contributors
#
# SPDX-License-Identifier: GPL-2.0-only
#
DEPLOY_DIR_SPDX ??= "${DEPLOY_DIR}/spdx"
# The product name that the CVE database uses. Defaults to BPN, but may need to
# be overriden per recipe (for example tiff.bb sets CVE_PRODUCT=libtiff).
CVE_PRODUCT ??= "${BPN}"
CVE_VERSION ??= "${PV}"
SPDXDIR ??= "${WORKDIR}/spdx"
SPDXDEPLOY = "${SPDXDIR}/deploy"
SPDXWORK = "${SPDXDIR}/work"
SPDXIMAGEWORK = "${SPDXDIR}/image-work"
SPDXSDKWORK = "${SPDXDIR}/sdk-work"
SPDXDEPS = "${SPDXDIR}/deps.json"
SPDX_TOOL_NAME ??= "oe-spdx-creator"
SPDX_TOOL_VERSION ??= "1.0"
SPDXRUNTIMEDEPLOY = "${SPDXDIR}/runtime-deploy"
SPDX_INCLUDE_SOURCES ??= "0"
SPDX_ARCHIVE_SOURCES ??= "0"
SPDX_ARCHIVE_PACKAGED ??= "0"
SPDX_UUID_NAMESPACE ??= "sbom.openembedded.org"
SPDX_NAMESPACE_PREFIX ??= "http://spdx.org/spdxdocs"
SPDX_PRETTY ??= "0"
SPDX_LICENSES ??= "${COREBASE}/meta/files/spdx-licenses.json"
SPDX_CUSTOM_ANNOTATION_VARS ??= ""
SPDX_ORG ??= "OpenEmbedded ()"
SPDX_SUPPLIER ??= "Organization: ${SPDX_ORG}"
SPDX_SUPPLIER[doc] = "The SPDX PackageSupplier field for SPDX packages created from \
this recipe. For SPDX documents create using this class during the build, this \
is the contact information for the person or organization who is doing the \
build."
def extract_licenses(filename):
import re
lic_regex = re.compile(rb'^\W*SPDX-License-Identifier:\s*([ \w\d.()+-]+?)(?:\s+\W*)?$', re.MULTILINE)
try:
with open(filename, 'rb') as f:
size = min(15000, os.stat(filename).st_size)
txt = f.read(size)
licenses = re.findall(lic_regex, txt)
if licenses:
ascii_licenses = [lic.decode('ascii') for lic in licenses]
return ascii_licenses
except Exception as e:
bb.warn(f"Exception reading {filename}: {e}")
return None
def is_work_shared_spdx(d):
return bb.data.inherits_class('kernel', d) or ('work-shared' in d.getVar('WORKDIR'))
def get_json_indent(d):
if d.getVar("SPDX_PRETTY") == "1":
return 2
return None
python() {
import json
if d.getVar("SPDX_LICENSE_DATA"):
return
with open(d.getVar("SPDX_LICENSES"), "r") as f:
data = json.load(f)
# Transform the license array to a dictionary
data["licenses"] = {l["licenseId"]: l for l in data["licenses"]}
d.setVar("SPDX_LICENSE_DATA", data)
}
def process_sources(d):
pn = d.getVar('PN')
assume_provided = (d.getVar("ASSUME_PROVIDED") or "").split()
if pn in assume_provided:
for p in d.getVar("PROVIDES").split():
if p != pn:
pn = p
break
# glibc-locale: do_fetch, do_unpack and do_patch tasks have been deleted,
# so avoid archiving source here.
if pn.startswith('glibc-locale'):
return False
if d.getVar('PN') == "libtool-cross":
return False
if d.getVar('PN') == "libgcc-initial":
return False
if d.getVar('PN') == "shadow-sysroot":
return False
# We just archive gcc-source for all the gcc related recipes
if d.getVar('BPN') in ['gcc', 'libgcc']:
bb.debug(1, 'spdx: There is bug in scan of %s is, do nothing' % pn)
return False
return True
def collect_direct_deps(d, dep_task):
current_task = "do_" + d.getVar("BB_CURRENTTASK")
pn = d.getVar("PN")
taskdepdata = d.getVar("BB_TASKDEPDATA", False)
for this_dep in taskdepdata.values():
if this_dep[0] == pn and this_dep[1] == current_task:
break
else:
bb.fatal(f"Unable to find this {pn}:{current_task} in taskdepdata")
deps = set()
for dep_name in this_dep.deps:
dep_data = taskdepdata[dep_name]
if dep_data.taskname == dep_task and dep_data.pn != pn:
deps.add((dep_data.pn, dep_data.hashfn, dep_name in this_dep.taskhash_deps))
return sorted(deps)
collect_direct_deps[vardepsexclude] += "BB_TASKDEPDATA"
collect_direct_deps[vardeps] += "DEPENDS"
python do_collect_spdx_deps() {
# This task calculates the build time dependencies of the recipe, and is
# required because while a task can deptask on itself, those dependencies
# do not show up in BB_TASKDEPDATA. To work around that, this task does the
# deptask on do_create_spdx and writes out the dependencies it finds, then
# do_create_spdx reads in the found dependencies when writing the actual
# SPDX document
import json
from pathlib import Path
spdx_deps_file = Path(d.getVar("SPDXDEPS"))
deps = collect_direct_deps(d, "do_create_spdx")
with spdx_deps_file.open("w") as f:
json.dump(deps, f)
}
# NOTE: depending on do_unpack is a hack that is necessary to get it's dependencies for archive the source
addtask do_collect_spdx_deps after do_unpack
do_collect_spdx_deps[depends] += "${PATCHDEPENDENCY}"
do_collect_spdx_deps[deptask] = "do_create_spdx"
do_collect_spdx_deps[dirs] = "${SPDXDIR}"
def get_spdx_deps(d):
import json
from pathlib import Path
spdx_deps_file = Path(d.getVar("SPDXDEPS"))
with spdx_deps_file.open("r") as f:
return json.load(f)
def collect_package_providers(d):
from pathlib import Path
import oe.sbom
import oe.spdx
import json
deploy_dir_spdx = Path(d.getVar("DEPLOY_DIR_SPDX"))
providers = {}
deps = collect_direct_deps(d, "do_create_spdx")
deps.append((d.getVar("PN"), d.getVar("BB_HASHFILENAME"), True))
for dep_pn, dep_hashfn, _ in deps:
localdata = d
recipe_data = oe.packagedata.read_pkgdata(dep_pn, localdata)
if not recipe_data:
localdata = bb.data.createCopy(d)
localdata.setVar("PKGDATA_DIR", "${PKGDATA_DIR_SDK}")
recipe_data = oe.packagedata.read_pkgdata(dep_pn, localdata)
for pkg in recipe_data.get("PACKAGES", "").split():
pkg_data = oe.packagedata.read_subpkgdata_dict(pkg, localdata)
rprovides = set(n for n, _ in bb.utils.explode_dep_versions2(pkg_data.get("RPROVIDES", "")).items())
rprovides.add(pkg)
if "PKG" in pkg_data:
pkg = pkg_data["PKG"]
rprovides.add(pkg)
for r in rprovides:
providers[r] = (pkg, dep_hashfn)
return providers
collect_package_providers[vardepsexclude] += "BB_TASKDEPDATA"
def spdx_get_src(d):
"""
save patched source of the recipe in SPDX_WORKDIR.
"""
import shutil
spdx_workdir = d.getVar('SPDXWORK')
spdx_sysroot_native = d.getVar('STAGING_DIR_NATIVE')
pn = d.getVar('PN')
workdir = d.getVar("WORKDIR")
try:
# The kernel class functions require it to be on work-shared, so we dont change WORKDIR
if not is_work_shared_spdx(d):
# Change the WORKDIR to make do_unpack do_patch run in another dir.
d.setVar('WORKDIR', spdx_workdir)
# Restore the original path to recipe's native sysroot (it's relative to WORKDIR).
d.setVar('STAGING_DIR_NATIVE', spdx_sysroot_native)
# The changed 'WORKDIR' also caused 'B' changed, create dir 'B' for the
# possibly requiring of the following tasks (such as some recipes's
# do_patch required 'B' existed).
bb.utils.mkdirhier(d.getVar('B'))
bb.build.exec_func('do_unpack', d)
# Copy source of kernel to spdx_workdir
if is_work_shared_spdx(d):
share_src = d.getVar('WORKDIR')
d.setVar('WORKDIR', spdx_workdir)
d.setVar('STAGING_DIR_NATIVE', spdx_sysroot_native)
src_dir = spdx_workdir + "/" + d.getVar('PN')+ "-" + d.getVar('PV') + "-" + d.getVar('PR')
bb.utils.mkdirhier(src_dir)
if bb.data.inherits_class('kernel',d):
share_src = d.getVar('STAGING_KERNEL_DIR')
cmd_copy_share = "cp -rf " + share_src + "/* " + src_dir + "/"
cmd_copy_shared_res = os.popen(cmd_copy_share).read()
bb.note("cmd_copy_shared_result = " + cmd_copy_shared_res)
git_path = src_dir + "/.git"
if os.path.exists(git_path):
shutils.rmtree(git_path)
# Make sure gcc and kernel sources are patched only once
if not (d.getVar('SRC_URI') == "" or is_work_shared_spdx(d)):
bb.build.exec_func('do_patch', d)
# Some userland has no source.
if not os.path.exists( spdx_workdir ):
bb.utils.mkdirhier(spdx_workdir)
finally:
d.setVar("WORKDIR", workdir)
spdx_get_src[vardepsexclude] += "STAGING_KERNEL_DIR"