Skip to content

Commit de0682b

Browse files
committed
Migrate debian importer to importer-improver model
Signed-off-by: Tushar Goel <tushar.goel.dav@gmail.com>
1 parent 75b2920 commit de0682b

File tree

13 files changed

+999
-147
lines changed

13 files changed

+999
-147
lines changed

requirements.txt

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -107,7 +107,7 @@ toml==0.10.2
107107
tomli==2.0.1
108108
traitlets==5.1.1
109109
typing_extensions==4.1.1
110-
univers==30.4.0
110+
univers==30.5.1
111111
urllib3==1.26.9
112112
wcwidth==0.2.5
113113
websocket-client==0.59.0

vulnerabilities/helpers.py

Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -26,6 +26,7 @@
2626
import logging
2727
import os
2828
import re
29+
from collections import defaultdict
2930
from functools import total_ordering
3031
from typing import List
3132
from typing import Optional
@@ -267,3 +268,14 @@ def _get_gh_response(gh_token, graphql_query):
267268
endpoint = "https://api.github.com/graphql"
268269
headers = {"Authorization": f"bearer {gh_token}"}
269270
return requests.post(endpoint, headers=headers, json=graphql_query).json()
271+
272+
273+
def get_affected_packages_by_patched_package(
274+
affected_packages: List[AffectedPackage],
275+
):
276+
affected_packages_by_patched_package = defaultdict(list)
277+
for package in affected_packages:
278+
affected_packages_by_patched_package[package.patched_package].append(
279+
package.vulnerable_package
280+
)
281+
return affected_packages_by_patched_package

vulnerabilities/importer.py

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -150,7 +150,9 @@ def get_fixed_purl(self):
150150
return fixed_purl
151151

152152
@classmethod
153-
def merge(cls, affected_packages: Iterable):
153+
def merge(
154+
cls, affected_packages: Iterable
155+
) -> Tuple[PackageURL, List[VersionRange], List[Version]]:
154156
"""
155157
Return a tuple with all attributes of AffectedPackage as a set
156158
for all values in the given iterable of AffectedPackage

vulnerabilities/importers/__init__.py

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -20,6 +20,7 @@
2020
# VulnerableCode is a free software code scanning tool from nexB Inc. and others.
2121
# Visit https://github.com/nexB/vulnerablecode/ for support and download.
2222
from vulnerabilities.importers import alpine_linux
23+
from vulnerabilities.importers import debian
2324
from vulnerabilities.importers import github
2425
from vulnerabilities.importers import nginx
2526
from vulnerabilities.importers import nvd
@@ -31,6 +32,7 @@
3132
github.GitHubAPIImporter,
3233
nvd.NVDImporter,
3334
openssl.OpensslImporter,
35+
debian.DebianImporter,
3436
]
3537

3638
IMPORTERS_REGISTRY = {x.qualified_name: x for x in IMPORTERS_REGISTRY}

vulnerabilities/importers/debian.py

Lines changed: 131 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -21,115 +21,194 @@
2121
# VulnerableCode is a free software tool from nexB Inc. and others.
2222
# Visit https://github.com/nexB/vulnerablecode/ for support and download.
2323

24-
import dataclasses
24+
import logging
2525
from typing import Any
26+
from typing import Iterable
2627
from typing import List
2728
from typing import Mapping
28-
from typing import Set
2929

3030
import requests
31-
from dateutil import parser as dateparser
31+
from django.db.models.query import QuerySet
3232
from packageurl import PackageURL
33+
from univers.version_range import DebianVersionRange
34+
from univers.versions import DebianVersion
3335

36+
from vulnerabilities.helpers import AffectedPackage as LegacyAffectedPackage
37+
from vulnerabilities.helpers import get_affected_packages_by_patched_package
38+
from vulnerabilities.helpers import get_item
3439
from vulnerabilities.helpers import nearest_patched_package
3540
from vulnerabilities.importer import AdvisoryData
41+
from vulnerabilities.importer import AffectedPackage
3642
from vulnerabilities.importer import Importer
3743
from vulnerabilities.importer import Reference
44+
from vulnerabilities.importer import UnMergeablePackageError
45+
from vulnerabilities.improver import MAX_CONFIDENCE
46+
from vulnerabilities.improver import Improver
47+
from vulnerabilities.improver import Inference
48+
from vulnerabilities.models import Advisory
3849

50+
logger = logging.getLogger(__name__)
3951

40-
class DebianImporter(Importer):
41-
def __enter__(self):
42-
if self.response_is_new():
43-
self._api_response = self._fetch()
52+
IGNORABLE_VERSIONS = {"3.8.20-4."}
4453

45-
else:
46-
self._api_response = {}
4754

48-
def updated_advisories(self) -> Set[AdvisoryData]:
49-
advisories = []
55+
class DebianImporter(Importer):
5056

51-
for pkg_name, records in self._api_response.items():
52-
advisories.extend(self._parse(pkg_name, records))
57+
spdx_license_expression = "MIT"
58+
license_url = "https://www.debian.org/license"
5359

54-
return self.batch_advisories(advisories)
60+
api_url = "https://security-tracker.debian.org/tracker/data/json"
5561

56-
def _fetch(self) -> Mapping[str, Any]:
57-
return requests.get(self.config.debian_tracker_url).json()
62+
def get_response(self):
63+
response = requests.get(self.api_url)
64+
if response.status_code == 200:
65+
return response.json()
66+
raise Exception(
67+
f"Failed to fetch data from {self.api_url!r} with status code: {response.status_code!r}"
68+
)
5869

59-
def _parse(self, pkg_name: str, records: Mapping[str, Any]) -> List[AdvisoryData]:
60-
advisories = []
61-
ignored_versions = {"3.8.20-4."}
70+
def advisory_data(self) -> Iterable[AdvisoryData]:
71+
response = self.get_response()
72+
for pkg_name, records in response.items():
73+
yield from self.parse(pkg_name, records)
6274

75+
def parse(
76+
self, pkg_name: str, records: Mapping[str, Any], ignored_versions=IGNORABLE_VERSIONS
77+
) -> Iterable[AdvisoryData]:
6378
for cve_id, record in records.items():
64-
impacted_purls, resolved_purls = [], []
79+
affected_versions, fixed_versions = [], []
6580
if not cve_id.startswith("CVE"):
81+
logger.error(f"Invalid CVE ID: {cve_id} in {record} in package {pkg_name}")
6682
continue
6783

6884
# vulnerabilities starting with something else may not be public yet
6985
# see for instance https://web.archive.org/web/20201215213725/https://security-tracker.debian.org/tracker/TEMP-0000000-A2EB44
7086
# TODO: this would need to be revisited though to ensure we are not missing out on anything
7187

7288
for release_name, release_record in record["releases"].items():
73-
if not release_record.get("repositories", {}).get(release_name):
74-
continue
89+
version = get_item(release_record, "repositories", release_name)
7590

76-
version = release_record["repositories"][release_name]
91+
if not version:
92+
logger.error(
93+
f"Version not found for {release_name} in {record} in package {pkg_name}"
94+
)
95+
continue
7796

7897
if version in ignored_versions:
98+
logger.error(f"Ignoring version {version} in {record} in package {pkg_name}")
7999
continue
80100

81101
purl = PackageURL(
82102
name=pkg_name,
83103
type="deb",
84104
namespace="debian",
85-
version=version,
86105
qualifiers={"distro": release_name},
87106
)
88107

89108
if release_record.get("status", "") == "resolved":
90-
resolved_purls.append(purl)
109+
fixed_versions.append(version)
91110
else:
92-
impacted_purls.append(purl)
111+
affected_versions.append(version)
93112

94113
if (
95114
"fixed_version" in release_record
96115
and release_record["fixed_version"] not in ignored_versions
97116
):
98-
resolved_purls.append(
99-
PackageURL(
100-
name=pkg_name,
101-
type="deb",
102-
namespace="debian",
103-
version=release_record["fixed_version"],
104-
qualifiers={"distro": release_name},
105-
)
106-
)
117+
fixed_versions.append(release_record["fixed_version"])
107118

108119
references = []
109120
debianbug = record.get("debianbug")
110121
if debianbug:
111122
bug_url = f"https://bugs.debian.org/cgi-bin/bugreport.cgi?bug={debianbug}"
112-
references.append(Reference(url=bug_url, reference_id=debianbug))
113-
advisories.append(
114-
AdvisoryData(
115-
vulnerability_id=cve_id,
116-
affected_packages=nearest_patched_package(impacted_purls, resolved_purls),
117-
summary=record.get("description", ""),
118-
references=references,
123+
references.append(Reference(url=bug_url, reference_id=str(debianbug)))
124+
affected_versions = list(dict.fromkeys(affected_versions))
125+
fixed_versions = list(dict.fromkeys(fixed_versions))
126+
affected_version_range = (
127+
DebianVersionRange.from_versions(affected_versions) if affected_versions else None
128+
)
129+
affected_packages = []
130+
for fixed_version in fixed_versions:
131+
affected_packages.append(
132+
AffectedPackage(
133+
package=purl,
134+
affected_version_range=affected_version_range,
135+
fixed_version=DebianVersion(fixed_version),
136+
)
119137
)
138+
yield AdvisoryData(
139+
aliases=[cve_id],
140+
summary=record.get("description", ""),
141+
affected_packages=affected_packages,
142+
references=references,
120143
)
121144

122-
return advisories
123145

124-
def response_is_new(self):
146+
class DebianBasicImprover(Improver):
147+
@property
148+
def interesting_advisories(self) -> QuerySet:
149+
return Advisory.objects.filter(created_by=DebianImporter.qualified_name)
150+
151+
def get_inferences(self, advisory_data: AdvisoryData) -> Iterable[Inference]:
125152
"""
126-
Return True if a request response is for new data likely changed or
127-
updated since we last checked.
153+
Yield Inferences for the given advisory data
128154
"""
129-
head = requests.head(self.config.debian_tracker_url)
130-
date_str = head.headers.get("last-modified")
131-
last_modified_date = dateparser.parse(date_str)
132-
if self.config.last_run_date:
133-
return self.config.last_run_date < last_modified_date
155+
if not advisory_data.affected_packages:
156+
return iter([])
157+
try:
158+
purl, affected_version_ranges, fixed_versions = AffectedPackage.merge(
159+
advisory_data.affected_packages
160+
)
161+
except UnMergeablePackageError:
162+
logger.error(f"Cannot merge with different purls {advisory_data.affected_packages!r}")
163+
return iter([])
164+
165+
pkg_type = purl.type
166+
pkg_namespace = purl.namespace
167+
pkg_name = purl.name
168+
pkg_qualifiers = purl.qualifiers
169+
fixed_purls = [
170+
PackageURL(
171+
type=pkg_type,
172+
namespace=pkg_namespace,
173+
name=pkg_name,
174+
version=str(version),
175+
qualifiers=pkg_qualifiers,
176+
)
177+
for version in fixed_versions
178+
]
179+
if not affected_version_ranges:
180+
for fixed_purl in fixed_purls:
181+
yield Inference.from_advisory_data(
182+
advisory_data,
183+
confidence=MAX_CONFIDENCE, # We are getting all valid versions to get this inference
184+
affected_purls=[],
185+
fixed_purl=fixed_purl,
186+
)
187+
else:
188+
aff_versions = set()
189+
for affected_version_range in affected_version_ranges:
190+
for constraint in affected_version_range.constraints:
191+
aff_versions.add(constraint.version.string)
192+
affected_purls = [
193+
PackageURL(
194+
type=pkg_type,
195+
namespace=pkg_namespace,
196+
name=pkg_name,
197+
version=version,
198+
qualifiers=pkg_qualifiers,
199+
)
200+
for version in aff_versions
201+
]
202+
affected_packages: List[LegacyAffectedPackage] = nearest_patched_package(
203+
vulnerable_packages=affected_purls, resolved_packages=fixed_purls
204+
)
134205

135-
return True
206+
for (fixed_package, affected_packages,) in get_affected_packages_by_patched_package(
207+
affected_packages=affected_packages
208+
).items():
209+
yield Inference.from_advisory_data(
210+
advisory_data,
211+
confidence=MAX_CONFIDENCE, # We are getting all valid versions to get this inference
212+
affected_purls=affected_packages,
213+
fixed_purl=fixed_package,
214+
)

vulnerabilities/importers/github.py

Lines changed: 2 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -37,6 +37,7 @@
3737
from vulnerabilities import helpers
3838
from vulnerabilities import severity_systems
3939
from vulnerabilities.helpers import AffectedPackage as LegacyAffectedPackage
40+
from vulnerabilities.helpers import get_affected_packages_by_patched_package
4041
from vulnerabilities.helpers import get_item
4142
from vulnerabilities.helpers import nearest_patched_package
4243
from vulnerabilities.importer import AdvisoryData
@@ -465,18 +466,10 @@ def get_inferences(self, advisory_data: AdvisoryData) -> Iterable[Inference]:
465466
vulnerable_packages=affected_purls, resolved_packages=unaffected_purls
466467
)
467468

468-
unique_patched_packages_with_affected_packages = {}
469-
for package in affected_packages:
470-
if package.patched_package not in unique_patched_packages_with_affected_packages:
471-
unique_patched_packages_with_affected_packages[package.patched_package] = []
472-
unique_patched_packages_with_affected_packages[package.patched_package].append(
473-
package.vulnerable_package
474-
)
475-
476469
for (
477470
fixed_package,
478471
affected_packages,
479-
) in unique_patched_packages_with_affected_packages.items():
472+
) in get_affected_packages_by_patched_package(affected_packages).items():
480473
yield Inference.from_advisory_data(
481474
advisory_data,
482475
confidence=100, # We are getting all valid versions to get this inference

vulnerabilities/improvers/__init__.py

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -28,6 +28,7 @@
2828
default.DefaultImprover,
2929
importers.nginx.NginxBasicImprover,
3030
importers.github.GitHubBasicImprover,
31+
importers.debian.DebianBasicImprover,
3132
]
3233

3334
IMPROVERS_REGISTRY = {x.qualified_name: x for x in IMPROVERS_REGISTRY}

vulnerabilities/templates/vulnerabilities.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -36,7 +36,7 @@ <h1 class="title">
3636
</tr>
3737
{% for vulnerability in vulnerabilities %}
3838
<tr>
39-
<td><a href="{% url 'vulnerability_view' vulnerability.pk %}">{{vulnerability.vulcoid}}</a></td>
39+
<td><a href="{% url 'vulnerability_view' vulnerability.pk %}">{{vulnerability.vulnerability_id}}</a></td>
4040
<td>{{vulnerability.vulnerable_package_count}}</td>
4141
<td>{{vulnerability.patched_package_count}}</td>
4242
</tr>

vulnerabilities/tests/conftest.py

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -46,7 +46,6 @@ def no_rmtree(monkeypatch):
4646
"test_archlinux.py",
4747
"test_data_source.py",
4848
"test_debian_oval.py",
49-
"test_debian.py",
5049
"test_elixir_security.py",
5150
"test_gentoo.py",
5251
"test_importer_yielder.py",

0 commit comments

Comments
 (0)