Merge pull request #1533 from aboutcode-org/917-ghost-remover

keshav-space · web-flow · commit 1e3afdcc830f · 2024-08-26T13:20:39.000+05:30
Add improver pipeline to flag ghost packages #644 #917 #1395
diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
@@ -9,7 +9,7 @@ jobs:
     strategy:
       max-parallel: 4
       matrix:
-        python-version: [3.8]
+        python-version: [3.9]
 
     steps:
       - name: Checkout code
diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
@@ -29,7 +29,7 @@ jobs:
     strategy:
       max-parallel: 4
       matrix:
-        python-version: ["3.8", "3.9", "3.10"]
+        python-version: ["3.9", "3.10", "3.11"]
 
     steps:
       - name: Checkout code
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -1,6 +1,14 @@
 Release notes
 =============
 
+Version (next)
+-------------------
+
+- Add Pipeline to flag ghost packages (#1533)
+- Add logging configuration (#1533)
+- Drop support for python 3.8 (#1533)
+
+
 Version v34.0.0
 -------------------
 
diff --git a/requirements.txt b/requirements.txt
@@ -1,3 +1,4 @@
+aboutcode.pipeline==0.1.0
 aiosignal==1.2.0
 alabaster==0.7.12
 asgiref==3.5.2
@@ -10,6 +11,7 @@ bcrypt==3.2.0
 beautifulsoup4==4.10.0
 binaryornot==0.4.4
 black==22.3.0
+bleach==6.1.0
 boolean.py==3.8
 certifi==2024.7.4
 cffi==1.15.0
@@ -49,6 +51,7 @@ jsonschema==3.2.0
 license-expression==21.6.14
 lxml==4.9.1
 Markdown==3.3.4
+markdown-it-py==3.0.0
 MarkupSafe==2.1.1
 matplotlib-inline==0.1.3
 multidict==6.0.2
diff --git a/setup.cfg b/setup.cfg
@@ -48,7 +48,7 @@ license_files =
     README.rst
 
 [options]
-python_requires = >=3.8
+python_requires = >=3.9
 
 packages=find:
 include_package_data = true
@@ -92,6 +92,9 @@ install_requires =
     requests>=2.25.1
     fetchcode>=0.3.0
 
+    #pipeline
+    aboutcode.pipeline>=0.1.0
+
     #vulntotal
     python-dotenv
     texttable
diff --git a/vulnerabilities/improvers/__init__.py b/vulnerabilities/improvers/__init__.py
@@ -10,6 +10,7 @@
 from vulnerabilities.improvers import valid_versions
 from vulnerabilities.improvers import vulnerability_kev
 from vulnerabilities.improvers import vulnerability_status
+from vulnerabilities.pipelines import flag_ghost_packages
 
 IMPROVERS_REGISTRY = [
     valid_versions.GitHubBasicImprover,
@@ -29,6 +30,7 @@
     valid_versions.GithubOSVImprover,
     vulnerability_status.VulnerabilityStatusImprover,
     vulnerability_kev.VulnerabilityKevImprover,
+    flag_ghost_packages.FlagGhostPackagePipeline,
 ]
 
 IMPROVERS_REGISTRY = {x.qualified_name: x for x in IMPROVERS_REGISTRY}
diff --git a/vulnerabilities/management/commands/improve.py b/vulnerabilities/management/commands/improve.py
@@ -14,6 +14,7 @@
 
 from vulnerabilities.improve_runner import ImproveRunner
 from vulnerabilities.improvers import IMPROVERS_REGISTRY
+from vulnerabilities.pipelines import VulnerableCodePipeline
 
 
 class Command(BaseCommand):
@@ -56,6 +57,13 @@ def improve_data(self, improvers):
 
         for improver in improvers:
             self.stdout.write(f"Improving data using {improver.qualified_name}")
+            if issubclass(improver, VulnerableCodePipeline):
+                status, error = improver().execute()
+                if status != 0:
+                    self.stdout.write(error)
+                    failed_improvers.append(improver.qualified_name)
+                continue
+
             try:
                 ImproveRunner(improver_class=improver).run()
                 self.stdout.write(
diff --git a/vulnerabilities/migrations/0062_package_is_ghost.py b/vulnerabilities/migrations/0062_package_is_ghost.py
@@ -0,0 +1,21 @@
+# Generated by Django 4.1.13 on 2024-08-23 12:47
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("vulnerabilities", "0061_alter_packagechangelog_software_version_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="package",
+            name="is_ghost",
+            field=models.BooleanField(
+                default=False,
+                help_text="True if the package does not exist in the upstream package manager or its repository.",
+            ),
+        ),
+    ]
diff --git a/vulnerabilities/models.py b/vulnerabilities/models.py
@@ -610,6 +610,11 @@ class Package(PackageURLMixin):
         db_index=True,
     )
 
+    is_ghost = models.BooleanField(
+        default=False,
+        help_text="True if the package does not exist in the upstream package manager or its repository.",
+    )
+
     objects = PackageQuerySet.as_manager()
 
     def save(self, *args, **kwargs):
diff --git a/vulnerabilities/pipelines/__init__.py b/vulnerabilities/pipelines/__init__.py
@@ -0,0 +1,34 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# VulnerableCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/vulnerablecode for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+import logging
+from datetime import datetime
+from datetime import timezone
+
+from aboutcode.pipeline import BasePipeline
+
+from vulnerabilities.utils import classproperty
+
+module_logger = logging.getLogger(__name__)
+
+
+class VulnerableCodePipeline(BasePipeline):
+    def log(self, message, level=logging.INFO):
+        """Log the given `message` to the current module logger and execution_log."""
+        now_local = datetime.now(timezone.utc).astimezone()
+        timestamp = now_local.strftime("%Y-%m-%d %H:%M:%S.%f")[:-3]
+        message = f"{timestamp} {message}"
+        module_logger.log(level, message)
+        self.append_to_log(message)
+
+    @classproperty
+    def qualified_name(cls):
+        """
+        Fully qualified name prefixed with the module name of the pipeline used in logging.
+        """
+        return f"{cls.__module__}.{cls.__qualname__}"
diff --git a/vulnerabilities/pipelines/flag_ghost_packages.py b/vulnerabilities/pipelines/flag_ghost_packages.py
@@ -0,0 +1,102 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# VulnerableCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/vulnerablecode for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+import logging
+from itertools import groupby
+from traceback import format_exc as traceback_format_exc
+
+from aboutcode.pipeline import LoopProgress
+from fetchcode.package_versions import SUPPORTED_ECOSYSTEMS as FETCHCODE_SUPPORTED_ECOSYSTEMS
+from fetchcode.package_versions import versions
+from packageurl import PackageURL
+
+from vulnerabilities.models import Package
+from vulnerabilities.pipelines import VulnerableCodePipeline
+
+
+class FlagGhostPackagePipeline(VulnerableCodePipeline):
+    """Detect and flag packages that do not exist upstream."""
+
+    @classmethod
+    def steps(cls):
+        return (cls.flag_ghost_packages,)
+
+    def flag_ghost_packages(self):
+        detect_and_flag_ghost_packages(logger=self.log)
+
+
+def detect_and_flag_ghost_packages(logger=None):
+    """Check if packages are available upstream. If not, mark them as ghost package."""
+    interesting_packages_qs = (
+        Package.objects.order_by("type", "namespace", "name")
+        .filter(type__in=FETCHCODE_SUPPORTED_ECOSYSTEMS)
+        .filter(qualifiers="")
+        .filter(subpath="")
+    )
+
+    distinct_packages_count = (
+        interesting_packages_qs.values("type", "namespace", "name")
+        .distinct("type", "namespace", "name")
+        .count()
+    )
+
+    grouped_packages = groupby(
+        interesting_packages_qs.paginated(),
+        key=lambda pkg: (pkg.type, pkg.namespace, pkg.name),
+    )
+
+    ghost_package_count = 0
+    progress = LoopProgress(total_iterations=distinct_packages_count, logger=logger)
+    for type_namespace_name, packages in progress.iter(grouped_packages):
+        ghost_package_count += flag_ghost_packages(
+            base_purl=PackageURL(*type_namespace_name),
+            packages=packages,
+            logger=logger,
+        )
+
+    if logger:
+        logger(f"Successfully flagged {ghost_package_count:,d} ghost Packages")
+
+
+def flag_ghost_packages(base_purl, packages, logger=None):
+    """
+    Check if `packages` are available upstream.
+    If not, update `is_ghost` to `True`.
+    Return the number of packages flagged as ghost.
+    """
+    known_versions = get_versions(purl=base_purl, logger=logger)
+    # Skip if encounter error while fetching known versions
+    if known_versions is None:
+        return 0
+
+    ghost_packages = 0
+    for pkg in packages:
+        pkg.is_ghost = False
+        if pkg.version.lstrip("vV") not in known_versions:
+            pkg.is_ghost = True
+            ghost_packages += 1
+
+            if logger:
+                logger(f"Flagging ghost package {pkg.purl!s}", level=logging.DEBUG)
+        pkg.save()
+
+    return ghost_packages
+
+
+def get_versions(purl, logger=None):
+    """Return set of known versions for the given purl."""
+    try:
+        return {v.value.lstrip("vV") for v in versions(str(purl))}
+    except Exception as e:
+        if logger:
+            logger(
+                f"Error while fetching known versions for {purl!s}: {e!r} \n {traceback_format_exc()}",
+                level=logging.ERROR,
+            )
+        return
diff --git a/vulnerabilities/templates/package_details.html b/vulnerabilities/templates/package_details.html
@@ -62,6 +62,21 @@
                                         {{ fixed_package_details.purl.to_string }}
                                     </td>
                                 </tr>
+                                {% if package.is_ghost %}
+                                <tr>
+                                    <td class="two-col-left">
+                                        Tags
+                                    </td>
+                                    <td class="two-col-right">
+                                        <span
+                                            class="tag is-warning is-hoverablem has-tooltip-multiline has-tooltip-black"
+                                            data-tooltip="This package does not exist in the upstream package manager or its repository."
+                                            style="margin-right: 8px;">
+                                            Ghost
+                                        </span>
+                                    </td>
+                                </tr>
+                                {% endif %}
                             </tbody>
                         </table>
                     </div>
diff --git a/vulnerabilities/tests/pipelines/__init__.py b/vulnerabilities/tests/pipelines/__init__.py
@@ -0,0 +1,20 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# VulnerableCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/vulnerablecode for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+import io
+
+
+class TestLogger:
+    buffer = io.StringIO()
+
+    def write(self, msg, level=None):
+        self.buffer.write(msg)
+
+    def getvalue(self):
+        return self.buffer.getvalue()
diff --git a/vulnerabilities/tests/pipelines/test_flag_ghost_packages.py b/vulnerabilities/tests/pipelines/test_flag_ghost_packages.py
@@ -0,0 +1,71 @@
+#
+# Copyright (c) nexB Inc. and others. All rights reserved.
+# VulnerableCode is a trademark of nexB Inc.
+# SPDX-License-Identifier: Apache-2.0
+# See http://www.apache.org/licenses/LICENSE-2.0 for the license text.
+# See https://github.com/nexB/vulnerablecode for support or download.
+# See https://aboutcode.org for more information about nexB OSS projects.
+#
+
+
+from pathlib import Path
+from unittest import mock
+
+from django.test import TestCase
+from fetchcode.package_versions import PackageVersion
+from packageurl import PackageURL
+
+from vulnerabilities.models import Package
+from vulnerabilities.pipelines import flag_ghost_packages
+from vulnerabilities.tests.pipelines import TestLogger
+
+
+class FlagGhostPackagePipelineTest(TestCase):
+    data = Path(__file__).parent.parent / "test_data"
+
+    @mock.patch("vulnerabilities.pipelines.flag_ghost_packages.versions")
+    def test_flag_ghost_package(self, mock_fetchcode_versions):
+        Package.objects.create(type="pypi", name="foo", version="2.3.0")
+        Package.objects.create(type="pypi", name="foo", version="3.0.0")
+
+        mock_fetchcode_versions.return_value = [
+            PackageVersion(value="2.3.0"),
+        ]
+        interesting_packages_qs = Package.objects.all()
+        base_purl = PackageURL(type="pypi", name="foo")
+
+        self.assertEqual(0, Package.objects.filter(is_ghost=True).count())
+
+        flagged_package_count = flag_ghost_packages.flag_ghost_packages(
+            base_purl=base_purl,
+            packages=interesting_packages_qs,
+        )
+        self.assertEqual(1, flagged_package_count)
+        self.assertEqual(1, Package.objects.filter(is_ghost=True).count())
+
+    @mock.patch("vulnerabilities.pipelines.flag_ghost_packages.versions")
+    def test_detect_and_flag_ghost_packages(self, mock_fetchcode_versions):
+        Package.objects.create(type="pypi", name="foo", version="2.3.0")
+        Package.objects.create(type="pypi", name="foo", version="3.0.0")
+        Package.objects.create(
+            type="deb",
+            namespace="debian",
+            name="foo",
+            version="3.0.0",
+            qualifiers={"distro": "trixie"},
+        )
+
+        mock_fetchcode_versions.return_value = [
+            PackageVersion(value="2.3.0"),
+        ]
+
+        self.assertEqual(3, Package.objects.count())
+        self.assertEqual(0, Package.objects.filter(is_ghost=True).count())
+
+        logger = TestLogger()
+
+        flag_ghost_packages.detect_and_flag_ghost_packages(logger=logger.write)
+        expected = "Successfully flagged 1 ghost Packages"
+
+        self.assertIn(expected, logger.getvalue())
+        self.assertEqual(1, Package.objects.filter(is_ghost=True).count())
diff --git a/vulnerablecode/settings.py b/vulnerablecode/settings.py
