Improve License Clarity at Top Package Level

setup.cfg

@@ -188,6 +188,7 @@ scancode_scan =
 # scan plugins and before the output plugins. See also plugincode.post_scan
 # module for details and doc.
 scancode_post_scan =
+    package_summary = packagedcode.plugin_package:PackageSummary
     summary = summarycode.summarizer:ScanSummary
     tallies = summarycode.tallies:Tallies
     tallies-with-details = summarycode.tallies:TalliesWithDetails

src/packagedcode/models.py

@@ -1530,13 +1530,18 @@ class Package(PackageData):
         label='datasource ids',
         help='List of the datasource ids used to create this package.'
     )
-
+
+    license_clarity_score = attr.ib(default=attr.Factory(dict))
+
     def __attrs_post_init__(self, *args, **kwargs):
         if not self.package_uid:
             self.package_uid = build_package_uid(self.purl)
 
-    def to_dict(self):
-        return super().to_dict(with_details=False)
+    def to_dict(self, package_summary= False):
+        data = super().to_dict(with_details=False)
+        if not package_summary:
+            data.pop("license_clarity_score")
+        return data
 
     def to_package_data(self):
         mapping = super().to_dict(with_details=True)

src/packagedcode/plugin_package.py

@@ -16,11 +16,13 @@
 
 from commoncode.cliutils import PluggableCommandLineOption
 from commoncode.cliutils import DOC_GROUP
-from commoncode.cliutils import SCAN_GROUP
+from commoncode.cliutils import SCAN_GROUP, POST_SCAN_GROUP
 from commoncode.resource import Resource
 from commoncode.resource import strip_first_path_segment
 from plugincode.scan import scan_impl
+from plugincode.post_scan import post_scan_impl
 from plugincode.scan import ScanPlugin
+from plugincode.post_scan import PostScanPlugin
 
 from licensedcode.cache import build_spdx_license_expression
 from licensedcode.cache import get_cache
@@ -200,7 +202,7 @@ def get_scanner(self, package=True, system_package=False, package_only=False, **
             package_only=package_only,
         )
 
-    def process_codebase(self, codebase, strip_root=False, package_only=False, **kwargs):
+    def process_codebase(self, codebase, strip_root=False, package_only=False, package_summary=False, **kwargs):
         """
         Populate the ``codebase`` top level ``packages`` and ``dependencies``
         with package and dependency instances, assembling parsed package data
@@ -260,7 +262,7 @@ def process_codebase(self, codebase, strip_root=False, package_only=False, **kwa
                 logger_debug(f'packagedcode: process_codebase: add_license_from_sibling_file: modified: {modified}')
 
         # Create codebase-level packages and dependencies
-        create_package_and_deps(codebase, strip_root=strip_root, **kwargs)
+        create_package_and_deps(codebase, package_summary, strip_root=strip_root, **kwargs)
         #raise Exception()
 
         if has_licenses:
@@ -272,6 +274,32 @@ def process_codebase(self, codebase, strip_root=False, package_only=False, **kwa
                 if TRACE_LICENSE and modified:
                     logger_debug(f'packagedcode: process_codebase: add_referenced_license_matches_from_package: modified: {modified}')
 
+@post_scan_impl
+class PackageSummary(PostScanPlugin):
+    """
+    Summary at the Package Level.
+    """
+    run_order = 11
+    sort_order= 11
+
+    options = [
+        PluggableCommandLineOption(('--package-summary',),
+        is_flag=True, default=False,
+        help='Summarize scans by providing License Clarity Score at the '
+        'Package attribute level.',
+        help_group=POST_SCAN_GROUP)
+    ]
+
+    def is_enabled(self, package_summary, **kwargs):
+        return package_summary
+
+    def process_codebase(self, codebase, package_summary, **kwargs):
+        """
+        Process the codebase.
+        """
+        if not self.is_enabled(package_summary):
+            return
+
 
 def add_license_from_file(resource, codebase):
     """
@@ -352,7 +380,7 @@ def get_installed_packages(root_dir, processes=2, **kwargs):
     yield from packages_by_uid.values()
 
 
-def create_package_and_deps(codebase, package_adder=add_to_package, strip_root=False, **kwargs):
+def create_package_and_deps(codebase, package_summary= False ,package_adder=add_to_package, strip_root=False, **kwargs):
     """
     Create and save top-level Package and Dependency from the parsed
     package data present in the codebase.
@@ -363,8 +391,10 @@ def create_package_and_deps(codebase, package_adder=add_to_package, strip_root=F
         strip_root=strip_root,
         **kwargs
     )
-
-    codebase.attributes.packages.extend(package.to_dict() for package in packages)
+    codebase.attributes.packages.extend(
+        package.to_dict(package_summary= package_summary)
+        for package in packages
+    )
     codebase.attributes.dependencies.extend(dep.to_dict() for dep in dependencies)
 
 

tests/formattedcode/data/common/manifests-expected.jsonlines

@@ -12,19 +12,19 @@
           "--package": true
         },
         "notice": "Generated with ScanCode and provided on an \"AS IS\" BASIS, WITHOUT WARRANTIES\nOR CONDITIONS OF ANY KIND, either express or implied. No content created from\nScanCode should be considered or used as legal advice. Consult an Attorney\nfor any legal advice.\nScanCode is a free software code scanning tool from nexB Inc. and others.\nVisit https://github.com/nexB/scancode-toolkit/ for support and download.",
-        "output_format_version": "3.0.0",
+        "output_format_version": "3.1.0",
         "message": null,
         "errors": [],
         "warnings": [],
         "extra_data": {
           "system_environment": {
             "operating_system": "linux",
             "cpu_architecture": "64",
-            "platform": "Linux-5.15.0-89-generic-x86_64-with-glibc2.29",
-            "platform_version": "#99~20.04.1-Ubuntu SMP Thu Nov 2 15:16:47 UTC 2023",
-            "python_version": "3.8.10 (default, Nov 22 2023, 10:22:35) \n[GCC 9.4.0]"
+            "platform": "Linux-5.15.0-112-generic-x86_64-with-glibc2.35",
+            "platform_version": "#122-Ubuntu SMP Thu May 23 07:48:21 UTC 2024",
+            "python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]"
           },
-          "spdx_license_list_version": "3.22",
+          "spdx_license_list_version": "3.23",
           "files_count": 4
         }
       }

tests/formattedcode/data/common/manifests-expected.yaml

@@ -21,18 +21,18 @@ headers:
       for any legal advice.
       ScanCode is a free software code scanning tool from nexB Inc. and others.
       Visit https://github.com/nexB/scancode-toolkit/ for support and download.
-    output_format_version: 3.0.0
+    output_format_version: 3.1.0
     message:
     errors: []
     warnings: []
     extra_data:
       system_environment:
         operating_system: linux
         cpu_architecture: 64
-        platform: Linux-5.15.0-89-generic-x86_64-with-glibc2.29
-        platform_version: '#99~20.04.1-Ubuntu SMP Thu Nov 2 15:16:47 UTC 2023'
-        python_version: "3.8.10 (default, Nov 22 2023, 10:22:35) \n[GCC 9.4.0]"
-      spdx_license_list_version: '3.22'
+        platform: Linux-5.15.0-112-generic-x86_64-with-glibc2.35
+        platform_version: '#122-Ubuntu SMP Thu May 23 07:48:21 UTC 2024'
+        python_version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
+      spdx_license_list_version: '3.23'
       files_count: 4
 summary:
   declared_license_expression: apache-2.0 AND cddl-1.0 AND mit

tests/formattedcode/data/yaml/package-and-licenses-expected.yaml

@@ -21,18 +21,18 @@ headers:
       for any legal advice.
       ScanCode is a free software code scanning tool from nexB Inc. and others.
       Visit https://github.com/nexB/scancode-toolkit/ for support and download.
-    output_format_version: 3.0.0
+    output_format_version: 3.1.0
     message:
     errors: []
     warnings: []
     extra_data:
       system_environment:
         operating_system: linux
         cpu_architecture: 64
-        platform: Linux-5.15.0-89-generic-x86_64-with-glibc2.29
-        platform_version: '#99~20.04.1-Ubuntu SMP Thu Nov 2 15:16:47 UTC 2023'
-        python_version: "3.8.10 (default, Nov 22 2023, 10:22:35) \n[GCC 9.4.0]"
-      spdx_license_list_version: '3.22'
+        platform: Linux-5.15.0-112-generic-x86_64-with-glibc2.35
+        platform_version: '#122-Ubuntu SMP Thu May 23 07:48:21 UTC 2024'
+        python_version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
+      spdx_license_list_version: '3.23'
       files_count: 4
 summary:
   declared_license_expression: apache-2.0