add dependencies to rpm

Added dependencies to rpm. Reference: #649 Signed-off-by: Alok Kumar <[email protected]> Signed-off-by: Alok Kumar <[email protected]>
aboutcode-org · Jan 10, 2025 · e16372a · e16372a
1 parent 1d0fe75
commit e16372a
Show file tree

Hide file tree

Showing 5 changed files with 251 additions and 11 deletions.
diff --git a/src/packagedcode/rpm.py b/src/packagedcode/rpm.py
@@ -15,6 +15,7 @@
 from collections import namedtuple
 from pathlib import Path
 
+from packageurl import PackageURL
 from packagedcode import models
 from packagedcode import nevra
 from packagedcode.licensing import RESOURCE_TO_PACKAGE_LICENSE_FIELDS
@@ -44,7 +45,6 @@ def logger_debug(*args):
 """
 Support for RPMs, installed databases and spec files.
 """
-# TODO: retrieve dependencies
 
 # TODO: parse spec files see:
 #  http://www.faqs.org/docs/artu/ch05s02.html#id2906931%29.)
@@ -125,7 +125,6 @@ def to_string(self):
         return vr
 
 
-# TODO: add dependencies!!!
 class BaseRpmInstalledDatabaseHandler(models.DatafileHandler):
 
     @classmethod
@@ -135,13 +134,45 @@ def parse(cls, location, package_only=False):
         loc_path = Path(location)
         rpmdb_loc = str(loc_path.parent)
 
+        rpm_tags = get_rpm_tags(location, include_desc=True)
+
+        if TRACE: logger_debug('recognize: rpm_tags', rpm_tags)
+        if not rpm_tags:
+            return
+
+        dependencies = []
+        name = rpm_tags.name
+        version = rpm_tags.version
+        is_pinned = version is not None and version != ""
+
+        # Construct PackageURL without '@version' if version is missing
+        purl = PackageURL(
+            type="rpm",
+            namespace=None,  # RPMs typically don't use namespaces
+            name=name,
+            version=version if is_pinned else None
+        )
+
+        # Prepare the dependent package model
+        dependencies.append(
+            models.DependentPackage(
+                purl=purl.to_string(),
+                scope="dependencies",
+                is_runtime=True,
+                is_optional=False,
+                is_pinned=is_pinned,
+                extracted_requirement=version,
+            )
+        )
+
         # dump and parse the rpmdb to XMLish
         xmlish_loc = collect_installed_rpmdb_xmlish_from_rpmdb_loc(rpmdb_loc=rpmdb_loc)
         package_data = parse_rpm_xmlish(
             location=xmlish_loc,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
             package_only=package_only,
+            dependencies=dependencies,
         )
         # TODO: package_data.namespace = cls.default_package_namespace
         return package_data
@@ -225,7 +256,6 @@ def assemble(cls, package_data, resource, codebase, package_adder):
             yield resource
 
 
-# TODO: add dependencies!!!
 class RpmInstalledNdbDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     # used by recent Suse
     datasource_id = 'rpm_installed_database_ndb'
@@ -237,8 +267,48 @@ class RpmInstalledNdbDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     description = 'RPM installed package NDB database'
     documentation_url = 'https://fedoraproject.org/wiki/Changes/NewRpmDBFormat'
 
+    @classmethod
+    def parse(cls, location, package_only=False):
+        rpm_tags = get_rpm_tags(location, include_desc=True)
+
+        if TRACE: logger_debug('recognize: rpm_tags', rpm_tags)
+        if not rpm_tags:
+            return
+
+        dependencies = []
+        name = rpm_tags.name
+        version = rpm_tags.version
+        is_pinned = version is not None and version != ""
+
+        # Construct PackageURL without '@version' if version is missing
+        purl = PackageURL(
+            type="rpm",
+            namespace=None,  # RPMs typically don't use namespaces
+            name=name,
+            version=version if is_pinned else None
+        )
+
+        # Prepare the dependent package model
+        dependencies.append(
+            models.DependentPackage(
+                purl=purl.to_string(),
+                scope="dependencies",
+                is_runtime=True,
+                is_optional=False,
+                is_pinned=is_pinned,
+                extracted_requirement=version,
+            )
+        )
+
+        package_data = dict(
+            datasource_id=cls.datasource_id,
+            type=cls.default_package_type,
+            dependencies=dependencies,
+        )
+
+        yield models.PackageData.from_data(package_data, package_only)    
+
 
-# TODO: add dependencies!!!
 class RpmInstalledSqliteDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     # used by newer RHEL/CentOS/Fedora/CoreOS
     # Filetype: SQLite 3.x database, ...
@@ -253,8 +323,48 @@ class RpmInstalledSqliteDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     description = 'RPM installed package SQLite database'
     documentation_url = 'https://fedoraproject.org/wiki/Changes/Sqlite_Rpmdb'
 
+    @classmethod
+    def parse(cls, location, package_only=False):
+        rpm_tags = get_rpm_tags(location, include_desc=True)
+
+        if TRACE: logger_debug('recognize: rpm_tags', rpm_tags)
+        if not rpm_tags:
+            return
+
+        dependencies = []
+        name = rpm_tags.name
+        version = rpm_tags.version
+        is_pinned = version is not None and version != ""
+
+        # Construct PackageURL without '@version' if version is missing
+        purl = PackageURL(
+            type="rpm",
+            namespace=None,  # RPMs typically don't use namespaces
+            name=name,
+            version=version if is_pinned else None
+        )
+
+        # Prepare the dependent package model
+        dependencies.append(
+            models.DependentPackage(
+                purl=purl.to_string(),
+                scope="dependencies",
+                is_runtime=True,
+                is_optional=False,
+                is_pinned=is_pinned,
+                extracted_requirement=version,
+            )
+        )
+
+        package_data = dict(
+            datasource_id=cls.datasource_id,
+            type=cls.default_package_type,
+            dependencies=dependencies,
+        )
+
+        yield models.PackageData.from_data(package_data, package_only)    
+
 
-# TODO: add dependencies!!!
 class RpmInstalledBdbDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     # used by legacy RHEL/CentOS/Fedora/Suse
     datasource_id = 'rpm_installed_database_bdb'
@@ -267,6 +377,47 @@ class RpmInstalledBdbDatabaseHandler(BaseRpmInstalledDatabaseHandler):
     description = 'RPM installed package BDB database'
     documentation_url = 'https://man7.org/linux/man-pages/man8/rpmdb.8.html'
 
+    @classmethod
+    def parse(cls, location, package_only=False):
+        rpm_tags = get_rpm_tags(location, include_desc=True)
+
+        if TRACE: logger_debug('recognize: rpm_tags', rpm_tags)
+        if not rpm_tags:
+            return
+
+        dependencies = []
+        name = rpm_tags.name
+        version = rpm_tags.version
+        is_pinned = version is not None and version != ""
+
+        # Construct PackageURL without '@version' if version is missing
+        purl = PackageURL(
+            type="rpm",
+            namespace=None,  # RPMs typically don't use namespaces
+            name=name,
+            version=version if is_pinned else None
+        )
+
+        # Prepare the dependent package model
+        dependencies.append(
+            models.DependentPackage(
+                purl=purl.to_string(),
+                scope="dependencies",
+                is_runtime=True,
+                is_optional=False,
+                is_pinned=is_pinned,
+                extracted_requirement=version,
+            )
+        )
+
+        package_data = dict(
+            datasource_id=cls.datasource_id,
+            type=cls.default_package_type,
+            dependencies=dependencies,
+        )
+
+        yield models.PackageData.from_data(package_data, package_only)    
+
 
 # TODO: implement me!!@
 class RpmSpecfileHandler(models.NonAssemblableDatafileHandler):
@@ -278,7 +429,6 @@ class RpmSpecfileHandler(models.NonAssemblableDatafileHandler):
     documentation_url = 'https://en.wikipedia.org/wiki/RPM_Package_Manager'
 
 
-# TODO: add dependencies!!!
 class RpmArchiveHandler(models.DatafileHandler):
     datasource_id = 'rpm_archive'
     path_patterns = ('*.rpm', '*.src.rpm', '*.srpm', '*.mvl', '*.vip',)
@@ -354,6 +504,30 @@ def parse(cls, location, package_only=False):
 
         description = build_description(summary=rpm_tags.summary, description=rpm_tags.description)
 
+        dependencies = []
+        name = rpm_tags.name
+        version = rpm_tags.version
+        is_pinned = version is not None and version != ""
+
+        # Construct PackageURL without '@version' if version is missing
+        purl = PackageURL(
+            type="rpm",
+            namespace=None,  # RPMs typically don't use namespaces
+            name=name,
+            version=version if is_pinned else None
+        )
+
+        # Prepare the dependent package model
+        dependencies.append(
+            models.DependentPackage(
+                purl=purl.to_string(),
+                scope="dependencies",
+                is_runtime=True,
+                is_optional=False,
+                is_pinned=is_pinned,
+                extracted_requirement=version,
+            )
+        )
         if TRACE:
             data = dict(
                 name=name,
@@ -363,6 +537,7 @@ def parse(cls, location, package_only=False):
                 parties=parties,
                 extracted_license_statement=rpm_tags.license or None,
                 source_packages=source_packages,
+                dependencies=dependencies,
             )
             logger_debug('recognize: data to create a package:\n', data)
 
@@ -377,6 +552,7 @@ def parse(cls, location, package_only=False):
             parties=parties,
             extracted_license_statement=rpm_tags.license or None,
             source_packages=source_packages,
+            dependencies=dependencies,
         )
 
         if TRACE:

diff --git a/tests/packagedcode/data/plugin/rpm-package-expected.json b/tests/packagedcode/data/plugin/rpm-package-expected.json
@@ -142,7 +142,19 @@
           "is_private": false,
           "is_virtual": false,
           "extra_data": {},
-          "dependencies": [],
+          "dependencies": [
+            {
+              "purl": "pkg:rpm/[email protected]",
+              "extracted_requirement": "2.0",
+              "scope": "dependencies",
+              "is_runtime": true,
+              "is_optional": false,
+              "is_pinned": true,
+              "is_direct": true,
+              "resolved_package": {},
+              "extra_data": {}
+            }
+          ],
           "repository_homepage_url": null,
           "repository_download_url": null,
           "api_data_url": null,

diff --git a/.../packagedcode/data/rpm/header/libproxy-bin-0.3.0-4.el6_3.x86_64.rpm-package-expected.json b/.../packagedcode/data/rpm/header/libproxy-bin-0.3.0-4.el6_3.x86_64.rpm-package-expected.json
@@ -69,7 +69,19 @@
     "is_private": false,
     "is_virtual": false,
     "extra_data": {},
-    "dependencies": [],
+    "dependencies": [
+      {
+        "purl": "pkg:rpm/[email protected]",
+        "extracted_requirement": "0.3.0",
+        "scope": "dependencies",
+        "is_runtime": true,
+        "is_optional": false,
+        "is_pinned": true,
+        "is_direct": true,
+        "resolved_package": {},
+        "extra_data": {}
+      }      
+    ],
     "repository_homepage_url": null,
     "repository_download_url": null,
     "api_data_url": null,

diff --git a/...agedcode/data/rpm/header/libproxy-bin-0.3.0-4.el6_3.x86_64.rpm-package-only-expected.json b/...agedcode/data/rpm/header/libproxy-bin-0.3.0-4.el6_3.x86_64.rpm-package-only-expected.json
@@ -46,7 +46,19 @@
     "is_private": false,
     "is_virtual": false,
     "extra_data": {},
-    "dependencies": [],
+    "dependencies": [
+      {
+        "purl": "pkg:rpm/[email protected]",
+        "extracted_requirement": "0.3.0",
+        "scope": "dependencies",
+        "is_runtime": true,
+        "is_optional": false,
+        "is_pinned": true,
+        "is_direct": true,
+        "resolved_package": {},
+        "extra_data": {}
+      }      
+    ],
     "repository_homepage_url": null,
     "repository_download_url": null,
     "api_data_url": null,

diff --git a/tests/scancode/data/rpm/fping-2.4-0.b2.rhfc1.dag.i386.rpm.expected.json b/tests/scancode/data/rpm/fping-2.4-0.b2.rhfc1.dag.i386.rpm.expected.json
@@ -82,7 +82,23 @@
       "purl": "pkg:rpm/[email protected]"
     }
   ],
-  "dependencies": [],
+  "dependencies": [
+    {
+      "purl": "pkg:rpm/[email protected]",
+      "extracted_requirement": "2.4",
+      "scope": "dependencies",
+      "is_runtime": true,
+      "is_optional": false,
+      "is_pinned": true,
+      "is_direct": true,
+      "resolved_package": {},
+      "extra_data": {},
+      "dependency_uid": "pkg:rpm/[email protected]?uuid=4c055d32-6a8c-465e-9c76-f2714b7ce23f",
+      "for_package_uid": "pkg:rpm/[email protected]?uuid=1903aca6-54da-49d6-a219-ff432eff4398",
+      "datafile_path": "fping-2.4-0.b2.rhfc1.dag.i386.rpm",
+      "datasource_id": "rpm_archive"
+    }
+  ],
   "files": [
     {
       "path": "fping-2.4-0.b2.rhfc1.dag.i386.rpm",
@@ -158,7 +174,19 @@
           "is_private": false,
           "is_virtual": false,
           "extra_data": {},
-          "dependencies": [],
+          "dependencies": [
+            {
+              "purl": "pkg:rpm/[email protected]",
+              "extracted_requirement": "2.4",
+              "scope": "dependencies",
+              "is_runtime": true,
+              "is_optional": false,
+              "is_pinned": true,
+              "is_direct": true,
+              "resolved_package": {},
+              "extra_data": {}
+            }
+          ],
           "repository_homepage_url": null,
           "repository_download_url": null,
           "api_data_url": null,