diff --git a/src/packagedcode/about.py b/src/packagedcode/about.py
index 3f33c6838db..5a1192fdd28 100644
--- a/src/packagedcode/about.py
+++ b/src/packagedcode/about.py
@@ -47,7 +47,7 @@ class AboutFileHandler(models.DatafileHandler):
     documentation_url = 'https://aboutcode-toolkit.readthedocs.io/en/latest/specification.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location`` pointing to a
         package archive, manifest or similar.
@@ -90,7 +90,7 @@ def parse(cls, location):
             file_references.append(models.FileReference(path=about_resource))
 
         # FIXME: we should put the unprocessed attributes in extra data
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=package_type,
             namespace=package_ns,
@@ -103,6 +103,7 @@ def parse(cls, location):
             download_url=download_url,
             file_references=file_references,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assemble(cls, package_data, resource, codebase, package_adder):
diff --git a/src/packagedcode/alpine.py b/src/packagedcode/alpine.py
index 59223cf72da..1b155a90c97 100644
--- a/src/packagedcode/alpine.py
+++ b/src/packagedcode/alpine.py
@@ -63,11 +63,12 @@ class AlpineInstalledDatabaseHandler(models.DatafileHandler):
     description = 'Alpine Linux installed package database'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield from parse_alpine_installed_db(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
     @classmethod
@@ -134,9 +135,14 @@ class AlpineApkbuildHandler(models.DatafileHandler):
     documentation_url = 'https://wiki.alpinelinux.org/wiki/APKBUILD_Reference'
 
     @classmethod
-    def parse(cls, location):
-        package_data = parse_apkbuild(location, strict=True)
-        cls.populate_license_fields(package_data)
+    def parse(cls, location, package_only=False):
+        package_data = parse_apkbuild(
+            location=location,
+            strict=True,
+            package_only=package_only
+        )
+        if not package_only:
+            cls.populate_license_fields(package_data)
         if package_data:
             yield package_data
 
@@ -165,7 +171,7 @@ def assign_package_to_resources(cls, package, resource, codebase, package_adder)
         )
 
 
-def parse_alpine_installed_db(location, datasource_id, package_type):
+def parse_alpine_installed_db(location, datasource_id, package_type, package_only=False):
     """
     Yield PackageData objects from an installed database file at `location`
     or None. Typically found at '/lib/apk/db/installed' in an Alpine
@@ -179,6 +185,7 @@ def parse_alpine_installed_db(location, datasource_id, package_type):
             package_fields=package_fields,
             datasource_id=datasource_id,
             package_type=package_type,
+            package_only=package_only,
         )
 
 
@@ -241,7 +248,7 @@ def get_alpine_installed_db_fields(location):
 ])
 
 
-def parse_apkbuild(location, strict=False):
+def parse_apkbuild(location, strict=False, package_only=False):
     """
     Return a PackageData object from an APKBUILD file at ``location`` or None.
 
@@ -256,6 +263,7 @@ def parse_apkbuild(location, strict=False):
         datasource_id=AlpineApkbuildHandler.datasource_id,
         package_type=AlpineApkbuildHandler.default_package_type,
         strict=strict,
+        package_only=package_only,
     )
 
 
@@ -732,7 +740,7 @@ def fix_apkbuild(text):
     return text
 
 
-def parse_apkbuild_text(text, datasource_id, package_type, strict=False):
+def parse_apkbuild_text(text, datasource_id, package_type, strict=False, package_only=False):
     """
     Return a PackageData object from an APKBUILD text context or None. Only
     consider variables with a name listed in the ``names`` set.
@@ -761,7 +769,8 @@ def parse_apkbuild_text(text, datasource_id, package_type, strict=False):
     package = build_package_data(
         variables,
         datasource_id=datasource_id,
-        package_type=package_type
+        package_type=package_type,
+        package_only=package_only,
     )
 
     if package and unresolved:
@@ -800,7 +809,7 @@ def parse_pkginfo(location):
     raise NotImplementedError
 
 
-def build_package_data(package_fields, datasource_id, package_type):
+def build_package_data(package_fields, datasource_id, package_type, package_only=False):
     """
     Return a PackageData object from a ``package_fields`` iterable of (name,
     value) tuples.
@@ -850,7 +859,16 @@ def build_package_data(package_fields, datasource_id, package_type):
 
             converted_fields.update(converted)
 
-    return models.PackageData.from_dict(converted_fields)
+    fields_not_required = ["current_file", "current_dir"]
+    for field in fields_not_required:
+        value = converted_fields.get(field)
+        if value:
+            converted_fields.pop(field)
+
+    return models.PackageData.from_data(
+        package_data=converted_fields,
+        package_only=package_only,
+    )
 
 #####################################
 # Note: all handlers MUST accept **kwargs as they also receive the current data
diff --git a/src/packagedcode/build.py b/src/packagedcode/build.py
index 7657f7c64b3..dc7a9cae742 100644
--- a/src/packagedcode/build.py
+++ b/src/packagedcode/build.py
@@ -55,7 +55,7 @@ class AutotoolsConfigureHandler(models.DatafileHandler):
     documentation_url = 'https://www.gnu.org/software/automake/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # we use the parent directory as a package name
         name = fileutils.file_name(fileutils.parent_directory(location))
         # we could use checksums as version in the future
@@ -67,12 +67,13 @@ def parse(cls, location):
         # there are dependencies we could use
         # dependencies = []
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
             version=version,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assign_package_to_resources(cls, package, resource, codebase, package_adder):
@@ -112,6 +113,7 @@ def assemble(cls, package_data, resource, codebase, package_adder):
             package = models.Package.from_package_data(
                 package_data=package_data,
                 datafile_path=resource.path,
+                package_only=True,
             )
 
             if TRACE:
@@ -143,8 +145,7 @@ def assemble(cls, package_data, resource, codebase, package_adder):
         yield resource
 
     @classmethod
-    def parse(cls, location):
-
+    def parse(cls, location, package_only=False):
         # Thanks to Starlark being a Python dialect, we can use `ast` to parse it
         with open(location, 'rb') as f:
             tree = ast.parse(f.read())
@@ -196,23 +197,28 @@ def parse(cls, location):
                     if TRACE:
                         logger_debug(f"build: parse: license_files: {license_files}")
 
-                    package_data = models.PackageData(
+                    package_data = dict(
                         datasource_id=cls.datasource_id,
                         type=cls.default_package_type,
                         name=name,
+                        extracted_license_statement=license_files,
+                    )
+                    # `package_only` is True as we do the license detection
+                    # on assembly
+                    yield models.PackageData.from_data(
+                        package_data=package_data,
+                        package_only=True,
                     )
-
-                    package_data.extracted_license_statement = license_files
-                    yield package_data
 
         else:
             # If we don't find anything in the pkgdata file, we yield a Package
             # with the parent directory as the name
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 type=cls.default_package_type,
                 name=fileutils.file_name(fileutils.parent_directory(location))
             )
+            yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assign_package_to_resources(cls, package, resource, codebase, package_adder, skip_name=None):
@@ -334,7 +340,7 @@ class BuckMetadataBzlHandler(BaseStarlarkManifestHandler):
     documentation_url = 'https://buck.build/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=True):
 
         with open(location, 'rb') as f:
             tree = ast.parse(f.read())
@@ -386,7 +392,7 @@ def parse(cls, location):
         ):
             # TODO: Create function that determines package type from download URL,
             # then create a package of that package type from the metadata info
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 type=metadata_fields.get('upstream_type', cls.default_package_type),
                 name=metadata_fields.get('name'),
@@ -396,6 +402,7 @@ def parse(cls, location):
                 homepage_url=metadata_fields.get('upstream_address', ''),
                 # TODO: Store 'upstream_hash` somewhere
             )
+            yield models.PackageData.from_data(package_data, package_only=True)
 
         if (
             'package_type'
@@ -409,7 +416,7 @@ def parse(cls, location):
             and 'vcs_commit_hash'
             in metadata_fields
         ):
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 type=metadata_fields.get('package_type', cls.default_package_type),
                 name=metadata_fields.get('name'),
@@ -422,6 +429,7 @@ def parse(cls, location):
                 sha1=metadata_fields.get('download_archive_sha1', ''),
                 extra_data=dict(vcs_commit_hash=metadata_fields.get('vcs_commit_hash', ''))
             )
+            yield models.PackageData.from_data(package_data, package_only=True)
 
     @classmethod
     def assign_package_to_resources(cls, package, resource, codebase, package_adder):
diff --git a/src/packagedcode/build_gradle.py b/src/packagedcode/build_gradle.py
index 308abe5bc62..9c4882ccf04 100644
--- a/src/packagedcode/build_gradle.py
+++ b/src/packagedcode/build_gradle.py
@@ -59,9 +59,9 @@ class BuildGradleHandler(models.DatafileHandler):
     description = 'Gradle build script'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         dependencies = get_dependencies(location)
-        return build_package(cls, dependencies)
+        return build_package(cls, dependencies, package_only)
 
     # TODO: handle complex cases of nested builds with many packages
     @classmethod
@@ -328,7 +328,7 @@ def get_dependencies(build_gradle_location):
     return list(get_dependencies_from_parse_tree(parse_tree))
 
 
-def build_package(cls, dependencies):
+def build_package(cls, dependencies, package_only=False):
     """
     Yield PackageData from a ``dependencies`` list of mappings.
     """
@@ -364,10 +364,11 @@ def build_package(cls, dependencies):
             )
         )
 
-    yield models.PackageData(
+    package_data = dict(
         datasource_id=cls.datasource_id,
         type=cls.default_package_type,
         primary_language=BuildGradleHandler.default_primary_language,
         dependencies=package_dependencies,
     )
+    yield models.PackageData.from_data(package_data, package_only)
 
diff --git a/src/packagedcode/cargo.py b/src/packagedcode/cargo.py
index ffe8dea5159..ca46e388383 100644
--- a/src/packagedcode/cargo.py
+++ b/src/packagedcode/cargo.py
@@ -29,7 +29,7 @@ class CargoTomlHandler(models.DatafileHandler):
     documentation_url = 'https://doc.rust-lang.org/cargo/reference/manifest.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         package_data = toml.load(location, _dict=dict)
 
         core_package_data = package_data.get('package', {})
@@ -67,7 +67,7 @@ def parse(cls, location):
         repository_download_url = name and version and f'https://crates.io/api/v1/crates/{name}/{version}/download'
         api_data_url = name and f'https://crates.io/api/v1/crates/{name}'
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -83,6 +83,7 @@ def parse(cls, location):
             api_data_url=api_data_url,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assemble(cls, package_data, resource, codebase, package_adder):
@@ -116,7 +117,7 @@ class CargoLockHandler(models.DatafileHandler):
     # ]
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         cargo_lock = toml.load(location, _dict=dict)
         dependencies = []
         package = cargo_lock.get('package', [])
@@ -137,12 +138,13 @@ def parse(cls, location):
                 )
             )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assemble(cls, package_data, resource, codebase, package_adder):
diff --git a/src/packagedcode/chef.py b/src/packagedcode/chef.py
index 5c378cdce4e..d5df6a2f6c4 100644
--- a/src/packagedcode/chef.py
+++ b/src/packagedcode/chef.py
@@ -183,14 +183,18 @@ def is_datafile(cls, location, filetypes=tuple()):
             return not parent.endswith('dist-info')
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location``
         pointing to a package archive, manifest or similar.
         """
         with io.open(location, encoding='utf-8') as loc:
             package_data = json.load(loc)
-        yield build_package(package_data, datasource_id=cls.datasource_id)
+        yield build_package(
+            package_data=package_data,
+            datasource_id=cls.datasource_id,
+            package_only=package_only,
+        )
 
 
 class ChefMetadataRbHandler(BaseChefMetadataHandler):
@@ -202,7 +206,7 @@ class ChefMetadataRbHandler(BaseChefMetadataHandler):
     documentation_url = 'https://docs.chef.io/config_rb_metadata/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with io.open(location, encoding='utf-8') as loc:
             file_contents = loc.read()
 
@@ -213,10 +217,14 @@ def parse(cls, location):
             ChefMetadataFormatter()
         )
         package_data = json.loads(formatted_file_contents)
-        yield build_package(package_data, datasource_id=cls.datasource_id)
+        yield build_package(
+            package_data=package_data,
+            datasource_id=cls.datasource_id,
+            package_only=package_only,
+        )
 
 
-def build_package(package_data, datasource_id):
+def build_package(package_data, datasource_id, package_only=False):
     """
     Return a PackageData object from a package_data mapping from a metadata.json
     or similar or None.
@@ -261,7 +269,7 @@ def build_package(package_data, datasource_id):
             )
         )
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=ChefMetadataJsonHandler.default_package_type,
         name=name,
@@ -275,3 +283,4 @@ def build_package(package_data, datasource_id):
         primary_language='Ruby',
         **get_urls(name, version),
     )
+    return models.PackageData.from_data(package_data, package_only)
diff --git a/src/packagedcode/cocoapods.py b/src/packagedcode/cocoapods.py
index 9773ae13638..8ca3617800b 100644
--- a/src/packagedcode/cocoapods.py
+++ b/src/packagedcode/cocoapods.py
@@ -216,7 +216,7 @@ class PodspecHandler(BasePodHandler):
     documentation_url = 'https://guides.cocoapods.org/syntax/podspec.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location``
         pointing to a package archive, manifest or similar.
@@ -258,7 +258,7 @@ def parse(cls, location):
             homepage_url=homepage_url,
             vcs_url=vcs_url)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -273,6 +273,7 @@ def parse(cls, location):
             parties=parties,
             **urls,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class PodfileHandler(PodspecHandler):
@@ -293,7 +294,7 @@ class PodfileLockHandler(BasePodHandler):
     documentation_url = 'https://guides.cocoapods.org/using/the-podfile.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield PackageData from a YAML Podfile.lock.
         """
@@ -337,12 +338,13 @@ def parse(cls, location):
                     )
                 )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class PodspecJsonHandler(models.DatafileHandler):
@@ -354,7 +356,7 @@ class PodspecJsonHandler(models.DatafileHandler):
     documentation_url = 'https://guides.cocoapods.org/syntax/podspec.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location) as psj:
             data = json.load(psj)
 
@@ -423,7 +425,7 @@ def parse(cls, location):
             name=name,
             version=version, homepage_url=homepage_url, vcs_url=vcs_url)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             primary_language=cls.default_primary_language,
             type=cls.default_package_type,
@@ -437,6 +439,7 @@ def parse(cls, location):
             download_url=download_url,
             **urls,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def get_urls(name=None, version=None, homepage_url=None, vcs_url=None, **kwargs):
diff --git a/src/packagedcode/conan.py b/src/packagedcode/conan.py
index 7d4b637f9bc..44321c581a7 100644
--- a/src/packagedcode/conan.py
+++ b/src/packagedcode/conan.py
@@ -123,7 +123,7 @@ class ConanFileHandler(models.DatafileHandler):
     documentation_url = "https://docs.conan.io/2.0/reference/conanfile.html"
 
     @classmethod
-    def _parse(cls, conan_recipe):
+    def _parse(cls, conan_recipe, package_only=False):
         try:
             tree = ast.parse(conan_recipe)
             recipe_class_def = next(
@@ -150,7 +150,7 @@ def _parse(cls, conan_recipe):
 
         dependencies = get_dependencies(parser.requires)
 
-        return models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -163,13 +163,14 @@ def _parse(cls, conan_recipe):
             extracted_license_statement=parser.license,
             dependencies=dependencies,
         )
+        return models.PackageData.from_data(package_data, package_only)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with io.open(location, encoding="utf-8") as loc:
             conan_recipe = loc.read()
 
-        yield cls._parse(conan_recipe)
+        yield cls._parse(conan_recipe, package_only)
 
 
 class ConanDataHandler(models.DatafileHandler):
@@ -184,7 +185,7 @@ class ConanDataHandler(models.DatafileHandler):
     )
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with io.open(location, encoding="utf-8") as loc:
             conan_data = loc.read()
 
@@ -203,7 +204,7 @@ def parse(cls, location):
             elif isinstance(source_urls, list):
                 url = source_urls[0]
 
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 type=cls.default_package_type,
                 primary_language=cls.default_primary_language,
@@ -212,6 +213,8 @@ def parse(cls, location):
                 download_url=url,
                 sha256=sha256,
             )
+            yield models.PackageData.from_data(package_data, package_only)
+            
 
     @classmethod
     def assemble(
@@ -245,7 +248,7 @@ def assemble(
                 ] = conanfile_package_data.get("extracted_license_statement")
 
         datafile_path = resource.path
-        pkg_data = models.PackageData.from_dict(package_data_dict)
+        pkg_data = models.PackageData.from_data(package_data_dict)
 
         if pkg_data.purl:
             package = models.Package.from_package_data(
diff --git a/src/packagedcode/conda.py b/src/packagedcode/conda.py
index 545b3a1fb53..225940d69fc 100644
--- a/src/packagedcode/conda.py
+++ b/src/packagedcode/conda.py
@@ -79,7 +79,7 @@ def assign_package_to_resources(cls, package, resource, codebase, package_adder)
         )
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         metayaml = get_meta_yaml_data(location)
         package_element = metayaml.get('package') or {}
         package_name = package_element.get('name')
@@ -118,7 +118,7 @@ def parse(cls, location):
                     )
                 )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=package_name,
@@ -131,6 +131,7 @@ def parse(cls, location):
             extracted_license_statement=extracted_license_statement,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def get_meta_yaml_data(location):
diff --git a/src/packagedcode/cran.py b/src/packagedcode/cran.py
index ef274105da6..827ff38f9d3 100644
--- a/src/packagedcode/cran.py
+++ b/src/packagedcode/cran.py
@@ -30,7 +30,7 @@ class CranDescriptionFileHandler(models.DatafileHandler):
     documentation_url = 'https://r-pkgs.org/description.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         cran_desc = get_cran_description(location)
 
         name = cran_desc.get('Package')
@@ -93,7 +93,7 @@ def parse(cls, location):
         # TODO: Let's handle the release date as a Date type
         # release_date = cran_desc.get('Date/Publication'),
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -105,6 +105,7 @@ def parse(cls, location):
             dependencies=package_dependencies,
             repository_homepage_url=f'https://cran.r-project.org/package={name}',
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 # FIXME: THIS IS NOT YAML but RFC 822
 
diff --git a/src/packagedcode/debian.py b/src/packagedcode/debian.py
index 860144b9723..6b45dc15a92 100644
--- a/src/packagedcode/debian.py
+++ b/src/packagedcode/debian.py
@@ -60,7 +60,7 @@ class DebianDebPackageHandler(models.DatafileHandler):
     documentation_url = 'https://manpages.debian.org/unstable/dpkg-dev/deb.5.en.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield build_package_data_from_package_filename(
             filename=fileutils.file_name(location),
             datasource_id=cls.datasource_id,
@@ -83,7 +83,7 @@ class DebianSourcePackageMetadataTarballHandler(models.DatafileHandler):
     documentation_url = 'https://manpages.debian.org/unstable/dpkg-dev/deb.5.en.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # strip extension
         filename, _, _ = location.rpartition('.tar')
         yield build_package_data_from_package_filename(
@@ -108,7 +108,7 @@ class DebianSourcePackageTarballHandler(models.DatafileHandler):
     documentation_url = 'https://manpages.debian.org/unstable/dpkg-dev/deb.5.en.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # strip extension
         filename, _, _ = location.rpartition('.tar')
         yield build_package_data_from_package_filename(
@@ -132,13 +132,14 @@ class DebianControlFileInExtractedDebHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/debian-policy/ch-controlfields.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # TODO: we cannot know the distro from the name only
         yield build_package_data(
             debian_data=get_paragraph_data_from_file(location=location),
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
             distro='debian',
+            package_only=package_only,
         )
 
     @classmethod
@@ -158,7 +159,7 @@ class DebianControlFileInSourceHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/debian-policy/ch-controlfields.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # NOTE: a control file in a source repo or debina.tar tarball can contain more than one package
         debian_packages = []
         for debian_data in get_paragraphs_data_from_file(location=location):
@@ -167,6 +168,7 @@ def parse(cls, location):
                     debian_data=debian_data,
                     datasource_id=cls.datasource_id,
                     package_type=cls.default_package_type,
+                    package_only=package_only,
                 )
             )
 
@@ -191,7 +193,7 @@ class DebianDscFileHandler(models.DatafileHandler):
     documentation_url = 'https://wiki.debian.org/dsc'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # this is typically signed
         debian_data = get_paragraph_data_from_file(
             location=location,
@@ -207,6 +209,7 @@ def parse(cls, location):
             debian_data=debian_data,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
         package_data.update_purl_fields(package_data=package_data_from_file)
         yield package_data
@@ -225,7 +228,7 @@ class DebianInstalledStatusDatabaseHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/debian-policy/ch-controlfields.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # note that we do not know yet the distro at this stage
         # we could get it... but we get that later during assemble()
         debian_packages = []
@@ -235,6 +238,7 @@ def parse(cls, location):
                     debian_data=debian_data,
                     datasource_id=cls.datasource_id,
                     package_type=cls.default_package_type,
+                    package_only=package_only,
                 )
             )
 
@@ -392,7 +396,7 @@ class DebianDistrolessInstalledDatabaseHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/debian-policy/ch-controlfields.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield installed PackageData objects given a ``location``
         var/lib/dpkg/status.d/<status> file as found in a distroless container
@@ -406,6 +410,7 @@ def parse(cls, location):
                     debian_data=debian_data,
                     datasource_id=cls.datasource_id,
                     package_type=cls.default_package_type,
+                    package_only=package_only,
                 )
             )
 
@@ -475,7 +480,7 @@ class DebianInstalledFilelistHandler(models.DatafileHandler):
     description = 'Debian installed file paths list'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         return parse_debian_files_list(
             location=location,
             datasource_id=cls.datasource_id,
@@ -501,7 +506,7 @@ class DebianInstalledMd5sumFilelistHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/manuals/debian-handbook/sect.package-meta-information.en.html#sect.configuration-scripts'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         return parse_debian_files_list(
             location=location,
             datasource_id=cls.datasource_id,
@@ -526,7 +531,7 @@ class DebianMd5sumFilelistInPackageHandler(models.DatafileHandler):
     documentation_url = 'https://www.debian.org/doc/manuals/debian-handbook/sect.package-meta-information.en.html#sect.configuration-scripts'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         return parse_debian_files_list(
             location=location,
             datasource_id=cls.datasource_id,
@@ -561,7 +566,7 @@ def build_package_data_from_package_filename(filename, datasource_id, package_ty
     if isinstance(version, DebVersion):
         version = str(version)
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         name=deb.name,
@@ -569,6 +574,7 @@ def build_package_data_from_package_filename(filename, datasource_id, package_ty
         version=version,
         qualifiers=qualifiers,
     )
+    return models.PackageData.from_data(package_data)
 
 
 def parse_debian_files_list(location, datasource_id, package_type):
@@ -606,16 +612,17 @@ def parse_debian_files_list(location, datasource_id, package_type):
     if not file_references:
         return
 
-    yield models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         name=name,
         qualifiers=qualifiers,
         file_references=file_references,
     )
+    yield models.PackageData.from_data(package_data) 
 
 
-def build_package_data(debian_data, datasource_id, package_type='deb', distro=None):
+def build_package_data(debian_data, datasource_id, package_type='deb', distro=None, package_only=False):
     """
     Return a PackageData object from a package_data mapping (from a dpkg status
     or similar file) or None.
@@ -691,7 +698,7 @@ def build_package_data(debian_data, datasource_id, package_type='deb', distro=No
 
         source_packages.append(source_pkg_purl)
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         namespace=distro,
@@ -706,6 +713,7 @@ def build_package_data(debian_data, datasource_id, package_type='deb', distro=No
         parties=parties,
         extra_data=extra_data,
     )
+    return models.PackageData.from_data(package_data, package_only)
 
 
 def populate_debian_namespace(packages):
diff --git a/src/packagedcode/debian_copyright.py b/src/packagedcode/debian_copyright.py
index 510abf620d6..7f08592fe8d 100644
--- a/src/packagedcode/debian_copyright.py
+++ b/src/packagedcode/debian_copyright.py
@@ -95,7 +95,7 @@ def is_datafile(cls, location, filetypes=tuple(), strict=False):
                 return True
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         debian_copyright = parse_copyright_file(location)
         license_fields = DebianLicenseFields.get_license_fields(
             debian_copyright=debian_copyright
@@ -111,20 +111,27 @@ def parse(cls, location):
             # no name otherwise for now
             name = None
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
-            extracted_license_statement=license_fields.extracted_license_statement,
-            declared_license_expression=license_fields.declared_license_expression,
-            declared_license_expression_spdx=license_fields.declared_license_expression_spdx,
-            license_detections=license_fields.license_detections,
-            other_license_expression=license_fields.other_license_expression,
-            other_license_expression_spdx=license_fields.other_license_expression_spdx,
-            other_license_detections=license_fields.other_license_detections,
-            copyright=debian_copyright.get_copyright(),
         )
 
+        if not package_only:
+            license_data = dict(
+                extracted_license_statement=license_fields.extracted_license_statement,
+                declared_license_expression=license_fields.declared_license_expression,
+                declared_license_expression_spdx=license_fields.declared_license_expression_spdx,
+                license_detections=license_fields.license_detections,
+                other_license_expression=license_fields.other_license_expression,
+                other_license_expression_spdx=license_fields.other_license_expression_spdx,
+                other_license_detections=license_fields.other_license_detections,
+                copyright=debian_copyright.get_copyright(),
+            )
+            package_data.update(license_data)
+
+        yield models.PackageData.from_data(package_data, package_only)
+
 
 @attr.s
 class DebianLicenseFields:
@@ -279,14 +286,14 @@ def assemble(cls, package_data, resource, codebase, package_adder):
         yield from super().assemble(package_data, resource, codebase, package_adder)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Gets license/copyright information from file like
         other copyright files, but also gets purl fields if
         present in copyright filename, if obtained from
         upstream metadata archive.
         """
-        package_data = list(super().parse(location)).pop()
+        package_data = list(super().parse(location, package_only)).pop()
         package_data_from_file = build_package_data_from_metadata_filename(
             filename=os.path.basename(location),
             datasource_id=cls.datasource_id,
diff --git a/src/packagedcode/distro.py b/src/packagedcode/distro.py
index 6336e6b3111..7e572948a03 100644
--- a/src/packagedcode/distro.py
+++ b/src/packagedcode/distro.py
@@ -25,7 +25,7 @@ class EtcOsReleaseHandler(models.NonAssemblableDatafileHandler):
     documentation_url = 'https://www.freedesktop.org/software/systemd/man/os-release.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         distro = Distro.from_os_release_file(location)
         distro_identifier = distro.identifier
         pretty_name = distro.pretty_name and distro.pretty_name.lower() or ''
@@ -53,13 +53,14 @@ def parse(cls, location):
 
         version = distro.version_id
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             namespace=namespace,
             name=name,
             version=version,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def find_linux_rootfs_root_resource(cls, resource, codebase):
diff --git a/src/packagedcode/freebsd.py b/src/packagedcode/freebsd.py
index c1f5342b562..d170e8f8222 100644
--- a/src/packagedcode/freebsd.py
+++ b/src/packagedcode/freebsd.py
@@ -52,7 +52,7 @@ class CompactManifestHandler(models.DatafileHandler):
     documentation_url = 'https://www.freebsd.org/cgi/man.cgi?pkg-create(8)#MANIFEST_FILE_DETAILS'
 
     @classmethod
-    def _parse(cls, yaml_data):
+    def _parse(cls, yaml_data, package_only=False):
         package_data = models.PackageData(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
@@ -97,7 +97,8 @@ def _parse(cls, yaml_data):
         # license_mapper needs multiple fields
         license_mapper(yaml_data, package_data)
 
-        cls.populate_license_fields(package_data)
+        if not package_only:
+            cls.populate_license_fields(package_data)
 
         if TRACE:
             logger_debug(
@@ -107,7 +108,7 @@ def _parse(cls, yaml_data):
         return package_data
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location`` pointing to a
         package archive, manifest or similar.
@@ -115,7 +116,7 @@ def parse(cls, location):
         with io.open(location, encoding='utf-8') as loc:
             yaml_data = saneyaml.load(loc)
 
-        yield cls._parse(yaml_data)
+        yield cls._parse(yaml_data, package_only)
 
     @staticmethod
     def get_license_detections_and_expression(package_data):
diff --git a/src/packagedcode/godeps.py b/src/packagedcode/godeps.py
index 68ff54787ff..c9d15a3934a 100644
--- a/src/packagedcode/godeps.py
+++ b/src/packagedcode/godeps.py
@@ -38,7 +38,7 @@ class GodepsHandler(models.NonAssemblableDatafileHandler):
     documentation_url = 'https://github.com/tools/godep'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         godeps = Godep(location)
 
         if godeps.import_path:
@@ -64,7 +64,7 @@ def parse(cls, location):
                 )
             )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             namespace=namespace,
@@ -72,6 +72,7 @@ def parse(cls, location):
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
          )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assign_package_to_resources(cls, package, resource, codebase, package_adder):
diff --git a/src/packagedcode/golang.py b/src/packagedcode/golang.py
index 6075c713e86..68651fa6cb1 100644
--- a/src/packagedcode/golang.py
+++ b/src/packagedcode/golang.py
@@ -49,7 +49,7 @@ class GoModHandler(BaseGoModuleHandler):
     documentation_url = 'https://go.dev/ref/mod'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         gomods = go_mod.parse_gomod(location)
 
         dependencies = []
@@ -89,7 +89,7 @@ def parse(cls, location):
         if namespace and name:
             repository_homepage_url = f'https://pkg.go.dev/{namespace}/{name}'
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -100,6 +100,7 @@ def parse(cls, location):
             dependencies=dependencies,
             primary_language=cls.default_primary_language,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class GoSumHandler(BaseGoModuleHandler):
@@ -111,7 +112,7 @@ class GoSumHandler(BaseGoModuleHandler):
     documentation_url = 'https://go.dev/ref/mod#go-sum-files'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         gosums = go_mod.parse_gosum(location)
         package_dependencies = []
         for gosum in gosums:
@@ -126,9 +127,10 @@ def parse(cls, location):
                 )
             )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             dependencies=package_dependencies,
             primary_language=cls.default_primary_language,
         )
+        yield models.PackageData.from_data(package_data, package_only)
diff --git a/src/packagedcode/haxe.py b/src/packagedcode/haxe.py
index 99f5be4ef7d..630d4e36175 100644
--- a/src/packagedcode/haxe.py
+++ b/src/packagedcode/haxe.py
@@ -45,11 +45,11 @@ class HaxelibJsonHandler(models.DatafileHandler):
     documentation_url = 'https://lib.haxe.org/documentation/creating-a-haxelib-package/'
 
     @classmethod
-    def _parse(cls, json_data):
+    def _parse(cls, json_data, package_only=False):
         name = json_data.get('name')
         version = json_data.get('version')
 
-        package_data = models.PackageData(
+        package_mapping = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -60,6 +60,7 @@ def _parse(cls, json_data):
             description=json_data.get('description'),
             primary_language=cls.default_primary_language,
         )
+        package_data = models.PackageData.from_data(package_mapping, package_only)
 
         if name and version:
             download_url = f'https://lib.haxe.org/p/{name}/{version}/download/'
@@ -91,7 +92,7 @@ def _parse(cls, json_data):
         return package_data
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location`` pointing to a
         package_data archive, manifest or similar.
@@ -111,4 +112,4 @@ def parse(cls, location):
         with io.open(location, encoding='utf-8') as loc:
             json_data = json.load(loc)
 
-        yield cls._parse(json_data)
+        yield cls._parse(json_data, package_only)
diff --git a/src/packagedcode/maven.py b/src/packagedcode/maven.py
index b5d521dc44a..1498add8cf0 100644
--- a/src/packagedcode/maven.py
+++ b/src/packagedcode/maven.py
@@ -132,13 +132,14 @@ class JavaJarManifestHandler(MavenBasePackageHandler):
     documentation_url = 'https://docs.oracle.com/javase/tutorial/deployment/jar/manifestindex.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         sections = parse_manifest(location)
         if sections:
             main_section = sections[0]
             manifest = get_normalized_java_manifest_data(main_section)
             if manifest:
-                yield models.PackageData(**manifest,)
+                package_data = dict(**manifest,)
+                yield models.PackageData.from_data(package_data, package_only)
 
 
 class JavaJarManifestHandlerMixin(models.DatafileHandler):
@@ -206,13 +207,14 @@ def is_datafile(cls, location, filetypes=tuple()):
                     return True
 
     @classmethod
-    def parse(cls, location, base_url='https://repo1.maven.org/maven2'):
+    def parse(cls, location, package_only=False, base_url='https://repo1.maven.org/maven2'):
         package_data = parse(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             base_url=base_url,
+            package_only=package_only,
         )
         if package_data:
             yield package_data
@@ -303,7 +305,7 @@ class MavenPomPropertiesHandler(models.NonAssemblableDatafileHandler):
     documentation_url = 'https://maven.apache.org/pom.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield PackageData from a pom.properties file (which is typically side-
         by-side with its pom file.)
@@ -313,10 +315,10 @@ def parse(cls, location):
             if TRACE:
                 logger.debug(f'MavenPomPropertiesHandler.parse: properties: {properties!r}')
             if properties:
-                yield from cls.parse_pom_properties(properties=properties) 
+                yield from cls.parse_pom_properties(properties=properties, package_only=package_only) 
 
     @classmethod
-    def parse_pom_properties(cls, properties):
+    def parse_pom_properties(cls, properties, package_only=False):
         namespace = properties.pop("groupId", None)
         name = properties.pop("artifactId", None)
         version = properties.pop("version", None)
@@ -325,7 +327,7 @@ def parse_pom_properties(cls, properties):
         else:
             extra_data = {}
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -334,6 +336,7 @@ def parse_pom_properties(cls, properties):
             version=version,
             extra_data=extra_data,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def build_url(
@@ -1189,6 +1192,7 @@ def parse(
     package_type,
     primary_language,
     base_url='https://repo1.maven.org/maven2',
+    package_only=False,
 ):
     """
     Return Packagedata objects from parsing a Maven pom file at `location` or
@@ -1199,7 +1203,8 @@ def parse(
         package_type=package_type,
         primary_language=primary_language,
         location=location,
-        base_url=base_url
+        base_url=base_url,
+        package_only=package_only,
     )
     if package:
         return package
@@ -1212,6 +1217,7 @@ def _parse(
     location=None,
     text=None,
     base_url='https://repo1.maven.org/maven2',
+    package_only=False,
 ):
     """
     Yield Packagedata objects from parsing a Maven pom file at `location` or
@@ -1283,7 +1289,7 @@ def _parse(
     ))
 
     # FIXME: there are still other data to map in a PackageData
-    return MavenPackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         primary_language=primary_language,
@@ -1300,6 +1306,7 @@ def _parse(
         bug_tracking_url=bug_tracking_url,
         **urls,
     )
+    return MavenPackageData.from_data(package_data, package_only)
 
 class MavenPackageData(models.PackageData):
 
diff --git a/src/packagedcode/models.py b/src/packagedcode/models.py
index a30a79af3c1..1b65497e590 100644
--- a/src/packagedcode/models.py
+++ b/src/packagedcode/models.py
@@ -728,6 +728,9 @@ def from_data(cls, package_data, package_only=False):
 
         Skip the license/copyright detection step if `package_only` is True.
         """
+        if "purl" in package_data:
+            package_data.pop("purl")
+
         package_data = cls(**package_data)
 
         if not package_only:
@@ -1048,13 +1051,16 @@ def is_datafile(cls, location, filetypes=tuple(), _bare_filename=False):
                     return any(ft in actual_type for ft in filetypes)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more PackageData objects given a package data file at
         ``location``.
 
         Subclasses must implement and are responsible for returning proper
         computed license fields and list of resources and files.
+
+        If `package_only`, skip the license/copyright detection on extracted
+        license/copyright data.
         """
         raise NotImplementedError
 
@@ -1515,8 +1521,6 @@ def __attrs_post_init__(self, *args, **kwargs):
         if not self.package_uid:
             self.package_uid = build_package_uid(self.purl)
 
-        self.populate_license_fields()
-
     def to_dict(self):
         return super().to_dict(with_details=False)
 
@@ -1528,7 +1532,7 @@ def to_package_data(self):
         return PackageData.from_dict(mapping)
 
     @classmethod
-    def from_package_data(cls, package_data, datafile_path):
+    def from_package_data(cls, package_data, datafile_path, package_only=False):
         """
         Return a Package from a ``package_data`` PackageData object
         or mapping. Or None.
@@ -1552,7 +1556,16 @@ def from_package_data(cls, package_data, datafile_path):
                 if not license_match['from_file']:
                     license_match['from_file'] = datafile_path
 
-        return cls.from_dict(package_data_mapping)
+        package = cls.from_dict(package_data_mapping)
+        
+        if not package.package_uid:
+            package.package_uid = build_package_uid(package.purl)
+        
+        if not package_only:
+            package.populate_license_fields()
+            package.populate_holder_field()
+
+        return package
 
     @classmethod
     def from_dict(cls, mapping):
diff --git a/src/packagedcode/msi.py b/src/packagedcode/msi.py
index 31deea3040b..5ce1ba534ed 100644
--- a/src/packagedcode/msi.py
+++ b/src/packagedcode/msi.py
@@ -124,6 +124,7 @@ def create_package_data_from_msiinfo_results(
     msiinfo_results,
     datasource_id='msi_installer',
     package_type='msi',
+    package_only=False,
 ):
     """
     Return PackageData from a mapping of `msiinfo_results`
@@ -150,7 +151,7 @@ def create_package_data_from_msiinfo_results(
     description = msiinfo_results.pop('Comments', '')
     keywords = msiinfo_results.pop('Keywords', [])
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         name=name,
@@ -160,11 +161,14 @@ def create_package_data_from_msiinfo_results(
         keywords=keywords,
         extra_data=msiinfo_results
     )
+    return models.PackageData.from_data(package_data, package_only)
 
 
-def msi_parse(location,
+def msi_parse(
+    location,
     datasource_id='msi_installer',
     package_type='msi',
+    package_only=False,
 ):
     """
     Return PackageData from ``location``
@@ -175,6 +179,7 @@ def msi_parse(location,
             msiinfo_results=info,
             datasource_id=datasource_id,
             package_type=package_type,
+            package_only=package_only,
         )
     else:
         return models.PackageData(
@@ -192,5 +197,5 @@ class MsiInstallerHandler(models.DatafileHandler):
     documentation_url = 'https://docs.microsoft.com/en-us/windows/win32/msi/windows-installer-portal'
 
     @classmethod
-    def parse(cls, location):
-        yield msi_parse(location)
+    def parse(cls, location, package_only=False):
+        yield msi_parse(location, package_only)
diff --git a/src/packagedcode/npm.py b/src/packagedcode/npm.py
index c831aa8e855..18ad27d7564 100644
--- a/src/packagedcode/npm.py
+++ b/src/packagedcode/npm.py
@@ -185,7 +185,7 @@ class NpmPackageJsonHandler(BaseNpmHandler):
     documentation_url = 'https://docs.npmjs.com/cli/v8/configuring-npm/package-json'
 
     @classmethod
-    def _parse(cls, json_data):
+    def _parse(cls, json_data, package_only=False):
         name = json_data.get('name')
         version = json_data.get('version')
         homepage_url = json_data.get('homepage', '')
@@ -200,7 +200,7 @@ def _parse(cls, json_data):
         namespace, name = split_scoped_package_name(name)
 
         urls = get_urls(namespace, name, version)
-        package = models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -211,6 +211,7 @@ def _parse(cls, json_data):
             homepage_url=homepage_url,
             **urls,
         )
+        package = models.PackageData.from_data(package_data, package_only)
         vcs_revision = json_data.get('gitHead') or None
 
         # mapping of top level package.json items to a function accepting as
@@ -249,7 +250,8 @@ def _parse(cls, json_data):
         lics = json_data.get('licenses')
         package = licenses_mapper(lic, lics, package)
 
-        package.populate_license_fields()
+        if not package_only:
+            package.populate_license_fields()
 
         if TRACE:
             logger_debug(f'NpmPackageJsonHandler: parse: package: {package.to_dict()}')
@@ -257,17 +259,17 @@ def _parse(cls, json_data):
         return package
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with io.open(location, encoding='utf-8') as loc:
             json_data = json.load(loc)
 
-        yield cls._parse(json_data)
+        yield cls._parse(json_data, package_only)
 
 
 class BaseNpmLockHandler(BaseNpmHandler):
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
 
         with io.open(location, encoding='utf-8') as loc:
             package_data = json.load(loc)
@@ -280,7 +282,7 @@ def parse(cls, location):
 
         extra_data = dict(lockfile_version=lockfile_version)
         # this is the top level element that we return
-        root_package_data = models.PackageData(
+        root_package_mapping = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -290,6 +292,7 @@ def parse(cls, location):
             extra_data=extra_data,
             **get_urls(root_ns, root_name, root_version)
         )
+        root_package_data = models.PackageData.from_data(root_package_mapping, package_only)
 
         # https://docs.npmjs.com/cli/v8/configuring-npm/package-lock-json#lockfileversion
         if lockfile_version == 1:
@@ -359,7 +362,7 @@ def parse(cls, location):
             integrity = dep_data.get('integrity')
             misc.update(get_algo_hexsum(integrity).items())
 
-            resolved_package = models.PackageData(
+            resolved_package_mapping = dict(
                 datasource_id=cls.datasource_id,
                 type=cls.default_package_type,
                 primary_language=cls.default_primary_language,
@@ -369,6 +372,7 @@ def parse(cls, location):
                 extracted_license_statement=extracted_license_statement,
                 **misc,
             )
+            resolved_package = models.PackageData.from_data(resolved_package_mapping, package_only)
             # these are paths t the root of the installed package in v2
             if dep:
                 resolved_package.file_references = [models.FileReference(path=dep)],
@@ -490,7 +494,7 @@ def is_datafile(cls, location, filetypes=tuple()):
         return super().is_datafile(location, filetypes=filetypes) and is_yarn_v2(location)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Parse a bew yarn.lock v2 YAML format which looks like this:
 
@@ -545,12 +549,13 @@ def parse(cls, location):
                 )
             top_dependencies.append(dependency)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=top_dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class YarnLockV1Handler(BaseNpmHandler):
@@ -569,7 +574,7 @@ def is_datafile(cls, location, filetypes=tuple()):
         return super().is_datafile(location, filetypes=filetypes) and not is_yarn_v2(location)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Parse a classic yarn.lock format which looks like this:
             "@babel/core@^7.1.0", "@babel/core@^7.3.4":
@@ -657,7 +662,7 @@ def parse(cls, location):
             misc.update(get_algo_hexsum(integrity).items())
 
             # we create a resolve package with the details
-            resolved_package_data = models.PackageData(
+            resolved_package_mapping = dict(
                 datasource_id=cls.datasource_id,
                 type=cls.default_package_type,
                 namespace=ns,
@@ -666,6 +671,7 @@ def parse(cls, location):
                 primary_language=cls.default_primary_language,
                 **misc,
             )
+            resolved_package_data = models.PackageData.from_data(resolved_package_mapping, package_only)
 
             # we add the sub-deps to the resolved package
             for subns, subname, subconstraint in sub_dependencies:
@@ -701,12 +707,13 @@ def parse(cls, location):
             )
             dependencies.append(dep)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def get_checksum_and_url(url):
diff --git a/src/packagedcode/nuget.py b/src/packagedcode/nuget.py
index dee1afb39d5..de584038d48 100644
--- a/src/packagedcode/nuget.py
+++ b/src/packagedcode/nuget.py
@@ -106,7 +106,7 @@ class NugetNuspecHandler(models.DatafileHandler):
     documentation_url = 'https://docs.microsoft.com/en-us/nuget/reference/nuspec'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location, 'rb') as loc:
             parsed = xmltodict.parse(loc)
 
@@ -163,7 +163,7 @@ def parse(cls, location):
         elif 'licenseUrl' in nuspec:
             extracted_license_statement = nuspec.get('licenseUrl')
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -177,4 +177,5 @@ def parse(cls, location):
             vcs_url=vcs_url,
             **urls,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
diff --git a/src/packagedcode/opam.py b/src/packagedcode/opam.py
index 94e2f4942d7..f2fde2e017e 100644
--- a/src/packagedcode/opam.py
+++ b/src/packagedcode/opam.py
@@ -31,7 +31,7 @@ def get_package_root(cls, resource, codebase):
         return resource.parent(codebase)
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         opams = parse_opam(location)
 
         package_dependencies = []
@@ -90,7 +90,7 @@ def parse(cls, location):
                 )
             )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -111,6 +111,7 @@ def parse(cls, location):
             repository_homepage_url=repository_homepage_url,
             primary_language=cls.default_primary_language
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def assign_package_to_resources(cls, package, resource, codebase, package_adder):
diff --git a/src/packagedcode/phpcomposer.py b/src/packagedcode/phpcomposer.py
index b729e0bfc25..d217e5283d4 100644
--- a/src/packagedcode/phpcomposer.py
+++ b/src/packagedcode/phpcomposer.py
@@ -58,7 +58,7 @@ class PhpComposerJsonHandler(BasePhpComposerHandler):
     documentation_url = 'https://getcomposer.org/doc/04-schema.md'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         """
         Yield one or more Package manifest objects given a file ``location``
         pointing to a package archive, manifest or similar.
@@ -69,7 +69,7 @@ def parse(cls, location):
         with io.open(location, encoding='utf-8') as loc:
             package_json = json.load(loc)
 
-        yield build_package_data(package_json)
+        yield build_package_data(package_json, package_only)
 
 
 def get_repository_homepage_url(namespace, name):
@@ -86,7 +86,7 @@ def get_api_data_url(namespace, name):
         return f'https://packagist.org/p/packages/{name}.json'
 
 
-def build_package_data(package_data):
+def build_package_data(package_data, package_only=False):
 
     # Note: A composer.json without name and description is not a usable PHP
     # composer package. Name and description fields are required but only for
@@ -103,7 +103,7 @@ def build_package_data(package_data):
     else:
         ns, _, name = ns_name.rpartition('/')
 
-    package = models.PackageData(
+    package_mapping = dict(
         datasource_id=PhpComposerJsonHandler.datasource_id,
         type=PhpComposerJsonHandler.default_package_type,
         namespace=ns,
@@ -112,6 +112,7 @@ def build_package_data(package_data):
         api_data_url=get_api_data_url(ns, name),
         primary_language=PhpComposerJsonHandler.default_primary_language,
     )
+    package = models.PackageData.from_data(package_mapping, package_only)
 
     # mapping of top level composer.json items to the Package object field name
     plain_fields = [
@@ -157,7 +158,9 @@ def build_package_data(package_data):
     vendor_mapper(package)
 
     # Per https://getcomposer.org/doc/04-schema.md#license this is an expression
-    package.populate_license_fields()
+    if not package_only:
+        package.populate_license_fields()
+
     return package
 
 
@@ -170,16 +173,16 @@ class PhpComposerLockHandler(BasePhpComposerHandler):
     documentation_url = 'https://getcomposer.org/doc/01-basic-usage.md#commit-your-composer-lock-file-to-version-control'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with io.open(location, encoding='utf-8') as loc:
             package_data = json.load(loc)
 
         packages = [
-            build_package_data(p)
+            build_package_data(p, package_only)
             for p in package_data.get('packages', [])
         ]
         packages_dev = [
-            build_package_data(p)
+            build_package_data(p, package_only)
             for p in package_data.get('packages-dev', [])
         ]
 
@@ -192,12 +195,13 @@ def parse(cls, location):
             for p in packages_dev
         ]
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=required_deps + required_dev_deps
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
         for package in packages + packages_dev:
             yield package
diff --git a/src/packagedcode/pubspec.py b/src/packagedcode/pubspec.py
index a6abd5b8642..df18d409ba4 100644
--- a/src/packagedcode/pubspec.py
+++ b/src/packagedcode/pubspec.py
@@ -60,11 +60,11 @@ class DartPubspecYamlHandler(BaseDartPubspecHandler):
     documentation_url = 'https://dart.dev/tools/pub/pubspec'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location) as inp:
             pubspec_data = saneyaml.load(inp.read())
 
-        package_data = build_package(pubspec_data)
+        package_data = build_package(pubspec_data, package_only)
         if package_data:
             yield package_data
 
@@ -78,18 +78,19 @@ class DartPubspecLockHandler(BaseDartPubspecHandler):
     documentation_url = 'https://web.archive.org/web/20220330081004/https://gpalma.pt/blog/what-is-the-pubspec-lock/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location) as inp:
             locks_data = saneyaml.load(inp.read())
 
         dependencies = list(collect_locks(locks_data))
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def collect_locks(locks_data):
@@ -238,7 +239,7 @@ def build_dep(name, version, scope, is_runtime=True, is_optional=False):
     return dep
 
 
-def build_package(pubspec_data):
+def build_package(pubspec_data, package_only=False):
     """
     Return a package object from a package data mapping or None
     """
@@ -315,7 +316,7 @@ def add_to_extra_if_present(_key):
     add_to_extra_if_present('executables')
     add_to_extra_if_present('publish_to')
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=DartPubspecYamlHandler.datasource_id,
         type=DartPubspecYamlHandler.default_primary_language,
         primary_language=DartPubspecYamlHandler.default_primary_language,
@@ -333,3 +334,4 @@ def add_to_extra_if_present(_key):
         api_data_url=api_data_url,
         repository_download_url=repository_download_url,
     )
+    return models.PackageData.from_data(package_data, package_only)
diff --git a/src/packagedcode/pypi.py b/src/packagedcode/pypi.py
index d90dacbabef..33f2ec394ea 100644
--- a/src/packagedcode/pypi.py
+++ b/src/packagedcode/pypi.py
@@ -79,11 +79,12 @@ class PythonEggPkgInfoFile(models.DatafileHandler):
     documentation_url = 'https://peps.python.org/pep-0376/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield parse_metadata(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
     @classmethod
@@ -103,11 +104,12 @@ class PythonEditableInstallationPkgInfoFile(models.DatafileHandler):
     documentation_url = 'https://peps.python.org/pep-0376/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield parse_metadata(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
     @classmethod
@@ -320,11 +322,12 @@ def is_datafile(cls, location):
         )
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield parse_metadata(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
 
@@ -337,11 +340,12 @@ class PythonInstalledWheelMetadataFile(models.DatafileHandler):
     documentation_url = 'https://packaging.python.org/en/latest/specifications/core-metadata/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield parse_metadata(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
     @classmethod
@@ -457,7 +461,7 @@ class PyprojectTomlHandler(models.NonAssemblableDatafileHandler):
 META_DIR_SUFFIXES = '.dist-info', '.egg-info', 'EGG-INFO',
 
 
-def parse_metadata(location, datasource_id, package_type):
+def parse_metadata(location, datasource_id, package_type, package_only=False):
     """
     Return a PackageData object from a PKG-INFO or METADATA file at ``location``
     which is a path string or pathlib.Path-like object (including a possible zip
@@ -487,7 +491,7 @@ def parse_metadata(location, datasource_id, package_type):
 
     file_references = list(get_file_references(dist))
 
-    return models.PackageData(
+    package_data = dict(
         datasource_id=datasource_id,
         type=package_type,
         primary_language='Python',
@@ -502,6 +506,7 @@ def parse_metadata(location, datasource_id, package_type):
         extra_data=extra_data,
         **urls,
     )
+    return models.PackageData.from_data(package_data, package_only)
 
 
 def urlsafe_b64decode(data):
@@ -551,7 +556,7 @@ class PypiWheelHandler(models.DatafileHandler):
     documentation_url = 'https://peps.python.org/pep-0427/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with zipfile.ZipFile(location) as zf:
             for path in ZipPath(zf).iterdir():
                 if not path.name.endswith(META_DIR_SUFFIXES):
@@ -564,6 +569,7 @@ def parse(cls, location):
                         location=metapath,
                         datasource_id=cls.datasource_id,
                         package_type=cls.default_package_type,
+                        package_only=package_only,
                     )
 
 
@@ -577,7 +583,7 @@ class PypiEggHandler(models.DatafileHandler):
     documentation_url = 'https://web.archive.org/web/20210604075235/http://peak.telecommunity.com/DevCenter/PythonEggs'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with zipfile.ZipFile(location) as zf:
             for path in ZipPath(zf).iterdir():
                 if not path.name.endswith(META_DIR_SUFFIXES):
@@ -591,6 +597,7 @@ def parse(cls, location):
                         location=metapath,
                         datasource_id=cls.datasource_id,
                         package_type=cls.default_package_type,
+                        package_only=package_only,
                     )
 
 
@@ -610,7 +617,7 @@ def is_datafile(cls, location, filetypes=tuple()):
             return True
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # FIXME: add dependencies
 
         try:
@@ -622,7 +629,7 @@ def parse(cls, location):
         version = sdist.version
         urls, extra_data = get_urls(metainfo=sdist, name=name, version=version)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -635,6 +642,7 @@ def parse(cls, location):
             extra_data=extra_data,
             **urls,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class PythonSetupPyHandler(BaseExtractedPythonLayout):
@@ -646,7 +654,7 @@ class PythonSetupPyHandler(BaseExtractedPythonLayout):
     documentation_url = 'https://docs.python.org/3.11/distutils/setupscript.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         setup_args = get_setup_py_args(location)
 
         # it may be legit to have a name-less package?
@@ -664,7 +672,7 @@ def parse(cls, location):
         python_requires = get_setup_py_python_requires(setup_args)
         extra_data.update(python_requires)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
@@ -678,6 +686,7 @@ def parse(cls, location):
             extra_data=extra_data,
             **urls,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class ResolvedPurl(NamedTuple):
@@ -694,7 +703,7 @@ class BaseDependencyFileHandler(models.DatafileHandler):
     """
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         file_name = fileutils.file_name(location)
 
         dependency_type = get_dparse2_supported_file_name(file_name)
@@ -705,12 +714,13 @@ def parse(cls, location):
             location=location,
             file_name=dependency_type,
         )
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class SetupCfgHandler(BaseExtractedPythonLayout):
@@ -722,7 +732,7 @@ class SetupCfgHandler(BaseExtractedPythonLayout):
     documentation_url = 'https://peps.python.org/pep-0390/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
 
         metadata = {}
         parser = ConfigParser()
@@ -801,7 +811,7 @@ def parse(cls, location):
                 extracted_license_statement = ''
             extracted_license_statement += f" license_files: {license_file_references}"
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=metadata.get('name'),
@@ -812,6 +822,7 @@ def parse(cls, location):
             dependencies=dependent_packages,
             extracted_license_statement=extracted_license_statement,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
     @classmethod
     def parse_reqs(cls, reqs, scope):
@@ -873,7 +884,7 @@ class PipfileLockHandler(BaseDependencyFileHandler):
     documentation_url = 'https://github.com/pypa/pipfile'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location) as f:
             content = f.read()
 
@@ -890,13 +901,14 @@ def parse(cls, location):
             file_name='Pipfile.lock',
         )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             sha256=sha256,
             dependencies=dependent_packages,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 class PipRequirementsFileHandler(BaseDependencyFileHandler):
@@ -919,15 +931,16 @@ class PipRequirementsFileHandler(BaseDependencyFileHandler):
     documentation_url = 'https://pip.pypa.io/en/latest/reference/requirements-file-format/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         dependencies, extra_data = get_requirements_txt_dependencies(location=location)
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             primary_language=cls.default_primary_language,
             dependencies=dependencies,
             extra_data=extra_data,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 # TODO: enable nested load
 
diff --git a/src/packagedcode/readme.py b/src/packagedcode/readme.py
index 63741e86648..a03886929ba 100644
--- a/src/packagedcode/readme.py
+++ b/src/packagedcode/readme.py
@@ -55,11 +55,11 @@ class ReadmeHandler(models.NonAssemblableDatafileHandler):
     documentation_url = ''
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location, encoding='utf-8') as loc:
             readme_manifest = loc.read()
 
-        package_data = build_package(readme_manifest)
+        package_data = build_package(readme_manifest, package_only)
 
         if not package_data.name:
             # If no name was detected for the Package, then we use the basename
@@ -71,7 +71,7 @@ def parse(cls, location):
         yield package_data
 
 
-def build_package(readme_manifest):
+def build_package(readme_manifest, package_only=False):
     """
     Return a Package object from a readme_manifest mapping (from a
     README.chromium file or similar) or None.
@@ -104,5 +104,8 @@ def build_package(readme_manifest):
             continue
         setattr(package, package_key, value)
 
-    package.populate_license_fields()
+    if not package_only:
+        package.populate_license_fields()
+        package.populate_holder_field()
+
     return package
diff --git a/src/packagedcode/rpm.py b/src/packagedcode/rpm.py
index bcbe8294070..a3314785fe2 100644
--- a/src/packagedcode/rpm.py
+++ b/src/packagedcode/rpm.py
@@ -124,7 +124,7 @@ def to_string(self):
 class BaseRpmInstalledDatabaseHandler(models.DatafileHandler):
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         # we receive the location of the Package database file and we need to
         # scan the parent which is the directory that contains the rpmdb
         loc_path = Path(location)
@@ -136,6 +136,7 @@ def parse(cls, location):
             location=xmlish_loc,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
         # TODO: package_data.namespace = cls.default_package_namespace
         return package_data
@@ -274,7 +275,7 @@ class RpmArchiveHandler(models.DatafileHandler):
     documentation_url = 'https://en.wikipedia.org/wiki/RPM_Package_Manager'
 
     @classmethod
-    def parse(cls, location):
+    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)
@@ -351,7 +352,7 @@ def parse(cls, location):
             )
             logger_debug('recognize: data to create a package:\n', data)
 
-        package = models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             # TODO: namespace=cls.default_package_namespace,
@@ -365,9 +366,9 @@ def parse(cls, location):
         )
 
         if TRACE:
-            logger_debug('recognize: created package:\n', package)
+            logger_debug('recognize: created package:\n', name)
 
-        yield package
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 ALGO_BY_ID = {
diff --git a/src/packagedcode/rpm_installed.py b/src/packagedcode/rpm_installed.py
index 5d01436d2e1..8e6a12261c5 100644
--- a/src/packagedcode/rpm_installed.py
+++ b/src/packagedcode/rpm_installed.py
@@ -36,7 +36,7 @@ def logger_debug(*args):
         return logger.debug(' '.join(isinstance(a, str) and a or repr(a) for a in args))
 
 
-def parse_rpm_xmlish(location, datasource_id, package_type):
+def parse_rpm_xmlish(location, datasource_id, package_type, package_only=False):
     """
     Yield PackageData built from an RPM XML'ish file at ``location``. This is a file
     created with the rpm CLI with the xml query option.
@@ -58,6 +58,7 @@ def parse_rpm_xmlish(location, datasource_id, package_type):
             rpm_tags=tags,
             datasource_id=datasource_id,
             package_type=package_type,
+            package_only=package_only,
         )
 
 
@@ -133,7 +134,7 @@ def collect_tags(raw_tags):
         yield name, value_type, value
 
 
-def build_package(rpm_tags, datasource_id, package_type, package_namespace=None):
+def build_package(rpm_tags, datasource_id, package_type, package_namespace=None, package_only=False):
     """
     Return a PackageData object from an ``rpm_tags`` iterable of (name,
     value_type, value) tuples.
@@ -157,8 +158,12 @@ def build_package(rpm_tags, datasource_id, package_type, package_namespace=None)
             except Exception as e:
                 raise Exception(value, converted) from e
             converted.update(handled)
+    
+    current_filerefs = converted.get("current_filerefs", None)
+    if current_filerefs:
+        converted.pop("current_filerefs")
 
-    package_data = models.PackageData.from_dict(converted)
+    package_data = models.PackageData.from_data(converted, package_only)
     return package_data
 
 ################################################################################
@@ -183,7 +188,10 @@ def handler(value, **kwargs):
 
 
 def size_handler(value, **kwargs):
-    return {'size': int(value)}
+    if not value == '0':
+        return {'size': int(value)}
+    else:
+        return {'size': None}
 
 
 def arch_handler(value, **kwargs):
diff --git a/src/packagedcode/rubygems.py b/src/packagedcode/rubygems.py
index 6b77d6945d6..704005c0d21 100644
--- a/src/packagedcode/rubygems.py
+++ b/src/packagedcode/rubygems.py
@@ -41,12 +41,13 @@ class GemArchiveHandler(models.DatafileHandler):
     )
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         metadata = extract_gem_metadata(location)
         metadata = saneyaml.load(metadata)
         yield build_rubygem_package_data(
             gem_data=metadata,
             datasource_id=cls.datasource_id,
+            package_only=package_only,
         )
 
 
@@ -84,13 +85,14 @@ class GemMetadataArchiveExtractedHandler(models.DatafileHandler):
     )
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location, 'rb') as met:
             metadata = met.read()
         metadata = saneyaml.load(metadata)
         yield build_rubygem_package_data(
             gem_data=metadata,
             datasource_id=cls.datasource_id,
+            package_only=package_only,
         )
 
     @classmethod
@@ -129,7 +131,7 @@ class GemspecHandler(models.DatafileHandler):
     documentation_url = 'https://guides.rubygems.org/specification-reference/'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         gemspec = spec.parse_spec(
             location=location,
             package_type=cls.default_package_type,
@@ -152,7 +154,7 @@ def parse(cls, location):
 
         urls = get_urls(name=name, version=version)
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -166,6 +168,7 @@ def parse(cls, location):
             dependencies=dependencies,
             **urls
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 class GemspecInExtractedGemHandler(GemspecHandler):
     datasource_id = 'gemspec_extracted'
@@ -234,7 +237,7 @@ class GemfileLockHandler(BaseGemProjectHandler):
     documentation_url = 'https://bundler.io/man/gemfile.5.html'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         gemfile_lock = GemfileLockParser(location)
         all_gems = list(gemfile_lock.all_gems.values())
         if not all_gems:
@@ -258,7 +261,7 @@ def parse(cls, location):
             ]
             urls = get_urls(primary_gem.name, primary_gem.version)
 
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 primary_language=cls.default_primary_language,
                 type=cls.default_package_type,
@@ -267,6 +270,7 @@ def parse(cls, location):
                 dependencies=deps,
                 **urls
             )
+            yield models.PackageData.from_data(package_data, package_only)
         else:
             deps = [
                 models.DependentPackage(
@@ -284,12 +288,13 @@ def parse(cls, location):
                 ) for gem in all_gems
             ]
 
-            yield models.PackageData(
+            package_data = dict(
                 datasource_id=cls.datasource_id,
                 type=cls.default_package_type,
                 dependencies=deps,
                 primary_language=cls.default_primary_language,
             )
+            yield models.PackageData.from_data(package_data, package_only)
 
 
 class GemfileLockInExtractedGemHandler(GemfileLockHandler):
@@ -417,7 +422,7 @@ def extract_gem_metadata(location):
             fileutils.delete(extract_loc)
 
 
-def build_rubygem_package_data(gem_data, datasource_id):
+def build_rubygem_package_data(gem_data, datasource_id, package_only=False):
     """
     Return a PackageData for ``datasource_id`` built from a Gem `gem_data`
     mapping or None. The ``gem_data`` can come from a .gemspec or .gem/metadata.
@@ -461,7 +466,7 @@ def build_rubygem_package_data(gem_data, datasource_id):
     dependencies = get_dependencies(gem_data.get('dependencies'))
     file_references = get_file_references(metadata.get('files'))
 
-    package_data = models.PackageData(
+    package_mapping = dict(
         datasource_id=datasource_id,
         type=GemArchiveHandler.default_package_type,
         primary_language=GemArchiveHandler.default_primary_language,
@@ -477,6 +482,7 @@ def build_rubygem_package_data(gem_data, datasource_id):
         dependencies=dependencies,
         **urls,
     )
+    package_data = models.PackageData.from_data(package_mapping, package_only)
 
     # we can have one singular or a plural list of authors
     authors = gem_data.get('authors') or []
diff --git a/src/packagedcode/win_pe.py b/src/packagedcode/win_pe.py
index a51ceb9496e..ce040eaedfe 100644
--- a/src/packagedcode/win_pe.py
+++ b/src/packagedcode/win_pe.py
@@ -276,7 +276,7 @@ def is_datafile(cls, location, filetypes=tuple()):
             return True
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         infos = pe_info(location)
 
         version = get_first(
@@ -328,7 +328,7 @@ def parse(cls, location):
             parties = [Party(type=party_org, role='author', name=cname)]
         homepage_url = get_first(infos, 'URL', 'WWW')
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -340,3 +340,4 @@ def parse(cls, location):
             parties=parties,
             homepage_url=homepage_url,
         )
+        yield models.PackageData.from_data(package_data, package_only)
diff --git a/src/packagedcode/win_reg.py b/src/packagedcode/win_reg.py
index e4e59cd3b1e..aac41691f01 100644
--- a/src/packagedcode/win_reg.py
+++ b/src/packagedcode/win_reg.py
@@ -77,6 +77,7 @@ def get_installed_dotnet_versions_from_hive(
     datasource_id,
     package_type,
     registry_path='\\Microsoft\\NET Framework Setup\\NDP',
+    package_only=False,
 ):
     """
     Yield PackageData for the installed versions of .NET framework from the
@@ -90,6 +91,7 @@ def get_installed_dotnet_versions_from_hive(
         registry_tree=registry_tree,
         datasource_id=datasource_id,
         package_type=package_type,
+        package_only=package_only,
     )
 
 
@@ -97,6 +99,7 @@ def get_installed_dotnet_versions_from_regtree(
     registry_tree,
     datasource_id,
     package_type,
+    package_only=False,
 ):
     """
     Yield PackageData for the installed versions of .NET framework from a
@@ -122,13 +125,14 @@ def get_installed_dotnet_versions_from_regtree(
             if key == 'InstallPath':
                 file_references.append(models.FileReference(path=value))
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=datasource_id,
             type=package_type,
             name='microsoft-dot-net-framework',
             version=version,
             file_references=file_references,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def get_installed_windows_programs_from_hive(
@@ -136,6 +140,7 @@ def get_installed_windows_programs_from_hive(
     datasource_id,
     package_type,
     registry_path='\\Microsoft\\Windows\\CurrentVersion\\Uninstall',
+    package_only=False,
 ):
     """
     Yield installed Windows PackageData from a Windows registry file at
@@ -151,6 +156,7 @@ def get_installed_windows_programs_from_hive(
         registry_tree=registry_tree,
         datasource_id=datasource_id,
         package_type=package_type,
+        package_only=package_only,
     )
 
 
@@ -158,6 +164,7 @@ def get_installed_windows_programs_from_regtree(
     registry_tree,
     datasource_id,
     package_type,
+    package_only=False,
 ):
     """
     Yield installed Windows PackageData from a Windows ``registry_tree``.
@@ -213,7 +220,7 @@ def get_installed_windows_programs_from_regtree(
         if uninstall_string:
             file_references.append(models.FileReference(path=uninstall_string))
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=datasource_id,
             type=package_type,
             name=name,
@@ -222,12 +229,14 @@ def get_installed_windows_programs_from_regtree(
             homepage_url=homepage_url,
             file_references=file_references,
         )
+        yield models.PackageData.from_data(package_data, package_only)
 
 
 def get_packages_from_registry_from_hive(
     location,
     datasource_id,
     package_type,
+    package_only=False,
 ):
     """
     Yield PackageData for Installed Windows Programs from the Windows registry
@@ -238,6 +247,7 @@ def get_packages_from_registry_from_hive(
         datasource_id=datasource_id,
         package_type=package_type,
         registry_path='\\Microsoft\\Windows\\CurrentVersion\\Uninstall',
+        package_only=package_only,
     )
 
     yield from get_installed_windows_programs_from_hive(
@@ -245,6 +255,7 @@ def get_packages_from_registry_from_hive(
         datasource_id=datasource_id,
         package_type=package_type,
         registry_path='\\Wow6432Node\\Microsoft\\Windows\\CurrentVersion\\Uninstall',
+        package_only=package_only,
     )
 
     yield from get_installed_dotnet_versions_from_hive(
@@ -252,10 +263,11 @@ def get_packages_from_registry_from_hive(
         datasource_id=datasource_id,
         package_type=package_type,
         registry_path='\\Microsoft\\NET Framework Setup\\NDP',
+        package_only=package_only,
     )
 
 
-def get_installed_packages(root_dir, is_container=True):
+def get_installed_packages(root_dir, is_container=True, package_only=False):
     """
     Yield PackageData for Installed Windows Programs for every detected
     installed program from Windows registry hive files found in well known
@@ -280,7 +292,7 @@ def get_installed_packages(root_dir, is_container=True):
     for software_reg_loc, root_prefix in root_prefixes_by_software_reg_locations.items():
         if not os.path.exists(software_reg_loc):
             continue
-        for package in get_packages_from_registry_from_hive(software_reg_loc):
+        for package in get_packages_from_registry_from_hive(software_reg_loc, package_only):
             package.populate_installed_files(root_dir, root_prefix=root_prefix)
             yield package
 
@@ -342,11 +354,12 @@ class BaseRegInstalledProgramHandler(models.DatafileHandler):
     root_path_relative_to_datafile_path = None
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         yield from get_packages_from_registry_from_hive(
             location=location,
             datasource_id=cls.datasource_id,
             package_type=cls.default_package_type,
+            package_only=package_only,
         )
 
     @classmethod
diff --git a/src/packagedcode/windows.py b/src/packagedcode/windows.py
index 09d62f4c2d9..6fdb07815e3 100644
--- a/src/packagedcode/windows.py
+++ b/src/packagedcode/windows.py
@@ -20,7 +20,7 @@ class MicrosoftUpdateManifestHandler(models.NonAssemblableDatafileHandler):
     description = 'Microsoft Update Manifest .mum file'
 
     @classmethod
-    def parse(cls, location):
+    def parse(cls, location, package_only=False):
         with open(location , 'rb') as loc:
             parsed = xmltodict.parse(loc)
 
@@ -47,7 +47,7 @@ def parse(cls, location):
                 )
             )
 
-        yield models.PackageData(
+        package_data = dict(
             datasource_id=cls.datasource_id,
             type=cls.default_package_type,
             name=name,
@@ -57,3 +57,4 @@ def parse(cls, location):
             parties=parties,
             copyright=copyrght,
         )
+        yield models.PackageData.from_data(package_data, package_only)
diff --git a/tests/packagedcode/data/build/buck/end2end-expected.json b/tests/packagedcode/data/build/buck/end2end-expected.json
index 70f7b321317..2b169491ec3 100644
--- a/tests/packagedcode/data/build/buck/end2end-expected.json
+++ b/tests/packagedcode/data/build/buck/end2end-expected.json
@@ -256,9 +256,7 @@
           "other_license_expression": null,
           "other_license_expression_spdx": null,
           "other_license_detections": [],
-          "extracted_license_statement": [
-            "LICENSE"
-          ],
+          "extracted_license_statement": "- LICENSE\n",
           "notice_text": null,
           "source_packages": [],
           "file_references": [],
diff --git a/tests/packagedcode/test_build.py b/tests/packagedcode/test_build.py
index 3d0f438e551..147a075b15b 100644
--- a/tests/packagedcode/test_build.py
+++ b/tests/packagedcode/test_build.py
@@ -85,48 +85,46 @@ def test_BuckPackage_recognize_with_license(self):
 
     def test_MetadataBzl_parse(self):
         test_file = self.get_test_loc('metadatabzl/METADATA.bzl')
-        result_packages = build.BuckMetadataBzlHandler.parse(test_file)
-        expected_packages = [
-            models.PackageData(
-                datasource_id=build.BuckMetadataBzlHandler.datasource_id,
-                type='github',
-                name='example',
-                version='0.0.1',
-                extracted_license_statement=['BSD-3-Clause'],
-                parties=[
-                    models.Party(
-                        type=models.party_org,
-                        name='oss_foundation',
-                        role='maintainer'
-                    )
-                ],
-                homepage_url='https://github.com/example/example',
-            ),
-        ]
+        result_packages = build.BuckMetadataBzlHandler.parse(test_file, package_only=True)
+        package_data = dict(
+            datasource_id=build.BuckMetadataBzlHandler.datasource_id,
+            type='github',
+            name='example',
+            version='0.0.1',
+            extracted_license_statement=['BSD-3-Clause'],
+            parties=[
+                models.Party(
+                    type=models.party_org,
+                    name='oss_foundation',
+                    role='maintainer'
+                )
+            ],
+            homepage_url='https://github.com/example/example',
+        )
+        expected_packages = [models.PackageData.from_data(package_data=package_data, package_only=True)]
         compare_package_results(expected_packages, result_packages)
 
     def test_MetadataBzl_recognize_new_format(self):
         test_file = self.get_test_loc('metadatabzl/new-format/METADATA.bzl')
-        result_packages = build.BuckMetadataBzlHandler.parse(test_file)
-        expected_packages = [
-            models.PackageData(
-                datasource_id=build.BuckMetadataBzlHandler.datasource_id,
-                type='github',
-                name='example/example',
-                version='0.0.1',
-                extracted_license_statement='BSD-3-Clause',
-                parties=[
-                    models.Party(
-                        type=models.party_org,
-                        name='example_org',
-                        role='maintainer'
-                    )
-                ],
-                download_url='',
-                sha1='',
-                homepage_url='https://github.com/example/example',
-                vcs_url='https://github.com/example/example.git',
-                extra_data=dict(vcs_commit_hash="deadbeef")
-            )
-        ]
+        result_packages = build.BuckMetadataBzlHandler.parse(test_file, package_only=True)
+        package_data = dict(
+            datasource_id=build.BuckMetadataBzlHandler.datasource_id,
+            type='github',
+            name='example/example',
+            version='0.0.1',
+            extracted_license_statement='BSD-3-Clause',
+            parties=[
+                models.Party(
+                    type=models.party_org,
+                    name='example_org',
+                    role='maintainer'
+                )
+            ],
+            download_url='',
+            sha1='',
+            homepage_url='https://github.com/example/example',
+            vcs_url='https://github.com/example/example.git',
+            extra_data=dict(vcs_commit_hash="deadbeef")
+        )
+        expected_packages = [models.PackageData.from_data(package_data=package_data, package_only=True)]
         compare_package_results(expected_packages, result_packages)
diff --git a/tests/packagedcode/test_package_models.py b/tests/packagedcode/test_package_models.py
index 98a3606093d..62369a9c115 100644
--- a/tests/packagedcode/test_package_models.py
+++ b/tests/packagedcode/test_package_models.py
@@ -76,7 +76,7 @@ def test_Package_creation_and_dump(self):
         assert list(pd.to_dict().items()) == expected
 
     def test_Package_simple(self):
-        package = PackageData(
+        package_mapping = dict(
             datasource_id = 'rpm_archive',
             type='rpm',
             name='Sample',
@@ -86,8 +86,9 @@ def test_Package_simple(self):
             vcs_url='git+https://somerepo.com/that.git',
             extracted_license_statement='apache-2.0',
         )
+        package_data = PackageData.from_data(package_data=package_mapping, package_only=False)
         expected_loc = 'models/simple-expected.json'
-        self.check_package_data(package, expected_loc, regen=REGEN_TEST_FIXTURES)
+        self.check_package_data(package_data, expected_loc, regen=REGEN_TEST_FIXTURES)
 
     def test_Package_model_qualifiers_are_serialized_as_mappings(self):
         package = models.PackageData(
@@ -108,7 +109,7 @@ def test_Package_model_qualifiers_are_converted_to_mappings(self):
         assert package.qualifiers == dict(this='that')
 
     def test_Package_full(self):
-        package = PackageData(
+        package_mapping = dict(
             type='rpm',
             datasource_id = 'rpm_archive',
             namespace='fedora',
@@ -136,8 +137,9 @@ def test_Package_full(self):
             notice_text='licensed under the apacche 2.0 \nlicense',
             source_packages=["pkg:maven/aspectj/aspectjtools@1.5.4?classifier=sources"],
         )
+        package_data = PackageData.from_data(package_data=package_mapping, package_only=False)
         expected_loc = 'models/full-expected.json'
-        self.check_package_data(package, expected_loc, regen=REGEN_TEST_FIXTURES)
+        self.check_package_data(package_data, expected_loc, regen=REGEN_TEST_FIXTURES)
 
     def test_package_data_datasource_id_are_unique(self):
         """
@@ -177,11 +179,12 @@ def test_package_data_file_patterns_are_tuples(self):
 
     def test_add_to_package(self):
         test_loc = self.get_test_loc('npm/electron')
-        test_package = models.Package(
+        test_package_data = dict(
             type='npm',
             name='electron',
             version='3.1.11',
         )
+        test_package = models.Package.from_data(test_package_data)
         test_package_uid = test_package.package_uid
         test_codebase = Codebase(
             location=test_loc,
@@ -241,12 +244,13 @@ def test_create_package_not_handled_by_packagedcode(self):
             'gpl',
             'GNU General Public License version 2.0 (GPLv2)',
         ]
-        package = PackageData(
+        package_mapping = dict(
             type='sourceforge',
             name='openstunts',
             copyright='Copyright (c) openstunts project',
             extracted_license_statement=extracted_license_statement,
         )
+        package = PackageData.from_data(package_data=package_mapping, package_only=False)
         # Test generated fields
         assert package.purl == 'pkg:sourceforge/openstunts'
         assert package.holder == 'openstunts project'
diff --git a/tests/scancode/data/help/help.txt b/tests/scancode/data/help/help.txt
index 3076354074c..340d674bdc5 100644
--- a/tests/scancode/data/help/help.txt
+++ b/tests/scancode/data/help/help.txt
@@ -12,6 +12,9 @@ Options:
     -p, --package     Scan <input> for application package and dependency
                       manifests, lockfiles and related data.
     --system-package  Scan <input> for installed system package databases.
+    --package-only    Only detect package information and skip license/copyright
+                      detection steps, in application package and dependency
+                      manifests, lockfiles and related data.
     -c, --copyright   Scan <input> for copyrights.
 
   other scans: