Feature/fair signposting

addons/base/views.py

@@ -1035,9 +1035,12 @@ def persistent_file_download(auth, **kwargs):
 
     query_params = request.args.to_dict()
 
-    return redirect(
-        file.generate_waterbutler_url(**query_params),
-        code=http_status.HTTP_302_FOUND
+    return make_response(
+        '', http_status.HTTP_302_FOUND, {
+            'Location': file.generate_waterbutler_url(**query_params),
+            'Link': f'<{settings.DOMAIN}metadata/{id_or_guid}/?format=linkset> ; rel="linkset" ; type="application/linkset",'
+                    f' <{settings.DOMAIN}metadata/{id_or_guid}/?format=linkset-json"> ; rel="linkset-json" ; type="application/linkset+json"',
+        }
     )
 
 

addons/osfstorage/tests/test_views.py

@@ -1551,11 +1551,19 @@ def test_download_file(self):
         # Test download works with path
         url = base_url.format(file._id)
         redirect = self.app.get(url, auth=self.user.auth)
+        link_header = (f'<{settings.DOMAIN}metadata/{file._id}/?format=linkset> ; rel="linkset" ; type="application/linkset", '
+                       f'<{settings.DOMAIN}metadata/{file._id}/?format=linkset-json"> ; rel="linkset-json" ; type="application/linkset+json"')
+        assert link_header == redirect.headers['Link']
         assert redirect.status_code == 302
 
         # Test download works with guid
-        url = base_url.format(file.get_guid(create=True)._id)
+        guid = file.get_guid(create=True)._id
+        url = base_url.format(guid)
         redirect = self.app.get(url, auth=self.user.auth)
+        link_header = (
+            f'<{settings.DOMAIN}metadata/{guid}/?format=linkset> ; rel="linkset" ; type="application/linkset", '
+            f'<{settings.DOMAIN}metadata/{guid}/?format=linkset-json"> ; rel="linkset-json" ; type="application/linkset+json"')
+        assert link_header == redirect.headers['Link']
         assert redirect.status_code == 302
 
         # Test nonexistent file 404's

api/cedar_metadata_records/views.py

@@ -19,8 +19,8 @@
     CedarMetadataRecordsDetailSerializer,
 )
 from framework.auth.oauth_scopes import CoreScopes
-
-from osf.models import CedarMetadataRecord
+from osf.models import CedarMetadataRecord, Node, Registration
+from website import settings
 
 logger = logging.getLogger(__name__)
 
@@ -99,5 +99,10 @@ def get_serializer_class(self):
 
     def get(self, request, *args, **kwargs):
         record = self.get_object()
+        is_referent_project_or_registration = isinstance(record.guid.referent, (Node, Registration))
         file_name = f'{record._id}-{record.get_template_name()}-v{record.get_template_version()}.json'
-        return Response(record.metadata, headers={'Content-Disposition': f'attachment; filename={file_name}'})
+        headers = {'Content-Disposition': f'attachment; filename={file_name}'}
+        if is_referent_project_or_registration:
+            guid_id = record.guid._id
+            headers['link'] = f'<{settings.DOMAIN}{guid_id}/>; rel="describes"; type="text/html"'
+        return Response(record.metadata, headers=headers)

api_tests/cedar_metadata_records/views/test_record_metadata_download_get.py

@@ -3,6 +3,8 @@
 from .test_record import TestCedarMetadataRecord
 from osf.utils.permissions import READ, WRITE
 from osf_tests.factories import AuthUserFactory
+from website import settings
+
 
 @pytest.mark.django_db
 class TestCedarMetadataRecordMetadataDownloadPrivateProjectPublishedMetadata(TestCedarMetadataRecord):
@@ -13,6 +15,7 @@ def test_record_metadata_download_for_node_with_admin_auth(self, app, node, user
         resp = app.get(f'/_/cedar_metadata_records/{cedar_record_for_node._id}/metadata_download/', auth=admin.auth)
         assert resp.status_code == 200
         assert resp.headers['Content-Disposition'] == f'attachment; filename={self.get_record_metadata_download_file_name(cedar_record_for_node)}'
+        assert resp.headers.get('Link') == f'<{settings.DOMAIN}{node._id}/>; rel="describes"; type="text/html"'
         assert resp.json == cedar_record_metadata_json
 
     def test_record_metadata_download_for_node_with_write_auth(self, app, node, cedar_record_for_node, cedar_record_metadata_json):
@@ -179,6 +182,7 @@ def test_record_metadata_download_for_registration_with_admin_auth(self, app, us
         resp = app.get(f'/_/cedar_metadata_records/{cedar_record_for_registration._id}/metadata_download/', auth=admin.auth)
         assert resp.status_code == 200
         assert resp.headers['Content-Disposition'] == f'attachment; filename={self.get_record_metadata_download_file_name(cedar_record_for_registration)}'
+        assert resp.headers.get('Link') == f'<{settings.DOMAIN}{cedar_record_for_registration.guid._id}/>; rel="describes"; type="text/html"'
         assert resp.json == cedar_record_metadata_json
 
     def test_record_metadata_download_for_registration_with_write_auth(self, app, registration, cedar_record_for_registration, cedar_record_metadata_json):
@@ -307,6 +311,7 @@ def test_record_metadata_download_for_node_with_admin_auth(self, app, user, ceda
         resp = app.get(f'/_/cedar_metadata_records/{cedar_draft_record_for_file_alt._id}/metadata_download/', auth=admin.auth)
         assert resp.status_code == 200
         assert resp.headers['Content-Disposition'] == f'attachment; filename={self.get_record_metadata_download_file_name(cedar_draft_record_for_file_alt)}'
+        assert not resp.headers.get('Link')
         assert resp.json == cedar_record_metadata_json
 
     def test_record_metadata_download_for_node_with_write_auth(self, app, node_alt, cedar_draft_record_for_file_alt, cedar_record_metadata_json):

framework/auth/decorators.py

@@ -76,6 +76,29 @@ def wrapped(*args, **kwargs):
 
     return wrapped
 
+
+def is_contributor_or_public_resource(resource_kw='resource'):
+    """
+    Require that user be contributor or resource be public.
+    """
+    def decorator(func):
+        @wraps(func)
+        def wrapped(*args, **kwargs):
+            from osf.models import BaseFileNode, Guid
+            referent = kwargs.get(resource_kw)
+            if isinstance(referent, Guid):
+                referent = referent.referent
+            target_resource = referent.target if isinstance(referent, BaseFileNode) else referent
+            if target_resource.is_public:
+                return func(*args, **kwargs)
+            auth = Auth.from_kwargs(request.args.to_dict(), {})
+            if auth.logged_in and target_resource.is_contributor(auth.user):
+                return func(*args, **kwargs)
+            raise HTTPError(http_status.HTTP_403_FORBIDDEN)
+        return wrapped
+    return decorator
+
+
 # TODO Can remove after Waterbutler is sending requests to V2 endpoints.
 # This decorator has been adapted for use in an APIv2 parser - HMACSignedParser
 def must_be_signed(func):

osf/metadata/osf_gathering.py

@@ -6,6 +6,7 @@
 
 from django.contrib.contenttypes.models import ContentType
 from django import db
+from mimetypes import MimeTypes
 import rdflib
 
 from api.caching.tasks import get_storage_usage_total
@@ -44,6 +45,8 @@
 
 logger = logging.getLogger(__name__)
 
+mime = MimeTypes()
+
 
 ##### BEGIN "public" api #####
 
@@ -373,7 +376,7 @@ def osf_iri(guid_or_model):
     return OSFIO[guid._id]
 
 
-def osfguid_from_iri(iri):
+def osfguid_from_iri(iri: str) -> str:
     if iri.startswith(OSFIO):
         return without_namespace(iri, OSFIO)
     raise ValueError(f'expected iri starting with "{OSFIO}" (got "{iri}")')
@@ -702,6 +705,16 @@ def gather_files(focus):
             yield (DCTERMS.requires, file_focus)
 
 
+@gather.er(DCAT.mediaType)
+def gather_file_mediatype(focus):
+    (mime_type, _) = mime.guess_type(focus.dbmodel.name)
+    yield (DCAT.mediaType, (
+        'application/octet-stream'
+        if mime_type is None
+        else mime_type
+    ))
+
+
 @gather.er(DCTERMS.hasPart, DCTERMS.isPartOf)
 def gather_parts(focus):
     if isinstance(focus.dbmodel, osfdb.AbstractNode):

osf/metadata/rdfutils.py

@@ -26,7 +26,7 @@
 PROV = rdflib.Namespace('http://www.w3.org/ns/prov#')                   # "provenance"
 # non-standard namespace for datacite terms (resolves to datacite docs)
 DATACITE = rdflib.Namespace('https://schema.datacite.org/meta/kernel-4/#')
-
+SCHEMA = rdflib.Namespace('https://schema.org/')
 
 # namespace prefixes that will be shortened by default
 # when serialized, instead of displaying the full iri
@@ -43,6 +43,49 @@
 }
 
 
+DATACITE_SCHEMA_RESOURCE_TYPE_GENERAL_MAPPING = {
+    DATACITE.Audiovisual: SCHEMA.MediaObject,
+    DATACITE.Book: SCHEMA.Book,
+    DATACITE.BookChapter: SCHEMA.Chapter,
+    DATACITE.Collection: SCHEMA.Collection,
+    DATACITE.ComputationalNotebook: SCHEMA.SoftwareSourceCode,
+    DATACITE.ConferencePaper: SCHEMA.Article,
+    DATACITE.ConferenceProceeding: SCHEMA.Periodical,
+    DATACITE.DataPaper: SCHEMA.Article,
+    DATACITE.Dataset: SCHEMA.Dataset,
+    DATACITE.Dissertation: SCHEMA.Thesis,
+    DATACITE.Event: SCHEMA.Event,
+    DATACITE.Image: SCHEMA.ImageObject,
+    DATACITE.InteractiveResource: SCHEMA.CreativeWork,
+    DATACITE.Journal: SCHEMA.Periodical,
+    DATACITE.JournalArticle: SCHEMA.ScholarlyArticle,
+    DATACITE.Model: SCHEMA.CreativeWork,
+    DATACITE.OutputManagementPlan: SCHEMA.HowTo,
+    DATACITE.PeerReview: SCHEMA.Review,
+    DATACITE.PhysicalObject: SCHEMA.Thing,
+    DATACITE.Preprint: SCHEMA.ScholarlyArticle,
+    DATACITE.Report: SCHEMA.Report,
+    DATACITE.Service: SCHEMA.Service,
+    DATACITE.Software: SCHEMA.SoftwareSourceCode,
+    DATACITE.Sound: SCHEMA.AudioObject,
+    DATACITE.Standard: SCHEMA.CreativeWork,
+    DATACITE.Text: SCHEMA.Text,
+    DATACITE.Workflow: SCHEMA.HowTo,
+    DATACITE.Other: SCHEMA.CreativeWork,
+    DATACITE.Instrument: SCHEMA.MeasurementMethodEnum,
+    DATACITE.StudyRegistration: SCHEMA.Text,
+    OSF.Project: SCHEMA.CreativeWork,
+    OSF.Preprint: SCHEMA.ScholarlyArticle,
+    OSF.Registration: SCHEMA.Text,
+    OSF.File: SCHEMA.DigitalDocument,
+    OSF.ProjectComponent: SCHEMA.CreativeWork,
+    OSF.RegistrationComponent: SCHEMA.Text,
+}
+
+
+DEFAULT_SCHEMADOTORG_RESOURCE_TYPE = SCHEMA.CreativeWork
+
+
 def contextualized_graph(graph=None) -> rdflib.Graph:
     '''bind default namespace prefixes to a new (or given) rdf graph
     '''
@@ -147,3 +190,6 @@ def smells_like_iri(maybe_iri: str) -> bool:
         isinstance(maybe_iri, str)
         and '://' in maybe_iri
     )
+
+def map_resource_type_general_datacite_to_scheme(_type_iri: rdflib.URIRef, resource_rdftype: rdflib.URIRef) -> str:
+    return DATACITE_SCHEMA_RESOURCE_TYPE_GENERAL_MAPPING.get(_type_iri) or DATACITE_SCHEMA_RESOURCE_TYPE_GENERAL_MAPPING.get(resource_rdftype, DEFAULT_SCHEMADOTORG_RESOURCE_TYPE)

osf/metadata/serializers/__init__.py

@@ -9,13 +9,16 @@
 from .datacite import DataciteJsonMetadataSerializer, DataciteXmlMetadataSerializer
 from .google_dataset_json_ld import GoogleDatasetJsonLdSerializer
 from .turtle import TurtleMetadataSerializer
+from .linkset import SignpostLinkset, SignpostLinksetJSON
 
 
 METADATA_SERIALIZER_REGISTRY = {
     'turtle': TurtleMetadataSerializer,
     'datacite-json': DataciteJsonMetadataSerializer,
     'datacite-xml': DataciteXmlMetadataSerializer,
     'google-dataset-json-ld': GoogleDatasetJsonLdSerializer,
+    'linkset': SignpostLinkset,
+    'linkset-json': SignpostLinksetJSON
 }