Cscfairmeta 642 draft field validation (#638)

* CSCFAIRMETA-642: [FIX] bug allowed new version from draft dataset * CSCFAIRMETA-642: [FIX] bug allowed user metadata adds via normal update - V2 api should only allow adding files and user metadata of those files via separate rpc api and not by updating the dataset with PUT or PATCH * CSCFAIRMETA-642: [FIX] crash when creating dataset without catalog * CSCFAIRMETA-642: [FIX] bug where v2 fields were visible in v1 model * CSCFAIRMETA-642: [ADD] more permissive validation for drafts - New JSON schema for draft datasets which requires only title in research_dataset in addition to metadata_provider_org and metadata_provider_user. - New draft data catalog that can be used only for draft datasets. This can be used in case where the file origin is not yet determined. * CSCFAIRMETA-642: [ADD] v2 api for datacatalogs - V2 datacatalogs allow different schemas for v1 and v2 apis. No need to save the schema in both v1/schemas for datacatalogs and v2 for datasets. * CSCFAIRMETA-642: [ADD] documentation of the new data catalog * CSCFAIRMETA-642: [ADD] catalog permissions to initialdata Co-authored-by: Katri Tegel <[email protected]> Co-authored-by: Evgenia Lyjina <[email protected]> Co-authored-by: genie9 <[email protected]>
CSCfi · Nov 5, 2020 · 13a4a77 · 13a4a77
1 parent 3d3bc2d
commit 13a4a77
Show file tree

Hide file tree

Showing 23 changed files with 1,991 additions and 43 deletions.
diff --git a/docs/v2/source/datasets.rst b/docs/v2/source/datasets.rst
@@ -106,12 +106,13 @@ The official Fairdata data catalogs with end user write access are:
 Other data catalogs where End Users can directly store dataset metadata:
 
 
-+---------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
-| Catalog | Purpose                                                                                                                                                                                                                       | Identifier                         |
-+---------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
-| Legacy  | Store legacy datasets that are published elsewhere. Published datasets may not have all of the required metadata to qualify as Fairdata-dataset. Identifiers are not generated by Metax: User has to provide any identifiers. | urn:nbn:fi:att:data-catalog-legacy |
-+---------+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
-
++---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
+| Catalog | Purpose                                                                                                                                                                                                                            | Identifier                         |
++---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
+| Legacy  | Store legacy datasets that are published elsewhere. Published datasets may not have all of the required metadata to qualify as Fairdata-dataset. Identifiers are not generated by Metax: User has to provide any identifiers.      | urn:nbn:fi:att:data-catalog-legacy |
++---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
+| Draft   | Store datasets which are in draft state and data catalog is not yet decided. This catalog is used as a default catalog when creating datasets. Note you must change catalog to one from above before file addition or publication. | urn:nbn:fi:att:data-catalog-dft    |
++---------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+------------------------------------+
 
 
 

diff --git a/src/metax_api/api/rest/base/serializers/catalog_record_serializer.py b/src/metax_api/api/rest/base/serializers/catalog_record_serializer.py
@@ -11,7 +11,7 @@
 from django.conf import settings as django_settings
 from rest_framework.serializers import ValidationError
 
-from metax_api.exceptions import Http403
+from metax_api.exceptions import Http400, Http403
 from metax_api.models import CatalogRecord, DataCatalog, Directory, Contract, Common, File
 from metax_api.services import (
     CatalogRecordService as CRS,
@@ -51,7 +51,7 @@
 
 END_USER_ALLOWED_DATA_CATALOGS = django_settings.END_USER_ALLOWED_DATA_CATALOGS
 LEGACY_CATALOGS = django_settings.LEGACY_CATALOGS
-
+DFT_CATALOG = django_settings.DFT_DATA_CATALOG_IDENTIFIER
 
 class CatalogRecordSerializer(CommonSerializer):
 
@@ -145,6 +145,9 @@ def is_valid(self, raise_exception=False):
             # in the database, and what the data catalog is being changed to.
             self._validate_research_dataset_uniqueness(self.instance.research_dataset)
 
+            # updating data catalog of published dataset to draft data catalog is restricted
+            self._validate_draft_data_catalog()
+
         # executes other validation related code, such as validate_research_dataset()
         super(CatalogRecordSerializer, self).is_valid(raise_exception=raise_exception)
 
@@ -164,6 +167,10 @@ def update(self, instance, validated_data):
         return super(CatalogRecordSerializer, self).update(instance, validated_data)
 
     def create(self, validated_data):
+        if 'V2' not in self.__class__.__name__ and \
+                validated_data['data_catalog'].catalog_json['identifier'] == DFT_CATALOG:
+            raise Http400({ 'detail': [ 'Draft catalog cannot be used in V1 API' ]})
+
         if self._migration_override_requested():
 
             # any custom stuff before create that might be necessary for migration purposes
@@ -390,8 +397,14 @@ def _populate_file_titles(self, ds):
                         files_from_db.pop(i)
                         break
 
-    def validate_research_dataset(self, value):
+    def _validate_draft_data_catalog(self):
+        # catalog object is not yet included to initial_data so have to fetch it
+        dc_pid = DataCatalog.objects.get(pk=self.initial_data['data_catalog']).catalog_json['identifier']
+
+        if dc_pid == DFT_CATALOG and (self.instance.is_published() or self.instance.draft_of):
+            raise ValidationError({ 'detail': ['Catalog cannot be changed back to draft'] })
 
+    def validate_research_dataset(self, value):
         self._populate_file_and_dir_titles(value)
 
         self.validate_json_schema(value)
@@ -422,7 +435,7 @@ def validate_json_schema(self, value):
 
         if self._operation_is_create:
             if not value.get('preferred_identifier', None):
-                if DataCatalogService.is_harvested(self.initial_data['data_catalog']):
+                if DataCatalogService.is_harvested(self.initial_data.get('data_catalog')):
                     raise ValidationError({ 'preferred_identifier':
                         ['harvested catalog record must have preferred identifier']})
 

diff --git a/src/metax_api/api/rest/v2/router.py b/src/metax_api/api/rest/v2/router.py
@@ -26,13 +26,13 @@
 from metax_api.api.rest.base.views import (
     ApiErrorViewSet,
     ContractViewSet,
-    DataCatalogViewSet,
     DirectoryViewSet,
     FileStorageViewSet,
     FileViewSet,
     SchemaViewSet
 )
 from .views import (
+    DataCatalogViewSet,
     DatasetViewSet,
 )
 
@@ -92,7 +92,6 @@ def __init__(self, *args, **kwargs):
 router_v1 = CustomRouter(trailing_slash=False)
 router_v1.register(r'apierrors/?', ApiErrorViewSet)
 router_v1.register(r'contracts/?', ContractViewSet)
-router_v1.register(r'datacatalogs/?', DataCatalogViewSet)
 router_v1.register(r'directories/?', DirectoryViewSet)
 router_v1.register(r'files/?', FileViewSet)
 router_v1.register(r'filestorages/?', FileStorageViewSet)
@@ -102,6 +101,7 @@ def __init__(self, *args, **kwargs):
 
 router_v2 = CustomRouterV2(trailing_slash=False)
 router_v2.register(r'datasets/?', DatasetViewSet)
+router_v2.register(r'datacatalogs/?', DataCatalogViewSet)
 router_v2.register(
     r'datasets/(?P<identifier>.+)/metadata_versions/(?P<metadata_version_identifier>.+)/?',
     DatasetViewSet