ref: duplicate definition and minor cleanup

ckanext/dcor_schemas/jobs.py

@@ -3,14 +3,34 @@
 import dclab
 from dcor_shared import (
     DC_MIME_TYPES, get_resource_path, get_dc_instance, s3, s3cc, sha256sum,
-    wait_for_resource
+    wait_for_resource, get_ckan_config_option
 )
 
 
 def admin_context():
     return {'ignore_auth': True, 'user': 'default'}
 
 
+def get_base_metadata(resource):
+    res_dict_base = {}
+    if not resource.get("mimetype"):
+        suffix = "." + resource["name"].rsplit(".", 1)[-1]
+        for mt in DC_MIME_TYPES:
+            if suffix in DC_MIME_TYPES[mt]:
+                res_dict_base["mimetype"] = mt
+                break
+
+    # Also make sure the resource has "url" defined.
+    if not resource.get("url"):
+        site_url = get_ckan_config_option("ckan.site_url")
+        meta_url = (f"{site_url}"
+                    f"/dataset/{resource['package_id']}"
+                    f"/resource/{resource['id']}"
+                    f"/download/{resource['name'].lower()}")
+        res_dict_base["url"] = meta_url
+    return res_dict_base
+
+
 def patch_resource_noauth(package_id, resource_id, data_dict):
     """Patch a resource using package_revise"""
     package_revise = logic.get_action("package_revise")
@@ -28,20 +48,17 @@ def set_resource_meta_base_data(resource):
     run before. So it makes sense to perform any additional calls
     to `package_revise` in those background jobs.
     """
-    data_dict = {}
-    if "url" in resource:
-        data_dict["url"] = resource["url"]
-    if "mimetype" in resource:
-        data_dict["mimetype"] = resource["mimetype"]
+    res_dict_base = get_base_metadata(resource)
     patch_resource_noauth(
         package_id=resource["package_id"],
         resource_id=resource["id"],
-        data_dict=data_dict)
+        data_dict=res_dict_base)
     return True
 
 
 def set_dc_config_job(resource):
     """Store all DC config metadata"""
+    resource.update(get_base_metadata(resource))
     if (resource.get('mimetype') in DC_MIME_TYPES
             and resource.get("dc:setup:channel width", None) is None):
         rid = resource["id"]

ckanext/dcor_schemas/plugin.py

@@ -403,33 +403,23 @@ def get_user_dataset_labels(self, user_obj):
 
     # IResourceController
     def before_resource_create(self, context, resource):
-        """Ran before creating the resource; changes will be in package dict"""
-        res_data_dict = {
-            "last_modified": datetime.datetime.now(datetime.timezone.utc)
-        }
-
-        if not resource.get("mimetype"):
-            suffix = "." + resource["name"].rsplit(".", 1)[-1]
-            for mt in DC_MIME_TYPES:
-                if suffix in DC_MIME_TYPES[mt]:
-                    res_data_dict["mimetype"] = mt
-                    break
-
-        # Also make sure the resource has "url" defined.
-        if not resource.get("url"):
-            site_url = get_ckan_config_option("ckan.site_url")
-            meta_url = (f"{site_url}"
-                        f"/dataset/{resource['package_id']}"
-                        f"/resource/{resource['id']}"
-                        f"/download/{resource['name'].lower()}")
-            res_data_dict["url"] = meta_url
-
-        resource.update(res_data_dict)
+        if "upload" in resource:
+            # set/override the filename
+            upload = resource["upload"]
+            if hasattr(upload, "filename"):
+                filename = upload.filename
+            elif hasattr(upload, "name"):
+                filename = pathlib.Path(upload.name).name
+            else:
+                raise ValueError(
+                    f"Could not determine filename for {resource}")
+            resource["name"] = filename
+        resource.update(jobs.get_base_metadata(resource))
 
     def after_resource_create(self, context, resource):
         """Add custom jobs"""
         # Make sure mimetype etc. are set properly
-        self.before_resource_create(context, resource)
+        resource.update(jobs.get_base_metadata(resource))
 
         # Make sure the resource has a mimetype if possible. This is a
         # workaround for data uploaded via S3.
@@ -539,19 +529,6 @@ def after_resource_create(self, context, resource):
         datapreview.add_views_to_resource(context={"ignore_auth": True},
                                           resource_dict=resource)
 
-    def before_resource_create(self, context, resource):
-        if "upload" in resource:
-            # set/override the filename
-            upload = resource["upload"]
-            if hasattr(upload, "filename"):
-                filename = upload.filename
-            elif hasattr(upload, "name"):
-                filename = pathlib.Path(upload.name).name
-            else:
-                raise ValueError(
-                    f"Could not determine filename for {resource}")
-            resource["name"] = filename
-
     # ITemplateHelpers
     def get_helpers(self):
         # Template helper function names should begin with the name of the