Add dct identifier to publisher #301
Conversation
hcvdwerf
force-pushed
the
add-dct-identifier-to-publisher
branch
from
83a1ddf
to
aefa22c
Compare
ckanext/dcat/converters.py
Outdated
| elif isinstance(dcat_publisher, dict) and dcat_publisher.get('name'): | ||
| package_dict['extras'].append({'key': 'dcat_publisher_name', 'value': dcat_publisher.get('name')}) | ||
| package_dict['extras'].append({'key': 'dcat_publisher_email', 'value': dcat_publisher.get('mbox')}) | ||
| if dcat_publisher.get('name'): |
There was a problem hiding this comment.
There's a double checl for `.get('name') which can be removed.
There was a problem hiding this comment.
I removed the second one
| @@ -66,6 +66,11 @@ dataset_fields: | |||
|
|
|||
| - field_name: type | |||
| label: Type | |||
|
|
|||
| - field_name: identifier | |||
There was a problem hiding this comment.
Can you add it to the dcat_ap_full.yaml file as well?
| @@ -123,7 +123,7 @@ def _parse_dataset_base(self, dataset_dict, dataset_ref): | |||
|
|
|||
| # Publisher | |||
| publisher = self._publisher(dataset_ref, DCT.publisher) | |||
| for key in ("uri", "name", "email", "url", "type"): | |||
| for key in ("uri", "name", "email", "url", "type", "identifier"): | |||
There was a problem hiding this comment.
This will take care of the DCAT -> CKAN mapping, but you also need to add the other way around, CKAN -> DCAT. This is done a bit further down in the file here and here.
Additionally, this will cover the legacy way of defining publisher fields based on namespaced publisher_* extra fields, but going forward we want to support scheming objects ("publisher": {"id": xx, "name": yy, "url": zz}). This is handled in the euro_dcat_ap_scheming profile here (for CKAN -> DCAT, the opposite should already be handled by the base profile). Do you mind adding support for the new field there as well? And if you extend this test to cover the new field, even better.
@hcvdwerf Let me know if all this makes sense
There was a problem hiding this comment.
Tnx for the comments. Verry helpful @amercader . Jut commit the changes
hcvdwerf
force-pushed
the
add-dct-identifier-to-publisher
branch
from
64ace0a
to
e0becf6
Compare
Co-authored-by: Mark <markusjanse@gmail.com>
|
Thanks @hcvdwerf ! |
The system previously lacked support for storing and serializing publisher identifiers, such as ROR IDs, ORCIDs, or other unique identifiers for publishers. While other metadata fields like publisher name, email, and URL were being correctly parsed and serialized, the identifier field (e.g., dct:identifier) was not being included in the publisher’s metadata. This omission resulted in incomplete RDF metadata and affected datasets that needed to capture a unique reference to the publisher.