|Status:||Closed||Start date:||14 May 2018|
|Assignee:||Angelo Quaglia||% Done:|
|Submitting Organisation:||NL||Knowledge-Base relevant?:|
|Country:||NL - The Netherlands||Keyword #2:|
|Originating UI:||Thematic Viewer||Keyword #3:|
We've noticed that there are WFS implementations in NL where the dataset (metadata) can't be mapped to the service, although the metadata URLs (for dataset metadata) per featuretype are provided. In these cases the WFS publishes a code and namespace and the metadata document publishes the dataset identifier with the MD_Identifier.
Download Services are required to publish spatialdataset identifiers in their Capabilities (WFS) or feeds (ATOM). These identifiers exist of either a code or a code+namespace. The TG Download Services maps this as follows (4.1.3 Mapping of Spatial Data Set Identifier parameter, https://inspire.ec.europa.eu/documents/technical-guidance-implementation-inspire-download-services):
#1 Updated by Ine de Visser about 2 years ago
It seems that the Thematic Viewer is quite strict: if a namespace is provided in the Capabilities, the MD_Identifier is ignored. So these WFS implementations are not found for the dataset. Would it be possible that the TV checks for either the MD_Identifier code or the RS_Identifier code+namespace?
I can if needed duscuss this issue Tuesday personal with you, I am in Ispra.
Best regards Ine
#2 Updated by Angelo Quaglia about 2 years ago
- Subject changed from NL MD_Identifier to NL: MD_Identifier
- Description updated (diff)
- Category set to Harvesting results
- Status changed from New to Assigned
- Assignee set to Angelo Quaglia
- Submitting Organisation set to NL
- Country set to NL - The Netherlands
- Originating UI set to Thematic Viewer
#3 Updated by Angelo Quaglia about 2 years ago
without a specific case my answer will have to be a general one.
The word "strict" is rather out of place and here is why:
the definition of an INSPIRE Unique Resource Identifier given by the Metadata Regulation is:
A value uniquely identifying the resource. The value domain of this metadata element is a mandatory character string code, generally assigned by the data owner, and a character string namespace uniquely identifying the context of the identifier code (for example, the data owner).
That means that when you define your naming scheme for your Spatial Data Unique Resource Identifiers, you are free to choose whether to mint them using:
1) a code
2) a code and a namespace
In scenario 1, the code will need to be a value capable of uniquely identifying the resource across the INSPIRE infrastructure, so usually, it is a Uniform Resource Identifier.
In scenario 2, the code can be any free text that uniquely identifies the resource inside the organisation uniquely identified by the namespace (which will likely take the form of a Uniform Resource Identifier).
Of course, if you choose case 2 (code+namespace) the identifier of the dataset is code+namespace, it cannot be just the code, it would not make sense because it would open the door to collisions within the INSPIRE infrastructure.
If, for any reason, you have chosen scenario 2 (code + namespace) but have minted the code in a way that it alone can uniquely identify the resource across the INSPIRE infrastructure (which would raise the question of what you need the namespace for), then I suggest you add an additional MD_identifier to the metadata of your dataset, with only the code.
The INSPIRE Geoportal is smart enough to consider all Unique Resource Identifiers as aliases.
As a sidenote, please be aware that the Thematic Viewer is a client-side application reading data from the INSPIRE Geoportal which performs the matching between Spatial Data (Sets and Series) and the Network Services.
P.S.: You can pass by my office anytime today but before 17.30, becuase today I have to leave early.
#5 Updated by Ine de Visser about 2 years ago
Thanx for your explanation.
On this moment, in the Netherlands most (all) metadata is still based on TG MD v1.3.
In this guideline the description of the Unique resource identifier is "This element is a value uniquely identifying the resource" where only the code property is required.
In the Netherlands most dataproviders provide an UUID in the code property. That seems in accordance with the TG MD.
In TG download services v3.1 the inspire_dls:SpatialDataSetIdentifier/inspire_common:Code is mapped to gmd:MD_Identifier/code in table 13.
the inspire_dls:SpatialDataSetIdentifier/inspire_common:Namespace is not used in case of MD_Identifier, although the namespace is mandatory in TG download requirement 13.
To be sure i understand you correct:
A couple of dataproviders have chosen scenario 2 (code + namespace) in the capabilities, but have minted the code in a way that it alone can uniquely identify the resource across the INSPIRE infrastructure, using an UUID.
Is it possible to map inspire_dls:SpatialDataSetIdentifier/inspire_common:Code, whitout using the provided inspire_dls:SpatialDataSetIdentifier/inspire_common:Namespace to the MD_identifier of the metadata of the dataset, with solely the code?
Or perhaps easier, is it allowed to drop the namespace from the capabilities?
Best regards Ine
#6 Updated by Angelo Quaglia about 2 years ago
Yes, it is possible to omit the namespace element because it is optional, as you can see from the XML schema definitions:
<documentation>Extended capabilities for OGC WFS 2.0 services</documentation>
<element name="SpatialDataSetIdentifier" type="inspire_common:uniqueResourceIdentifier" minOccurs="1" maxOccurs="unbounded">
<documentation>For each spatial data resource served by the Download Service, the most current Unique Resource Identifier shall be listed here </documentation>
<xs:element name="Code" type="notEmptyString"/>
<xs:element name="Namespace" type="notEmptyString" minOccurs="0"/>
<xs:attribute name="metadataURL" type="xs:anyURI"/>