Validation of CRS element from ISO 19115 community profile XML document

[enricoboldrini]

Error returned: The metadata record does not specify the coordinate reference system(s) used in the dataset.

I'm validating a SeaDataNet XML document, that is a community profile of ISO 19115 metadata (https://www.seadatanet.org/Standards/Metadata-formats/CDI), drafted according to official ISO 19115 rules.
Actually the CRS is specified in the document, at XPath: gmd:referenceSystemInfo/gmd:MD_ReferenceSystem/gmd:referenceSystemIdentifier/gmd:RS_Identifier/gmd:code/sdn:SDN_CRSCode
Most likely, the test expects gco:CharacterString instead of the codelist sdn:SDN_CRSCode that is a community defined code list.

• Test Report: http://inspire-sandbox.jrc.ec.europa.eu/etf-webapp/v2/TestRuns/EIDcb80f9ea-9d37-463a-9aaf-b45806194b35.html
• Assertion URI: http://inspire-sandbox.jrc.ec.europa.eu/etf-webapp//v2/TestRuns/EIDcb80f9ea-9d37-463a-9aaf-b45806194b35.html?lang=en#EID4b3bf91c-2498-4b4f-bf1e-d8bbb22b5264
• Referenced file: https://s3.amazonaws.com/essi.documents/seadatanet-test-2.xml

[enricoboldrini]

Hello, I've noticed the planned tag has been added. Can you elaborate more on the planning schedule for this issue? Thank you in advance. Please let me know if more info are needed.

[enricoboldrini]

In the meantime an updated version of SeaDataNet community profile has been released. Here are the updated test reports:

• Test Report: http://inspire-sandbox.jrc.ec.europa.eu/etf-webapp/v2/TestRuns/EID1f2bbb8d-7789-42d8-b9b6-27ff5a23337c.html
• File used for validation: https://www.seadatanet.org/content/download/2602/file/CDI_ISO19139_full_example_11.0.0.xml

[MarieLambois]

Again an issue with extensions. There is no interest in testing that the type is gco:CharacterString. It could be many other things (such as Anchor). The test should be made on:
gmd:referenceSystemInfo//gmd:referenceSystemIdentifier//gmd:code/*

[AntoRot]

No reference to the use of gco:CharacterString is made in the TG req 2.1, but only to the child elements of gmd:RS_Identifier 'gmd:code' (mandatory) and 'gmd:codeSpace' (optional). So, I agree with @MarieLambois that the test should be made on gmd:referenceSystemInfo//gmd:referenceSystemIdentifier//gmd:code/* and in case on gmd:referenceSystemInfo//gmd:referenceSystemIdentifier//gmd:codeSpace/* consistently with the mentioned TG req 2.1.

[enricoboldrini]

This is now solved, thank you! However the value in the metadata document is: "World Geodetic System 84", and as I understand only URI identifiers from the table at https://inspire.ec.europa.eu/documents/Data_Specifications/INSPIRE_DataSpecification_Template_v3.0rc3.pdf are considered valid (e.g. http://www.opengis.net/def/crs/EPSG/0/4258). Are there any possibilities that this constraint will be relaxed in the near future, in particular to accomodate EPSG:4326?

[michellutz]

I am a bit surprised that the ATS and ETS for the interoperability metadata still has the checks for specific CRSs in. We agreed a long time ago (see https://themes.jrc.ec.europa.eu/discussion/view/64808/what-crss-to-document-in-metadata-when-serving-a-dataset-through-a-wfs) that the obligation on metadata is only to provide the CRS metadata element (with the source CRS the data set is originally published in).

The obligation to publish the data in an INSPIRE CRS is checked in this data test: https://github.com/inspire-eu-validation/data/blob/3.0rc3/reference-systems/spatial.md

Also in the MD v2.0 tests, the CRSs are not checked for this element. So I think the test in https://github.com/inspire-eu-validation/data/blob/master/interoperability-metadata/coordinate-reference-system.md and the corresponding ETS should be updated (to remove the check for specific CRSs).

[AntoRot]

I checked the file used for validation and I noticed that the error returned on the CRS metadata element doesn't concern the value of the identifier (see Assertion URI http://staging-inspire-validator.eu-west-1.elasticbeanstalk.com/etf-webapp/v2/TestRuns/EID94ada6ca-9754-42f5-a86c-4bfe4904f99e.html?lang=it#EIDea254806-f3f2-402a-b5ab-ce559989163a), but it concerns the test https://github.com/inspire-eu-validation/metadata/blob/2.0/isdss/crs.md on the existence of at least one RS_Identifier element and one Code element (see Assertion URI http://staging-inspire-validator.eu-west-1.elasticbeanstalk.com/etf-webapp/v2/TestRuns/EID94ada6ca-9754-42f5-a86c-4bfe4904f99e.html?lang=it#EIDd06cc736-39c8-4828-88dc-30a62d4ee09f), although those elements are included in the record.
I think that is due to the check on the presence of gco:CharacterString or gmx:Anchor as child elements of the Code element, while in the record a code list is used. So, the test should be relaxed consequently.