search

inspire-eu-validation / ets-repository

Repository for Executable Test Suites of the INSPIRE validator (under development)
European Union Public License 1.2
13 stars 25 forks source link

Validation of metadata for interoperability - some tests just apply to specific themes #95

Open dgterritorio opened 7 years ago

dgterritorio commented 7 years ago

Some of the tests listed for validation of metadata for interoperability just apply to some themes, like "Topological Consistency" just apply to TN or HY networks. This could lead to confusion on users less informed.

cportele commented 7 years ago

The problem is that the requirement for "Topological Consistency" applies whenever "the data set includes types from the Generic Network Model and does not assure centreline topology (connectivity of centrelines) for the network."

There is no way to reliably determine this information from the metadata record. As suggested, one could attempt to determine the theme via the keywords compare the keywords against all theme names in all official languages, but this also does not identify all the datasets to which the pre-condition applies (e.g. it does not apply for a hydrography dataset with physical waters).

This is a question to the MIG whether the abstract test case should be amended to (try to) determine the theme and report the manual check only for the themes that include types derived from the Generic Network Model.

michellutz commented 7 years ago

It might be enough to specify explicitly (in the ATS) the themes, application schemas (or even better: Spatial Object Types) that would need to be included in the data set to require this metadata element. There are not that many types / application schemas.

That might already be sufficient to limit the confusion.

cportele commented 7 years ago

@michellutz - that list would not really help as the test is not against data (where we could check the spatial object types included), but against a metadata record and the only aspect that the metadata covers (not sure how reliable in practice) is the themes covered by the dataset.

michellutz commented 7 years ago

@cportele I know. I meant to include the list as a note for testers to understand when they have to do the manual test, rather than to use it in an automatic procedure (to decide when to flag this for manual testing or not).

cportele commented 7 years ago

Ah ok, I think for now we could do this already just in ETS and add this as a comment in the ATS for future consideration (probably as part of the ATS for the MD TG v2.0).