Dear WG members,
The Metadata Standards Catalog has reached another milestone, which we're marking by bumping the version number up to 2.1. The significance of this is that the overhaul of the Catalog code is complete, and it can now do everything the v1 Catalog could do, but in a way that is more complete, more robust, more maintainable, more adaptable, and easier for other systems to work with.
The main changes from v2.0 are in the API:
- The GET endpoints at the collection level for the main entities (metadata schemes, organizations, tools, crosswalks , endorsements), entity relationship records, and folksonomic vocabularies (data types, entity types, location types, and ID schemes) now accept a q parameter which filters the results using a subset of Lucene syntax:
- Use q=term to get back only records that contain “term” somewhere in them.
- Use q=title:"term phrase" to get back only records that contain “term phrase” in the “title” field.
- Use q=versions.issued:[2015-01 TO 2016-06] to get back only records where one of the versions has an issue date in the range of January 2015 to June 2016 inclusive.
- Parentheses and boolean AND, OR and NOT are supported for more complex searches, e.g. q=title:(term AND (NOT "some terms")) OR description:standard.
- There are new endpoints at /api2/user/token for requesting a time-limited access token and /api2/user/reset-password for changing an existing password. You need to have a dedicated API account set up to use these.
- With an API account you can
- add new main entity records and folksonomic terms with POST requests to /api2/m, /api2/g, /api2/dataype, and so on;
- update or delete (subject to restrictions) a main entity record or folksonomic term with PUT or DELETE requests to /api2/m1, /api2/datatype1, and so on;
- add or update the relationships between main entity records with POST or PUT requests to /api2/rel/m1, api2/rel/t1, /api2/rel/c1, and so on.
- incrementally edit the relationships between main entity records using JSON Patch syntax with PATCH requests to /api2/rel/m1, /api2/rel/t1, /api2/invrel/g1, and so on.
In response to a user request, we have also started collecting information about Linked Data namespaces used by metadata schemes. The record for Dublin Core shows this in action.
Now these features are in place, our focus can return to making sure the content is complete, up-to-date and relevant. But as a community-maintained resource we do welcome and rely on your contributions, so if you spot a problem you can fix, don't be shy! Also if you have feature requests for the Catalog, please let me know or leave an issue on GitHub and I'll see what we can do.