The clipped collections/processes are loaded with the identifier in the id property, and the algorithm filters on these. But when one wants to know more about the results and expands the Backend component's Collections or Processes section, the full dataset is fetched from the API, which returns the raw data. In case of openEO v0.3.x backends, this raw data has the identifiers in the name property. This causes the filtered result to disappear from the results list because the filter condition (id contains one of the wanted identifiers) isn't met anymore (because the identifier is now contained in name).
This was the quick fix. The nicer way would be to always serve migrated data from the API - but to do so, all components have to be told that the data is already migrated...
This applies only to openEO API v0.3.x backends.
The clipped collections/processes are loaded with the identifier in the
id
property, and the algorithm filters on these. But when one wants to know more about the results and expands theBackend
component's Collections or Processes section, the full dataset is fetched from the API, which returns the raw data. In case of openEO v0.3.x backends, this raw data has the identifiers in thename
property. This causes the filtered result to disappear from the results list because the filter condition (id
contains one of the wanted identifiers) isn't met anymore (because the identifier is now contained inname
).