The Standard API expects to find a metadata.yaml file in the data repository, but it does not add entries to the file when data is written through the pipeline. It would be very convenient to be able to populate a repository through the Python API in a way which allowed the Python API to read the data back. This would require writing new entries to the metadata.yaml file. This could be made optional (if necessary). When data products are replaced, the entries should probably be updated.
The Standard API expects to find a metadata.yaml file in the data repository, but it does not add entries to the file when data is written through the pipeline. It would be very convenient to be able to populate a repository through the Python API in a way which allowed the Python API to read the data back. This would require writing new entries to the metadata.yaml file. This could be made optional (if necessary). When data products are replaced, the entries should probably be updated.