To avoid publication pipelines having to interface with parts of the CDDS code we should add a CLI support the extraction of files for a single dataset. The interface should be something similar to:
cdds_retrieve_dataset action dataset_id destination [--create-directories] [--mass-root=MASS_ROOT]
action = get or ls. “ls” will return location, list files and sizes plus checksums (JSON?)
dataset_id = dataset id
destination = location on disk
--create-directories = directory structure corresponding to DRS is created
--mass-root = root path in MASS (defaults to usual location)
Return codes
0: success
1: location not found
2: mass credentials expired
3: other error
Where the dataset_id would be of the form
MIP-DRS7.CMIP7.CMIP.UKNCSP.UKESM1-3-LL.historical.r1i1p1f1.glb.mon.tas.tavg-h2m-hxy-u.g108.v20260815
To avoid publication pipelines having to interface with parts of the CDDS code we should add a CLI support the extraction of files for a single dataset. The interface should be something similar to:
cdds_retrieve_dataset action dataset_id destination [--create-directories] [--mass-root=MASS_ROOT]
action = get or ls. “ls” will return location, list files and sizes plus checksums (JSON?)
dataset_id = dataset id
destination = location on disk
--create-directories = directory structure corresponding to DRS is created
--mass-root = root path in MASS (defaults to usual location)
Return codes
0: success
1: location not found
2: mass credentials expired
3: other error
Where the dataset_id would be of the form
MIP-DRS7.CMIP7.CMIP.UKNCSP.UKESM1-3-LL.historical.r1i1p1f1.glb.mon.tas.tavg-h2m-hxy-u.g108.v20260815