ceda-datapoint is a Python package which provides python-based search/access tools for using data primarily from the CEDA Archive. For some time we've been generating so-called Cloud Formats which act as representations, references or mappers to data stored in the CEDA Archive. Most of our data is in archival formats like NetCDF/HDF which makes them great for use with the HPC architecture on which the archive resides (see the JASMIN homepage for more details), but not so good for open access outside of JASMIN.
See the documentation at https://cedadev.github.io/datapoint for more information.
The DataPoint module is now an installable module with pip!
pip install ceda-datapoint
See the documentation for a more in-depth description of how to run a search query and access data.
from ceda_datapoint import DataPointClient
client = DataPointClient(org='CEDA)
# Continue to perform searches and access data