Access via API the connected data of a variable

[treerink]

In the Airtable v1.2.1 I can click a tab like Variables , Experiments or MIPs and I see columns for each entry. I would like to be able to access this data via the python API. For instance for each experiment I am able request the minum of years to run this experiment (that's nice). For each MIP I can request its name, but not its longname (error) while longname is an attribute. For each experiment I can not request the MIP it belongs to (it does even not have MIP as attribute currently), but this is what I would appreciate (ok I made now manually a dictionary for that to circumvent, but that is maintenance unfriendly).

So it would be nice to be able to request via the python API:

• Given an experiment, return the corresponding MIP

• Given an experiment, return the (default) parent experiment

• Given a MIP, return the list of experiments for this MIP

• Given a MIP, return th elist of experiments of this MIP

• In general: being able to reach all the metadata / attributes / info provided in the AIRtable given
  a selected entity like: experiment, MIP, Physical Parameter, etc. Maybe all is a lot? But starting with the most frequent / handiest ones?

Or:

• Possibe to have a full dump of the AIRTables in a json or xml file

Below a short code which demonstrates the issue mentioned above:

#!/usr/bin/env python

# Based on the ../notebooks/HowTo-0*ipynb examples

from data_request_api.content import dump_transformation as dt
from data_request_api.query import data_request as dr

content_dic = dt.get_transformed_content()
DR = dr.DataRequest.from_separated_inputs(**content_dic)

# View attributes of a given Experiment
all_experiments = DR.get_experiments()
all_mips        = DR.get_mips()

print('\n CMIP7 MIPs and attributes:')
nr_of_mips = 0
for mip in all_mips:
 if nr_of_mips < 2:                   # Reducing the output for now
  for k, v in mip.attributes.items():
   print('  {:30} {}'.format(k, v))
  print('\n')
  nr_of_mips = nr_of_mips + 1

for mip in all_mips:
 print('  {:14}      '.format(str(mip.attributes['name'])                                                                ))
#print('  {:14} {} {}'.format(str(mip.attributes['name']), mip.attributes['mip_long_name'], mip.attributes['mip_website']))  # DOES NOT WORK currently

print('\n CMIP7 experiment name            minimum duration (year):')
for experiment in all_experiments:
 print('  {:34} {:4}   '.format(str(experiment.attributes['name']), experiment.attributes['size_(years,_minimum)']                              ))
#print('  {:34} {:4} {}'.format(str(experiment.attributes['name']), experiment.attributes['size_(years,_minimum)'], experiment.attributes['MIP']))  # DOES NOT WORK currently
print('\n')

[JamesAnstey]

Hi (again) @treerink - the dreq content export to a json file, which is what the software loads, is in fact a full dump of the airtable content, so all this info is available - but I guess your question is about how easy it is to find and use it.

One way is via the specific query functions illustrated in scripts/workflow_example_2.py and in the technical documentation. We can try to functions supporting specific queries if they aren't already in there.

Another is to use objects from data_request_api.query.dreq_classes, which render the Airtable tables & records as python objects. The records have attributes corresponding to the table columns. E.g. a code snippet getting experiments for an opportunity in dq.get_opp_expts:

    for link in opp.experiment_groups:
        expt_group = expt_groups.records[link.record_id]

where opp is an object representing a record from the Opportunities table in Airtable, opp.experiment_groups lists its links to Experiment Groups, and expt_group is an object representing a record in the Experiment Groups table.