Read data from ixmp v4.0

[danielhuppmann]

Per our discussion on the development of ixmp v4.0, we agreed to implement a new method pyam.read_ixmp() that queries data and meta indicators (and other dimensions of the data model in the future) from an ixmp Platform instance.

The function should take the following arguments:

• An ixmp.Platform instance or arguments to instantiate an instance (connect to a database)
• Arguments to select runs (model, scenario, version, run-id's) - if not given, read all runs
• Arguments to select IAMC-format datapoints (variable, region, time)
  • fail if not provided (*)
  • require variable = * to read all datapoints
• Arguments to select meta indicators - retrieve none if not provided

These arguments can also be provided as a yaml-file, so there should also be a kwarg config (or similar).

() An IamDataFrame cannot be instantiated without data*.

fyi @meksor @peterkolp

[danielhuppmann]

The implementation strategy should be something like

@classmethod
def read_ixmp(pyam.IamDataFrame, platform, **kwargs):

    platform = platform if isinstance(mp, ixmp.Platform) else ixmp.Platform(** platform)

    run_ids = < select run-id's >
    data = < query data based on run_id's and kwargs >

    if meta in kwargs:
        meta =  < query meta based on run_id's and kwargs >
    else:
        meta = None

    return IamDataFrame(data, meta=meta)

[danielhuppmann]

Closing as implemented