Regression on Carboncycle output

[rgieseke]

@znicholls Do we have a regression on https://github.com/openclimatedata/pymagicc/issues/84?

I don't think i can read them in the latest version and i don't understand the xfail introduced in 548ee587

@pytest.mark.xfail(
    reason="Not sure how best to fail reading CARBONCYCLE.OUT at the moment"
)

[rgieseke]

In [1]: import pymagicc                                                                                                                                                                                     

In [2]: pymagicc.__version__                                                                                                                                                                                
Out[2]: '2.0.0-alpha.1+9.gd922396'

In [3]: out = pymagicc.run(pymagicc.rcp26, out_carboncycle=1)      
In [5]: out.variable.unique()                                                                                                                                                                               

Out[5]: 
array(['Atmospheric Concentrations|CO2 Equivalent',
       'Emissions|CO2|MAGICC Fossil and Industrial',
       'Atmospheric Concentrations|CH4', 'Surface Temperature',
       'HEATUPTAKE_EBALANCE_TOTAL', 'Emissions|CO2|MAGICC AFOLU',
       'Atmospheric Concentrations|CO2', 'Atmospheric Concentrations|N2O',
       'Emissions|BC|MAGICC Fossil and Industrial',
       'Radiative Forcing|Greenhouse Gases', 'Radiative Forcing',
       'Emissions|BC|MAGICC AFOLU', 'KYOTOGHGS_GWPEMIS'], dtype=object)

[znicholls]

Do we have a regression on #84?

I think I got rid of it

I don't think i can read them in the latest version and i don't understand the xfail introduced in 548ee58

The problem with the carbon cycle output files is that they don't have units and I don't know what most of the output variables actually are (hence can't translate them to OpenSCM variables). I'll send an email now to see if I can find out.

[znicholls]

In MAGICC7, we're adding the various carbon cycle outputs back in, but as individual outputs (e.g. you can request DAT_CO2AIR2LANDFLUX as an output variable) instead of using CARBONCYCLE.OUT. Hence this should be sorted with the release of MAGICC7.