The recent climatology updates (!184) had a bug when there are multiple history file for a component. e.g.
- type: 'atmos_scalar'
sources:
- history_file: "atmos_global_cmip"
- history_file: "atmos_scalar"
climatology:
- frequency: yr
interval_years: *CLIMO_YEARS
- frequency: mon
interval_years: *CLIMO_YEARS
postprocess_on: True
In an input configuration like that, lines 77-100 in get_climatology_info.py (https://github.com/NOAA-GFDL/fre-workflows/blob/main/Jinja2Filters/get_climatology_info.py#L77-L100) has two bugs.
First, the closing graph string graph += f"\"\"\"\n" should be outside the for loop, not within.
Second, the opening task for a second (or more) history file for the same component should not start with &.
The recent climatology updates (!184) had a bug when there are multiple history file for a component. e.g.
In an input configuration like that, lines 77-100 in get_climatology_info.py (https://github.com/NOAA-GFDL/fre-workflows/blob/main/Jinja2Filters/get_climatology_info.py#L77-L100) has two bugs.
First, the closing graph string
graph += f"\"\"\"\n"should be outside the for loop, not within.Second, the opening task for a second (or more) history file for the same component should not start with
&.