Open mnlevy1981 opened 4 years ago
@matt-long points out that we could do
[path to catalog1]:
experiment: experiment1
member_id: ###
[path to catalog2]:
experiment: experiment1
member_id: ###
[path to catalog3]:
experiment: experiment2
member_id: ###
Note that this requires us leaving case
in when addressing NCAR/cesm-catalog#2
This leaves the individual catalogs we've created as the primary keys, which makes sense from a "building upon existing catalogs" viewpoint
maybe above would need to allow something like
[path to catalog1]:
experiment: experiment1
member_id: ###
[path to catalog2]:
experiment: experiment1
member_id: ###
[path to catalog3]:
experiment: experiment2
member_id: ###
ctrl_experiment: experiment1
ctrl_member_id: ###
to override trying to explicitly determine the ctrl_
values
Is this approach assuming that the paths to catalogs are going to be unique?
Yes.
I think a YAML file like
would define the ensemble well enough. Each individual catalog would be read in, the experiment and member_id columns would be added, and the
ctrl_case
column would be replaced withctrl_experiment
andctrl_member_id
(assumingctrl_case
is also a member of the ensemble). Then all the individual catalogs would be concatenated into one giant file (we'd need to replace relative path names with absolute ones at this stage).