I am not sure how useful this is, but I'd like to update the list in default_diagnostics.py. Currently, it contains the content of available_diagnostic.log of an LL90 run at some point, so it cannot be complete. I am at least as lazy as everyone else, so I don't want to generate the list of diagnostics "by hand". Instead, I suggest to use something like this in MITgcm/verification (after running testreport):
cat */*run*/available_diagnostics.log | awk -F'|' '!seen[$2]++' | sort -u -t "|" -k 2
(ChatGPT gave me the awk command ). This gives an alphabetical list of all diagnostics in the verification experiments, approximately 1136. default_diagnostics.py has about 311 lines. It would probably be slower to parse the longer list, also the long list is not in any order that MITgcm users are used to and the numbering no is correct.
Does this make sense at all?
I am not sure how useful this is, but I'd like to update the list in
default_diagnostics.py. Currently, it contains the content ofavailable_diagnostic.logof anLL90run at some point, so it cannot be complete. I am at least as lazy as everyone else, so I don't want to generate the list of diagnostics "by hand". Instead, I suggest to use something like this inMITgcm/verification(after runningtestreport):(ChatGPT gave me the
awkcommand ). This gives an alphabetical list of all diagnostics in the verification experiments, approximately 1136.default_diagnostics.pyhas about 311 lines. It would probably be slower to parse the longer list, also the long list is not in any order that MITgcm users are used to and the numbering no is correct.Does this make sense at all?