If the commit hash of the the last commit in a previous experiment (when the experiment ID last changed) were saved in metadata.yaml it would make it trivially easy to specify the range of commits that constitute an experiment. The refA..refB syntax used for selecting a range of git commits automatically excludes the initial commit, refA in this example.
Otherwise you would need to walk the commit history of metadata.yaml in a repo looking for changes.
Would not be saved if not known (so field would be absent).
Automated
metadata.yaml
creation is added in this PR https://github.com/payu-org/payu/pull/384If the commit hash of the the last commit in a previous experiment (when the experiment ID last changed) were saved in
metadata.yaml
it would make it trivially easy to specify the range of commits that constitute an experiment. TherefA..refB
syntax used for selecting a range of git commits automatically excludes the initial commit,refA
in this example.Otherwise you would need to walk the commit history of
metadata.yaml
in a repo looking for changes.Would not be saved if not known (so field would be absent).