byexistingruns split type
Basic idea: take the base dataset, subtract snapshot of used files, pick a file from it, get its run number from metadata, make dataset defname:base_dataset and run_number x.
Use the snapshot-based state we are using in drainingx and stagedfiles.
[Also, make by_run skip empty runs.]
#1 Updated by Marc Mengel over 1 year ago
- Subject changed from samerun split type to byexistingruns split type
- Status changed from New to Work in progress
- Target version changed from v4_1_0 to v4_0_2