Our diffs have enough information to be reusable. The same diff format could be used to patch a dataset, and this might be an elegant way of recording a set of custom changes against an official version of a dataset.
Ideally, you wouldn't need to specify the set of columns used in the patch. Perhaps the diff format could include the column names in a key property.
Our diffs have enough information to be reusable. The same diff format could be used to patch a dataset, and this might be an elegant way of recording a set of custom changes against an official version of a dataset.
Ideally, you wouldn't need to specify the set of columns used in the patch. Perhaps the diff format could include the column names in a
key
property.