We could handle the dataset management by implementing two features:
[ ] add a "cache dataset" button, that allows to access a given dataset from every folder. This would resemble the "fav" button allowing the add datasets to the permanent favorite list. The difference here would be that the dataset would be copied to a temporary folder, that is emptied every time HAL starts. This way, we make sure that the list of "cached" dataset does not grow too much with time. In this case, the "fav" button is only meant to keep datasets that are really useful on a long term basis (for instance, calibration runs or a collection of runs that should be updated from time to time to keep track of some experimental parameter)
[ ] allow the user to display datasets created from another day. This would allow to access all datasets from, say, the day before, in order to compare the current data with some reference. The list of displayed days would be reset at startup
We could handle the dataset management by implementing two features: