You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think a user should be able to supplement the metadata that rloadest::loadReg collects with any slot that loadflex::metadata acknowledges that isn't already known to loadReg. This is especially important now that metadata has been extended to include many more slots.
The trick is to do this tidily, so that the user doesn't have to do a whole lot of retyping if they're fitting multiple models. One less-tidy option is to add explicit parameters for all of the metadata slots that aren't already collected by loadReg, e.g.,
loadReg2<-function(load.reg,
pred.format=c("flux","conc"),
store=c("data","fitting.function"),
site.id, lat, lon, basin.area, # etc. - add all the slots that loadReg doesn't have...) {
but this could get unweildy. So another option might be to accept a partial metadata object in lieu of all those independent objects, from which only the non-loadReg slots will be used...
but this will require careful communication to the user and/or some way of preventing them from trying to override the slots that loadReg does know about. So this option requires some thinking, too.
For any solution we choose, the idea should be to get the non-loadReg slots optionally filled in.
I think a user should be able to supplement the metadata that
rloadest::loadReg
collects with any slot thatloadflex::metadata
acknowledges that isn't already known toloadReg
. This is especially important now thatmetadata
has been extended to include many more slots.The trick is to do this tidily, so that the user doesn't have to do a whole lot of retyping if they're fitting multiple models. One less-tidy option is to add explicit parameters for all of the metadata slots that aren't already collected by
loadReg
, e.g.,but this could get unweildy. So another option might be to accept a partial metadata object in lieu of all those independent objects, from which only the non-loadReg slots will be used...
but this will require careful communication to the user and/or some way of preventing them from trying to override the slots that loadReg does know about. So this option requires some thinking, too.
For any solution we choose, the idea should be to get the non-loadReg slots optionally filled in.
Here are the slots that
loadReg
tracks:https://github.com/USGS-R/loadflex/blob/master/R/loadReg.R#L31-L39
and here are all the slots in
loadflex::metadata
:https://github.com/USGS-R/loadflex/blob/master/R/metadata.R#L68-L94
The text was updated successfully, but these errors were encountered: