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
Updating just documentation for data objects currently requires:
Updating docs in epidatasets.
Re-documenting downstream packages with latest epidatasets, as downstream packages inherit documentation.
Every developer pulling epidatasets to not reverse doc updates.
(Maybe some other things? Adjusting version bounds doesn't seem strictly necessary since it's just documentation, but might assist in forcing developers to stay in sync.)
Ideas for simplifying:
Replace downstream docs with just a single data re-exports topic with links to epidatasets documentation for each object. We get to keep the note about data(<obj name>) (and could add one about needing epidatasets attached for listing its data sets with just data()).
Remove downstream docs; just @export and that's it. ?/help should route to epidatasets documentation [since we Imports: it], assuming we have identical object naming. We lose the notes about not being able to use data(<obj name>), but that's an dated paradigm anyway, so hopefully not a huge loss.
Don't re-export at all downstream. This requires adding epidatasets:: and/or library(epidatasets) in various downstream examples and vignettes.
The text was updated successfully, but these errors were encountered:
I'm trying out the only-@export approach in cmu-delphi/epiprocess#564 because CHECK was complaining about {epiprocess} in the copied documentation. (Trying to parse it as .Rmd stuff.)
Migrating from memory of meeting.
Updating just documentation for data objects currently requires:
Ideas for simplifying:
data(<obj name>)
(and could add one about needing epidatasets attached for listing its data sets with justdata()
).@export
and that's it.?
/help
should route to epidatasets documentation [since weImports:
it], assuming we have identical object naming. We lose the notes about not being able to usedata(<obj name>)
, but that's an dated paradigm anyway, so hopefully not a huge loss.epidatasets::
and/orlibrary(epidatasets)
in various downstream examples and vignettes.The text was updated successfully, but these errors were encountered: