Multiple breaking changes:
mutate loses the index argument to match
the exact semantics of Spark SQL.mutate is single-threaded. Users are advised to use the
tidyr::nest + parallel map (furrr::future_map)
+ tidyr::unnest pattern when there are many groups and
parallelization is necessary.Enhancements:
tidier‘s mutate now supports same syntax over ’dbplyr’
tbls..complete argument in
tidier::mutatemutate can now modify a column (same name) in
sliding operation.NEWS.md file to track changes to the
package.