These are variants of existing functions that are designed to retain the pibble status of the object, as well as its .i, .t, and .d attributes.
# S3 method for tbl_pb
mutate(.data, ...)# S3 method for tbl_pb
distinct(.data, ..., .keep_all = FALSE)
# S3 method for tbl_pb
group_by(.data, ...)
# S3 method for tbl_pb
ungroup(x, ...)
# S3 method for tbl_pb
select(.data, ...)
# S3 method for tbl_pb
rename(.data, ...)
# S3 method for tbl_pb
summarize(.data, ...)
# S3 method for tbl_pb
summarise(.data, ...)
# S3 method for tbl_pb
transmute(.data, ...)
These functions take a tbl_pb (i.e. pibble) object as input
Other parameters to be passed to the relevant functions
Some functions that already preserve pibble status and so don't need special methods include:
dplyr::add_row(), tibble:add_column(), dplyr::arrange(), dplyr::bind_cols(), dplyr::filter(), dplyr::sample_frac(), dplyr::sample_n(), dplyr::slice(), dplyr::top_n
as well as all scoped variants (_all, _if, _at) of dplyr functions.
dplyr::bind_rows() is currently not supported. If you use dplyr::bind_rows() you should pipe it to as_pibble().
Any function that takes two data frames/tibbles as inputs will retain the panel structure of the first argument.
If a function is not on the above list or elsewhere in this help file, then you may need to re-as_pibble your object after using the function.