Skip to contents

Keep only rows whose period is complete as of a cutoff

Source: R/period_completeness.R
filter_complete_periods.Rd

[Experimental]

Filters data to the rows whose period (per period_is_complete()) is complete as of as_of. Useful for dropping the in-progress trailing period from a summary or year-over-year comparison.

Usage

filter_complete_periods(
  data,
  date_col,
  unit = "month",
  as_of = NULL,
  calendar = "WeekendsOnly",
  week_start = getOption("lubridate.week.start", 7)
)

Arguments

data

A data frame.

date_col

<data-masked> The date column whose period completeness is tested.

unit

Period size, one of "week", "month", "quarter", or "year".

as_of

Reference cutoff date, a single Date (or coercible). Defaults to NULL, in which case it is set to the maximum of date_col so completeness is judged against how current the data actually is. An explicit as_of earlier than the maximum of date_col is an error, because it would drop periods the data demonstrably completes; use period_is_complete() directly for point-in-time completeness.

calendar

(character) A QuantLib calendar id (see qlcal::calendars). Defaults to "WeekendsOnly".

week_start

Integer 1-7 giving the first day of the week (only used when unit = "week"). Defaults to getOption("lubridate.week.start", 7).

Value

data filtered to rows in complete periods.

See also

period_is_complete()

Examples

df <- data.frame(day = seq(as.Date("2025-04-01"), as.Date("2025-06-12"), by = "day"))
# as_of defaults to max(day) = 2025-06-12, so the partial June is dropped
nrow(filter_complete_periods(df, day, unit = "month"))
#> [1] 61