group_by_period: Aggregate a numeric column over a specified time grouping.
In louis-vines/tidyutils: An Implementation of Common Patterns in the Tidyverse

group_by_period takes a date column and a numeric column and aggregates the numeric column over specified date period. This is a common pattern in data analysis in business where we want to convert transactional data to daily, weekly or monthly data.

1 2	group_by_period(.data, .date_col, .summarise_col, unit, agg_fun = sum, na.rm = FALSE)

`.data`	A `data.frame` to apply `group_by_period` to
`.date_col`	A date column that will be used for grouping the date. This should be of type `date` or `datetime`
`.summarise_col`	The column which will be aggregated
`unit`	A character string specifying a time unit or a multiple of a unit to be rounded to. See `?lubridate::floor_date` for more details
`agg_fun`	A function used to aggregate the numeric column
`na.rm`	`NA` handling - this is passed to the `na.rm` argument of the aggregating function

library(nycflights13)
library(tidyr)
library(dplyr)

data("flights")

flights_w_date <- flights %>%
  unite(flight_date, year, month, day, sep = "-") %>%
  mutate(flight_date = ymd(flight_date))

flights_w_date %>%
  group_by_period(flight_date, distance, "month")