Expand monthly end-of-month values to all days or hours of the same month — fill_month_to

For each row in df, this function assumes that the time column contains a monthly value at (or near) the end of a month (e.g. "2002-01-31" or "2002-01-31 23:00:00") and expands it backwards to all days or all hours from the 1st of that month up to the given timestamp.

Usage

fill_month_to_start(
  df,
  date_col = "date",
  datetime_col = "datetime",
  temporal_res = c("auto", "days", "hours"),
  date_min = NULL,
  date_max = NULL
)

Arguments

df: A data frame or tibble.
date_col: Name of the daily date column (default: "date").
datetime_col: Name of the hourly datetime column (default: "datetime").
temporal_res: One of "auto", "days", "hours". If "auto", the presence of date_col / datetime_col decides.
date_min: Optional lower bound of the output time window. For daily mode this is interpreted as Date (or converted via as.Date()); for hourly mode it can be a POSIXct or Date. If NULL, no lower cropping is applied.
date_max: Optional upper bound of the output time window. Same interpretation as date_min. If NULL, no upper cropping is applied.

Value

A tibble where each original row is expanded so that all days (for "days") or all hours (for "hours") from the 1st of the respective month up to the original timestamp are present. All other columns are copied to the created rows. The result is then cropped to [date_min, date_max] if those are supplied. The name of the time column is preserved ("date" or "datetime").

Details

If temporal_res = "auto" (default), the function chooses daily expansion when a column date is present and hourly expansion when a column datetime is present.

The resulting series can optionally be cropped to a user-defined time window via date_min and date_max.