aweek (version 1.0.1)

as.aweek: Convert characters or dates to aweek objects

Description

Convert characters or dates to aweek objects

Usage

as.aweek(x, week_start = get_week_start(), ...)

# S3 method for default as.aweek(x, week_start = NULL, ...)

# S3 method for `NULL` as.aweek(x, week_start = NULL, ...)

# S3 method for character as.aweek(x, week_start = get_week_start(), start = week_start, ...)

# S3 method for factor as.aweek(x, week_start = get_week_start(), ...)

# S3 method for Date as.aweek(x, week_start = get_week_start(), ...)

# S3 method for POSIXt as.aweek(x, week_start = get_week_start(), ...)

# S3 method for aweek as.aweek(x, week_start = NULL, ...)

Arguments

x

a Date, POSIXct, POSIXlt, or a correctly formatted (YYYY-Www-d) character string that represents the year, week, and weekday.

week_start

a number indicating the start of the week based on the ISO 8601 standard from 1 to 7 where 1 = Monday OR an abbreviation of the weekdate in an English or current locale. Note: using a non-English locale may render your code non-portable. Defaults to the value of get_week_start()

...

arguments passed on to date2week() and as.POSIXlt()

start

an integer (or character) vector of days that the weeks start on for each corresponding week. Defaults to the value of get_week_start(). Note that these will not determine the final week.

Value

an aweek object

Details

The as.aweek() will coerce character, dates, and datetime objects to aweek objects. Dates are trivial to convert to weeks because there is only one correct way to convert them with any given week_start.

There is a bit of nuance to be aware of when converting characters to aweek objects:

  • The characters must be correctly formatted as YYYY-Www-d, where YYYY is the year relative to the week, Www is the week number (ww) prepended by a W, and d (optional) is the day of the week from 1 to 7 where 1 represents the week_start. This means that characters formatted as dates will be rejected.

  • By default, the week_start and start parameters are identical. If your data contains heterogeneous weeks (e.g. some dates will have the week start on Monday and some will have the week start on Sunday), then you should use the start parameter to reflect this. Internally, the weeks will first be converted to dates with their respective starts and then converted back to weeks, unified under the week_start parameter.

See Also

"aweek-class" for details on the aweek object, get_aweek() for converting numeric weeks to weeks or dates, date2week() for converting dates to weeks, week2date() for converting weeks to dates.

Examples

Run this code
# NOT RUN {
# aweek objects can only be created from valid weeks:

as.aweek("2018-W10-5", week_start = 7) # works!
try(as.aweek("2018-10-5", week_start = 7)) # doesn't work :(

# you can also convert dates or datetimes
as.aweek(Sys.Date())
as.aweek(Sys.time())

# all functions get passed to date2week, so you can use any of its arguments:
as.aweek("2018-W10-5", week_start = 7, floor_day = TRUE, factor = TRUE) 
as.aweek(as.Date("2018-03-09"), floor_day = TRUE, factor = TRUE)

# If you have a character vector where different elements begin on different
# days of the week, you can use the "start" argument to ensure they are
# correctly converted.
as.aweek(c(mon = "2018-W10-1", tue = "2018-W10-1"), 
         week_start = "Monday", 
         start = c("Monday", "Tuesday"))

# you can convert aweek objects to aweek objects:
x <- get_aweek()
as.aweek(x)
as.aweek(x, week_start = 7)
# }

Run the code above in your browser using DataLab