Re_order based on month and date while ignoring year in R

Lubridate doesn't have a month-date col type it appears and I would like to make a histogram across several years that focuses just on the month and day while ignoring the year. But when it comes time to sort it will think 1-10 comes after 1-1 and so on. I'm sure there's an easy fix to this but I can't seem to find it.

Here's some sample code if necessary:

structure(list(date = structure(c(19512, 13135, 14242), class = "Date"), 
    sport = c("MLB", "NFL", "NHL")), row.names = c(NA, -3L), class = c("tbl_df", 
"tbl", "data.frame"))

Solution

You can arrange using month and date functions:

library(tidyverse)

df = structure(list(date = structure(c(19512, 13135, 14242), class = "Date"), 
                    sport = c("MLB", "NFL", "NHL")), row.names = c(NA, -3L), 
               class = c("tbl_df", "tbl", "data.frame"))

df %>% arrange(month(date), day(date))
#> # A tibble: 3 × 2
#>   date       sport
#>   <date>     <chr>
#> 1 2023-06-04 MLB  
#> 2 2005-12-18 NFL  
#> 3 2008-12-29 NHL

^{Created on 2024-07-17 with reprex v2.0.2}

UPDATE

To arrange the a plot by month and day without considering year you can make another date with the same year and use month and day only to label the breaks.

library(tidyverse)

df = structure(list(date = structure(c(19512, 13135, 14242), class = "Date"), 
                    sport = c("MLB", "NFL", "NHL")), row.names = c(NA, -3L), 
               class = c("tbl_df", "tbl", "data.frame"))

df$dateMD = df$date
year(df$dateMD) = 2001
df
#> # A tibble: 3 × 3
#>   date       sport dateMD    
#>   <date>     <chr> <date>    
#> 1 2023-06-04 MLB   2001-06-04
#> 2 2005-12-18 NFL   2001-12-18
#> 3 2008-12-29 NHL   2001-12-29

ggplot(df, aes(dateMD, 1:3)) +
  geom_col() +
  scale_x_date(name = "date", date_labels = "%b-%d", date_break = "1 month")

^{Created on 2024-07-17 with reprex v2.0.2}

UPDATE 2

If you want to change the year within dplyr chain check this question.