Edit string value based on a condition / another column?

Question

I'm cleaning a large dataset and have a Comments column where I am commenting on changes made to the data. I've provided a dummy sample set below as an example of what I'm trying to achieve. I'm using tidyverse packages.

Data:

structure(list(Date = structure(c(17199, 17226, 17263, 17300, 
17346, 17504, 17508), class = "Date"), Skipper = c("Agatha", 
"Gertrude", "Julio", "Dylis", "Agatha", "Dylis", "Julio"), Success = c("No", 
"Yes", "Yes", "Yes", "No", "Yes", "No"), Time = c(60L, 50L, 120L, 
30L, 100L, 120L, 40L), Comments = c("Pirates spotted.", "Illegal fishers spotted.", 
"Engine troubles.", "Lost fishing line.", NA, "Pirates spotted.", 
"Lost fishing line.")), class = "data.frame", row.names = c(NA, 
-7L))

I'm looking to add text to string values in the Comments in relation to Date, without deleting values already present.

So for 2017-04-07 and 2017-12-04 I would like to add Iceberg spotted. to the Comments for the respected Date's.

        Date  Skipper Success Time                 Comments
1 2017-02-02   Agatha      No   60         Pirates spotted.
2 2017-03-01 Gertrude     Yes   50 Illegal fishers spotted.
3 2017-04-07    Julio     Yes  120         Engine troubles.
4 2017-05-14    Dylis     Yes   30       Lost fishing line.
5 2017-06-29   Agatha      No  100                     <NA>
6 2017-12-04    Dylis     Yes  120         Pirates spotted.
7 2017-12-08    Julio      No   40       Lost fishing line.

Using stringr and str_c R_example$Comments %>% str_c("Iceberg spotted")

[1] "Pirates spotted.Iceberg spotted"

How can I select what dates to apply the above ^ code so that I can see changes made to the dataset for specified dates. Do I need to supply a filter or if_else function?

I have been trying with case_when but this replaces the existing values. I also could do this by creating another column then binding two columns together but I rather not do this.

R_example %>% 
  mutate(Comments = case_when(Date == "2017-04-07" & Date == "2017-12-04" ~ "Iceberg spotted.", 
                              TRUE ~ as.character(Comments)))

Thank you.

EDIT:

I forgot a little detail in my dataset. If I have multiple rows for a fishing trip and in the Comments column <NA> values, how can the string replace the <NA> values instead of adding to the <NA> value? Like so:

What my data looks like:

        Date  Skipper Success Time         Comments
1 2017-02-02   Agatha      No   60 Pirates spotted.
2 2017-02-02   Agatha      No   60             <NA>

What I would like to achieve:

        Date  Skipper Success Time                          Comments
1 2017-02-02   Agatha      No   60 Pirates spotted. Iceberg spotted.
2 2017-02-02   Agatha      No   60                  Iceberg spotted.

What I currently get from the code in answers below:

        Date  Skipper Success Time                          Comments
1 2017-02-02   Agatha      No   60 Pirates spotted. Iceberg spotted.
2 2017-02-02   Agatha      No   60                NAIceberg spotted.

Do you mean that you want additional rows with the new comments, or append the "Iceberg spotted" string to the existing comments as longer strings? Also, there is no "2017-04-17" in your sample data. Is that a typo, or do you mean you want a new row in that case? — Z.Lin
Hi @Z.Lin I would like to append "Iceberg spotted" string to the existing strings for 2017-04-07 and 2017-12-04. Making them longer :). That was a typo, sorry I've made an edit now! — JL_sey
There is a problem with your case_when example, as a date cannot be 2017-04-07 and 2017-12-04 simultaneously. You need an OR statement (| instead of &). — Freguglia

Bertil Baron Bertil Baron · Accepted Answer · 2020-08-24T13:30:48

I would use ifelse in this case

dta %>% 
  mutate(
    Comments = if_else(Date %in% c(as.Date("2017-04-07"), as.Date("2017-12-04")), 
                       paste0(Comments, "Iceberg spotted"), 
                       Comments)
  )

Edit string value based on a condition / another column?

2 Answers