If/else if on a text column in R

user_16481732 0
Viewed 73

I'm a new to R and need help with the following.

Have Need
Male_18_24_pn 18_24
Male_25_39_pn 25_39
Male_40_64_pn 40_64
Male_65_84_pn 65_84
Male_85_plus_pn 85_plus
Female_18_24_pn 18_24

I need to create the "Need" column using the "Have" column, wondering how I can achieve this in R. As a initial effort, I tried the following code to test but got warning message and every cell of "Need" populated with "18_24":

if (str_detect(pe_1P_new$Have,"18_24")) {pe_1P_new$Need= "18_24"}
Warning message:
In if (str_detect(pe_1P_new$Have, "18_24")) { :
  the condition has length > 1 and only the first element will be used

Your help is greatly appreciated. Thanks in advance!

r
4 Answers

You can do:

require(data.table)

dt = data.table(
  have = c("Male_18_24_pn", "Male_25_39_pn",
           "Male_40_64_pn", "Male_65_84_pn",
           "Male_85_plus_pn", "Female_18_24_pn")
)

dt[ , need := gsub('(Male|Female)_(.+)(_pn)', '\\2', have) ]

Base R solution:

dt$need = gsub('(Male|Female)_(.+)(_pn)', '\\2', dt$have)

No need for a loop or any conditional statements. You can extract the needed information with the help of a vectorized function, such as gsub() and a simple regex.

[Volunteer edit] This is an attempt to explain the regular expression logic of that substitution pattern (see ?regex for a very terse but complete description). You need to understand what a capture class can do.

gsub('(Male|Female)#This matches "Male" or "Female", the first "capture classes" 
       _(.+)# Second capture class, matching anything after an underscore
       (_pn)',# ... up to but not including an "_pn"
       '\\2', # replace anything matched with only the second capture class
       dt$have)

You will not be able to run this version because the carriage retruns and spaces get in the way of the regex engine process.

user_47428890

We could use trimws as well

dt[, need := trimws(have, whitespace = '[[:alpha:]]+_|_[[:alpha:]]+')]
user_37322710

We can try gsub like below

> dt[, need := gsub(".*?_(.*)_.*", "\\1", have)][]
              have    need
1:   Male_18_24_pn   18_24
2:   Male_25_39_pn   25_39
3:   Male_40_64_pn   40_64
4:   Male_65_84_pn   65_84
5: Male_85_plus_pn 85_plus
6: Female_18_24_pn   18_24
user_121587570

You can use the package {unglue}

df <- data.frame(
  Have = c("Male_18_24_pn", "Male_25_39_pn",
           "Male_40_64_pn", "Male_65_84_pn",
           "Male_85_plus_pn", "Female_18_24_pn")
)

library(unglue)
unglue_unnest(df, Have, "{}_{Need}_pn", remove = FALSE)
#>              Have    Need
#> 1   Male_18_24_pn   18_24
#> 2   Male_25_39_pn   25_39
#> 3   Male_40_64_pn   40_64
#> 4   Male_65_84_pn   65_84
#> 5 Male_85_plus_pn 85_plus
#> 6 Female_18_24_pn   18_24

Created on 2021-07-20 by the reprex package (v0.3.0)

user_22704750
Related