Delete duplicate word, comma and whitespace-CodePudding

How can I delete all the duplicate words alongside the following comma and whitespace using Regex in R?

So far I have come up with the following regular expression, that matches the duplicate, however not the comma and whitespace. :

    (\b\w \b)(?=[\S\s]*\b\1\b)

An example list would be:

    blue, red, blue, yellow, green, blue

The output should look like:

    blue, red, yellow, green

So it would have to match two of the "blue" in this case, as well as the following comma and whitespace (if there is any).

CodePudding user response：

Depends if your list is truly a list or if it is a string with comma's

# your data is actually already a list/vector
v <- c("blue", "red", "blue", "yellow", "green", "blue")

unique(v)
[1] "blue"   "red"    "yellow" "green"

# if your data is actually a comma seperated string
s <- "blue, red, blue, yellow, green, blue"

# if output needs to be a vector
unique(strsplit(s, ", ")[[1]])
[1] "blue"   "red"    "yellow" "green" 

# if output needs to be a string again
paste(unique(strsplit(s, ", ")[[1]]), collapse = ", ")
[1] "blue, red, yellow, green"

Example based on the list column in a data.table or data.frame

dt <- data.table(
  id = rep(1:5),
  colors = list(
    c("blue", "red", "blue", "yellow", "green", "blue"),
    c("blue", "blue", "yellow", "green", "blue"),
    c("blue", "red", "blue", "yellow"),
    c("red", "red", "yellow", "yellow", "green", "blue"),
    c("black")
  )
)

## using data.table
library(data.table)
setDT(dt)
# use colors instead of clean_list to just fix the existing column
dt[, clean_list := lapply(colors, function(x) unique(x))]

## using dplyr
library(dplyr)
# use colors instead of clean_list to just fix the existing column
dt %>% mutate(clean_list = lapply(colors, function(x) unique(x)))

dt
#    id                           colors            clean_list
# 1:  1  blue,red,blue,yellow,green,blue blue,red,yellow,green
# 2:  2      blue,blue,yellow,green,blue     blue,yellow,green
# 3:  3             blue,red,blue,yellow       blue,red,yellow
# 4:  4 red,red,yellow,yellow,green,blue red,yellow,green,blue
# 5:  5                            black                 black

# or just simply in base
dt$colors <- lapply(dt$colors, function(x) unique(x))

CodePudding user response：

We could use paste with unique and collapse:

paste(unique(string), collapse= (", "))

[1] "blue, red, yellow, green"

data:

string <- c("blue", "red", "blue", "yellow", "green", "blue")