CODEWITHSUNDEEP
rregex
chippycentra
chippycentra

Reputation: 3432

Using regex pattern in gsub r

Hello I hve a df such as

COL1         
BLOC1.1_3_10-355(+)Sp_3
BLOC2.1_10-355(-)SSp_4
BLOC3.1_10-355(+)SP_32
BLOC1_3_10-355(+)SP4_2

How can I find a regex that can replace the _ here > _[Number]-[Number]( by

:[Number]-[Number](

Here I should get

COL1         
BLOC1.1_3:10-355(+)Sp_3
BLOC2.1:10-355(-)SSp_4
BLOC3.1:10-355(+)SP_32
BLOC1_3:10-355(+)SP4_2

I tried : gsub("_[0-9]-[0-9](",":[0-9]-[0-9](",df$COL1)

Upvotes: 1

Views: 221

Answers (3)

The fourth bird
The fourth bird

Reputation: 163207

You can use

_([0-9]+-[0-9]+\()

And replace with : and capture group 1.

Regex demo

COL1 <- c("BLOC1.1_3_10-355(+)Sp_3", "BLOC2.1_10-355(-)SSp_4", "BLOC3.1_10-355(+)SP_32", "BLOC1_3_10-355(+)SP4_2")
gsub("_([0-9]+-[0-9]+\\()", ":\\1", COL1)

Output

[1] "BLOC1.1_3:10-355(+)Sp_3" "BLOC2.1:10-355(-)SSp_4" 
[3] "BLOC3.1:10-355(+)SP_32"  "BLOC1_3:10-355(+)SP4_2"

Upvotes: 4

Tim Biegeleisen
Tim Biegeleisen

Reputation: 520898

A solution using string splitting:

output <- sapply(COL1, function(x) {
    parts <- strsplit(x, "_(?=\\d+-)", perl=TRUE)
    paste(parts[[1]][1], parts[[1]][2], sep=":")
})
names(output) <- c(1:4)
output

                        1                         2                         3 
"BLOC1.1_3:10-355(+)Sp_3"  "BLOC2.1:10-355(-)SSp_4"  "BLOC3.1:10-355(+)SP_32" 
                        4 
 "BLOC1_3:10-355(+)SP4_2"

Data:

COL1 <- c("BLOC1.1_3_10-355(+)Sp_3", "BLOC2.1_10-355(-)SSp_4",
          "BLOC3.1_10-355(+)SP_32",  "BLOC1_3_10-355(+)SP4_2")

Upvotes: 3

Wimpel
Wimpel

Reputation: 27732

COL1 <- c("BLOC1.1_3_10-355(+)Sp_3",
"BLOC2.1_10-355(-)SSp_4",
"BLOC3.1_10-355(+)SP_32",
"BLOC1_3_10-355(+)SP4_2")

gsub( "(.*[0-9]+)(_)([0-9]+-.*)", "\\1:\\3", COL1)

[1] "BLOC1.1_3:10-355(+)Sp_3" "BLOC2.1:10-355(-)SSp_4"  "BLOC3.1:10-355(+)SP_32" 
[4] "BLOC1_3:10-355(+)SP4_2"

Upvotes: 4

Related Questions