Remove substring from string between special characters in R

Question

I was looking for an answer but I failed. My question is a bit peculiar and I'm still learing regex. I'd like to achieve, from this:

str1 <- 'txt1/txt2/123|abc_def_123|1991-02-11'

something like this:

txt1|abc_def_123|1991-02-11

so everything from / (included) to first occurence of | should be removed. So far I wrote this one:

sub("\\/.*\\|", "|", str1 )

but it removes everything to the last occurence of |

"txt1|1991-02-11"

How can I point out that the substring should be removed to he first occurence of | ?

Answer 1

You can use /[^|]* which takes the first / and everything after this as long as it is not a | .

sub("/[^|]*", "", str1)
#[1] "txt1|abc_def_123|1991-02-11"

Answer 2

Following your attempt, you can make your regex lazy

sub("/.*?\\|", "|", str1 )
#[1] "txt1|abc_def_123|1991-02-11"

Answer 3

You could use a negated character class with a positive lookahead

/[^|]*(?=\|)

To not cross newlines, you could extend the negated character class:

/[^|\r\n]*(?=\|)

regex demo

str1 <- 'txt1/txt2/123|abc_def_123|1991-02-11'
sub("/[^|]*(?=\\|)", "", str1, perl=TRUE)

Output

[1] "txt1|abc_def_123|1991-02-11"

Answer 4

Another approach is by using backreference:

sub("(^.*)/.*/.*?(\\|.*$)", "\\1\\2", str1)
[1] "txt1|abc_def_123|1991-02-11"

Here, the double backreference \\1\\2 'recalls' the strings in the two capturing groups (...) , while the parts not included in capturing groups get removed.