Extracting substring ending with some sign but without this sign

Question

I have some string

string <- "Shakira - Wolf - 02.Hips don't lie.mp3"

I want only the first part, so the name of the artist. I use regex like this

stri_extract_all_regex(string, "^.*?-")

The output: "Shakira -". But I don't want " -". How to write regex which allows me to take only the substring which stands before the first " -"?

Starred answer works: http://stackoverflow.com/questions/2013124/regex-matching-up-to-the-first-occurrence-of-a-character — Vlo, Dec 10 '14 at 21:08
Nothing works the way I want, maybe because I have several "-" in my string. — jjankowiak, Dec 10 '14 at 21:24

David Arenburg · Answer 1 · 2014-12-10T22:43:15.640

5

I think you just need (without using external packages)

sub(" -.*", "", string)
## [1] "Shakira"

Explanation

This simply matches " -" and everything after it until the end of the string and replaces it with nothing. Which basically leaves you with everything before the first " -"

If you insist on stringi package (for speed) you could use stri_extract_first with the simple regex of

stri_extract_first(string, regex = "[A-Za-z]+")
## [1] "Shakira"

edited Dec 10 '14 at 22:43

answered Dec 10 '14 at 21:34

David Arenburg

91,361
17
137
196

3

`sub`, the small and humble solution that I'm sure Shakira prefers. – thelatemail Dec 10 '14 at 22:33
@thelatemail, `gsub` will also work with this specific regex, but `sub` seems safer – David Arenburg Dec 10 '14 at 22:37
1

I'll +1 yours (deleted mine) I prefer `sub` in this case. – hwnd Dec 10 '14 at 22:39
@hwnd, you should probably leave it in order to make this Q/A richer – David Arenburg Dec 10 '14 at 22:40
1

@DavidArenburg no that is ok. Btw you can remove the `$` anchor it is not necessary really =) – hwnd Dec 10 '14 at 22:41

IRTFM · Accepted Answer · 2014-12-10T22:35:34.197

4

The negated character-class method succeeds:

> stri_extract_all_regex(string, "^[^-]+")
[[1]]
[1] "Shakira "

Challenged by The Other David I'm attempting now to select only the spaces between alpha characters but to also use that function and therefore do it with a "positive" selection strategy:

string <- "Shakira and Friends - Wolf - 02.Hips don't lie.mp3"
stri_extract_all_regex(string, "^[[:alpha:]]+( *[[:alpha:]])*")
[[1]]
[1] "Shakira and Friends"

edited Dec 10 '14 at 22:35

answered Dec 10 '14 at 21:41

IRTFM

258,963
21
364
487

1

This still has a space in it. It should be `stri_extract_all_regex(string, "^[^ -]+")[[1]]` – David Arenburg Dec 10 '14 at 22:06

score 2 · Answer 3 · answered Dec 10 '14 at 21:34

2

How about using strsplit?

strsplit(string, split = " -")[[1]][1]

answered Dec 10 '14 at 21:34

Kara Woo

3,595
19
31

score 0 · Answer 4 · answered Dec 12 '14 at 21:24

0

Using rex may make this type of task a little simpler.

string <- "Shakira - Wolf - 02.Hips don't lie.mp3"

library(rex)
re_matches(string,
  rex(capture(zero_or_more(any, type='lazy')), spaces, "-"))$'1'

#> [1] "Shakira"

answered Dec 12 '14 at 21:24

Jim

4,687
29
30

Extracting substring ending with some sign but without this sign

4 Answers4