regular expression in R: how to extract characters from a string

Question

I have a vector of strings each containing last and first name of one or more authors. I would like to extract the last names of each author in each string. What I know is that the name that comes first is always the last name of an author (the first author), and the last names of the other authors are everything that is between between a ; and a ,. For example, in the following string:

tutu <- "goulenok, tiphaine miquel; meune, christophe; gossec, laure; dougados, maxime; kahan, andre; allanore, yannick"

I would like to extract:

"goulenok" "meune" "gossec" "dougados" "kahan" "allanore"

The last name may include punctuation characters such as ' or - but always be between a ; and a ,

Any idea?

Is your input all in one character string (suggested by your code), or is it a vector of strings (suggested by your question)? — Blue Magister, Jan 14 '13 at 21:23
it is a vector of strings actually. I won't extract the same number of author names and I'd need to put them in a data.frame in order of appearance. Thanks, — jejuba, Jan 14 '13 at 21:40

score 2 · Accepted Answer · answered Jan 14 '13 at 21:23

2

> sub(",.*$", "", strsplit(tutu, ";[ ]+")[[1]])
[1] "goulenok" "meune"    "gossec"   "dougados" "kahan"    "allanore"

answered Jan 14 '13 at 21:23

Arun

116,683
26
284
387

score 1 · Answer 2 · answered Jan 15 '13 at 17:38

1

Here is an approach that uses the gsubfn package:

library(gsubfn)

unlist(strapplyc(tutu, "(?:^|;) *([^,]+)"))

answered Jan 15 '13 at 17:38

Greg Snow

48,497
6
83
110

score 0 · Answer 3 · answered Jan 15 '13 at 01:50

0

This is a bit more blunt but also works:

sapply(unlist(lapply(strsplit(tutu, ";"), strsplit, ",")), "[", 1)

answered Jan 15 '13 at 01:50

Tyler Rinker

108,132
65
322
519

regular expression in R: how to extract characters from a string

3 Answers3

Linked