R：通过正则表达式提取字符串匹配部分的列表

Question

Let's say that I need to extract different parts from a string as list, for example I would like to divide the string "aaa12xxx" in three parts.

One possibility is to do three gsub calls:

parts = c()
parts[1] = gsub('([[:alpha:]]+)([0-9]+)([[:alpha:]]+)', '\\1', "aaa12xxx")
parts[2] = gsub('([[:alpha:]]+)([0-9]+)([[:alpha:]]+)', '\\2', "aaa12xxx")
parts[3] = gsub('([[:alpha:]]+)([0-9]+)([[:alpha:]]+)', '\\3', "aaa12xxx")

Of course this seems quite a waste (even if it's inside a for loop). Isn't there a function that simply returns the list of parts from a regex and a test string?

Answer 1

Just split the input string through strsplit and get the parts you want..

> x <- "aaa12xxx"
> strsplit(x,"(?<=[[:alpha:]])(?=\\d)|(?<=\\d)(?=[[:alpha:]])", perl=TRUE)
[[1]]
[1] "aaa" "12"  "xxx"

Get the parts by specifying the index number..

> m <- unlist(strsplit(x,"(?<=[[:alpha:]])(?=\\d)|(?<=\\d)(?=[[:alpha:]])", perl=TRUE))
> m[1]
[1] "aaa"
> m[2]
[1] "12"
> m[3]
[1] "xxx"

• (?<=[[:alpha:]])(?=\\\\d) Matches all the boundaries which are preceded by an alphabet and followed by a digit.

• | OR

• (?<=\\\\d)(?=[[:alpha:]]) Matches all the boundaries which are preceded by a digit and followed by an alphabet.

• Splitting your input according to the matched boundaries will give you the desired output.

Answer 2

(\\d+)|([a-zA-Z]+)

or

([[:alpha:]]+)|([0-9]+)

You can just grab the capture.use str_match_all() from library(stringr) .See demo.

https://regex101.com/r/fA6wE2/8