从字符串中提取特定单词

Question

I need to extract all the single words that appear after the "join" word in the below string in R.

db<- c("select *
        FROM a
        left join bd on bd.id=a.id
        left join ca on ca.id=a.id
        left join dc on dc.id=a.id
        where a.names != NULL")

My Result should be "bd" "ca" "dc"

Is there any best possible approach for above mentioned query.

Answer 1

Use stringr::str_match_all(.) to match your string to a regular expression. You can use a feature of regular expression called "capture groups". That means, you can group parts of the regular expression using ( and ) . Their contents then show up separately in the results:

library(stringr)
res <- str_match_all(db, "join ([a-z]+)")
res
[[1]]
     [,1]      [,2]
[1,] "join bd" "bd"
[2,] "join ca" "ca"
[3,] "join dc" "dc"

You see that the result is a list of one element (can be seen by the [[1]] output part or if you use str(res) ). If you would have provided a vector of strings instead of only one string, this list would have more elements. Inside each list element, there is a matrix of strings. Each row of the matrix is one place of the input string where the regular expression matched. Each column of the matrix is one of the capture groups (...) of the regular expression. The first column is always the whole match. Therefore, the second column holds what you're looking for.

As mentioned in the comments, look into R for Data Science > Strings and stringr documentation > Regular expressions to learn about the topic.