R 中的正则表达式：有效地提取模式之前的文本和模式之后的数字

Question

Suppose that in R, I have many strings composed of a mixture of words and numbers, with one thing in common: in all strings, there is always a pattern zzzz followed by a space then a number of unknown digits.

For instance:

x <- "many words, some number like 908, then zzzz 145 and some other numbers like 377 and so on"

Then, what I am trying to do is to extract both the numbers that come after the recurring pattern zzzz , but also the text that comes before it.

Following this answer , I know how to extract the numbers after the pattern:

regmatches(x, gregexpr("zzzz \\K\\d+", x, perl=TRUE))

That returns "145" , with x example above. What I am trying to find is the most efficient way (since I have millions of strings to evaluate) to retrieve both the number after the pattern but also the content before it, which means returning the vector:

"many words, some number like 908, then " "145"

What would be the most efficient way of achieving that in R?

Answer 1

We can extract data in two groups.

1. Everything till zzzz pattern
2. Number followed by zzzz .

strcapture('(.*) zzzz (\\d+)', x, list(col1 = character(), col2 = numeric()))

#                                    col1 col2
#1 many words, some number like 908, then  145       

Answer 2

Here is a base R option using strsplit and sub :

x <- "many words, some number like 908, then zzzz 145 and some other numbers like 377 and so on"
parts <- strsplit(x, "\\bzzzz\\s+")[[1]]
parts[2] <- sub("\\s+.*$", "", parts[2])
parts

[1] "many words, some number like 908, then "
[2] "145"