R separate numbers and characters in a string-CodePudding

I have a string

rule <- "X[,16] <= -0.664 & X[,11] <= -2.1891 & X[,17] >= -0.4138"

I want to split a string into two parts

num_part = c(16 , -0.664 , 11 , -2.1891 , 17 , -0.4138)
char_part = c("X[,]<= & X[,] <= & X[,] >=")

I found a similar question but I couldn't apply any of the eight examples to my data

CodePudding user response：

We may use

library(stringr)
 scan(text = str_remove_all(rule, "\\[|\\]|\\>=|<=|&|X|,"), what = numeric(), quiet = TRUE)
[1] 16.0000 -0.6640 11.0000 -2.1891 17.0000 -0.4138
str_remove_all(rule, "[0-9.-] ")
[1] "X[,] <=  & X[,] <=  & X[,] >= "

CodePudding user response：

You can do this in one go:

library(tidyverse)
data.frame(rule) %>%
  mutate(num = str_extract_all(rule, "[\\d.-] "),
         char = str_extract_all(rule, "[^\\d.-] "))
                                                      rule                                  num
1 X[,16] <= -0.664 & X[,11] <= -2.1891 & X[,17] >= -0.4138 16, -0.664, 11, -2.1891, 17, -0.4138
                                      char
1 X[,, ] <= ,  & X[,, ] <= ,  & X[,, ] >=

CodePudding user response：

library(stringr)

rule <- "X[,16] <= -0.664 & X[,11] <= -2.1891 & X[,17] >= -0.4138"
num_part <- stringr::str_extract_all(rule, pattern = "[- ]?[0-9]\\d*(\\.\\d )?")
char_part <- stringr::str_remove_all(rule, "[0-9.-] ")