How do I remove letters from numeric cells so that I can make the column entirely numeric? (R)-CodePudding

I've got a dataframe with a column full of pixel coordinates. I want to remove the 'px' from these values so that I can make the entire column numeric without introducing NAs.

    > print(data_exp_59965_v11_task_j84b$`X Coordinate`)
   [1] NA           NA           NA           NA           NA           NA           NA           NA           NA          
  [10] NA           NA           NA           NA           NA           NA           NA           NA           NA          
  [19] NA           NA           "-401.222px" "401.222px"  "-200.611px" "347.458px"  "200.611px"  "347.458px"  "-200.611px"
  [28] "-347.458px" "200.611px"  "-347.458px" NA

CodePudding user response：

library(tidyverse)

data_exp_59965_v11_task_j84b %>% 
  mutate(`X Coordinate` = as.numeric(str_replace_all(`X Coordinate`, "px$", "")))

Output

  X Coordinate
1     -401.222
2      401.222
3     -200.611
4           NA
5           NA

Data

data_exp_59965_v11_task_j84b <- structure(list(`X Coordinate` = c("-401.222px", "401.222px", 
"-200.611px", NA, NA)), class = "data.frame", row.names = c(NA, 
-5L))

CodePudding user response：

You could use sub:

 df$`X Coordinate` <- as.numeric(sub("px$", "", df$`X Coordinate`, fixed=TRUE))

More generally, you might try:

df$`X Coordinate` <- as.numeric(sub(".*?(-?\\d (?:\\.\\d )?).*", "\\1", df$`X Coordinate`))

This option would capture every number, excluding any remaining content.

CodePudding user response：

Perfect use case for parse_number from readr it is in tidyverse: Data from @AndrewGB (many thanks)

library(dplyr)
library(readr)

data_exp_59965_v11_task_j84b %>% 
  mutate(`X Coordinate` = parse_number(`X Coordinate`))

  X Coordinate
1     -401.222
2      401.222
3     -200.611
4           NA
5           NA