Using gsubfn to replace many instances within a string

Question

I wrote a function that transforms a string representing a number (magrittr loaded in my system):

adjust_perc_format <- function(x, n=3){
  gsub(",", ".", x, perl = T) %>% as.numeric() %>% format(nsmall=n, decimal.mark = ",")
}

So that:

adjust_perc_format("2,5", 3) 
[1] "2,500"

The goal is to transform the ocurrences defined by a regex within a string (see here). For this purpose I tried gsubfn:

str <- "20MG/ML (2,5%)+0,5%"
gsubfn("[\\d,]+(?=%)", function(x) adjust_perc_format(x, n=3),str)

The expected result is "20MG/ML (2,500%)+0,500%". Instead, I got the same input string ("20MG/ML (2,5%)+0,5%").

I also tried to set the engine as below, no success:

options(gsubfn.engine = "R")

What am I missing here? Thank you.

score 1 · Accepted Answer · answered May 27 '21 at 21:25

You need to tell gsubfn to use the whole match by passing the backref=0 argument:

gsubfn("[\\d,]+(?=%)", function(x) adjust_perc_format(x, n=3),str, backref=0)

R test:

> library(gsubfn)
> str <- "20MG/ML (2,5%)+0,5%"
> gsubfn("[\\d,]+(?=%)", function(x) adjust_perc_format(x, n=3),str, backref=0)
[1] "20MG/ML (2,500%)+0,500%"

In case you want to make your pattern more reliable, you may use

gsubfn("\\d+(?:,\\d+)*(?=%)", function(x) adjust_perc_format(x, n=3),str, backref=0)

which will match one or more digits followed with zero or more occurrences of a comma and one or more digits (followed with % that is not consumed as it is in a positive lookahead).

Great! Thanks for the robustness sugestion. – Fabio Correa May 27 '21 at 21:41 — Fabio Correa, May 27 '21 at 21:41

Using gsubfn to replace many instances within a string

1 Answers1