Replace specific pattern (shortening notations) by full notation in R-CodePudding

I have a data frame of short forms like

Ann-e/i is the short form for Anne and Anni

How can I replace the pattern -e/i in the data frame by the full notations? Another example is Matt-e/i for Matte and Matti.

Thanks in advance for any help!

CodePudding user response：

x <- c("Ann-e/i", "Matt-e/i")
gsub("(^[a-zA-Z] ?)-([a-z])/([a-z])$", "\\1\\2 and \\1\\3", x)
[1] "Anne and Anni"   "Matte and Matti"

CodePudding user response：

Wimpel's suggestion using gsub from base R works well and is quite flexible. Another approach is provided by the package stringr from the tidyverse, which might be more intuitive.

library(stringr)

strings <- c("Ann-e/i", "Annerl", "Matt-e/i")
str_replace(strings, "(\\w )-e/i", "\\1i or \\1e")
#> [1] "Anni or Anne"   "Annerl"         "Matti or Matte"

^{Created on 2021-11-08 by the reprex package (v2.0.1)}

You'll find it helpful to learn about regular expressions (regex), if you're not already familiar with them. Since there are several varieties of regex with different syntax, here's a link that is specific to using it with stringr. https://stringr.tidyverse.org/articles/regular-expressions.html

CodePudding user response：

If you have comma-separated values you can do either of this depending on your desired outcome:

Data:

string <- c("Annerl,Ann-e/i", "Matt-e/i")

First solution:

sub("(^\\w )-(\\w)/(\\w)$", "\\1\\2 and \\1\\3", unlist(strsplit(string, ",")))
# [1] "Annerl"          "Anne and Anni"   "Matte and Matti"

Second:

c(sub("(^\\w ),(\\w )-(\\w)/(\\w)$|", "\\1, \\2\\3 and \\2\\4", string[grepl(",", string)]),
  sub("(^\\w )-(\\w)/(\\w)$", "\\1\\2 and \\1\\3", string[grep(",", string, invert = TRUE)]))
# [1] "Annerl, Anne and Anni" "Matte and Matti"