I have a dataset with German special charaters (ä, ö, ü, ß) and other special characters (brackets (), /, , etc.):

df <- data.frame(
id = 1:8,
date = seq.Date(as.Date("2024-12-01"), as.Date("2024-12-08"), "day"),
group = rep(LETTERS[1:2], 4),
text1 = c("Österreichische Botschaft", "Außenministerium", "Bundesheer", "Präsident",
          "Bundesregierung / Bundeskanzler", "Parteien", "Justiz", 
          "Erwerbstätig (unselbständig)")
)

I have another dataset containing the original strings and the replacement strings:

df_replace <- data.frame(
original = c("Österreichische Botschaft", "Außenministerium", "Präsident",
             "Bundesregierung / Bundeskanzler", "Parteien", "Justiz", 
             "Erwerbstätig (unselbständig)"),
replacement = c("Austrian embassy", "Foreign ministry", "President",
                "Government / Chancellor", "Parties", "Judiciary", "Working (employed)")
)

I now want to replace all the strings in df using the df_replace dataset.

Using stringi::stri_replace_all_fixed works fine, as long as the number of rows in the df and the df_replace datasets are the same:

df |> mutate_all((x) stringi::stri_replace_all_fixed(x, df_replace$original,
                                           df_replace$replacement))

But this is not the case for my df.

Hence, I receive the following error-message:

Caused by warning in "stringi::stri_replace_all_fixed()": ! longer object length is not a multiple of shorter object length

In addition, this solution only works, if the terms are in the same order. For example, if I switch the first terms (“Österreichische Botschaft”, “Austrian embassy”) to the last position, there are no replacements anymore.

Using the following code did not work either:

df |> mutate_all((x) stringr::str_replace_all(x, df_replace$original,
                                                  df_replace$replacement))

Because with stringr::str_replace_all I run into problems because of the special characters (especially the brackets).

Is there another way how to solve this problem?