Replace non-ASCII characters with unicode escapes in clessnverse::translate_text()

[judith-bourque]

Issue

> tools::showNonASCIIfile("R/dev_t.R")
988:     text <- stringr::str_replace_all(text, "\\<c2><ab>", "")
989:     text <- stringr::str_replace_all(text, "\\<c2><bb>", "")
990:     text <- stringr::str_replace_all(text, "\\<c2><ab>", "")
991:     text <- stringr::str_replace_all(text, "\\<c2><bb>", "")
992:     text <- stringr::str_replace_all(text, "\\<e2><80><99>", "\\\\'")

Source code

 # There atr characters that need to be escaped (or even removed) in order for the translator to
    # be able to take them
    text <- stringr::str_replace_all(text, "\\'", "\\\\'")
    text <- stringr::str_replace_all(text, "\\«", "")
    text <- stringr::str_replace_all(text, "\\»", "")
    text <- stringr::str_replace_all(text, "\\«", "")
    text <- stringr::str_replace_all(text, "\\»", "")
    text <- stringr::str_replace_all(text, "\\’", "\\\\'")

[judith-bourque]

990 and 991 seem to be a duplicate of 988 and 989. Remove?

[judith-bourque]

To get the unicode escape equivalent: stringi::stri_escape_unicode("«")

Reprex to test the code:

# Initial code
remove_char_for_translator <- function(text) {
  text <- stringr::str_replace_all(text, "\\'", "\\\\'")
  text <- stringr::str_replace_all(text, "\\«", "")
  text <- stringr::str_replace_all(text, "\\»", "")
  text <- stringr::str_replace_all(text, "\\«", "")
  text <- stringr::str_replace_all(text, "\\»", "")
  text <- stringr::str_replace_all(text, "\\’", "\\\\'")
  
  return(text)
}

# Unicode escapes
remove_char_for_translator <- function(text) {
  text <- stringr::str_replace_all(text, "\\'", "\\\\'")
  text <- stringr::str_replace_all(text, "\\\u00ab", "")
  text <- stringr::str_replace_all(text, "\\\u00bb", "")
  text <- stringr::str_replace_all(text, "\\\u00ab", "")
  text <- stringr::str_replace_all(text, "\\\u00bb", "")
  text <- stringr::str_replace_all(text, "\\\u2019", "\\\\'")
  
  return(text)
}

test_that("removing characters for translator works", {
  expect_equal(remove_char_for_translator("\\' Hello World.\\'"), "\\\\' Hello World.\\\\'")
  expect_equal(remove_char_for_translator("\\« Hello World. \\»"), "\\ Hello World. \\")
  expect_equal(remove_char_for_translator("\\’ Hello World.\\’"), "\\\\' Hello World.\\\\'")
  }
)

This could be made into a function.

Tests pass with unicode.