account for the presence of numeric values in a vector of characters

R/numberize.R

@@ -119,11 +119,24 @@ number_from <- function(digits) {
 #'
 # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 .numberize <- function(text, lang = c("en", "fr", "es")) {
-  digits <- digits_from(text, lang)
-  if (anyNA(digits)) {
+  # return NA if the input is NA
+  if (is.na(text)) {
     return(NA)
   }
-  number_from(digits)
+
+  # convert to numeric. Numeric values will pass and non numeric values will be
+  # coerced to NA and converted into numbers.
+  tmp_text <- suppressWarnings(as.numeric(text))
+  if (!is.na(tmp_text)) {
+    return(tmp_text)
+  } else {
+    # when the text does not correspond to a number, digits_from() returns NA
+    digits <- digits_from(text, lang)
+    if (anyNA(digits)) {
+      return(NA)
+    }
+    number_from(digits)
+  }
 }
 
 # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

README.Rmd

@@ -68,7 +68,7 @@ numberize("veintiuno", lang = "es")
 
 # convert a vector of written values
 numberize(
-  text = c("dix", "soixante-cinq", "deux mille vingt-quatre", NA),
+  text = c(17, "dix", "soixante-cinq", "deux mille vingt-quatre", NA),
   lang = "fr"
 )
 ```

README.md

@@ -53,36 +53,28 @@ These examples illustrate the current functionality.
 ``` r
 # numberize a French string
 numberize("zéro", lang = "fr")
-#> zéro 
-#>    0
+#> [1] 0
 
 # numberize a Spanish string
 numberize("Siete mil quinientos cuarenta y cinco", lang = "es")
-#> Siete mil quinientos cuarenta y cinco 
-#>                                  7545
+#> [1] 7545
 
 # numberize the English string "nine hundred and ninety-nine trillion, nine hundred and ninety-nine billion, nine hundred and ninety-nine million, nine hundred and ninety-nine thousand, nine hundred and ninety-nine" # nolint: line_length_linter.
 formatC(numberize("nine hundred and ninety-nine trillion, nine hundred and ninety-nine billion, nine hundred and ninety-nine million, nine hundred and ninety-nine thousand, nine hundred and ninety-nine"), big.mark = ",", format = "fg") # nolint: line_length_linter.
-#> nine hundred and ninety-nine trillion, nine hundred and ninety-nine billion, nine hundred and ninety-nine million, nine hundred and ninety-nine thousand, nine hundred and ninety-nine 
-#>                                                                                                                                                                  "999,999,999,999,999"
+#> [1] "999,999,999,999,999"
 
 # some edge cases
 numberize("veintiún", lang = "es")
-#> veintiún 
-#>       21
+#> [1] 21
 numberize("veintiuno", lang = "es")
-#> veintiuno 
-#>        21
+#> [1] 21
 
 # convert a vector of written values
 numberize(
-  text = c("dix", "soixante-cinq", "deux mille vingt-quatre", NA),
+  text = c(17, "dix", "soixante-cinq", "deux mille vingt-quatre", NA),
   lang = "fr"
 )
-#>                     dix           soixante-cinq deux mille vingt-quatre 
-#>                      10                      65                    2024 
-#>                    <NA> 
-#>                      NA
+#> [1]   17   10   65 2024   NA
 ```
 
 ## Related packages and Limitations

tests/testthat/test-numberize.R

@@ -88,3 +88,11 @@ test_that("non digit word returns NA", {
   res <- numberize("epiverse", lang = "en")
   expect_true(is.na(res))
 })
+
+test_that("vector with number and words and NA is properly handled", {
+  res <- numberize(
+    c(17, "dix", "soixante-cinq", "deux mille vingt-quatre", NA),
+    lang = "fr"
+  )
+  expect_identical(res, c(17, 10, 65, 2024, NA))
+})