Removing diacritics in Scala - regex

Question

The problem is trivial, taking a string in some language remove the diacritics symbols. For example taking "téléphone" produces the result "telephone".
In Java I can use such method:
public static String removeAccents(String str){
return Normalizer.normalize(str, Normalizer.Form.NFD).replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
}
and it works fine but in scala it doesn't... I tried the code as follows:
val str = Normalizer.normalize("téléphone",Normalizer.Form.NFD)
val exp = "\\p{InCombiningDiacriticalMarks}+".r
exp.replaceAllIn(str,"")
it does't work!
I think, I'm missing something in using Regex in Scala, so any help would be appreciated.

Answer 1

I came across this same issue using Normalizer. Found a solution from Apache Commons StringUtils in the form of stripAccents, which removes diacitics from a String.
val str = stripAccents("téléphone")
println(str)
This will yield "telephone". Hope this helps someone!

Answer 2

You can use this, create a function to return the value of stripAccents.
val spark=SparkBase.getSparkSession()
val sc=spark.sparkContext
import spark.implicits._
val str = stripAccents("téléphone")
println(str)
val str2 = stripAccents("SERNAQUE ARGÜELLO NORMA ELIZABETH")
println(str2)
case class Fruits(name: String, quantity: Int)
val sourceDS = Seq(("YÁBAR ARRIETA JENSON", 1), ("SERNAQUE ARGÜELLO NORMA ELIZABETH", 2)).toDF("text","num")
val check = udf((colValue: String) => {
stripAccents(colValue)
})
sourceDS.select(col("text"),check(col("text"))).show(false)
->OUTPUT
+---------------------------------+---------------------------------+
|text |UDF(text) |
+---------------------------------+---------------------------------+
|YÁBAR ARRIETA JENSON |YABAR ARRIETA JENSON |
|SERNAQUE ARGÜELLO NORMA ELIZABETH|SERNAQUE ARGUELLO NORMA ELIZABETH|
+---------------------------------+---------------------------------+