/**
* Removes duplicate words from a free-text string containing Unicode characters.
* Preserves word order and the case of the first occurrence.
*
* @param input The input free-text string containing punctuation and Unicode words.
* @return A space-separated string of unique words.
*/
fun removeDuplicateWords(input: String?): String {
// Built-in Kotlin extension function checking for null, empty, or whitespace-only strings
if (input.isNullOrBlank()) return ""
// 1. \p{L} matches any Unicode letter, \p{N} matches digits.
val wordRegex = Regex("""[\p{L}\p{N}_]+""")
// 2. HashSet for O(1) duplicate tracking.
val seenWords = HashSet<String>()
// 3. Extract matches as a Sequence and keep first occurrence (case-insensitive).
return wordRegex.findAll(input)
.map { it.value }
.filter { word ->
// HashSet.add returns true if the element was NOT already in the set
seenWords.add(word.lowercase())
}
.joinToString(" ")
}
fun main() {
val input = "Hello! こんにちは, ,hello こんにちは Bună ziua; Γεια σας Bună ziua *HELLO* Γεια σας"
val result = removeDuplicateWords(input)
println(result)
}
/*
run:
Hello こんにちは Bună ziua Γεια σας
*/
