/*
A clean, idiomatic Kotlin program that removes duplicate words
from free-text while preserving original order and casing.
*/
fun splitWords(text: String): List<String> {
/*
Splits free text into words using a Unicode-aware regex.
Kotlin/Java regex supports Unicode properties:
\p{L} → any Unicode letter
[^\p{L}]+ → any sequence of NON-letters
This gives correct splitting for multilingual free text.
*/
val trimmed: String = text.trim()
// Split on any sequence of non-letter characters
val parts: List<String> = trimmed.split(Regex("[^\\p{L}]+"))
// Filter out empty tokens
return parts.filter { it.isNotEmpty() }
}
fun removeDuplicateWords(text: String): String {
/*
Removes duplicate words while preserving:
- original order
- original casing of first occurrence
- case-insensitive comparison
Uses HashSet<String> for O(1) lookup.
*/
val words: List<String> = splitWords(text)
val seen: HashSet<String> = HashSet()
val unique: MutableList<String> = mutableListOf()
for (word: String in words) {
val key: String = word.lowercase() // Unicode-aware lowercase
if (!seen.contains(key)) {
seen.add(key)
unique.add(word) // preserve original casing
}
}
// Reassemble into a space-separated string
return unique.joinToString(" ")
}
fun main() {
val input: String =
"Hello, hello! This is a test. A TEST, hello universe... " +
"UNIVERSE! Hello; *** Is Anybody There?"
val output: String = removeDuplicateWords(input)
println(output)
}
/*
run:
Hello This is a test universe Anybody There
*/
