import Foundation
/*
splitWords
Splits free text into words using Unicode-aware logic.
Swift's CharacterSet.letters includes ALL Unicode letters.
So we split on ANY sequence of NON-letter characters by using
components(separatedBy:) with the inverted set.
This is fully Unicode-aware and avoids regex portability issues.
*/
func splitWords(_ text: String) -> [String] {
let trimmed: String = text.trimmingCharacters(in: .whitespacesAndNewlines)
// Split on any sequence of non-letter characters
let parts: [String] = trimmed
.components(separatedBy: CharacterSet.letters.inverted)
.filter { !$0.isEmpty }
return parts
}
/*
removeDuplicateWords
Removes duplicate words while preserving:
- original order
- original casing of first occurrence
- case-insensitive comparison
Uses Set<String> for O(1) lookup.
*/
func removeDuplicateWords(_ text: String) -> String {
let words: [String] = splitWords(text)
var seen: Set<String> = Set()
var unique: [String] = []
for word in words {
let key: String = word.lowercased() // Unicode-aware lowercase
if !seen.contains(key) {
seen.insert(key)
unique.append(word) // preserve original casing
}
}
// Reassemble into a space-separated string
return unique.joined(separator: " ")
}
// ------------------------------------------------------------
// Program entry point
// ------------------------------------------------------------
let input: String =
"Hello, hello! This is a test. A TEST, hello universe... " +
"UNIVERSE! Hello; *** Is Anybody There?"
let output: String = removeDuplicateWords(input)
print(output)
/*
run:
Hello This is a test universe Anybody There
*/
