/*
Remove duplicate case‑insensitive words separated by multiple delimiters.
Features:
- Case‑insensitive comparison (mb_strtolower)
- Preserves original casing of the first occurrence
- Trims whitespace around tokens
- Supports ANY number of delimiters (including multi‑character)
- Preserves original order
- Efficient O(n) lookup using associative array as a hash set
Algorithm:
1. Build a regex that matches ANY delimiter.
2. Split the string using preg_split.
3. Trim each token.
4. Convert to lowercase for comparison.
5. Keep only first occurrence.
6. Reassemble using a chosen delimiter.
*/
function removeDuplicatesMultiDelimiterCI(string $input, array $delimiters, string $outputDelimiter): string
{
// Step 1: Build a regex that matches ANY delimiter
// preg_quote ensures delimiters like "|" or "*" are treated literally
$escaped = array_map('preg_quote', $delimiters);
$pattern = '/(?:' . implode('|', $escaped) . ')+/u';
// Step 2: Split by ANY delimiter sequence
$tokens = preg_split($pattern, $input, -1, PREG_SPLIT_NO_EMPTY);
// Step 3: Remove duplicates (case‑insensitive)
$seen = [];
$unique = [];
foreach ($tokens as $token) {
$trimmed = trim($token);
if ($trimmed === '') {
continue;
}
// Case‑insensitive key
$key = mb_strtolower($trimmed, 'UTF-8');
if (!isset($seen[$key])) {
$seen[$key] = true;
$unique[] = $trimmed; // preserve original casing
}
}
// Step 4: Reassemble
return implode($outputDelimiter, $unique);
}
// Main
$s = "AAA | aaa , aAA * aaA | AAa | AAA | BBB | ccc ---- CCC | AAA ; aaa | bbb";
// Multiple delimiters
$delimiters = [" ", "|", ",", "*", "-", ";"];
$result = removeDuplicatesMultiDelimiterCI($s, $delimiters, " | ");
echo $result;
/*
run:
AAA | BBB | ccc
*/
