feat(domain-formatter): decode punycode hosts to Cyrillic with fallbacks

2026-04-17 22:03:33 +05:00
parent 6aa1ef41e6
commit 6068a6a54f
2 changed files with 96 additions and 0 deletions
@@ -0,0 +1,43 @@
+<?php
+declare(strict_types=1);
+
+namespace Cf7stg;
+
+final class DomainFormatter
+{
+    public static function humanize(string $input): string
+    {
+        $input = trim($input);
+        if ($input === '') {
+            return '';
+        }
+
+        // Normalize: strip scheme, userinfo, path, query, fragment
+        $host = $input;
+        if (preg_match('#^[a-z][a-z0-9+\-.]*://#i', $host)) {
+            $parsed = parse_url($host);
+            $host = $parsed['host'] ?? $host;
+        }
+        // Strip any remaining path/query
+        $host = preg_replace('~[/?#].*$~', '', $host) ?? $host;
+        // Strip leading www.
+        $host = preg_replace('/^www\./i', '', $host) ?? $host;
+
+        if ($host === '') {
+            return '';
+        }
+
+        // Punycode → Unicode per label (.xn-- pieces may appear on only one side)
+        if (function_exists('idn_to_utf8') && strpos($host, 'xn--') !== false) {
+            $variant = defined('INTL_IDNA_VARIANT_UTS46')
+                ? INTL_IDNA_VARIANT_UTS46
+                : 0;
+            $decoded = @idn_to_utf8($host, IDNA_DEFAULT, $variant);
+            if (is_string($decoded) && $decoded !== '') {
+                return $decoded;
+            }
+        }
+
+        return $host;
+    }
+}