Merge pull request #100 from frzsombor/token-patch-2

Fix problems with special Unicode characters

Author: Stichoza

src/Stichoza/GoogleTranslate/Tokens/GoogleTokenGenerator.php

@@ -36,16 +36,16 @@ private function TL($a)
         $tkk = $this->TKK();
         $b = $tkk[0];
 
-        for ($d = [], $e = 0, $f = 0; $f < mb_strlen($a, 'UTF-8'); $f++) {
-            $g = $this->charCodeAt($a, $f);
+        for ($d = [], $e = 0, $f = 0; $f < $this->JS_length($a); $f++) {
+            $g = $this->JS_charCodeAt($a, $f);
             if (128 > $g) {
                 $d[$e++] = $g;
             } else {
                 if (2048 > $g) {
                     $d[$e++] = $g >> 6 | 192;
                 } else {
-                    if (55296 == ($g & 64512) && $f + 1 < mb_strlen($a, 'UTF-8') && 56320 == ($this->charCodeAt($a, $f + 1) & 64512)) {
-                        $g = 65536 + (($g & 1023) << 10) + ($this->charCodeAt($a, ++$f) & 1023);
+                    if (55296 == ($g & 64512) && $f + 1 < $this->JS_length($a) && 56320 == ($this->JS_charCodeAt($a, $f + 1) & 64512)) {
+                        $g = 65536 + (($g & 1023) << 10) + ($this->JS_charCodeAt($a, ++$f) & 1023);
                         $d[$e++] = $g >> 18 | 240;
                         $d[$e++] = $g >> 12 & 63 | 128;
                     } else {
@@ -138,23 +138,27 @@ private function unsignedRightShift($a, $b)
     }
 
     /**
-     * Get the Unicode of the character at the specified index in a string.
+     * Get JS charCodeAt equivalent result with UTF-16 encoding
      *
      * @param string $str
      * @param int    $index
      *
-     * @return null|number
+     * @return number
      */
-    private function charCodeAt($str, $index)
-    {
-        $char = mb_substr($str, $index, 1, 'UTF-8');
-        if (mb_check_encoding($char, 'UTF-8')) {
-            $ret = mb_convert_encoding($char, 'UTF-32BE', 'UTF-8');
-            $result = hexdec(bin2hex($ret));
-
-            return $result;
-        }
+    private function JS_charCodeAt($str, $index) {
+        $utf16 = mb_convert_encoding($str, 'UTF-16LE', 'UTF-8');
+        return ord($utf16[$index*2]) + (ord($utf16[$index*2+1]) << 8);
+    }
 
-        return;
+    /**
+     * Get JS equivalent string length with UTF-16 encoding
+     *
+     * @param string $str
+     *
+     * @return number
+     */
+    private function JS_length($str) {
+        $utf16 = mb_convert_encoding($str, 'UTF-16LE', 'UTF-8');
+        return strlen($utf16)/2;
     }
 }

tests/TranslationTest.php

@@ -19,6 +19,14 @@ public function testTranslationEquality()
         $this->assertEquals($resultOne, $resultTwo, 'გამარჯობა');
     }
 
+    public function testUTF16Translation()
+    {
+        $resultOne = TranslateClient::translate('en', 'de', 'yes 👍🏽');
+        $resultTwo = $this->tr->setSource('en')->setTarget('de')->translate('yes 👍🏽');
+
+        $this->assertEquals($resultOne, $resultTwo, 'ja 👍🏽');
+    }
+
     public function testArrayTranslation()
     {
         $this->tr->setSource('en')->setTarget('ka');