Attempts to fix UTF-16 issues

app/Facades/Util.php

@@ -0,0 +1,13 @@
+<?php
+
+namespace App\Facades;
+
+use Illuminate\Support\Facades\Facade;
+
+class Util extends Facade
+{
+    protected static function getFacadeAccessor()
+    {
+        return 'Util';
+    }
+}

app/Models/Artist.php

@@ -3,6 +3,7 @@
 namespace App\Models;
 
 use Illuminate\Database\Eloquent\Model;
+use App\Facades\Util;
 
 /**
  * @property int id The model ID
@@ -26,6 +27,8 @@ class Artist extends Model
      * This makes sure they are always sane.
      *
      * @param $value
+     *
+     * @return string
      */
     public function getNameAttribute($value)
     {
@@ -42,6 +45,11 @@ class Artist extends Model
      */
     public static function get($name)
     {
+        // Remove the BOM from UTF-8/16/32, as it will mess up the database constraints.
+        if ($encoding = Util::detectUTFEncoding($name)) {
+            $name = iconv($encoding, 'UTF-8//IGNORE', $name);
+        }
+
         $name = trim($name) ?: self::UNKNOWN_NAME;
 
         return self::firstOrCreate(compact('name'), compact('name'));

app/Providers/UtilServiceProvider.php

@@ -0,0 +1,31 @@
+<?php
+
+namespace App\Providers;
+
+use App\Services\Util;
+use Illuminate\Support\ServiceProvider;
+
+class UtilServiceProvider extends ServiceProvider
+{
+    /**
+     * Bootstrap the application services.
+     *
+     * @return void
+     */
+    public function boot()
+    {
+        //
+    }
+
+    /**
+     * Register the application services.
+     *
+     * @return void
+     */
+    public function register()
+    {
+        app()->singleton('Util', function () {
+            return new Util();
+        });
+    }
+}

app/Services/Util.php

@@ -0,0 +1,46 @@
+<?php
+
+namespace App\Services;
+
+class Util
+{
+    public function __construct()
+    {
+        defined('UTF8_BOM') or define('UTF8_BOM', chr(0xEF).chr(0xBB).chr(0xBF));
+        defined('UTF16_LITTLE_ENDIAN_BOM') or define('UTF16_LITTLE_ENDIAN_BOM', chr(0xFF).chr(0xFE));
+        defined('UTF16_BIG_ENDIAN_BOM') or define('UTF16_BIG_ENDIAN_BOM', chr(0xFE).chr(0xFF));
+        defined('UTF32_LITTLE_ENDIAN_BOM') or define('UTF32_LITTLE_ENDIAN_BOM', chr(0xFF).chr(0xFE).chr(0x00).chr(0x00));
+        defined('UTF32_BIG_ENDIAN_BOM') or define('UTF32_BIG_ENDIAN_BOM', chr(0x00).chr(0x00).chr(0xFE).chr(0xFF));
+    }
+
+    /**
+     * Detects higher UTF encoded strings.
+     * 
+     * @param string $str
+     *
+     * @return string|false
+     */
+    public function detectUTFEncoding($str)
+    {
+        switch (substr($str, 0, 2)) {
+            case UTF16_BIG_ENDIAN_BOM:
+                return 'UTF-16BE';
+            case UTF16_LITTLE_ENDIAN_BOM:
+                return 'UTF-16LE';
+        }
+
+        switch (substr($str, 0, 3)) {
+            case UTF8_BOM:
+                return 'UTF-8';
+        }
+
+        switch (substr($str, 0, 4)) {
+            case UTF32_BIG_ENDIAN_BOM:
+                return 'UTF-32BE';
+            case UTF32_LITTLE_ENDIAN_BOM:
+                return 'UTF-32LE';
+        }
+
+        return false;
+    }
+}

config/app.php

@@ -149,6 +149,7 @@ return [
         App\Providers\EventServiceProvider::class,
         App\Providers\RouteServiceProvider::class,
         App\Providers\MediaServiceProvider::class,
+        App\Providers\UtilServiceProvider::class,
 
     ],
 
@@ -200,6 +201,7 @@ return [
         'View'      => Illuminate\Support\Facades\View::class,
 
         'Media' => App\Facades\Media::class,
+        'Util' => App\Facades\Util::class,
 
     ],
 

tests/ArtistTest.php

@@ -1,7 +1,6 @@
 <?php
 
 use App\Models\Artist;
-use App\Models\User;
 use Illuminate\Foundation\Testing\DatabaseTransactions;
 
 class ArtistTest extends TestCase
@@ -34,12 +33,12 @@ class ArtistTest extends TestCase
         $this->assertEquals(Artist::UNKNOWN_NAME, Artist::get('')->name);
     }
 
-    public function testNameWithWeirdCharacters()
+    public function testUtf16Names()
     {
-        // Don't really think this is even necessary if the user has set a proper utf8 encoding
+        $name = file_get_contents(dirname(__FILE__) . '/stubs/utf16');
-        // for the database.
+
-        $name = '<27><>Ой°Ы&囧rz';
+        $artist = Artist::get($name);
-        $artist = factory(Artist::class)->create(['name' => $name]);
+        $artist = Artist::get($name); // to make sure there's no constraint exception
 
         $this->assertEquals($artist->id, Artist::get($name)->id);
     }