Allow valid XML characters outside of BMP when storing property if

jackrabbit version is good enough

CHANGELOG.md

@@ -1,6 +1,15 @@
 Changelog
 =========
 
+1.4.2
+-----
+
+* Added the repository factory option `jackalope.jackrabbit_version` to allow
+  specifying the version of the Jackrabbit backend. The version is used to
+  decide if the backend supports full UTF-8 for node names (so you can use
+  emojis in node and property names). The minimum Jackrabbit version supporting
+  emojis is 2.18.0.
+
 1.4.1
 -----
 

src/Jackalope/RepositoryFactoryJackrabbit.php

@@ -45,6 +45,7 @@ class RepositoryFactoryJackrabbit implements RepositoryFactoryInterface
         Session::OPTION_AUTO_LASTMODIFIED => 'boolean: Whether to automatically update nodes having mix:lastModified. Defaults to true.',
         'jackalope.jackrabbit_force_http_version_10' => 'boolean: Force HTTP version 1.0, this can in solving problems with curl such as https://github.com/jackalope/jackalope-jackrabbit/issues/89',
         'jackalope.jackrabbit_curl_options' => 'array: Additional global curl-options',
+        'jackalope.jackrabbit_version' => 'string: Set the version of the jackrabbit server to allow the client to offer better functionality if possible',
     );
 
     /**
@@ -96,6 +97,9 @@ public function getRepository(array $parameters = null)
         if (isset($parameters['jackalope.check_login_on_server'])) {
             $transport->setCheckLoginOnServer($parameters['jackalope.check_login_on_server']);
         }
+        if (isset($parameters['jackalope.jackrabbit_version'])) {
+            $transport->setVersion($parameters['jackalope.jackrabbit_version']);
+        }
         if (isset($parameters['jackalope.logger'])) {
             $transport = $factory->get(
                 'Transport\Jackrabbit\LoggingClient',

src/Jackalope/Transport/Jackrabbit/Client.php

@@ -69,10 +69,19 @@
 class Client extends BaseTransport implements JackrabbitClientInterface
 {
     /**
-     * minimal version needed for the backend server
+     * Minimal version requirement for the Jackrabbit backend server.
      */
     const VERSION = "2.3.6";
 
+    /**
+     * Minimal version of the Jackrabbit backend server known to support
+     * storing unicode symbols outside of UTF-8 basic multilingual plane.
+     *
+     * Note: We are sure which exact version of Jackrabbit introduced support
+     * for full UTF-8 symbols. This is the lowest version known to work.
+     */
+    const UTF8_SUPPORT_MINIMAL_VERSION = "2.18.0";
+
     /**
      * Description of the namspace to be used for communication with the server.
      * @var string
@@ -214,6 +223,13 @@ class Client extends BaseTransport implements JackrabbitClientInterface
      */
     private $curlOptions = array();
 
+    /**
+     * Version of the Jackrabbit server as declared in the configuration.
+     *
+     * @var string|null
+     */
+    private $version = null;
+
     /**
      * Create a transport pointing to a server url.
      *
@@ -455,6 +471,25 @@ public function getRepositoryDescriptors()
                     $this->descriptors['jcr.repository.version'].
                     '". Need at least "'.self::VERSION.'"');
             }
+
+
+            if ($this->version) {
+                // Sanity check if the configured version has the same major and minor number as the version reported by the backend.
+                $serverVersion = implode('.', array_slice(explode('.', $this->descriptors['jcr.repository.version']), 0, 2));
+                $configuredVersion = implode('.', array_slice(explode('.', $this->version), 0, 2));
+
+                if (!version_compare($serverVersion, $configuredVersion, '==')) {
+                    trigger_error(
+                        sprintf(
+                            'Version mismatch between configured version %s and version %s reported by the backend at %s.',
+                            $this->version,
+                            $this->descriptors['jcr.repository.version'],
+                            $this->server
+                        ),
+                        E_USER_NOTICE
+                    );
+                }
+            }
         }
 
         return $this->descriptors;
@@ -1319,7 +1354,7 @@ private function storeProperty(Property $property)
      * If occurrence is found, returns false, otherwise true.
      * Invalid characters were taken from this list: http://en.wikipedia.org/wiki/Valid_characters_in_XML#XML_1.0
      *
-     * Uses regexp mentioned here: http://stackoverflow.com/a/961504
+     * Uses regexp built upon: http://stackoverflow.com/a/961504, https://stackoverflow.com/a/30240915
      *
      * @param $string string value
      * @return bool true if string is OK, false otherwise.
@@ -1328,6 +1363,11 @@ protected function isStringValid($string)
     {
         $regex = '/[^\x{9}\x{a}\x{d}\x{20}-\x{D7FF}\x{E000}-\x{FFFD}]+/u';
 
+        if ($this->version && version_compare($this->version, self::UTF8_SUPPORT_MINIMAL_VERSION, '>=')) {
+            // unicode symbols outside of bmp such as emojis are supported only by recent jackrabbit versions
+            $regex = '/[^\x{9}\x{a}\x{d}\x{20}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]+/u';
+        }
+
         return (preg_match($regex, $string, $matches) === 0);
     }
 
@@ -2171,4 +2211,9 @@ protected function getMimePart($name, $value, $mime_boundary)
 
         return $data;
     }
+
+    public function setVersion($version)
+    {
+        $this->version = $version;
+    }
 }

tests/Jackalope/Transport/Jackrabbit/ClientTest.php

@@ -555,14 +555,15 @@ public function deleteNodesProvider(): array
     /**
      * @dataProvider provideTestOutOfRangeCharacters
      */
-    public function testOutOfRangeCharacterOccurrence($string, $isValid): void
+    public function testOutOfRangeCharacterOccurrence($string, $version, $isValid): void
     {
         if (false === $isValid) {
             $this->expectException(ValueFormatException::class);
             $this->expectExceptionMessage('Invalid character found in property "test". Are you passing a valid string?');
         }
 
         $t = $this->getTransportMock();
+        $t->setVersion($version);
 
         $factory = new Factory;
         $session = $this->createMock(Session::class);
@@ -596,18 +597,20 @@ public function provideTestOutOfRangeCharacters(): array
     {
         // use http://rishida.net/tools/conversion/ to convert problematic utf-16 strings to code points
         return array(
-            array('This is valid too!'.$this->translateCharFromCode('\u0009'), true),
-            array('This is valid', true),
-            array($this->translateCharFromCode('\uD7FF'), true),
-            array('This is on the edge, but valid too.'. $this->translateCharFromCode('\uFFFD'), true),
-            array($this->translateCharFromCode('\u10000'), true),
-            array($this->translateCharFromCode('\u10FFFF'), true),
-            array($this->translateCharFromCode('\u0001'), false),
-            array($this->translateCharFromCode('\u0002'), false),
-            array($this->translateCharFromCode('\u0003'), false),
-            array($this->translateCharFromCode('\u0008'), false),
-            array($this->translateCharFromCode('\uFFFF'), false),
-            array($this->translateCharFromCode('Sporty Spice at Sporty spice @ work \uD83D\uDCAA\uD83D\uDCAA\uD83D\uDCAA'), false),
+            array('This is valid too!'.$this->translateCharFromCode('\u0009'), null, true),
+            array('This is valid', null, true),
+            array($this->translateCharFromCode('\uD7FF'), null, true),
+            array('This is on the edge, but valid too.'. $this->translateCharFromCode('\uFFFD'), null, true),
+            array($this->translateCharFromCode('\u10000'), null, true),
+            array($this->translateCharFromCode('\u10FFFF'), null, true),
+            array($this->translateCharFromCode('\u0001'), null, false),
+            array($this->translateCharFromCode('\u0002'), null, false),
+            array($this->translateCharFromCode('\u0003'), null, false),
+            array($this->translateCharFromCode('\u0008'), null, false),
+            array($this->translateCharFromCode('\uFFFF'), null, false),
+            array($this->translateCharFromCode('Sporty Spice at Sporty spice @ work \uD83D\uDCAA\uD83D\uDCAA\uD83D\uDCAA'), null, false),
+            array($this->translateCharFromCode('Sporty Spice at Sporty spice @ work \uD83D\uDCAA\uD83D\uDCAA\uD83D\uDCAA'), '2.8.0', false),
+            array($this->translateCharFromCode('Sporty Spice at Sporty spice @ work \uD83D\uDCAA\uD83D\uDCAA\uD83D\uDCAA'), '2.18.1', true),
         );
     }