Fixed encoding problems when PHP INI setting default_charset is not…

… `utf-8` [#2154]
getgrav · Feb 10, 2020 · e55b239 · e55b239
1 parent c2f374f
commit e55b239
Show file tree

Hide file tree

Showing 10 changed files with 55 additions and 27 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,11 @@
+# v1.6.21
+## mm/dd/2020
+
+1. [](#bugfix)
+    * Fixed encoding problems when PHP INI setting `default_charset` is not `utf-8` [#2154](https://github.com/getgrav/grav/issues/2154)
+
 # v1.6.20
-## 03/02/2020
+## 02/03/2020
 
 1. [](#bugfix)
     * Fixed incorrect routing caused by `str_replace()` in `Uri::init()` [#2754](https://github.com/getgrav/grav/issues/2754)

diff --git a/bin/gpm b/bin/gpm
@@ -28,6 +28,13 @@ if (!ini_get('date.timezone')) {
     date_default_timezone_set('UTC');
 }
 
+// Set internal encoding.
+if (!\extension_loaded('mbstring')) {
+    die("'mbstring' extension is not loaded.  This is required for Grav to run correctly");
+}
+@ini_set('default_charset', 'UTF-8');
+mb_internal_encoding('UTF-8');
+
 if (!file_exists(GRAV_ROOT . '/index.php')) {
     exit('FATAL: Must be run from ROOT directory of Grav!');
 }

diff --git a/bin/grav b/bin/grav
@@ -25,6 +25,17 @@ if (version_compare($ver = PHP_VERSION, $req = GRAV_PHP_MIN, '<')) {
     exit(sprintf("You are running PHP %s, but Grav needs at least PHP %s to run.\n", $ver, $req));
 }
 
+if (!ini_get('date.timezone')) {
+    date_default_timezone_set('UTC');
+}
+
+// Set internal encoding.
+if (!\extension_loaded('mbstring')) {
+    die("'mbstring' extension is not loaded.  This is required for Grav to run correctly");
+}
+@ini_set('default_charset', 'UTF-8');
+mb_internal_encoding('UTF-8');
+
 $climate = new League\CLImate\CLImate;
 $climate->arguments->add([
     'environment' => [
@@ -42,10 +53,6 @@ $environment = $climate->arguments->get('environment');
 $grav = Grav::instance(array('loader' => $autoload));
 $grav->setup($environment);
 
-if (!ini_get('date.timezone')) {
-    date_default_timezone_set('UTC');
-}
-
 if (!file_exists(GRAV_ROOT . '/index.php')) {
     exit('FATAL: Must be run from ROOT directory of Grav!');
 }

diff --git a/bin/plugin b/bin/plugin
@@ -32,6 +32,13 @@ if (!ini_get('date.timezone')) {
     date_default_timezone_set('UTC');
 }
 
+// Set internal encoding.
+if (!\extension_loaded('mbstring')) {
+    die("'mbstring' extension is not loaded.  This is required for Grav to run correctly");
+}
+@ini_set('default_charset', 'UTF-8');
+mb_internal_encoding('UTF-8');
+
 if (!file_exists(GRAV_ROOT . '/index.php')) {
     exit('FATAL: Must be run from ROOT directory of Grav!');
 }

diff --git a/index.php b/index.php
@@ -20,6 +20,16 @@
     die("PHP webserver requires a router to run Grav, please use: <pre>php -S {$_SERVER['SERVER_NAME']}:{$_SERVER['SERVER_PORT']} system/router.php</pre>");
 }
 
+// Set timezone to default, falls back to system if php.ini not set
+date_default_timezone_set(@date_default_timezone_get());
+
+// Set internal encoding.
+if (!\extension_loaded('mbstring')) {
+    die("'mbstring' extension is not loaded.  This is required for Grav to run correctly");
+}
+@ini_set('default_charset', 'UTF-8');
+mb_internal_encoding('UTF-8');
+
 // Ensure vendor libraries exist
 $autoload = __DIR__ . '/vendor/autoload.php';
 if (!is_file($autoload)) {
@@ -32,15 +42,6 @@
 use Grav\Common\Grav;
 use RocketTheme\Toolbox\Event\Event;
 
-// Set timezone to default, falls back to system if php.ini not set
-date_default_timezone_set(@date_default_timezone_get());
-
-// Set internal encoding if mbstring loaded
-if (!\extension_loaded('mbstring')) {
-    die("'mbstring' extension is not loaded.  This is required for Grav to run correctly");
-}
-mb_internal_encoding('UTF-8');
-
 // Get the Grav instance
 $grav = Grav::instance(
     array(

diff --git a/system/src/Grav/Common/Assets/Traits/AssetUtilsTrait.php b/system/src/Grav/Common/Assets/Traits/AssetUtilsTrait.php
@@ -138,9 +138,9 @@ protected function renderAttributes()
             }
 
             if (\in_array($key, $no_key, true)) {
-                $element = htmlentities($value, ENT_QUOTES, 'UTF-8', false);
+                $element = htmlentities($value, ENT_QUOTES | ENT_HTML5, 'UTF-8', false);
             } else {
-                $element = $key . '="' . htmlentities($value, ENT_QUOTES, 'UTF-8', false) . '"';
+                $element = $key . '="' . htmlentities($value, ENT_QUOTES | ENT_HTML5, 'UTF-8', false) . '"';
             }
 
             $html .= ' ' . $element;

diff --git a/system/src/Grav/Common/Page/Markdown/Excerpts.php b/system/src/Grav/Common/Page/Markdown/Excerpts.php
@@ -68,7 +68,7 @@ public function fireInitializedEvent($markdown): void
      */
     public function processLinkExcerpt(array $excerpt, string $type = 'link'): array
     {
-        $url = htmlspecialchars_decode(rawurldecode($excerpt['element']['attributes']['href']));
+        $url = htmlspecialchars_decode(rawurldecode($excerpt['element']['attributes']['href']), ENT_QUOTES | ENT_HTML5);
 
         $url_parts = $this->parseUrl($url);
 
@@ -152,7 +152,7 @@ static function ($carry, $item) {
      */
     public function processImageExcerpt(array $excerpt): array
     {
-        $url = htmlspecialchars_decode(urldecode($excerpt['element']['attributes']['src']));
+        $url = htmlspecialchars_decode(urldecode($excerpt['element']['attributes']['src']), ENT_QUOTES | ENT_HTML5);
         $url_parts = $this->parseUrl($url);
 
         $media = null;

diff --git a/system/src/Grav/Common/Page/Page.php b/system/src/Grav/Common/Page/Page.php
@@ -608,12 +608,12 @@ public function summary($size = null, $textOnly = false)
                 return $content;
             }
 
-            return mb_strimwidth($content, 0, $size, '...', 'utf-8');
+            return mb_strimwidth($content, 0, $size, '...', 'UTF-8');
         }
 
         $summary = Utils::truncateHtml($content, $size);
 
-        return html_entity_decode($summary,ENT_COMPAT | ENT_HTML401, 'utf-8');
+        return html_entity_decode($summary, ENT_COMPAT | ENT_HTML5, 'UTF-8');
     }
 
     /**
@@ -1713,7 +1713,7 @@ public function metadata($var = null)
                         $this->metadata[$prop_key] = [
                             'name' => $prop_key,
                             'property' => $prop_key,
-                            'content' => htmlspecialchars($prop_value, ENT_QUOTES, 'UTF-8')
+                            'content' => htmlspecialchars($prop_value, ENT_QUOTES | ENT_HTML5, 'UTF-8')
                         ];
                     }
                 } else {
@@ -1722,16 +1722,16 @@ public function metadata($var = null)
                         if (\in_array($key, $header_tag_http_equivs, true)) {
                             $this->metadata[$key] = [
                                 'http_equiv' => $key,
-                                'content' => htmlspecialchars($value, ENT_QUOTES, 'UTF-8')
+                                'content' => htmlspecialchars($value, ENT_QUOTES | ENT_HTML5, 'UTF-8')
                             ];
                         } elseif ($key === 'charset') {
-                            $this->metadata[$key] = ['charset' => htmlspecialchars($value, ENT_QUOTES, 'UTF-8')];
+                            $this->metadata[$key] = ['charset' => htmlspecialchars($value, ENT_QUOTES | ENT_HTML5, 'UTF-8')];
                         } else {
                             // if it's a social metadata with separator, render as property
                             $separator = strpos($key, ':');
                             $hasSeparator = $separator && $separator < strlen($key) - 1;
                             $entry = [
-                                'content' => htmlspecialchars($value, ENT_QUOTES, 'UTF-8')
+                                'content' => htmlspecialchars($value, ENT_QUOTES | ENT_HTML5, 'UTF-8')
                             ];
 
                             if ($hasSeparator && !Utils::startsWith($key, 'twitter')) {
@@ -2718,7 +2718,7 @@ public function collection($params = 'content', $pagination = true)
                         }
                         foreach ($items as $item) {
                             $item = rawurldecode($item);
-                            if (empty($page->taxonomy[$taxonomy]) || !\in_array(htmlspecialchars_decode($item, ENT_QUOTES), $page->taxonomy[$taxonomy], true)
+                            if (empty($page->taxonomy[$taxonomy]) || !\in_array(htmlspecialchars_decode($item, ENT_QUOTES | ENT_HTML5), $page->taxonomy[$taxonomy], true)
                             ) {
                                 $collection->remove($page->path());
                             }

diff --git a/system/src/Grav/Common/Twig/TwigExtension.php b/system/src/Grav/Common/Twig/TwigExtension.php
@@ -1055,7 +1055,7 @@ public function nonceFieldFunc($action, $nonceParamName = 'nonce')
      */
     public function jsonDecodeFilter($str, $assoc = false, $depth = 512, $options = 0)
     {
-        return json_decode(html_entity_decode($str), $assoc, $depth, $options);
+        return json_decode(html_entity_decode($str, ENT_COMPAT | ENT_HTML5, 'UTF-8'), $assoc, $depth, $options);
     }
 
     /**

diff --git a/system/src/Grav/Common/Uri.php b/system/src/Grav/Common/Uri.php
@@ -306,7 +306,7 @@ public function params($id = null, $array = false)
     public function param($id)
     {
         if (isset($this->params[$id])) {
-            return html_entity_decode(rawurldecode($this->params[$id]));
+            return html_entity_decode(rawurldecode($this->params[$id]), ENT_COMPAT | ENT_HTML5, 'UTF-8');
         }
 
         return false;