[BrowserKit] Adds support for meta refresh

jhedstrom · fabpot · commit 1c64c8267ddd · 2018-05-18T03:41:27.000+02:00
diff --git a/src/Symfony/Component/BrowserKit/Client.php b/src/Symfony/Component/BrowserKit/Client.php
@@ -40,6 +40,7 @@ abstract class Client
     protected $insulated = false;
     protected $redirect;
     protected $followRedirects = true;
+    protected $followMetaRefresh = false;
 
     private $maxRedirects = -1;
     private $redirectCount = 0;
@@ -68,6 +69,14 @@ public function followRedirects($followRedirect = true)
         $this->followRedirects = (bool) $followRedirect;
     }
 
+    /**
+     * Sets whether to automatically follow meta refresh redirects or not.
+     */
+    public function followMetaRefresh(bool $followMetaRefresh = true)
+    {
+        $this->followMetaRefresh = $followMetaRefresh;
+    }
+
     /**
      * Returns whether client automatically follows redirects or not.
      *
@@ -367,7 +376,16 @@ public function request(string $method, string $uri, array $parameters = array()
             return $this->crawler = $this->followRedirect();
         }
 
-        return $this->crawler = $this->createCrawlerFromContent($this->internalRequest->getUri(), $this->internalResponse->getContent(), $this->internalResponse->getHeader('Content-Type'));
+        $this->crawler = $this->createCrawlerFromContent($this->internalRequest->getUri(), $this->internalResponse->getContent(), $this->internalResponse->getHeader('Content-Type'));
+
+        // Check for meta refresh redirect
+        if ($this->followMetaRefresh && null !== $redirect = $this->getMetaRefreshUrl()) {
+            $this->redirect = $redirect;
+            $this->redirects[serialize($this->history->current())] = true;
+            $this->crawler = $this->followRedirect();
+        }
+
+        return $this->crawler;
     }
 
     /**
@@ -563,6 +581,21 @@ public function followRedirect()
         return $response;
     }
 
+    /**
+     * @see https://dev.w3.org/html5/spec-preview/the-meta-element.html#attr-meta-http-equiv-refresh
+     */
+    private function getMetaRefreshUrl(): ?string
+    {
+        $metaRefresh = $this->getCrawler()->filter('head meta[http-equiv="refresh"]');
+        foreach ($metaRefresh->extract(array('content')) as $content) {
+            if (preg_match('/^\s*0\s*;\s*URL\s*=\s*(?|\'([^\']++)|"([^"]++)|([^\'"].*))/i', $content, $m)) {
+                return str_replace("\t\r\n", '', rtrim($m[1]));
+            }
+        }
+
+        return null;
+    }
+
     /**
      * Restarts the client.
      *
diff --git a/src/Symfony/Component/BrowserKit/Tests/ClientTest.php b/src/Symfony/Component/BrowserKit/Tests/ClientTest.php
@@ -594,6 +594,39 @@ public function testFollowRedirectDropPostMethod()
         }
     }
 
+    /**
+     * @dataProvider getTestsForMetaRefresh
+     */
+    public function testFollowMetaRefresh(string $content, string $expectedEndingUrl, bool $followMetaRefresh = true)
+    {
+        $client = new TestClient();
+        $client->followMetaRefresh($followMetaRefresh);
+        $client->setNextResponse(new Response($content));
+        $client->request('GET', 'http://www.example.com/foo/foobar');
+        $this->assertEquals($expectedEndingUrl, $client->getRequest()->getUri());
+    }
+
+    public function getTestsForMetaRefresh()
+    {
+        return array(
+            array('<html><head><meta http-equiv="Refresh" content="4" /><meta http-equiv="refresh" content="0; URL=http://www.example.com/redirected"/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content="0;URL=http://www.example.com/redirected"/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content="0;URL=\'http://www.example.com/redirected\'"/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content=\'0;URL="http://www.example.com/redirected"\'/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content="0; URL = http://www.example.com/redirected"/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content="0;URL= http://www.example.com/redirected  "/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><meta http-equiv="refresh" content="0;url=http://www.example.com/redirected  "/></head></html>', 'http://www.example.com/redirected'),
+            array('<html><head><noscript><meta http-equiv="refresh" content="0;URL=http://www.example.com/redirected"/></noscript></head></head></html>', 'http://www.example.com/redirected'),
+            // Non-zero timeout should not result in a redirect.
+            array('<html><head><meta http-equiv="refresh" content="4; URL=http://www.example.com/redirected"/></head></html>', 'http://www.example.com/foo/foobar'),
+            array('<html><body></body></html>', 'http://www.example.com/foo/foobar'),
+            // Invalid meta tag placement should not result in a redirect.
+            array('<html><body><meta http-equiv="refresh" content="0;url=http://www.example.com/redirected"/></body></html>', 'http://www.example.com/foo/foobar'),
+            // Valid meta refresh should not be followed if disabled.
+            array('<html><head><meta http-equiv="refresh" content="0;URL=http://www.example.com/redirected"/></head></html>', 'http://www.example.com/foo/foobar', false),
+        );
+    }
+
     public function testBack()
     {
         $client = new TestClient();
