Added trim() to XML input when importing an XML or HTML string. Allow…

…s for slight errors in output XML found in the wild where extra space can be introduced in error.
mnapoli · Aug 4, 2012 · 327061e · 327061e
1 parent 984d799
commit 327061e
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 3 deletions.
diff --git a/library/Zend/Feed/Reader/Reader.php b/library/Zend/Feed/Reader/Reader.php
@@ -271,14 +271,15 @@ public static function importString($string)
         $libxml_errflag = libxml_use_internal_errors(true);
         libxml_disable_entity_loader(true);
         $dom = new DOMDocument;
-        $status = $dom->loadXML($string);
+        $status = $dom->loadXML(trim($string));
         libxml_disable_entity_loader(false);
         libxml_use_internal_errors($libxml_errflag);
 
         if (!$status) {
             // Build error message
             $error = libxml_get_last_error();
             if ($error && $error->message) {
+                $error->message = trim($error->message);
                 $errormsg = "DOMDocument cannot parse XML: {$error->message}";
             } else {
                 $errormsg = "DOMDocument cannot parse XML: Please check the XML document's validity";
@@ -298,7 +299,7 @@ public static function importString($string)
             $reader = new Feed\Atom($dom, $type);
         } else {
             throw new Exception\RuntimeException('The URI used does not point to a '
-            . 'valid Atom, RSS or RDF feed that Zend_Feed_Reader can parse.');
+            . 'valid Atom, RSS or RDF feed that Zend\Feed\Reader can parse.');
         }
         return $reader;
     }
@@ -340,13 +341,14 @@ public static function findFeedLinks($uri)
         $libxml_errflag = libxml_use_internal_errors(true);
         libxml_disable_entity_loader(true);
         $dom = new DOMDocument;
-        $status = $dom->loadHTML($responseHtml);
+        $status = $dom->loadHTML(trim($responseHtml));
         libxml_disable_entity_loader(false);
         libxml_use_internal_errors($libxml_errflag);
         if (!$status) {
             // Build error message
             $error = libxml_get_last_error();
             if ($error && $error->message) {
+                $error->message = trim($error->message);
                 $errormsg = "DOMDocument cannot parse HTML: {$error->message}";
             } else {
                 $errormsg = "DOMDocument cannot parse HTML: Please check the XML document's validity";

diff --git a/tests/Zend/Feed/Reader/ReaderTest.php b/tests/Zend/Feed/Reader/ReaderTest.php
@@ -37,6 +37,13 @@ public function tearDown()
         Reader\Reader::reset();
     }
 
+    public function testStringImportTrimsContentToAllowSlightlyInvalidXml()
+    {
+        $feed = Reader\Reader::importString(
+            '   ' . file_get_contents($this->_feedSamplePath.'/Reader/rss20.xml')
+        );
+    }
+
     public function testDetectsFeedIsRss20()
     {
         $feed = Reader\Reader::importString(