EZP-29370: ezxmltext -> ricktext conversion: links inside links (#55)

lib/FieldType/XmlText/Converter/RichText.php

@@ -347,13 +347,15 @@ protected function checkEmptyEmbedTags(DOMDocument $inputDocument, $contentField
     }
 
     /**
+     * We need to lookup object_id/node_id for links which refers to object_remote_id or node_object_id.
+     *
      * Before calling this function, make sure you are logged in as admin, or at least have access to all the objects
      * being linked to in the $document.
 
      * @param DOMDocument $document
      * @param $contentFieldId
      */
-    protected function checkLinkTags(DOMDocument $document, $contentFieldId)
+    protected function fixLinksWithRemoteIds(DOMDocument $document, $contentFieldId)
     {
         $xpath = new DOMXPath($document);
 
@@ -394,6 +396,34 @@ protected function checkLinkTags(DOMDocument $document, $contentFieldId)
         }
     }
 
+    /**
+     * ezxmltext may contain link elements below another link element. This method flattens such structure.
+     *
+     * @param DOMDocument $document
+     * @param $contentFieldId
+     */
+    protected function flattenLinksInLinks(DOMDocument $document, $contentFieldId)
+    {
+        $xpath = new DOMXPath($document);
+
+        // Get all link elements which are child of a link
+        $xpathExpression = '//link[parent::link]';
+
+        $links = $xpath->query($xpathExpression);
+
+        foreach ($links as $link) {
+            // Move link to parent
+            $targetElement = $link->parentNode->parentNode;
+            $parentLink = $link->parentNode;
+            $targetElement->insertBefore($link, $parentLink);
+
+            // We want parent link to be listed first.
+            $targetElement->insertBefore($parentLink, $link);
+
+            $this->logger->warning("Found nested links. Flatten links where contentobject_attribute.id=$contentFieldId");
+        }
+    }
+
     /**
      * Before calling this function, make sure you are logged in as admin, or at least have access to all the objects
      * being embedded and linked to in the $inputDocument.
@@ -409,7 +439,8 @@ public function convert(DOMDocument $inputDocument, $checkDuplicateIds = false,
     {
         $this->removeComments($inputDocument);
         $this->checkEmptyEmbedTags($inputDocument, $contentFieldId);
-        $this->checkLinkTags($inputDocument, $contentFieldId);
+        $this->fixLinksWithRemoteIds($inputDocument, $contentFieldId);
+        $this->flattenLinksInLinks($inputDocument, $contentFieldId);
 
         try {
             $convertedDocument = $this->getConverter()->convert($inputDocument);

tests/lib/FieldType/Converter/_fixtures/richtext/input/047-link_in_link.xml

@@ -0,0 +1,11 @@
+<?xml version="1.0" encoding="utf-8"?>
+<section
+        xmlns:image="http://ez.no/namespaces/ezpublish3/image/"
+        xmlns:xhtml="http://ez.no/namespaces/ezpublish3/xhtml/"
+        xmlns:custom="http://ez.no/namespaces/ezpublish3/custom/">
+    <paragraph>paragraph text and (
+        <link url_id="8">
+            <link url_id="42">some link text</link>
+        </link>), after link.
+    </paragraph>
+</section>

tests/lib/FieldType/Converter/_fixtures/richtext/log/047-link_in_link.log

@@ -0,0 +1 @@
+warning:Found nested links. Flatten links where contentobject_attribute.id=

tests/lib/FieldType/Converter/_fixtures/richtext/output/047-link_in_link.xml

@@ -0,0 +1,10 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<section
+        xmlns="http://docbook.org/ns/docbook"
+        xmlns:xlink="http://www.w3.org/1999/xlink"
+        xmlns:ezxhtml="http://ez.no/xmlns/ezpublish/docbook/xhtml"
+        xmlns:ezcustom="http://ez.no/xmlns/ezpublish/docbook/custom" version="5.0-variant ezpublish-1.0">
+  <para>paragraph text and (
+        <link xlink:href="ezurl://8" xlink:show="none"/><link xlink:href="ezurl://42" xlink:show="none">some link text</link>), after link.
+    </para>
+</section>