TYPO3 Core

While indexing records via the class.crawler.php in the function indexSingleRecord() the content of the fields is simply passed through strip_tags() - that's not sufficient.

In class.indexer.php in the function splitHTMLContent() it's solved like this:

// remove tags, but first make sure we don't concatenate words by doing it
        $contentArr['body'] = str_replace('<',' <',$contentArr['body']);
        $contentArr['body'] = trim(strip_tags($contentArr['body']));

This has to be done here too:

$theContent = '';
        foreach($fieldList as $k => $v)    {
            if (!$k)    {
                $theTitle = $r[$v];
            } else {
                $theContent.= $r[$v].' ';
            }
        }

// add the following lines to prevent concatenated words
                $theTitle= str_replace('<',' <',$theTitle);
        $theContent= str_replace('<',' <',$theContent);

// Indexing the record as a page (but with parameters set, see >backend_setFreeIndexUid())
        $indexerObj>backend_indexAsTYPO3Page(
            strip_tags($theTitle),
            '',
            '',
            strip_tags($theContent),
            $GLOBALS['LANG']->charSet,    // Requires that
            $r[$GLOBALS['TCA'][$cfgRec['table2index']]['ctrl']['tstamp']],
            $r[$GLOBALS['TCA'][$cfgRec['table2index']]['ctrl']['crdate']],
            $r['uid']
        );

(issue imported from #M9229)