TYPO3 Core

/**

* Extracts the "base href" from content string.

*

* @param string Content to analyze

* @return string The base href or an empty string if not found

*/

public function extractBaseHref($string) {

if (!is_object($this->htmlParser)) {

$this->htmlParser = t3lib_div::makeInstance('t3lib_parseHtml');

}

$parts = $this->htmlParser->splitTags('base', $string);

foreach ($parts as $k => $v) {

if ($k % 2) {

$params = $this->htmlParser->get_tag_attributes($v, 1);

$firstTagName = $this->htmlParser->getFirstTagName($v); // The 'name' of the first tag

switch (strtolower($firstTagName)) {

case 'base':

$href = $params[0]['href'];

if ($href) {

// Return the first "base href" found (a single one should be present anyway)

return $href;

}

}

}

}

return '';

}

$baseHref = $indexerObj->extractBaseHref($indexerObj->indexExternalUrl_content);

if (!$baseHref) {

// Extract base href from current URL

$baseHref = $url_qParts['scheme'] . '://' . $url_qParts['host'];

$baseHref .= substr($url_qParts['path'], 0, strrpos($url_qParts['path'], '/'));

}

$baseHref = rtrim($baseHref, '/');

$subUrl = $baseHref . '/' . t3lib_div::resolveBackPath($subUrl);