However, because not all entries contain the same number of children elements I can't simply rotate the element ID to access the CVSS score for the entire document. How can I access the document via element names instead? Is there a better approach to parsing out the three values I need?