The eXtensible Markup Language (XML) is becoming a new standard for information representation and exchange over the Internet. How to index XML data for efficient query processing is an important subject in the XML community. In this paper, based on the extended preorder indexing method, path information is added as part of the index. It is shown that the number of path joins can be reduced to the number of the "interested points", but not related to the length of the path expression in a query. The extra space needed is about the same as the extended preorder labeling method. Experimental results showed significant improvements by using the new indexing method over naive extended preorder labeling method.