In the paper we revise the current approaches to indexing XML data and introduce a novel approach to implementation of an XPath subset. Our approach enables an efficient accomplishment of querying text content of an element or an attribute value as well as of queries based on regular path expressions and XPath axes.