Algorithm for Matching Ancestor-Descendant Relationship on XML Stream

In most ancestor-descendant matching algorithm, it is more often to parse and split XML document with XML parser into element (or attribute) lists to store, and then make structural join according to the ancestor-descendant relationship between the element (or attribute) lists split. In the algorithm SSD proposed here, this paper directly inputs XML document, instead of splitting them into element (or attribute) lists to store. First it produces XML stream with SAX, and then attains ancestor-descendant matching based on XML stream. In term of analysis, SSD is more efficient with wider applicability, only once-scan, lower system resource occupancy and high-rate matching than others.