Index-based XPath query mode tree matching method

Abstract

The invention discloses an index-based XPath query mode tree matching method. The method comprises the following steps of: establishing an index for a document set in an extensible markup language (XML) database; analyzing an XPath query statement, and constructing a query mode tree corresponding to the query statement; dividing the query mode tree into a plurality of sub-queries, and acquiring sub-query results through the index; and restoring result node stream data acquired by the sub-queries, and performing mode tree matching on node stream data. On the basis of the index, by using a data restoration method, the scale of data to be processed by the mode tree matching is optimized, input / output (I / O) overhead is reduced, and matching performance is improved. Furthermore, the invention provides a query mode tree matching method using a position index, and the problem of position query is effectively solved.

Description

technical field [0001] The invention relates to the field of database query, in particular to an index-based XPath query pattern tree matching method in an XML database. Background technique [0002] With the rapid development of information technology, XML has gradually become the standard and carrier of data expression and data exchange, and has been widely used in many important fields such as e-commerce, network services and digital libraries. The high-performance query processing of the increasingly large amount of XML data has also become an important issue. [0003] XPath is a query language for locating nodes in XML files, and its relationship to XML databases is equivalent to that of SQL to relational databases. The core grammatical form of XPath is a path expression. A path expression is a set of steps from one node in an XML file to another node or a collection of nodes. These steps are separated by " / " characters, and each step consists of three steps. three bu...

AI Technical Summary

Problems solved by technology

[0006] The first type of method is based on the idea of ​​navigation, and matches the query mode by navigating in the XML document. The disadvantage is that it is very inefficient when processing large documents, and it is only suitable for linear mode queries, not for tree mode queries with branches.

[0007] The second type of method is based on the idea of ​​overall matching, mapping the document tree and query pattern tree of XML data into a specific sequence, and then matching based on the sequence. The usual method is to map to a string sequence. Although this method is concise, it also has its disadvantages. Defects, a large number of connection and matching operations on strings increase performance overhead, and a large number of data scans also increase the burden of I / O

[0008] The third type of method is based on the idea of ​​decomposing and then connecting. The query pattern is decomposed into several fragments, and the query results of each fragment are obtained separately and then merged. too many

For such location queries, or queries combining location and content, the existing pattern matching methods have made few attempts, and the solutions are not efficient enough

Images