Database Full-text Search Method And Its Application

Database technology has been 40 years of history, the new database technology are endless, there have been like SQL SERVER, ORACLE, MySQL, DB2 and many more excellent database management system, and as the upgrade version can also provide better full-text search function. For example, MySQL full text search function can only support the English fields, and can not support a large domain of full-text search field. As well as in SQL SERVER 7 version does not support full-text search, even in the new version was only one table can support a full-text retrieval.Now many of the forming of database applications, most of the construction of at least the age of 5 years or more, and after a long period of data accumulation, data from each database are generally recorded in the millions of lines above. For these database systems forming the database, to provide them with full-text search function there are many obstacles. Therefore, the need to find a can not affect the existing database design, and existing databases provide full-text search method. As a result, there has been such products as Sphinx, Sphinx can only support MySQL database, but we need more of is a database can provide this functionality to the software. Therefore, this paper's analysis of Sphinx, and learn from its design, combined with the use Lucene, an open source software such efforts to establish a more general database full-text retrieval system, database, data source and gives a general definition of the template, and design a set based on this template regardless of the type of database and database reptiles, and designed a database record local storage management model and quaternary structure of the data source (to the greatest extent possible is a shared data source, speed up the indexing speed), followed by design a general incremental record database monitoring methods, and the method used and the police platform project.Police platform through the practical application of the project, the design of this full-text indexing and retrieval methods are in line with the expected incremental efficiency index, and can support a large field domain of full-text search, database interface designed to handle correctly from MySQL and SQL SEVER full-text index data sources, and users can freely extended to support the database type.