As the developing of search technology, original web search was restrained by IE, so that the capability about application is not very well。 Desktop search engine equals the application mode combined of client and database, especially for the new function called drag search, hence, it make the various and personal search possible。According to the conception of traditional search engine and the recently coming desktop search engine, this thesis post a solution for whole series of search from client to server and implement those。 This solution includes Spider, multithread downloader, Unicode file storage, HTML/XML parser,Tokenization,Hash 2 indexed database,Web service, Pagerank, Drag search。Spider is charge of gathering all the link information from internet, including normal web page, office document, picture, multimedia, flash animation。 Spider put their link information and interrelated information (update time, source website) into database, and record the link from URL to be prepared for pagerank 。... |