Font Size: a A A

Study And Implementation On Information Storage And Retrieval Technique For Program Understanding

Posted on:2008-07-16Degree:MasterType:Thesis
Country:ChinaCandidate:H H YeFull Text:PDF
GTID:2178360242470581Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
As the development of software engineering field, the problem of software maintenance has become an important question for discussion currentlly in software industry. In the process of software maintenance, program understanding is a bald, time-consuming and difficult task, which needs program understanding aided systems to help. And that, the information storing and querying efficiency is one of the key factors that influence on the efficiency of program understanding aided systems, so much as program understanding. At present, most program understanding tools adopt professional database management systems (DBMS) to realize the function of storing and querying. However, professional DBMS take a lot of other factors into account, thus, the efficiency of storing and querying is reduced. In addition, professional DBMS demand to use special query languages which increaseds the process time of storing and the difficulty of program understanding.Aimed at the problems mentioned above, the main task of this paper is to research and design a program database with better efficiency of storing and querying, sustaining larger programs and incremented update. Firstly, according to the symbol types of programming language, put forward the idea of information sorted storing. According to the types, put the information into different base files, then binding all the files together to form a program database. Secordly, on the basis of professional DBMS' storing and querying mechanism, analyze the capability and index mechanism of the B+-tree, put forward the idea of using the structure of B+-tree to design the program database model. Optimize the structure of B+-tree from three aspects: inserting batch keywords from bottom to up, storing keywords with compressing technique, choosing a suit way-number, and design index within and among nodes. Thirdly, according to the needs of program understanding, design four query methods, satisfying the demands of four windows in program understanding aided systems. At last, do experiments with a series of files composing of different amount of information records, test the program database system's storing space and querying time. The result shows that under the circumstance of sacrificing some space, the program database achieves quickly storing and querying capability.
Keywords/Search Tags:program database, B+-tree, sorted storing, prefix compression, program understanding
PDF Full Text Request
Related items