Detection and extraction method of repeated fragments in software codes

Abstract

The invention provides a detection and extraction method of repeated fragments in software codes. The method comprises steps 1-8 in figures, hierarchical information in a syntax tree is adopted for extraction of code fragments, and grammar information in the code fragments is considered, so that the extracted code fragments are significant. Besides, the extraction process of the code fragments is controlled by a duplicate checking mechanism based on the reverse index; low-level extraction is not performed if high-level repetition is found. Compared with the manner that repetition checking is performed on the smallest fragments and then the fragments are combined in most techniques at present, the extraction method can reduce a large calculated amount. In the process, the size of a detection window can be adjusted automatically according to whether repeated contexts actually exist, the performance is improved, the detection speed is high, and the method can be applied to real scene detection. In addition, the method combines grammatical structure information, so that the misjudgment rate is very low.

Description

technical field [0001] The invention belongs to the field of computer programs, in particular to a method for detecting repeated segments in software codes. Background technique [0002] Code duplication detection is very important in software development. First of all, code duplication detection can improve the maintainability of software. If duplicate codes are allowed to be scattered everywhere, then if one code needs to evolve or undergo defect repair, the code in other places will also evolve or undergo defect repair, which will affect maintainability. Through code duplication detection to find duplication in the code, they can be extracted into functions in time to improve maintainability. Secondly, it can reduce the legal risk in software development. There are different licenses in software development. If the developer accidentally copies infectious license information (such as the GNU license) due to the negligence of the developer, it will bring serious harm to ...

AI Technical Summary

Problems solved by technology

This algorithm is slow, and the effect is not very good, and there are few practical applications

[0008] 5. Method based on code quality measurement (K.Kontogiannis, R.DeMori, E.Merlo, M.Galler, M.Bernstein, Pattern matching for clone and concept detection, Journal of Automated Software Engineering 3(1–2)(1996) 77–108), this method is inefficient and computationally intensive, so it is rarely used in practice

[0009] 6. Index-based (also known as inverted index) approach (Benjamin Hummel, Elmar Juergens, Lars Heinemann, and Michael Conradt. Index-based code clone detection: Incremental, distributed, scalable. In the International Conference on Software Maintenance, pages 1 –9, sept.2010), the index-based method is very efficient, but the only index-based sliding window scheme currently has poor performance and detection quality, is not stable enough, and has poor robustness

Images