Optimization of Single Pattern Matching Based on Sunday Algorithm

There are two problems exist in the application of Sunday algorithm to UTF-8 encoding in Chinese environment. For one thing, if the first word or the last word has high probability of occurrence, the number of matches will increase and time performance will decrease since matching strings apply left-to-right or right-to-left matching method when using original matching rule. For the other, directly using Chinese characters to generate auxiliary jump table will cause the expansion of space. While Splitting Chinese characters into 3 bytes for processing can reduce the space consumption, the execution speed of matching will be affected. This paper utilizes feature characters and uses internal correlation of Chinese unit in UTF-8 to solve the problem of the degradation of time performance mentioned above. The modified algorithm optimized the auxiliary jump table and matching rules of original algorithm, improving time performance in this environment in theory. Finally, we compare and test the time performance of the optimized algorithm in practice.