Abstract

ABSTRACT. Pattern matching is the problem of finding all occurrences of a pattern in a text. For a long period of time significant progress has been made in solving increasingly more generalized and dynamic versions of this problem. In this paper we introduce a fully dynamic generalization of the pattern matching problem. We show how to maintain a family of strings under split and concatenation operations. Given a string in the family, all occurrences of it in the family are reported within time occ time, where is the total size of the strings and occ is the number of occurrences. Updates are performed in time. These bounds are competitive or improve former results for less generalized versions of the problem. As an intermediate result of independent interest, we provide an almost quadratic improvement of the time bounds for the dynamic string equality problem due to Mehlhorn, Sundar and Uhrig.

Citations