2006/04/13 by Lifshits, Yury
#Computational Complexity (cs.CC) #Data Structures and Algorithms (cs.DS) #E.4 #F.2.2 #FOS: Computer and information sciences #I.7
paper · doi:10.48550/arxiv.cs/0604058
Here we study the complexity of string problems as a function of the size of a program that generates input. We consider straight-line programs (SLP), since all algorithms on SLP-generated strings could be applied to processing LZ-compressed texts. The main result is a new algorithm for pattern matching when both a text T and a pattern P are presented by SLPs (so-called fully compressed pattern matching problem). We show how to find a first occurrence, count all occurrences, check whether any given position is an occurrence or not in time O(n2m). Here m,n are the sizes of straight-line programs generating correspondingly P and T. Then we present polynomial algorithms for computing fingerprint table and compressed representation of all covers (for the first time) and for finding periods of a given compressed string (our algorithm is faster than previously known). On the other hand, we show that computing the Hamming distance between two SLP-generated strings is NP- and coNP-hard.