Errata for “Faster index for property matching”

Abstract

In [C.S. Iliopoulos, M.S. Rahman, Faster index for property matching, Information Processing Letters 105 (2008) 218–223], Iliopoulos and Rahman proposed a data structure called IDS-PIP for solving the property indexing pattern matching problem. IDS-PIP can be constructed in O ( n ) time, where n is the length of the text. Then, based on ID-PIP, each query takes O ( m log | Σ | + K ) time, where m is the length the pattern, Σ is the alphabet, and K is the output size. They assume that all intervals in property π are disjoint. If the intervals in property π are not disjoint, then they create an equivalent set of disjoint intervals π ′ to replace π. However, property π ′ is not equivalent to property π at all. In this erratum, we propose a way for finding correct property π ′ and modify IDS-PIP slightly.