Speculative data-parallel algorithms for language recognition have been widely experimented for various types of finitestate automata (FA), deterministic (DFA) and nondeterministic (NFA), often derived fromregular expressions (RE). Such an algorithm cuts the input string into chunks, independently recognizes each chunk in parallel by means of identical FAs, and at last joins the chunk results and checks the overall consistency. In chunk recognition, it is necessary to speculatively start the FAs in any state, thus causing an overhead that reduces the speedup over a serial algorithm. The existing data-parallel DFA-based recognizers suffer from an excessive number of starting states, and the NFA-based ones suffer from the number of nondeterministic transitions.

Summary In this paper we study the performance of POSIX submatch extraction algorithms based on nondeterministic finite automata (NFA). We propose an algorithm that combines Laurikari tagged NFA and extended Okui‐Suzuki disambiguation. The algorithm works in worst‐case O(n m2 t) time and O(m2) space (including preprocessing), where n is the length of input, m is the size of the regular expression with bounded repetition expanded and t is the number of capturing groups and subexpressions that contain them. On real‐world benchmarks our algorithm performs close to the O(n m t) complexity of leftmost‐greedy matching, although on artificial benchmarks it can be significantly slower. We propose a lazy version of the algorithm that runs much faster, but requires O(n m2) space. We show that the Kuklewicz algorithm is slower in practice, and the backward matching algorithm proposed by Cox is incorrect.