Transducer-based Rewriting Games for Active XML

TL;DR

This paper explores transducer-based rewriting games on nested XML-like structures, analyzing their properties and computational complexity, with some scenarios being computationally feasible and others highly complex or undecidable.

Contribution

It introduces transducer models for nested words in context-free games and studies their properties and complexity, a novel approach in Active XML rewriting.

Findings

Complexity ranges from NP-complete to undecidable depending on the scenario.

Identifies some restrictions where the winning problem becomes tractable.

Provides a formal framework for transducer-based rewriting in nested structures.

Abstract

Context-free games are two-player rewriting games that are played on nested strings representing XML documents with embedded function symbols. These games were introduced to model rewriting processes for intensional documents in the Active XML framework, where input documents are to be rewritten into a given target schema by calls to external services. This paper studies the setting where dependencies between inputs and outputs of service calls are modelled by transducers, which has not been examined previously. It defines transducer models operating on nested words and studies their properties, as well as the computational complexity of the winning problem for transducer-based context-free games in several scenarios. While the complexity of this problem is quite high in most settings (ranging from NP-complete to undecidable), some tractable restrictions are also identified.