The Timed Concurrent Constraint programming language (tccp in short),  was introduced for modeling reactive systems. This language allows one to model in a very intuitive way typical ingredients of  these systems such as timeouts, preemptions, etc. However, there is no natural way for modeling other desirable features such as functional computations, for example for calculating arithmetic results. In fact, although it is certainly possible to implement such kind of operations,  each single step of the computation takes time in tccp, and avoiding interferences with the intended overall behavior of the (reactive) system is quite involved.

In this paper, we propose an extension of tccp for modeling instantaneous computations which  improves the  expressiveness of the language, in the sense that operations that are cumbersome to implement in pure tccp, are executed by calling an efficient, external functional engine, while the tccp programmer can focus on the pure, and usually more complex, reactive part of the system. We also describe a case study which motivates the work, and discuss how the new capability presented here can also be used as a new tool for developers from the verification point of view.