COOSy (Curry Object Observation System)

This package contains the implementation of a tool to observe the execution of programs written in the declarative multi-paradigm language Curry

Details about COOSy and its implementation can be found in this paper:

B. Brassel, O. Chitil, M. Hanus, F. Huch: Observing Functional Logic Computations, Proc. of the Sixth International Symposium on Practical Aspects of Declarative Languages (PADL’04), Springer LNCS 3057, pp. 193-208, 2004

Important note

In contrast to the version described in the original paper, the current version also considers the fact that free variables can only occur in expressions with a Data context (see the paper Adding Data to Curry for a detailed explanation). Therefore, the current observation tool distinguishes the observation of expressions with or without a Data context: the operation Observe.observe can observe expressions containing free variables which are instantiated during the evaluation (which is the main innovation of COOSy), whereas the operation Observe.observeG can also observe expressions without a Data context but requires that they are always instantiated (otherwise, the observation might suspend). Since functional values have no Data context, one has to distinguish how to observe the arguments. For instance, observers with observer type

(oList oInt ~> oList oInt)

can observe functions which are possibly invoked with free variables and might return a free variable, whereas

(oList oInt ~~> oList oInt ~~> oInt)

observe functions which are possibly invoked with free variables but return a non-free results and

(oList oInt ~~~> oList oInt ~~~> oMaybe oInt)

observe functions which are called with non-free arguments and return non-free results.

The examples directory contains various programs containing observers.

Developed by
* Bernd Brassel (CAU Kiel, Germany, bbr@informatik.uni-kiel.de) * Olaf Chitil (University of Kent, UK, O.Chitil@kent.ac.uk) * Michael Hanus (CAU Kiel, Germany, mh@informatik.uni-kiel.de) * Frank Huch (CAU Kiel, Germany, fhu@informatik.uni-kiel.de)
Version of February 27, 2023