Innermost Strategy Research Articles

Faithful (meta-)encodings of programmable strategies into term rewriting systems

We consider the problem of inferring a grammar describing the output of a functional program given a grammar describing its input. Solutions to this problem are helpful for detecting bugs or proving safety properties of functional programs, and several rewriting tools exist for solving this problem. However, known grammar inference techniques are not able to take evaluation strategies of the program into account. This yields very imprecise results when the evaluation strategy matters. In this work, we adapt the Tree Automata Completion algorithm to approximate accurately the set of terms reachable by rewriting under the innermost strategy. We formally prove that the proposed technique is sound and precise w.r.t. innermost rewriting. We show that those results can be extended to the leftmost and rightmost innermost case. The algorithms for the general innermost case have been implemented in the Timbuk reachability tool. Experiments show that it noticeably improves the accuracy of static analysis for functional programs using the call-by-value evaluation strategy.

Needed Computations Shortcutting Needed Steps

We define a compilation scheme for a constructor-based, strongly-sequential, graph rewriting system which shortcuts some needed steps. The object code is another constructor-based graph rewriting system. This system is normalizing for the original system when using an innermost strategy. Consequently, the object code can be easily implemented by eager functions in a variety of programming languages. We modify this object code in a way that avoids total or partial construction of the contracta of some needed steps of a computation. When computing normal forms in this way, both memory consumption and execution time are reduced compared to ordinary rewriting computations in the original system.

Normalization properties for Shallow TRS and Innermost Rewriting

Computation with a term rewrite system (TRS) consists in the application of rules from a given starting term until a normal form is reached. Two natural questions arise from this definition: whether all terms can reach at least one normal form (weak normalization property), and whether all terms can reach at most one normal form (unique normalization property). Innermost rewriting is one of the most studied and used strategies for rewriting, since it corresponds to the “call by value” computation of programming languages. Henceforth, it is meaningful to study whether weak and unique normalization are indeed decidable for a significant class of TRS with the use of the innermost strategy. In this paper we study weak and unique normalization for shallow TRS and innermost rewriting. We show decidability of unique normalization in polynomial time, and decidability of weak normalization in doubly exponential time. We obtain an exptime-hardness proof for the latter case. Nevertheless, when a TRS satisfies the unique normalization property, we conclude the decidability of the weak normalization property in polynomial time, too. Hence, whether all terms reach one and only one normal form is decidable in polynomial time. All of these results are obtained assuming that the maximum arity of the signature is fixed for the problem, which is a usual assumption.

Using Context-Sensitive Rewriting for Proving Innermost Termination of Rewriting

Computational systems based on reducing expressions usually have a predefined reduction strategy to break down the nondeterminism which is inherent to reduction relations. The innermost strategy corresponds to call by value or eager computation, that is, the computational mechanism of several programming languages like Maude, OBJ, etc. where the arguments of a function call are always evaluated before calling the function. This strategy usually fails to terminate when nonterminating computations are possible in the programs and many eager programming languages also admit the explicit specification of a particular class of strategy annotations to (try to) avoid them. Context-Sensitive Rewriting provides an abstract model to describe and analyze the operational behavior of such programs. This paper aims at contributing to the development of appropriate techniques and tools for the verification of program termination in the aforementioned programming languages, so we focus on termination of innermost (context-sensitive) rewriting. We adapt the notion of usable argument introduced by Fernandez to prove innermost termination by proving termination of context-sensitive rewriting. Thanks to our recent developments for proving termination of (innermost) context-sensitive rewriting using dependency pairs, now we can also relax monotonicity requirements for proving innermost termination of (context-sensitive) rewriting. We have implemented these new improvements in the termination tool mu-term and evaluated the results with some benchmarks.

Closure of Tree Automata Languages under Innermost Rewriting

Preservation of regularity by a term rewriting system (TRS) states that the set of reachable terms from a tree automata (TA) language (aka regular term set) is also a TA language. It is an important and useful property, and there have been many works on identifying classes of TRS ensuring it; unfortunately, regularity is not preserved for restricted classes of TRS like shallow TRS. Nevertheless, this property has not been studied for important strategies of rewriting like the innermost strategy – which corresponds to the call by value computation of programming languages. We prove that the set of innermost-reachable terms from a TA language by a shallow TRS is not necessarily regular, but it can be recognized by a TA with equality and disequality constraints between brothers. As a consequence we conclude decidability of regularity of the reachable set of terms from a TA language by innermost rewriting and shallow TRS. This result is in contrast with plain (not necessarily innermost) rewriting for which we prove undecidability. We also show that, like for plain rewriting, innermost rewriting with linear and right-shallow TRS preserves regularity.

Compact Normalisation Trace via Lazy Rewriting

Innermost strategies are usually used in compiling term rewriting systems (TRSs) since they allow to efficiently build result terms in a bottom-up fashion. However, innermost strategies do not always give the shortest normalising derivation. In many cases, using an appropriate laziness annotation on the arguments of function symbols, we evaluate lazy arguments only if it is necessary and hence, get a shorter derivation to normal forms while avoiding non-terminating reductions. We provide in this work a transformation of annotated TRSs, that allows to compute normal forms using an innermost strategy and to extract lazy derivations in the original TRS from normalising derivations in the transformed TRS. We apply our result to improve the efficiency of equational reasoning in Coq proof assistant using ELAN as an external rewriting engine.

A rewriting approach to binary decision diagrams

Binary decision diagrams (BDDs) provide an established technique for propositional formula manipulation. In this paper, we present the basic BDD theory by means of standard rewriting techniques. Since a BDD is a DAG instead of a tree we need a notion of shared rewriting and develop appropriate theory. A rewriting system is presented by which canonical reduced ordered BDDs (ROBDDs) can be obtained and for which uniqueness of ROBDD representation is proved. Next, an alternative rewriting system is presented, suitable for actually computing ROBDDs from formulas. For this rewriting system a layerwise strategy is defined, and it is proved that when replacing the classical apply-algorithm by layerwise rewriting, roughly the same complexity bound is reached as in the classical algorithm. Moreover, a layerwise innermost strategy is defined and it is proved that the full classical algorithm for computing ROBDDs can be replaced by layerwise innermost rewriting without essentially affecting the complexity. Finally a lazy strategy is proposed sometimes performing much better than the traditional algorithm.

Call-by-value Solvability

The notion of solvability in the call-by-value λ -calculus is defined and completely characterized, both from an operational and a logical point of view. The operational characterization is given through a reduction machine, performing the classical β -reduction, according to an innermost strategy. In fact, it turns out that the call-by-value reduction rule is too weak for capturing the solvability property of terms. The logical characterization is given through an intersection type assignment system, assigning types of a given shape to all and only the call-by-value solvable terms.