G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming, Proceedings of the Tenth International Conference on Logic Programming, pages 354--373. MIT Press, 1993. 26  Home/Search   Document Details and Download   Summary   Related Articles   Check

....One of them is the possibility of tackling critical foundation problems of meta programming within a framework with a strong theoretical basis. Another is the surprising ease of programming. These reasons motivated an intensive research on meta programming inside the logic programming community [1, 2, 4, 7, 8]. On the other hand, termination analysis is one of the most intensive research areas in logic programming as well [3, 5] In [10] we present a novel approach to termination analysis of meta programs. Our research has been motivated by the famous vanilla meta interpreter: solve(true) ....

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming, Proc. of the 10th Int. Conference, pp. 354-373. MIT Press, 1993.

....In fact, the approach allows mixed object level and metalevel rules. Again, the application of the linking rules (which coincides, in practice, with the invocation of Demo) is left to the user, i.e. reflection is explicit. The semantics of this approach is, however, not easy to define (see e.g. [41, 26, 48, 53, 64]) and holds only if the metatheory and the linking rules provide an extension to the basic Horn clause language which is conservative, i.e. only if Demo is a faithful representation of Horn clause provability. Although the amalgamated language is far more expressive than the object language ....

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming -- Proc. 10th Intl. Conf. on Logic Programming, pages 354--373, Cambridge, 1993. MIT Press. 40

....One of them is the possibility of tackling critical foundation problems of meta programming within a framework with a strong theoretical basis. Another is the surprising ease of programming. These reasons motivated an intensive research on metaprogramming inside the logic programming community [4, 16, 19, 22, 23]. On the other hand, termination analysis is one of the most intensive research areas in logic programming as well. See [12] for the survey. More recent work on this topic can be found among others in [14, 18, 20, 24, 30] Traditionally, termination analysis of logic programs have been done ....

....sound and complete: solve(A) fail: It is sound, since there are no computed answers for solve(G) It is also complete, since s 1 ; s n required in Definition 7 do not exist. The vanilla meta interpreter M 0 (Example 1) is also both sound and complete, as shown by Levi and Ramundo in [19]. The following meta interpreter M 2 , mimicking the LD refutation with bounded depth (the depth provided as the second argument) is sound, but is not complete. solve(true; 0) solve( A; B) N) solve(A; N) solve(B; N) solve(A; s(N) clause(A; B) solve(B; N) It is intuitively clear why ....

[Article contains additional citation context not shown here]

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming, Proceedings of the Tenth International Conference on Logic Programming, pages 354--373. MIT Press, 1993.

....One of them is the possibility of tackling critical foundation problems of meta programming within a framework with a strong theoretical basis. Another is the surprising ease of programming. These reasons motivated an intensive research on meta programming inside the logic programming community [1, 8, 10, 11]. See also [6] for a survey. On the other hand, termination analysis is one of the most intensive research areas in logic programming as well. See [3] for the survey. More recent work on this topic can be found among others in [4, 5, 7, 9, 12, 14 16] Traditionally, termination analysis of ....

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming, Proceedings of the Tenth International Conference on Logic Programming, pages 354--373. MIT Press, 1993.

....work concerning the procedural and declarative interpretation of the logic programming counterpart of this program, that is the program without cuts. The early references include Kowalski [Kow79] Bowen and Kowalski [BK82] and Hill and Lloyd [HL88] The more recent references are Levi and Ramundo [LR93] Martens and de Schreye [MS95b, MS95a] and Kalsbeek [Kal95] Pedreschi and Ruggieri [PR96] studied various correctness aspects of the META INTERPRETER program and of its extensions. Most of the other programs presented here are taken from the books of Bratko [Bra86] Clocksin and Mellish [CM84] ....

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Proceedings ICLP'93, pages 354--373. The MIT Press, Cambridge, MA, 1993.

....in facts of the form clause(A; A 1 : An ) for every clause A A 1 ; An (n 1) in P , and facts of the form clause(A; true) for every fact A in P . 58 The predicate demo is the Horn clause representation of the entailment relation for definite logic programs, that is [LR93] p(t 1 ; t n ) 2 TP , demo(p(t 1 ; t n ) 2 T VP where VP is the program demo augmented with the definition of clause, and T is the fix point S semantics. Exists a variant, namely the multi vanilla metacircular interpreter for positive logic programming [Ste84, Ste85] ....

Giorgio Levi and Davide Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Proceedings of the 10th International Conference on Logic Programming, pages 354--373. MIT Press, 1993.

....for which the least Herbrand model semantics still holds for the vanilla metainterpreter and a limited form of amalgamation. This is the class of language independent programs, where language independence extends both domain independence and range restrictedness. As already noted by Levi Ramundo [19], however, the class of language independent programs is too small: it includes deductive database programs, but rules out any logic program computing partially determined data structures. Instead they point out another possibility for giving a declarative semantics to the vanilla metainterpreter ....

Levi, G. and Ramundo, D., A Formalization of Metaprogramming for Real, in: D. S. Warren (ed.), Logic Programming --- Proc. 10th Intl. Conf. on Logic Programming, MIT Press, Cambridge, 1993.

....by partial evaluation of an interpreter using a proper ground representation. In addition, De Schreye Martens has presented a least Herbrand model semantics for what they call untyped vanilla metaprogramming, including a form of amalgamation [23] These results are extended by Levi Ramundo [57]. Enhancing metainterpreters. Sterling has proposed a methodology primarily intended for building knowledge systems [82] where a given vanilla interpreter (essentially the three line interpreter described previously) is enhanced with features such as generation of explanations (proofs) depth ....

Levi, G. and Ramundo, D., A Formalization of Metaprogramming for Real, in: D. S. Warren (ed.), Proc. 10th Intl. Conf. on Logic Programming, MIT Press, Cambridge, Mass., 1993.

....of the object program and atoms demo(pp(t 1 ; t n )q) in the model of the meta program. They also considered limited forms of amalgamation through various textual combinations of the object program and meta program. Levi and Ramundo generalised these results to definite object programs in [11] by using the S semantics [7] of programs rather than the least Herbrand model semantics. Levi and Ramundo also consider an amalgamated vanilla metainterpreter which has a clause for dealing with demo atoms in the object program obtaining a version of their results for this case also. The ....

Giorgio Levi and Davide Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Proceedings of the Tenth International Conference on Logic Programming, pages 354--473. MIT Press, 1993.

.... is coped with by building the theories: the theory of an object can be constructed by some simple list building operations (on the names of clauses) What the current METAL implementation does in practice seem very similar to the description of inheritance in differential logic programming [7, 20]. The procedural semantics of METAL can also be based on the procedural semantics of DM although currently METAL is not implemented in this way. A direct implementation of the procedural semantics of DM has been done and is not too difficult. There are three remaining problems for a complete ....

Giorgio Levi and Davide Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Proceedings of the Tenth International Conference on Logic Programming, pages 354--473. MIT Press, 1993.

....vanilla and of the inheritance metainterpreter) that for any atomic goal, the success derivations in the specialized program have exactly the same length of the success derivations of the corresponding object level program. All the proofs of the new theorems reported in the paper can be found in [23]. 2 The semantics 2.1 The vanilla metainterpreter Let V denote the vanilla metainterpreter, i.e. the Horn clause representation of the proof procedure of definite logic programs, using the non ground representation for object level variables. 1: demo(true) 2: demo(X Y ) demo(X) demo(Y ) ....

G. Levi and D. Ramundo. A formalization of metaprogramming for real. Technical report, Dipartimento di Informatica, Universit`a di Pisa, November 1992.

....that this is exactly the class of language independent programs as defined in [37] Definition 3.9 [37] A program P with underlying language LP is language independent iff, for any extension L 0 of LP , its least L 0 Herbrand model is equal to its least LP Herbrand model. Theorem 3. 10 [85] Let P be a program. Then P is language independent iff O(P ) O 1 (P ) A program P belongs to this class only if any goal in P returns ground answers. It is therefore essentially the class of allowed positive programs [88] and does not contain any program able to compute partial data ....

....let us denote by O L (P ) the denotation for a given language L. If LP is the language underlying program P , the following theorem shows the language independence property. Note that the same property does not hold for other variable based semantics, such as those in [30, 49] Theorem 3. 11 [85] If P is a positive program, then O LP (P ) O L 0 (P ) for any extension L 0 of LP . As we will show in section 7.4, this is the key property which makes the s semantics adequate to formalize metaprogramming with the non ground metalevel representation of object level variables. 3.2 ....

[Article contains additional citation context not shown here]

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Proc. Tenth Int'l Conf. on Logic Programming, pages 354--373. The MIT Press, Cambridge, Mass., 1993.

No context found.

G. Levi and D. Ramundo. A formalization of metaprogramming for real. In D. S. Warren, editor, Logic Programming, Proceedings of the Tenth International Conference on Logic Programming, pages 354--373. MIT Press, 1993. 26

No context found.

Levi, G., Ramundo, D.: A formalization of metaprogramming for real. In Warren, D.S., ed.: Logic Programming - Procs. of the Tenth International Conference, Cambridge, Mass., The MIT Press (1993) 354--373

No context found.

G. Levi and D. Ramundo. A Formalization of Metaprogramming for Real. In D.S. Warren, editor, Proceedings Tenth International Conference on Logic Programming, pages 354-373. The MIT Press, 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC