B. C. Smith. Re ection and semantics in a procedural language. Technical Report MIT-LCSTR -2721, MIT, 1982.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....rst version of a prototype implementation in which only omission failures occur. 1 Introduction Most existing literature on re ective architectures is concerned with language interpreters and objectoriented programming methods, e.g. a Lisp interpreter written in Lisp. Examples include 3Lisp [27] and object oriented re ection [28] Their purpose is typically to make the internal operation of components inspectable and modi able by a user, to enable more exibility in implementation and experimentation with language extensions. In contrast, there is little work on re ective architectures ....

B. C. Smith. Re ection and semantics in a procedural language. Technical Report MIT-LCSTR -2721, MIT, 1982.

....of e 0 . These constructs are similar to those proposed in (Moggi et al. 1999; Benaissa et al. 1999; Taha Sheard, 1997; Davies Pfenning, 1996) In MiniML meta ref the meta level is re ected in the object level, thus one gets an in nite 6 Calcagno, Moggi, Sheard tower of levels (Smith, 1982): 0 is the meta level, 1 is the re ection of the meta level (i.e. the object level) 2 is the re ection of the re ection of the meta level, The main technical results are: type safety for MiniML meta ref , i.e. evaluation of well typed programs does not cause a run time error (Section ....

Smith, Brian Cantwell. (1982). Re ection and semantics in a procedural language. Ph.D.

....x e 1 y x e 2 y x e 3 y x (tagE:e) y E x e y x (untagE:e) y E 1 x e y x (tagF:e) y F x e y x (untagF:e) y F 1 x e y It is easy to see that x p e q y e. Such encoding decoding pairs have sometimes been called reify and re ect. This terminology was promoted by Brian Cantwell Smith [13]. Reify provides us with a way of materializing or representing 10 terms within the language, and re ect provides us with a way of interpreting an internal representation back into a (meta level) term. Note that all these functions exist at the meta level, and that expressing them within the ....

Brian Cantwell Smith. Re ection and Semantics in a Procedural Language. PhD thesis, Massachusetts Institute of Technology, January 1982.

....0 , and run e corresponds to the execution of e 0 . These constructs are similar to those proposed in (Moggi et al. 1999; Benaissa et al. 1999; Taha Sheard, 1997; Davies Pfenning, 1996) In MiniML meta ref the meta level is re ected in the object level, thus one gets a tower of levels (Smith, 1982), indexed by natural numbers: 0 is the meta level, 1 the re ection of the meta level, 2 the re ection of the re ection of the meta level, and so on. The main technical results are: type safety for MiniML meta ref , i.e. evaluation of well typed programs does not cause a run time error (Section ....

Smith, Brian Cantwell. (1982). Re ection and semantics in a procedural language. Ph.D.

....1 y x e 2 y x e 3 y x (tagE:e) y E x e y x (untagE:e) y E 1 x e y x (tagF:e) y F x e y x (untagF:e) y F 1 x e y It is easy to see that x p e q y e. Such encoding decoding pairs have sometimes been called reify and re ect. This terminology was promoted by Brian Cantwell Smith [12]. Reify provides us with a way of materializing or representing terms within the language, and re ect provides us with a way of interpreting an internal representation back into a (meta level) term. Note that all these 7 The second requirement is not hard: it is enough to know that D can ....

Brian Cantwell Smith. Re ection and Semantics in a Procedural Language. PhD thesis, Massachusetts Institute of Technology, 1982.

....1 y x e 2 y x e 3 y x (tagE:e) y E x e y x (untagE:e) y E 1 x e y x (tagF:e) y F x e y x (untagF:e) y F 1 x e y It is easy to see that x p e q y e. Such encoding decoding pairs have sometimes been called reify and re ect. This terminology was promoted by Brian Cantwell Smith [13]. Reify provides us with a way of materializing or representing terms within the language, and re ect provides us with a way of interpreting an internal representation back into a (meta level) term. Note that all these functions exist at the meta level, and that expressing them within the ....

Brian Cantwell Smith. Re ection and Semantics in a Procedural Language. PhD thesis, Massachusetts Institute of Technology, January 1982.

....the outer redex is reduced rst. It is worth noting that the counter example is a program. 4 Related Work While there are a large number of studies of LISP with which we share either our general methodology (e.g. Gor75, Car76, Mas86, MT89, FFKD86, Fel88] or our general subject matter (e.g. [Pit80, MP80, Smi82] or [Smi84, dRS84, FW84, FW86, DM88, Baw88] we are unfamiliar with any axiomatic treatments of LISP s metalinguistic power. The rst structured operational semantics for LISP was de ned for M expression LISP by Gordon [Gor75] This was used as the basis for an induction principle for proving ....

B. Smith. Re ection and Semantics in a Procedural Language. PhD thesis, Massachusetts Institute of Technology, 1982.

.... v , it is obviously not the case that reify [ 0 v is an extension (conservative or otherwise) of v . 23 5 Related Work While there are many studies of LISP with which we share either our general methodology [Gor75, BM75, Car76, Mas86, MT89, FFKD86, Fel88] or our general subject matter [Pit80, MP80, Smi82, Smi84, RS84, FW84, FW86, DM88, Baw88], we are unfamiliar with any attempts to axiomatize LISP s metalinguistic facilities. We are also unfamiliar with any reference in the literature to the problematic nature of equation (12) in McCarthy s original representation function. 5.1 Gordon The rst structured operational semantics for ....

Brian Smith. Re ection and Semantics in a Procedural Language. PhD thesis, MIT, Cambridge, MA, 1982.

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