D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67--99, January 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... [8] The seminal work on the CPS transformation and on shift reset is Danvy s and Filinski s 1992 paper [9] Filinski shows how to implement shift reset via call cc and a mutable reference [13] Danvy and Filinski note that shift reset coincides with Sitaram s and Felleisen s F operationally [24, 8]. Moreau and Queinnec also employ marks on the stack to define marker call pc, another pair of control operators for composable continuations [23] Gunter, R emy, and Riecke investigate another alternative approach to composable continuations via named prompts [15] They also mention the ....

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67-- 99, January 1990.

....a fresh prompt (with the type of the expression) and set s; all callcc s are then done via cupto s to this fresh prompt. To determine whether named, typed prompts are useful in programming requires some experience in writing programs. In the untyped case, prompts add signi cant expressive power [20, 21]; we believe the examples of [20] could be typed in our system. We also conjecture that many applications that currently uses callcc (such as various threads packages or CML) could bene t for instance, the explicit prompt mechanism may simplify the implementation of threads in a interactive ....

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. LISP and Symbolic Computation, 3:67-99, 1990.

....to understand whether the exchange is cost e ective, and if it can be further optimized by better box technology. Our proofnet technology might be extended to languages with functional control operators, such as Danvy and Filinski s shift and reset, and Sitaram and Felleisen s control and prompt [7, 30]. While shift and reset are de ned in terms of a CPS transformation, they do not have return type ; the DS transformation is then inapplicable. Both shift and reset, and control and prompt can be de ned in terms of call cc and a reference cell [11, 30] Bawden has shown how to implement reference ....

....and Sitaram and Felleisen s control and prompt [7, 30] While shift and reset are de ned in terms of a CPS transformation, they do not have return type ; the DS transformation is then inapplicable. Both shift and reset, and control and prompt can be de ned in terms of call cc and a reference cell [11, 30]. Bawden has shown how to implement reference cells using sharing graphs [4] so this strategy may still lead to an interesting proofnet implementation. 6 Conclusions We have shown how to implement various languages with explicit control using graph reduction, where the structure of the graphs ....

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67{ 99, January 1990. 13

....to understand whether the exchange is cost e ective, and if it can be further optimized by better box technology. Our proofnet technology might be extended to languages with functional control operators, such as Danvy and Filinski s shift and reset, and Sitaram and Felleisen s control and prompt [6, 27]. While shift and reset are de ned in terms of a CPS transformation, continuations do not have return type ; the DS transformation is then inapplicable. Both shift and reset, and control and prompt can be de ned in terms of call cc and a reference cell [10, 27] Bawden has shown how to implement ....

....and Felleisen s control and prompt [6, 27] While shift and reset are de ned in terms of a CPS transformation, continuations do not have return type ; the DS transformation is then inapplicable. Both shift and reset, and control and prompt can be de ned in terms of call cc and a reference cell [10, 27]. Bawden has shown how to implement reference cells using sharing graphs [3] so this strategy may still lead to an interesting proofnet implementation. 6 Conclusions We have shown how to implement various languages with explicit control using graph reduction, where the structure of the graphs ....

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67-99, January 1990.

....a fresh prompt (with the type of the expression) and set s; all callcc s are then done via cupto s to this fresh prompt. To determine whether named, typed prompts are useful in programming requires some experience in writing programs. In the untyped case, prompts add significant expressive power [20, 21]; we believe the examples of [20] could be typed in our system. We also conjecture that many applications that currently uses callcc (such as various threads packages or CML) could benefit for instance, the explicit prompt mechanism may simplify the implementation of threads in a interactive ....

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. LISP and Symbolic Computation, 3:67--99, 1990.

....a fresh prompt (with the type of the expression) and set s; all callcc s are then done via cupto s to this fresh prompt. To determine whether named, typed prompts are useful in programming requires some experience in writing programs. In the untyped case, prompts add significant expressive power [16, 15]; we believe the examples of [15] could be typed in our system. We also conjecture that many applications that currently uses callcc (such as various threads packages or CML) could benefit for instance, the explicit prompt mechanism may simplify the implementation of threads in a interactive ....

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. LISP and Symbolic Computation, 3:67--99, 1990.

....directly, without involving monads at all. Examples include uses of basic call cc for thread packages [Wan80] and imperative backtracking [HDM93] simple composable control [FWFD88, DF92] for nondeterminism and other basic effects, and a number of proposals for hierarchical control [DF90, SF90, GRR95] to represent general layered e ects. Most of these are based on operational de nitions of the control operators in terms of their actions on evaluation contexts (although many also include sample implementations in terms of Scheme primitives) Again, we take a more minimalist approach ....

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67{ 99, January 1990.

....in a later paper by Queinnec and Serpette [22] Our work is based on work by Felleisen, et al. 11, 9, 12] Johnson and Duggan [18] have developed a notion of partial continuations that also extends traditional continuation control in a similar manner. In a related work, Sitaram and Felleisen [23] introduce techniques to constrain the effects of prompts and functional continuations. They do so, however, by developing complicated protocols on top of primitive control structures, and they do not address concurrency issues. Miller [21] does address the issue of using continuations in a ....

Sitaram, Dorai and Felleisen, Matthias. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3, 1 (January 1990) 67-- 99.

....Prompts Both A and callcc allow unrestricted transfers of control. Denotational models for languages with such control operators naturally include elements that delimit control actions [80, 95] In addition to their theoretical importance, these delimiters present a useful programming paradigm [93, 94]. In this section, we study one traditional control delimiter: # (prompt) 24] Intuitively, in an expression (# M ) the prompt treats its subexpression as a complete program by evaluating it in the initial continuation, forwarding the result to the current continuation. To specify this behavior ....

Sitaram, D. and Felleisen, M. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3, 1 (1990) 67--100.

.... more externally towards implementation of an abstract machine that captures control categorically in its varied forms go tos, raising exceptions, handling exceptions, while loops, and more exotic and powerful contexttransformation mechanisms such as Felleisen s prompt and control operators [Fel88, FWFD88, SF90]. A keystone work in this area is Filinksi s SCL category (SCL for symmetric combinatory logic) that expresses several dualities: totality versus strictness of functions, call by name versus callby value evaluation, and values versus continuations [Fil89a, Fil89b] His categorical view of a ....

D. Sitaram and M. Felleisen, Control Delimiters and their Hierarchies, Lisp and Symbolic Computation 3, 1990.

.... implement multiprocessing [Wand 80] coroutines [Haynes et al. 84; Haynes et al. 86] timed preemption [Haynes Friedman 84; Haynes Friedman 87; Dybvig Hieb 89] and user level threads [Cooper Morrisett 90] Recent work of interest to operating systems in the areas of control delimiters [Sitaram Felleisen 90] and functional continuations [Felleisen et al. 88] addresses short comings of the usual first class continuation formulations in the areas of modularity and embedding. 2.3.1 Abstract Continuations Denotational semantics [Milne Strachey 76] ascribes meanings to programs in such a way that the ....

Dorai Sitaram and Matthias Felleisen. Control Delimiters and Their Hierarchies. Lisp and Symbolic Computation, 3:67-99, 1990.

....too. And so on. In other words, a language should provide the programmer with control operators able to capture partial continuations between any two points, whatever the marks appearing inside. For this purpose, hierarchies of control operators were introduced by Sitaram and Felleisen [17] and Danvy and Filinski [3] Control operators are now indexed by a number which is their control level , and an operator of level m is able to reify the partial continuation delimited by the first mark of level n if n m. Although this solution provides the possibility to reify any part of the ....

.... continuation: should we name the current block by a new name not given by the user or should this block be merged with the first block of the partial continuation In order to be able to reify a partial continuation between the current point and any mark (not the last one) Felleisen and Sitaram [17] and Danvy and Filinski [3] introduce a hierarchy of control operators. Therefore, their languages have 2n times control operators instead of 2. But, what is the intuitive difference between control operators of level n or level n 1 Moreau and Ribbens [11] also introduce a mechanism of prompt ....

Dorai Sitaram and Matthias Felleisen. Control Delimiters and Their Hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

....a fresh prompt (with the type of the expression) and set s; all callcc s are then done via cupto s to this fresh prompt. To determine whether named, typed prompts are useful in programming requires some experience in writing programs. In the untyped case, prompts add significant expressive power [23, 21]; we believe the examples of [21] could be typed in our system. We also conjecture that many applications that currently uses callcc (such as various threads packages or CML) could benefit for instance, the explicit prompt mechanism may simplify the implementation of threads in a interactive ....

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. LISP and Symbolic Computation, 3:67--99, 1990.

....] cupto [ a 1 ] fn x = a 2 ] The other constructs of our language can be translated directly into SML NJ, e.g. and 1 [ x: a] fn x = a] Appendix A gives an implementation in SML NJ. Our encoding is clearly of the same flavor as the untyped encoding of shift and reset [22] into Scheme with callcc, but it is not easy to relate them in a precise way, since the languages that they encode are also different. Notice that the signature involves weak type variables (cf. 10, 26] If SML were modified to have value only polymorphism, the signature of this module would be ....

....we have: the paper [17] does not state the theorem nor attempt to prove it, but using our proof technique it is easy to carry out. Appart from this significant difference, Queinnec and Serpette s splitter also comes closest to ours in adding multiple prompts. Others, notably Sitaram and Felleisen [22] and Danvy and Filinski [2] have added multiple prompts and control operations to languages to obtain more control. The difference between these operations and our language (and Queinnec and Serpette s) is important: prompts in our proposal are hidden in an abstract type that only the compiler ....

[Article contains additional citation context not shown here]

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

....this extent. MacLisp [42] and Common Lisp [64] catch and throw, and Eulisp let cc [50] are other examples of exception like control operators with a dynamic extent. More generally, control delimiters are used to create partial continuations whose semantics allow various degrees of dynamicness [8, 32, 47, 54, 62]. 7 exception foo; fun gee f = f ( handle foo = dynamic ; let fun bar ( raise foo in gee bar end handle foo = lexical ; Figure 3. Exception Handlers with a Dynamic Extent in Standard ML 2.4. Parallelism and Distribution Parallelism and distribution are usually considered as a possible ....

Dorai Sitaram and Matthias Felleisen. Control Delimiters and Their Hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

....transformation, are similar, differing primarily in that captured continuations include a prompt. Subcontinuations generalize Felleisen s single prompt to multiple nested prompts and allow continuations to be used to control tree structured concurrency [17] In related work, Sitaram and Felleisen [27] show how nested prompts may be obtained from single prompts in a sequential setting. Splitter [26] extends the notion of a continuation in a manner similar to subcontinuations in a sequential setting but separates the continuation capture mechanism from the continuation abort mechanism. Gunter, ....

Sitaram, Dorai and Felleisen, Matthias. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3, 1 (January 1990) 67-- 99.

....Prompts Both A and callcc allow unrestricted transfers of control. Denotational models for languages with such control operators naturally include elements that delimit control actions [80, 95] In addition to their theoretical importance, these delimiters present a useful programming paradigm [93, 94]. In this section, we study one traditional control delimiter: # (prompt) 24] Intuitively, in an expression (# M ) the prompt treats its subexpression as a complete program by evaluating it in the initial continuation, forwarding the result to the current continuation. To specify this behavior ....

Sitaram, D. and Felleisen, M. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3, 1 (1990) 67--100.

....delimiter endows the programmer with the ability to run a subcomputation as an independent program, in so far as control actions are concerned. The uses of the prompt are diverse: it allows cleaner realizations of existing control abstractions and makes possible completely new programming styles [16, 46]. As a radically different model for higher order control, we use Felleisen and Cartwright s technique for extending the basic direct model to our control sublanguages. Despite these models different approach to modeling control, they too are found to contain prompts This, combined with the ....

....acts do not create continuations. Furthermore, control handlers are true higher order versions of the popular exception handling mechanisms. 1.5 Publications of results Components of this thesis and some related results were published separately. In Control Delimiters and Their Hierarchies [46], we showed that control delimiters allow cleaner realizations of existing control abstractions and make possible completely new programming styles. In Reasoning with Continuations II: How to Get Full Abstraction for Models of Control [47] we presented the full abstraction result for the basic ....

[Article contains additional citation context not shown here]

D. Sitaram and M. Felleisen. Control Delimiters and Their Hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

....variant of the prompt is called run, to borrow a term used for an operator that runs programs [26] The prompt and run are equivalent: either can be seen as syntactic sugar for the other. Together with higher order control reifiers like call cc, the prompt supports powerful programming idioms [7, 20]. It has several successors specially suited to various practical settings, e.g. spawn [14] reset [3] and splitter [18] However, none of these constructs handles control objects the corresponding control reifier continues to double as handler. In this work, we continue the process of ....

....predicate returns false. 2.4 Tagged run and fcontrol To avoid interference between control actions arising from logically different uses of run fcontrol , we should identify matching pairs of these control operators. In an earlier approach, we suggested a hierarchically ordered set of delimiters [20]. For prompts with handlers, it is natural to continue our extrapolation from Lisp s catch and throw, giving tagged versions of run and fcontrol , invoked respectively as: run tagged htagi hthunki hhandleri) and (fcontrol tagged htagi hobjecti) One tagging protocol others are possible ....

D. Sitaram and M. Felleisen. Control Delimiters and Their Hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

No context found.

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67--99, January 1990.

No context found.

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67--99, January 1990.

No context found.

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67--99, 1990.

No context found.

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation, 3(1):67-- 99, January 1990.

No context found.

D. Sitaram and M. Felleisen. Control delimiters and their hierarchies. LISP and Symbolic Computation, 3:67--99, 1990.

No context found.

Dorai Sitaram and Matthias Felleisen. Control delimiters and their hierarchies. Lisp and Symbolic Computation 3(1):67--99, January 1990.

First 50 documents

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