## CASL: From Semantics to Tools (2000)

Venue: | TACAS 2000, LNCS 1785 |

Citations: | 16 - 10 self |

### BibTeX

@INPROCEEDINGS{Mossakowski00casl:from,

author = {Till Mossakowski},

title = {CASL: From Semantics to Tools},

booktitle = {TACAS 2000, LNCS 1785},

year = {2000},

pages = {93--108},

publisher = {Springer-Verlag}

}

### Years of Citing Articles

### OpenURL

### Abstract

CASL, the common algebraic specification language, has been developed as a language that subsumes many previous algebraic specification frameworks and also provides tool interoperability. CASL is a complex language with a complete formal semantics. It is therefore a challenge to build good tools for CASL. In this work, we present and discuss the Bremen HOL-CASL system, which provides parsing, static checking, conversion to LaTeX and theorem proving for CASL specifications. To make tool construction manageable, we have followed some guidelines: re-use of existing tools, interoperability of tools developed at different sites, and construction of generic tools that can be used for several languages. We describe the structure of and the experiences with our tool and discuss how the guidelines work in practice.

### Citations

3843 |
Introduction to Automata Theory, Languages and Computability
- Hopcroft, Ullman, et al.
- 2001
(Show Context)
Citation Context ...mically according to the mixfix declarations, and formulas and terms are parsed with the generic Isabelle parser, which uses the well-known Cocke-Younger-Kasami algorithm for context-free recognition =-=[11]-=-. This grammar-parameterised algorithm has a complexity of O(n 3 ), which is quite acceptable, since formulas and terms in CASL specifications are not that long (however, it would be much too slow to ... |

169 |
General logics
- Meseguer
- 1989
(Show Context)
Citation Context ...se the Isabelle logic HOL, for which already sophisticated support is available, with the help of the encoding described in section 6. 3 Formally, they are institution representations in the sense of =-=[17, 27]-=- 6 This encoding has a clear semantical basis due to the borrowing (most other encodings into Isabelle/HOL do not have an explicit model-theoretic counterpart) . However, a good semantic basis does no... |

136 |
Isabelle: A Generic Theorem
- Paulson
- 1994
(Show Context)
Citation Context ...em provers across different logics. 7 The interface to Isabelle/HOL Using the encoding described in the previous section, we have built an interface from CASL to Isabelle/HOL. We have chosen Isabelle =-=[22]-=- because it has a very small core guaranteeing correctness. Furthermore, there is over ten years of experience with it (several mathematical textbooks have been verified with Isabelle), Last but not l... |

50 | Moving between logical systems
- Tarlecki
- 1996
(Show Context)
Citation Context ...se the Isabelle logic HOL, for which already sophisticated support is available, with the help of the encoding described in section 6. 3 Formally, they are institution representations in the sense of =-=[17, 27]-=- 6 This encoding has a clear semantical basis due to the borrowing (most other encodings into Isabelle/HOL do not have an explicit model-theoretic counterpart) . However, a good semantic basis does no... |

46 |
I borrow your logic? (transporting logical structures along maps
- Cerioli, Meseguer
(Show Context)
Citation Context ...implicit soundness and completeness proof for the re-use of HOL-theorem provers for theorem proving in the CASL logic SubPCFOL. This is also known as the "borrowing" technique of Cerioli and=-= Meseguer [4]-=- that allows to borrow theorem provers across different logics. 7 The interface to Isabelle/HOL Using the encoding described in the previous section, we have built an interface from CASL to Isabelle/H... |

43 | The Common Framework Initiative for algebraic specification and development, electronic archives. Notes and Documents accessible by WWW at http://www.brics.dk/Projects/CoFI - CoFI |

43 | CoFI: The Common Framework Initiative for Algebraic Specification and Development
- Mosses
- 1997
(Show Context)
Citation Context ...ial, to re-use tools and to get algebraic methods adopted in industry. Therefore, in 1995, an initiative, CoFI 1 , to design a Common Framework for Algebraic Specification and Development was started =-=[20]-=-. The goal of CoFI is to get a common agreement in the algebraic specification community about basic concepts, and to provide a family of specification languages at different levels, a development met... |

29 |
The Maude System
- Clavel, FJ, et al.
- 1999
(Show Context)
Citation Context ... steps of the encoding (see section 6) separately available, so that one can choose to "encode out" just partiality and keep the subsorting (this will be useful, for example, in connection w=-=ith Maude [5]-=- which supports subsorting). The Unix interface works quite well when using the tool in combination with other tools, although we plan to provide a fully-fledged applications programmer interface (API... |

27 | Towards heterogeneous specifications
- Tarlecki
- 2000
(Show Context)
Citation Context ...ty of the generic analysis of CASL-in-the-large. It is possible to extend CASL to a heterogeneous specification language, where one can combine specifications written in several different logics, see =-=[28]-=- for some first ideas. Tool support for such a language would extend the generic analysis of CASL-in-the-large with an analysis of structuring mechanisms for moving specifications between different lo... |

22 | Static semantic analysis and theorem proving for Casl
- Mossakowski, Kolyang, et al.
- 1997
(Show Context)
Citation Context ... interoperability using a common interchange format is discussed. Section 5 describes the problems with parsing CASL's mixfix syntax. Section 6 recalls the encoding of CASL in higher-order logic from =-=[18]-=-, while section 7 reports our practical experiences when using this encoding to create an interface from CASL to Isabelle/HOL. Section 8 describes a way how to make the static analysis of CASL structu... |

20 | The UniForM Workbench, a Universal Development Environment for Formal Methods
- Krieg-Brückner, Peleska, et al.
- 1999
(Show Context)
Citation Context ...poral logic, or just your own favourite logic). In the future, we hope that more tools will be integrated to work with our tools. An even better integration can be achieved with the UniForM workbench =-=[13]-=-, which also provides library management and access to a generic transformation application system [16, 15] that will be instantiated to CASL. Future work will turn our tool into a theorem proving env... |

18 | Functional design and implementation of graphical user interfaces for theorem provers
- Lüth, Wolff
- 1999
(Show Context)
Citation Context ...ly-fledged applications programmer interface (API) in the future. The full stand-alone version of the tool 6 also provides the Isabelle theorem prover, and the generic graphical user interface IsaWin =-=[16, 15]-=-, which has been built on top of Isabelle. We have instantiated IsaWin with our HOL-CASL encoding of CASL into Isabelle/HOL. In Fig. 5, you can see a typical IsaWin window. The icons labelled with (\S... |

14 |
Specifications in an arbitrary institution with symbols
- Mossakowski
- 1999
(Show Context)
Citation Context ...ons-in-the-small (= basic specifications), while syntax and semantics of specification-in-the-large remains the same (see Fig. 3). This is achieved by working with so-called institutions with symbols =-=[19]-=-, a notion that allows to mathematically formalize the essential aspects of a logic that are needed for CASL structured specifications. Indeed, it is actually possible to use this logic-independent se... |

13 | Permissive subsorted partial logic in Casl
- Cerioli, Haxthausen, et al.
- 1997
(Show Context)
Citation Context ...eclarations and axioms representing theories of a first-order logic in which predicates, total as well as partial functions, and subsorts are allowed. Predicate and function symbols may be overloaded =-=[3]-=-. Datatype declarations allow to shortly describe the usual datatypes occurring in programming languages. Structured specifications allow to rename or hide parts of specifications, unite, extend and n... |

13 |
Subsorted partial higher-order logic as an extension of Casl, Recent Trends
- Mossakowski, Haxthausen, et al.
- 2000
(Show Context)
Citation Context ...ture of CASL and Iasbelle/HOL. In future work, we will study the semantic relationships between the different input and display syntaxes and a true combination of CASL and HOL (like higher-order CASL =-=[10]-=-) in more detail. 8 Generic static analysis of CASL-in-the-large Number of rules Static semantics Model Semantics Altogether Basic specifications 109/0 1/0 110/0 Structured specifications 19/29 1/20 2... |

12 | Precedences in specifications and implementations of programming languages
- Aasa
- 1995
(Show Context)
Citation Context ...ain a concise output, not all pretty-printed forms of the parse trees are shown, but only the local places at which they actually differ. 5 The definition of precedence correctness follows the one of =-=[1]-=-, generalized to CASL's pre-order based precedences ([1] uses number based precedences). Concerning static semantic analysis, the treatment of subsorts and overload resolution needs a careful algorith... |

12 |
Aterms: Exchanging data between heterogeneous tools for CASL. Note T-3
- Brand, Klint, et al.
- 1998
(Show Context)
Citation Context ...s for them. CASL was designed with the goal of providing a common language for better 4 tool interoperability. This is reflected by having a common interchange format for CASL tools, the ATerm format =-=[30]-=-. ATerms are an easy-to-handle format with libraries in several languages (C, Java, ML) available. They are used as low level tool format for data exchange between CASL tools. Based on this format, se... |

10 | TAS and IsaWin: Tools for transformational program developkment and theorem proving
- Lüth, Tej, et al.
- 1999
(Show Context)
Citation Context ...ly-fledged applications programmer interface (API) in the future. The full stand-alone version of the tool 6 also provides the Isabelle theorem prover, and the generic graphical user interface IsaWin =-=[16, 15]-=-, which has been built on top of Isabelle. We have instantiated IsaWin with our HOL-CASL encoding of CASL into Isabelle/HOL. In Fig. 5, you can see a typical IsaWin window. The icons labelled with (\S... |

8 | Development of Parsing Tools for CASL using Generic Language Technology
- Brand, Scheerder
- 2000
(Show Context)
Citation Context ...n the tool architecture, the Bremen HOL-CASL system can be used as a front-end or back-end in combination with other tools. Actually, it has been combined as a back-end with the Amsterdam CASL parser =-=[31]-=-, and as a front-end with several theorem proving tools: ELAN [23], PVS [2] and Isabelle (see section 7). See also the CoFI Tools Group home page [9]. 5 Parsing and static semantic analysis Apart from... |

6 |
UniForM Workbench --- Universelle Entwicklungsumgebung fur formale Methoden
- Krieg-Bruckner, Peleska, et al.
- 1995
(Show Context)
Citation Context ...ed entities into files that are distributed over the Internet. Here, a general mechanism maintaining officially registered CASL libraries needs to be implemented. We plan to use the UniForM Workbench =-=[12]-=- for library management, but this is future work. 9 Encoding of CASL structured specifications When encoding CASL structured specification into Isabelle, the problem arises that the structuring mechan... |

4 |
Formatting Casl specifications using L
- Mosses
- 1998
(Show Context)
Citation Context ...ication-specific information such as the signature. The L A T E X formatter allows to pretty print CASL specifications (which are input in ASCII format), using the L A T E X package from Peter Mosses =-=[21]-=-. Finally, the encoding is a bridge from CASL to first- or higher-order logic (FOL/HOL). It throws out subsorting and partiality by encoding it [18], and thus allows to re-use existing theorem proving... |

4 |
Second-order proof systems for algebraic specification languages
- Schobbens
- 1994
(Show Context)
Citation Context ...construction in the model categories of the logic. All that one can expect here is to simulate the semantics of free specifications in a particular logic within higher-order logic, along the lines of =-=[26, 25]-=-. Encoding of architectural specifications is beyond the scope of this paper -- it will be dealt with elsewhere. Fig. 4. The web interface of the HOL-CASL system As described in the previous section, ... |

3 |
Extensions of Initial Models and their Second-order Proof Systems
- Schobbens
- 1994
(Show Context)
Citation Context ...construction in the model categories of the logic. All that one can expect here is to simulate the semantics of free specifications in a particular logic within higher-order logic, along the lines of =-=[26, 25]-=-. Encoding of architectural specifications is beyond the scope of this paper -- it will be dealt with elsewhere. Fig. 4. The web interface of the HOL-CASL system As described in the previous section, ... |

1 |
Proving theorems about CASL specifications. Talk at the
- Baillie
- 1999
(Show Context)
Citation Context ...nd or back-end in combination with other tools. Actually, it has been combined as a back-end with the Amsterdam CASL parser [31], and as a front-end with several theorem proving tools: ELAN [23], PVS =-=[2]-=- and Isabelle (see section 7). See also the CoFI Tools Group home page [9]. 5 Parsing and static semantic analysis Apart from having a relatively complex grammar, CASL has several features that cause ... |

1 |
VDM and proof rules for underdetermined functions
- Larsen
- 1999
(Show Context)
Citation Context ...ndustry. This holds even stronger since CASL is a language with a formal semantics: many people believe that such a language cannot or will not be used in practice: "The best semantics will not w=-=in." [14]-=- Since CASL was designed with the goal to subsume many previous frameworks, it has become a powerful and quite complex language. This complexity makes it harder to build tools covering the whole langu... |

1 |
Demonstration of ELAN for rewriting in CASL specifications
- Ringeissen
- 1999
(Show Context)
Citation Context ... a front-end or back-end in combination with other tools. Actually, it has been combined as a back-end with the Amsterdam CASL parser [31], and as a front-end with several theorem proving tools: ELAN =-=[23]-=-, PVS [2] and Isabelle (see section 7). See also the CoFI Tools Group home page [9]. 5 Parsing and static semantic analysis Apart from having a relatively complex grammar, CASL has several features th... |

1 |
Basic datatypes
- Roggenbach, Mossakowski
- 1999
(Show Context)
Citation Context ...rs and proof tactics for CASL will have to be developed an tested. A first case study will be the verification of proof obligations and intended consequences for the libraries of CASL basic datatypes =-=[24]-=-. A further topic that would be interesting is the study of the semantic relationships between the encoding of CASL into HOL and a true combination of CASL and HOL (like higher-order CASL [10]). Anoth... |

1 |
CasFix -- mapping from the concrete CASL to the abstract syntax in ATerms format. http://adam.wins.uva.nl/markvdb/cofi/casl.html
- Brand
- 1998
(Show Context)
Citation Context ... in several languages (C, Java, ML) available. They are used as low level tool format for data exchange between CASL tools. Based on this format, several formats have been designed: the CasFix format =-=[29]-=- for abstract syntax trees, and a format for the global environment, containing the static semantic information. By providing conversions from and to ATerms at all intermediate points in the tool arch... |