Concern-Based (de)composition of Model-Driven Software Development Processes

Johannes, Jendrik; Aßmann, Uwe

doi:10.1007/978-3-642-16129-2_5

Cited by 9 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method as presented considers tree merging for the composition. Recently, in order to deal with graphical languages the method was extended to support typed graphs in [15], this method was implemented in the ReuseWare framework. This last implementation is consistent with the description of models as graphs in our Coq4MDE framework.…”

Section: Isc and Reuseware Approachmentioning

confidence: 99%

“…This is implemented in [15] using the remove operator which is automatically applied after composition execution to make the component understandable by tools where addressable points semantics is not defined. ElimInterf ace : M odel → M odel, such as:…”

Section: Component Interface Extractionmentioning

confidence: 99%

“…In this section, we present the implementation in our framework of the two basic operators of ISC (bind and extend) presented in [1] [15] . The difference between these operators is that "the bind applied to the hook replaces the hook (i.e., it removes the hook from its containing fragment) while extend applied on a hook does not modify the hook itself but uses it as a position for extension (i.e., the hook remains in its containing fragment) ".…”

Section: Components Compositionmentioning

confidence: 99%

See 2 more Smart Citations

A Proof Assistant Based Formalization of MDE Components

Kezadri

Combemale

Pantel

et al. 2012

Formal Aspects of Component Software

View full text Add to dashboard Cite

Abstract. Model driven engineering (MDE) now plays a key role in the development of safety critical systems through the use of early validation and verification of models, and the automatic generation of software and hardware artifacts from the validated and verified models. In order to ease the integration of formal specification and verification technologies, various formalizations of the MDE technologies were proposed by different authors using term or graph rewriting, proof assistants, logical frameworks, etc.The use of components is also mandatory to improve the efficiency of system development. Invasive Software Composition (ISC) has been proposed by Aßman in [1] to add a generic component structure to existing Domain Specific Modeling Languages in MDE. This approach is the basis of the ReuseWare toolset.We present in this paper an extension of a formal embedding of some key aspects of MDE in set theory in order to formalize ISC and prove the correctness of the proposed approach with respect to the conformance relation with the base metamodel. The formal embedding we rely on was developed by some of the authors, presented in [25] and then implemented using the Calculus of Inductive Construction and the Coq proof-assistant. This work 1 is a first step in the formalization of composable verification technologies in order to ease its integration for DSML extended with component features using ISC.

show abstract

Section: Isc and Reuseware Approachmentioning

confidence: 99%

Section: Component Interface Extractionmentioning

confidence: 99%

Section: Components Compositionmentioning

confidence: 99%

See 1 more Smart Citation

A Proof Assistant Based Formalization of MDE Components

Kezadri

Combemale

Pantel

et al. 2012

Formal Aspects of Component Software

View full text Add to dashboard Cite

show abstract

“…Tools assist developers at defining the language grammar and then automatically generate editors and views for documents that the domain experts edit. As a result it has become easier to define specialized languages for custom domains, for example for software composition with the Universal Composition Language [Johannes, 2010] or Aircraft design with CPACS [Nagel et al , 2012].…”

Section: Iterative Component Model Developmentmentioning

confidence: 99%

“…In software development the idea of component-based development is wellknown [McIlroy, 1968] and composition systems exist [Assmann, 2003;Johannes, 2010] which allow the composition of programs from code fragments. For the modeling of physical systems, the separation into model components and composition of larger systems based on model instances is even more intuitive.…”

Section: Component-based Developmentmentioning

confidence: 99%

An Integrated Development Environment with Enhanced Domain-Specific Interactive Model Validation

Samlaus¹

2015

Linköping Studies in Science and Technology. Dissertations

View full text Add to dashboard Cite

Physical modeling plays an important role in many engineering domains. For engineers it is important to model real world objects in order to simulate and assess their behavior. This is particularly true for the development of new product components such as for example wind turbines. The suitability and durability of components needs to be investigated early in the design process to avoid costs that arise when faulty designs appear during component testing. Another domain where models are frequently used is model-driven software development. Data structures and data flow are modeled, for example using UML diagrams. Code is automatically generated from the models and is refined by user-written code. Advantages of model-driven software development are the high abstraction of the code by visualization in diagrams and specialized views which display details of design aspects. Additionally, code generators are usually well tested and design patterns are used in the generated code which leads to good quality and readability. Models for simulation of wind turbines are developed at Fraunhofer Institute for Wind Energy and Energy System Technology.Models with different levels of detail are used during the design process, starting with simple models for basic design decisions and evaluations to high-detailed models (such as Computational Fluid Dynamics (CFD) models for detailed flow analysis of rotor blades). The models defined are parameter-based and do not define a tool specific syntax. Hence, these models must be transferred to a tool-specific format for simulation. Design flaws that are encountered in low-level models have an impact on the detailed models. But equally important, problems encountered in highly detailed models may require changes in the basic models. This leads to the problem that all models that have been defined for a design study, including the transformation to simulation tool models, may need to be altered. To remedy this problem, model-driven software development is used. Models are created once in our simulation environment and simulation models for the different tools are derived from those models. This dissertation focuses on the development of components with different levels of detail with the language Modelica. In order to develop models quickly, the developer needs to be assisted in creating models that are valid regarding language compliance and structural constraints and that show the expected behaviour during simulation. The Modelica Integrated Development Environment (IDE) OneModelica is introduced that provides a data model for Modelica models with the same technology as the general wind turbine models described above. Hence, Modelica code can be generated from the parametric models using existing tools. Features like project-based development allow separating models with different levels of detail, allowing the user to focus on one topic and to create libraries for certain functionality. Interactive Model validation has been implemented to assist the user during development. Hence, the...

show abstract