Code generation method supporting complex model structure conversion

Abstract

The invention provides a code generation method supporting complex model structure conversion, which uses a KARMA language to describe code generation, and supports a complex model structure formed by a plurality of model elements (including objects, relationships, points and roles) as a whole to be inquired and converted by defining a model mode. On one hand, generation of some target codes in specific formats is facilitated, on the other hand, through the mode of the model mode, a user only needs to describe a model structure needing to be converted without paying attention to details and processes of specific query, user friendliness is improved, and conversion capacity and efficiency are enhanced. Besides, according to the KARMA code generation language provided by the invention, six GOPPRR elements serve as elements for underlying query and conversion, specific details of domain models are shielded, and for various domain models (such as a SysML model and an AADL model), GOPPRR can be used as underlying elements, query is carried out by describing a model mode, and conversion from the model to a target code is realized.

Description

technical field [0001] The invention belongs to the field of system engineering and complex system design, in particular to a code generation method supporting complex model structure conversion. Background technique [0002] Model Based Systems Engineering (MBSE, Model Based System Engineering) is an emerging technology that supports the development of complex systems. MBSE uses a formalized model to describe functional activities such as requirements, design, analysis, verification and confirmation of the system development life cycle, and uses the model as the core to drive system development. Code generation is a key technology in the development of model-driven systems, also known as model-to-text conversion. It obtains element information in the source model through model query, and maps source model elements to target code or text (such as XML text) Rules, transformations generate object code or text. Through code generation, the system architecture model with a hig...

AI Technical Summary

Problems solved by technology

[0005] Existing code generators are highly proprietary, usually oriented to a specific domain model or data file to build a code generator for a specific target language, the scalability and versatility are not strong, and there is a lack of multi-domain models involved in the complex product development process. Code Generation Capabilities

The code generation language with strong flexibility can support the generation of multiple target language codes based on templates, but the mainstream code generation language can usually support the description of the query and conversion of a single element in the model, but lacks The ability to describe the query and transformation of a model structure with multiple model elements, while the data in some forms of object code is derived from the element information in the complete model structure, nested queries through the combination of layer by layer individual elements Although the model structure composed of multiple elements can also be obtained, it increases the complexity of the description and reduces the conversion efficiency

Images