Code generation for salvaging legacy business logic when moving to the cloud

Scania is transitioning from legacy application to a modern cloud based platform. The existing system depends on special functions called guides which help technicians perform complex tasks like diagnostics: checking fuel pumps, evaluating air pressure, updating embedded software in engine etc. These guides are important as they contain embedded business logic necessary for various technical procedures. In the new cloud-native platform, using the existing embedded business logic within the functional guides is difficult due to differences in source environments and architecture. Therefore, a method is required to preserve and convert the existing logic into a format suitable for the cloud-native environment. Existing studies on migrating business logic from functional guides in desktop applications to cloud-native platforms primarily focus on recreating the old source code to fit the new environment. However, there is limited information available on transforming existing source code directly into a new format which is more flexible with the frameworks of the cloud-native environment and how an automation can be done to make this process of converting one source to another source in large codebases effectively with minimal human intervention. This study examines an appropriate method for migrating a large, interdependent legacy application. The method identifies how the existing business logic, represented in XML, can be transformed without data loss while maintaining compatibility with the pre-existing dependencies. The proposed approach focuses on selecting a framework to convert the XML nto a cloud-native language that aligns with the identified migration strategy and integrates seamlessly with the dependencies of Scania’s new cloud application. Additionally, the study suggests using batch processing to automate the conversion of a large codebase with similar yet distinct code files, where applying a single rule-based template would be either impractical or overly complex.