HarmonyOS NEXT is a distributed operating system developed to support HarmonyOS native apps. To support the new and independent Harmony ecosystem, developers are required to migrate their applications from Android to HarmonyOS. However, HarmonyOS utilizes ArkTS, a superset of TypeScript, as the programming language for application development. Hence, migrating applications to HarmonyOS requires translating programs across different program languages, e.g., Java, which is known to be very challenging, especially for concurrency programs. Java utilizes shared memory to implement concurrency programs, while ArkTS relies on message passing (i.e., Actor model). This paper presents an LLMbased concurrent Java program to ArkTS converter. This converter leverages the sophisticated code comprehension and generation capabilities of large language models (LLMs) to streamline the translation process. By integrating the SharedArrayBuffer API native to ArkTS, we’ve crafted ThreadBridge, a shared library that replicates Java’s shared memory paradigm. Employing this library and LLM’s Chain-of-thought mechanism, our approach divides the translation challenge into specialized chains: the TypeScript (TS) chain, the concurrency chain, and the synchronization chain. These targeted chains are designed to efficiently translate Java’s source code into its ArkTS equivalent, handling TypeScript-specific syntax, concurrency patterns, and synchronization logic with precision. In particular, this study provides a set of effective solutions to overcome the differences in concurrency models between Java and ArkTS. With this converter, developers can reduce manual code rewriting and accelerate application adaptation and deployment on HarmonyOS NEXT. Experimental results show that our converter successfully compiles 66% of the 53 test samples, with an accuracy rate of 69% for the successfully compiled results. In conclusion, our approach demonstrates promising potential in handling the conversion of concurrent Java programs to ArkTS, providing a foundation for further improvement and optimization.
First you need to install LangChain and its package of openai
pip install -r requirements.txt
Set environment variables OPENAI_API_KEY to your api_key of openai.
To run the AST, you need to install Java 17.
To perform compilation check, you need to install tsc.
cd LangChain
streamlit run start.py