As the veteran workforce retires, COBOL mainframes are getting harder to understand. These codes, most of which lack proper documentation, are harder to understand by novice programmers. It becomes essential to extract Business Rules (BRs) from these systems in order to comprehend their core functionality. Existing state-of-the-art COBREX uses a rule-based approach (control flow graphs) to extract BRs from COBOL programs. We introduce COBRAIN, a tool which leverages large language models (LLMs) via few-shot prompting to extract and summarize BRs from legacy COBOL code. This work seeks to determine the viability of LLMs in accurately and comprehensively capturing business logic embedded within legacy COBOL systems. The research evaluates COBRAIN across three dimensions: (1) precision and recall in business rule extraction, using COBREX’s output as a benchmark; (2) accuracy, as measured by comparison to a manually curated ground-truth dataset; and (3) ease of comprehension and suitability for documentation, particularly for non-technical stakeholders, evaluated through a user-comprehension study. We use mixed-method study to evaluate the tool. COBRAIN achieved a precision of 1.0 and a recall of 0.746 when compared with COBREX. It achieved an F1 score of 0.73 when evaluated with ground truth, compared to COBREX’s F1 score of 0.59. In the comprehension study including 28 participants, over 80% chose COBRAIN over COBREX to have more understandable BRs.

Legacy modernization is a significant task in the software industry to maintain the relevance of legacy but critical software systems that are still widely used across various domains. COBOL, a programming language developed in the 1950s is tightly coupled to modern-day transactions and is the underlying base for the majority of these legacy systems. Researchers and industry have developed a range of approaches and tools to translate these systems to Java. On the other hand, despite the wide adoption of Python in AI and non-AI based applications over the last decade, there is limited research that focuses on Python, COBOL and legacy modernization. Hence, in this paper, we present COB2PY, one of the first tools that facilitates translation of COBOL to Python. COB2PY is a rulebased tool based on COBOL85 grammar for automatically converting procedurally-driven COBOL source code to objectoriented driven Python. It generates the Abstract Syntax Tree of the COBOL source code and translates it to Python code, while preserving the original COBOL code logic and functionality. The tool is evaluated using the Computational Accuracy (CA) on 103 COBOL programs from the CodeNet dataset, and achieved an accuracy of$\mathbf{9 8. 3 5 \%}$. To make translation from COBOL to Python reliable, we avoided Artificial Intelligence (AI) in this version, but a hybrid approach (rule-based + AI) could be explored as future work. We hope that our work could motivate researchers to leverage the best of non-AI and rulebased approaches for software engineering tasks before delving into AI or hybrid approaches. The tool and demo video can be found at https://rishalab.github.io/COB2PY/.