Data-Driven Program Completion

We introduce program splicing, a programming methodology that aims to automate the commonly used workƒow of copying, pasting, and modifying code available online. Here, the programmer starts by writing a “dra‰” that mixes un€nished code, natural language comments, and correctness requirements in the form of test cases or API call sequence constraints. A program synthesizer that interacts with a large, searchable database of program snippets is used to automatically complete the dra‰ into a program that meets the requirements. Œe synthesis process happens in two stages. First, the synthesizer identi€es a small number of programs in the database that are relevant to the synthesis task. Next it uses an enumerative search to systematically €ll the dra‰ with expressions and statements from these relevant programs. Œe resulting program is returned to the programmer, who can modify it and possibly invoke additional rounds of synthesis. We present an implementation of program splicing for the Java programming language. Œe implementation uses a corpus of over 3.5 million procedures from an open-source so‰ware repository. Our evaluation uses the system in a suite of everyday programming tasks, and includes a comparison with a state-of-the-art competing approach as well as a user study. Œe results point to the broad scope and scalability of program splicing and indicate that the approach can signi€cantly boost programmer productivity.