High quality source code is often paired with high level summaries of the computation it performs, for example in code documentation or in descriptions posted in online forums. Such summaries are extremely useful for applications such as code search but are expensive to manually author, hence only done for a small fraction of all code that is produced. In this paper, we present the first completely datadriven approach for generating high level summaries of source code. Our model, CODE-NN , uses Long Short Term Memory (LSTM) networks with attention to produce sentences that describe C# code snippets and SQL queries. CODE-NN is trained on a new corpus that is automatically collected from StackOverflow, which we release. Experiments demonstrate strong performance on two tasks: (1) code summarization, where we establish the first end-to-end learning results and outperform strong baselines, and (2) code retrieval, where our learned model improves the state of the art on a recently introduced C# benchmark by a large margin.

We present an approach to rapidly and easily build natural language interfaces to databases for new domains, whose performance improves over time based on user feedback, and requires minimal intervention. To achieve this, we adapt neural sequence models to map utterances directly to SQL with its full expressivity, bypassing any intermediate meaning representations. These models are immediately deployed online to solicit feedback from real users to flag incorrect queries. Finally, the popularity of SQL facilitates gathering annotations for incorrect predictions using the crowd, which is directly used to improve our models. This complete feedback loop, without intermediate representations or database specific engineering, opens up new ways of building high quality semantic parsers. Experiments suggest that this approach can be deployed quickly for any new target domain, as we show by learning a semantic parser for an online academic database from scratch.

No abstract

No abstract

SQL is the de facto language for manipulating relational data. Though powerful, SQL queries can be difficult to write due to their highly expressive constructs. Using the programming-by-example paradigm to help users write SQL queries presents an attractive proposition, as evidenced by online help forums such as Stack Overflow. However, developing techniques to synthesize SQL queries from input-output (I/O) examples has been difficult due to SQL's rich set of operators. In this paper, we present a new scalable and efficient algorithm to synthesize SQL queries from I/O examples. Our key innovation is the development of a language for abstract queries, i.e., queries with uninstantiated operators, that can express a large space of SQL queries efficiently. Using abstract queries to represent the search space nicely decomposes the synthesis problem into two tasks: (1) searching for abstract queries that can potentially satisfy the given I/O examples, and (2) instantiating the found abstract queries and ranking the results. We implemented the algorithm in a new tool, called SCYTHE, and evaluated it on 193 benchmarks collected from Stack Overflow. Our results showed that SCYTHE efficiently solved 74% of the benchmarks, most in just a few seconds. Queries synthesized by SCYTHE range from simple ones involving a single selection to complex ones with six levels of nested queries.