Existing Code Search Research Articles

C2B: A Semantic Source Code Retrieval Model using CodeT5 and Bi-LSTM

To enhance the software implementation process, developers frequently leverage preexisting code snippets by exploring an extensive codebase. Existing code search tools often rely on keyword- or syntactic-based methods and struggle to fully grasp the semantics and intent behind code snippets. In this paper, we propose a novel hybrid C2B model that combines CodeT5 and bidirectional long short-term memory (Bi-LSTM) for source code search and recommendation. Our proposed C2B hybrid model leverages CodeT5’s domain-specific pretraining and Bi-LSTM’s contextual understanding to improve code representation and capture sequential dependencies. As a proof-of-concept application, we implemented the proposed C2B hybrid model as a deep neural code search tool and empirically evaluated the model on the large-scale dataset of CodeSearchNet. The experimental findings showcase that our methodology proficiently retrieves pertinent code snippets and surpasses the performance of prior state-of-the-art techniques.

A Survey of Source Code Search: A 3-Dimensional Perspective

(Source) code search is widely concerned by software engineering researchers because it can improve the productivity and quality of software development. Given a functionality requirement usually described in a natural language sentence, a code search system can retrieve code snippets that satisfy the requirement from a large-scale code corpus, e.g., GitHub. To realize effective and efficient code search, many techniques have been proposed successively. These techniques improve code search performance mainly by optimizing three core components, including query understanding component, code understanding component, and query-code matching component. In this paper, we provide a 3-dimensional perspective survey for code search. Specifically, we categorize existing code search studies into query-end optimization techniques, code-end optimization techniques, and match-end optimization techniques according to the specific components they optimize. These optimization techniques are proposed to enhance the performance of specific components, and thus the overall performance of code search. Considering that each end can be optimized independently and contributes to the code search performance, we treat each end as a dimension. Therefore, this survey is 3-dimensional in nature, and it provides a comprehensive summary of each dimension in detail. To understand the research trends of the three dimensions in existing code search studies, we systematically review 68 relevant literatures. Different from existing code search surveys that only focus on the query end or code end or introduce various aspects shallowly (including codebase, evaluation metrics, modeling technique, etc.), our survey provides a more nuanced analysis and review of the evolution and development of the underlying techniques used in the three ends. Based on a systematic review and summary of existing work, we outline several open challenges and opportunities at the three ends that remain to be addressed in future work.

XCoS: Explainable Code Search Based on Query Scoping and Knowledge Graph

When searching code, developers may express additional constraints (e.g., functional constraints and nonfunctional constraints) on the implementations of desired functionalities in the queries. Existing code search tools treat the queries as a whole and ignore the different implications of different parts of the queries. Moreover, these tools usually return a ranked list of candidate code snippets without any explanations. Therefore, the developers often find it hard to choose the desired results and build confidence on them. In this article, we conduct a developer survey to better understand and address these issues and induct some insights from the survey results. Based on the insights, we propose XCoS, an explainable code search approach based on query scoping and knowledge graph. XCoS extracts a background knowledge graph from general knowledge bases like Wikidata and Wikipedia. Given a code search query, XCoS identifies different parts (i.e., functionalities, functional constraints, nonfunctional constraints) from it and use the expressions of functionalities and functional constraints to search the codebase. It then links both the query and the candidate code snippets to the concepts in the background knowledge graph and generates explanations based on the association paths between these two parts of concepts together with relevant descriptions. XCoS uses an interactive user interface that allows the user to better understand the associations between candidate code snippets and the query from different aspects and choose the desired results. Our evaluation shows that the quality of the extracted background knowledge and the concept linkings in codebase is generally high. Furthermore, the generated explanations are considered complete, concise, and readable, and the approach can help developers find the desired code snippets more accurately and confidently.

Fine-Tuning Pre-Trained CodeBERT for Code Search in Smart Contract

Smart contracts, which automatically execute on decentralized platforms like Ethereum, require high security and low gas consumption. As a result, developers have a strong demand for semantic code search tools that utilize natural language queries to efficiently search for existing code snippets. However, existing code search models face a semantic gap between code and queries, which requires a large amount of training data. In this paper, we propose a fine-tuning approach to bridge the semantic gap in code search and improve the search accuracy. We collect 80 723 different pairs of <comment, code snippet> from Etherscan.io and use these pairs to fine-tune, validate, and test the pre-trained CodeBERT model. Using the fine-tuned model, we develop a code search engine specifically for smart contracts. We evaluate the Recall@k and Mean Reciprocal Rank (MRR) of the fine-tuned CodeBERT model using different proportions of the fine-tuned data. It is encouraging that even a small amount of fine-tuned data can produce satisfactory results. In addition, we perform a comparative analysis between the fine-tuned CodeBERT model and the two state-of-the-art models. The experimental results show that the fine-tuned CodeBERT model has superior performance in terms of Recall@k and MRR. These findings highlight the effectiveness of our fine-tuning approach and its potential to significantly improve the code search accuracy.

Incorporating Code Structure and Quality in Deep Code Search

Developers usually search for reusable code snippets to improve software development efficiency. Existing code search methods, including methods based on full-text or deep learning, have two disadvantages: (1) ignoring structural information of code snippets, such as conditional statements and loop statements, and (2) ignoring quality information of code snippets, such as naming clarity and logical correctness. These disadvantages limit the performance of existing code search methods. In this paper, we propose a novel code search method named Structure and Quality based Deep Code Search (SQ-DeepCS). SQ-DeepCS introduces a code representation method called program slice to represent structual information as well as API usage of code snippets. Meanwhile, SQ-DeepCS introduces a novel deep neural network named Method-Description-Joint Embedding Neural Network (MD-JEnn) to weight the quality of code snippets. To evaluate the proposed methods, we train MD-JEnn and evaluate SQ-DeepCS by searching for code snippets with respect to the top-rated questions from Stack Overflow. We use four evaluation indicators to measure the effectiveness of SQ-DeepCS: FRank, SuccessRate@k, PrecisionRate@k, and Mean Reciprocal Rank (MRR). The experimental results show that our approach can provide better results than existing techniques when searching for relevant code snippets.

Opportunities and Challenges in Code Search Tools

Code search is a core software engineering task. Effective code search tools can help developers substantially improve their software development efficiency and effectiveness. In recent years, many code search studies have leveraged different techniques, such as deep learning and information retrieval approaches, to retrieve expected code from a large-scale codebase. However, there is a lack of a comprehensive comparative summary of existing code search approaches. To understand the research trends in existing code search studies, we systematically reviewed 81 relevant studies. We investigated the publication trends of code search studies, analyzed key components, such as codebase, query, and modeling technique used to build code search tools, and classified existing tools into focusing on supporting seven different search tasks. Based on our findings, we identified a set of outstanding challenges in existing studies and a research roadmap for future code search research.

FACER: An API usage-based code-example recommender for opportunistic reuse

To save time, developers often search for code examples that implement their desired software features. Existing code search techniques typically focus on finding code snippets for a single given query, which means that developers need to perform a separate search for each desired functionality. In this paper, we propose FACER (F eature-driven A PI usage-based C ode E xamples R ecommender), a technique that avoids repeated searches through opportunistic reuse. Specifically, given the selected code snippet that matches the initial search query, FACER finds and suggests related code snippets that represent features that the developer may want to implement next. FACER first constructs a code fact repository by parsing the source code of open-source Java projects to obtain methods’ textual information, call graphs, and Application Programming Interface (API) usages. It then detects unique features by clustering methods based on similar API usages, where each cluster represents a feature or functionality. Finally, it detects frequently co-occurring features across projects using frequent pattern mining and recommends related methods from the mined patterns. To evaluate FACER, we run it on 120 Java Android apps from GitHub. We first manually validate that the detected method clusters represent methods with similar functionality. We then perform an automated evaluation to determine the best parameters (e.g., similarity threshold) for FACER. We recruit 10 professional developers along with 39 experienced students to judge FACER’s recommendation of related methods. Our results show that, on average, FACER’s recommendations are 80% precise. We also survey a total of 20 professional Android and Java developers to understand their code search and reuse experiences, and also to obtain their feedback on the usability and usefulness of FACER. The survey results show that 95% of our surveyed professional developers find the idea of related method recommendations useful during code reuse.

Leveraging Official Content and Social Context to Recommend Software Documentation

For an unfamiliar Application Programming Interface (API), software developers often access the official documentation to learn its usage, and post questions related to this API on social question and answering (Q&A) sites to seek solutions. The official software documentation often captures the information about functionality and parameters, but lacks detailed descriptions in different usage scenarios. On the contrary, the discussions about APIs on social Q&A sites provide enriching usages. Moreover, existing code search engines and information retrieval systems cannot effectively return relevant software documentation when the issued query does not contain code snippets or API-like terms. In this paper, we present $\mathsf{CnCxL2R}$ CnCxL 2 R , a software documentation recommendation strategy incorporating the content of official documentation and the social context on Q&A into a learning-to-rank schema. In the proposed strategy, the content, local context and global context of documentation are considered to select candidate documents. Then four types of features are extracted to learn a ranking model. We conduct a large-scale automatic evaluation on Java documentation recommendation. The results show that $\mathsf{CnCxL2R}$ CnCxL 2 R achieves state-of-the-art performance over the eight baseline models. We also compare the $\mathsf{CnCxL2R}$ CnCxL 2 R with Google search. The results show that $\mathsf{CnCxL2R}$ CnCxL 2 R can recommend more relevant software documentation, and can effectively capture the semantic between the high-level intent in developers’ queries and the low-level implementation in software documentation.

Search for Compatible Source Code

Third-party libraries always evolve and produce multiple versions. Lucene, for example, released ten new versions (from version 7.7.0 to 8.4.0) in 2019. These versions confuse the existing code search methods to retrieve the source code that is not compatible with local programming language. To solve this issue, we propose DCSE, a deep code search model based on evolving information (i.e. evolved code tokens and evolution description). DCSE first deeply excavates evolved code tokens and evolution description in the code evolution process; then it takes evolved code tokens and evolution description as one feature of source code and code description, respectively. With such fuller representation, DCSE embeds source code and its code description into a high-dimensional shared vector space, and makes the cosine distance of their vectors closer. For the ever-evolving third-party libraries like Lucene, the experimental results show that DCSE could retrieve the source code that is compatible with local programming language, it outperforms the state-of-the-art methods (e.g. CODEnn) by 56.9–60.9[Formula: see text] in RFVersion. For the rarely-evolving third-party libraries, DCSE outperforms the state-of-the-art methods (e.g. CODEnn) by 4–11[Formula: see text] in Precision.

Interactive Query Reformulation for Source-Code Search With Word Relations

Searching source code is a common activity in many software engineering tasks. To some extent, the quality of the query determines the accuracy of query results. In practice, it is difficult for developers to provide a high-quality query, especially for the novice who just takes over the software project with a short time. What is more, existing code search techniques using queries expressed in natural language offer little support to help developers determine whether the search results are relevant or not. When a query preforms poorly, it has to be reformulated. In this paper, we present a novel approach, INQRES, to interactively reformulate the search query considering the relations between words in the source code to optimize the query quality. INQRES analyzes the keyword relations in the source code and builds AND and OR relations in an interactive way for developer to select suitable words for query reformulation. To evaluate the effectiveness of INQRES, we perform an empirical study on the jEdit project. Empirical results show that INQRES can effectively reformulate the search query, and the quality of the reformulated query of INQRES is better than that of the state-of-art technique, i.e., QReformu.

Portfolio

Different studies show that programmers are more interested in finding definitions of functions and their uses than variables, statements, or ordinary code fragments. Therefore, developers require support in finding relevant functions and determining how these functions are used. Unfortunately, existing code search engines do not provide enough of this support to developers, thus reducing the effectiveness of code reuse. We provide this support to programmers in a code search system called Portfolio that retrieves and visualizes relevant functions and their usages. We have built Portfolio using a combination of models that address surfing behavior of programmers and sharing related concepts among functions. We conducted two experiments: first, an experiment with 49 C/C++ programmers to compare Portfolio to Google Code Search and Koders using a standard methodology for evaluating information-retrieval-based engines; and second, an experiment with 19 Java programmers to compare Portfolio to Koders. The results show with strong statistical significance that users find more relevant functions with higher precision with Portfolio than with Google Code Search and Koders. We also show that by using PageRank, Portfolio is able to rank returned relevant functions more efficiently.