Principles and Practices of Large-Scale Code Analysis at Ant Group: A Data- and Logic-Oriented Approach
This program is tentative and subject to change.
In the domain of large-scale software development, there is a need for dynamic and multifaceted static code analysis that exceed the capabilities of traditional tools. Existing code analysis tools like CodeQL lack the capability of cross-language analysis. It is also time-consuming and resource-consuming. To bridge this gap, we present CodeFuse-Query, a large data system tailored for large- scale code analysis. Firstly, CodeFuse-Query adopts Logic Oriented Computation Design. The system’s logic-oriented facet employs Datalog, utilizing a unique two-tiered schema, COREF, to convert source code into data facts. At the same time, the system’s use of Gödel, a variant of Datalog, allows for the expression of complex analysis tasks in logical terms, leveraging the declarative nature of the language. Furthermore, CodeFuse-Query adopts Domain Opti- mized System Design. This approach optimizes resource utilization, prioritizes data reusability, applies incremental code extraction, and introduces tasks type characteristics specially for Code Change, underscoring its domain-optimized design. To demonstrate the capabilities of CodeFuse-Query, including its robustness, scalability, and efficiency, we present the empirical results and detail its application in real-world scenarios at Ant Group and examine how it contributes to the advancement of static code analysis in large-scale software development. Also, CodeFuse- Query is capable of processing up to 10 billion lines of code per day across over 300,000 distinct analysis tasks tailored to the needs of Ant Group. The results show that CodeFuse-Query has good capability in large-scale software analysis. CodeFuse-Query has been open-sourced
This program is tentative and subject to change.
Fri 17 AprDisplayed time zone: Brasilia, Distrito Federal, Brazil change
11:00 - 12:30 | Testing and Analysis 16Research Track / SE In Practice (SEIP) at Oceania II Chair(s): Andreas Zeller CISPA Helmholtz Center for Information Security | ||
11:00 15mTalk | Parse this! Summoning Context-Sensitive Inputs with Goblin Research Track Robert Lorch The University of Iowa, Muhammad Daniyal Pirwani Dar Stony Brook University, Cesare Tinelli University of Iowa, Omar Chowdhury Stony Brook University | ||
11:15 15mTalk | Context-Free Property Oriented Fuzzing Research Track Jiaqiang Yao College of Computer, National University of Defense Technology, Meixi Liu National University of Defense Technology, Changsha, China, Zhenbang Chen College of Computer, National University of Defense Technology, Yongchao Xing College of Computer, National University of Defense Technology, Jinjian Luo College of Computer, National University of Defense Technology, Yunlai Luo National University of Defense Technology, Guofeng Zhang College of Computer, National University of Defense Technology, Yufeng Zhang Hunan University, Ji Wang National University of Defense Technology | ||
11:30 15mTalk | Metamorphic Fuzzing for Multi-Agent Path Finding Algorithms Research Track Luxia Lin Institute of Software, Chinese Academy of Sciences, China, xudong zhang , Shihao Zhu State Key Laboratory of Computer Science,Institute of Software,Chinese Academy of Sciences,China, Yan Cai Institute of Software at Chinese Academy of Sciences | ||
11:45 15mTalk | Fuzzing JavaScript Engines by Fusing JavaScript and WebAssembly Research Track Jiayi Lin The University of Hong Kong, Changhua Luo The University of Hong Kong; Wuhan University, Mingxue Zhang Zhejiang University, Lanteng Lin The University of Hong Kong, Penghui Li Columbia University, Chenxiong Qian University of Hong Kong | ||
12:00 15mTalk | TypeJinja: Static Type Checking of Jinja Templates at dbt Labs SE In Practice (SEIP) Cheng Ding The University of Texas at Austin, Zhong Xu dbt Labs, Michael Levin dbt Labs, Wolfram Schulte dbt Labs, Milos Gligoric The University of Texas at Austin | ||
12:15 15mTalk | Principles and Practices of Large-Scale Code Analysis at Ant Group: A Data- and Logic-Oriented Approach SE In Practice (SEIP) Xiaoheng Xie Ant Group, Gang Fan Huawei Hong Kong Research Centre, Xiaojun Lin Ant Group, Ang Zhou Ant Group, Shijie Li Ant Group, Xunjin Zheng Ant Group, Yinan Liang Ant Group, Yu Zhang Ant Group, Na Yu Ant Group, Haokun Li Ant Group, Xinyu Chen Ant Group, Yingzhuang Chen Ant Group, Yi Zhen Ant Group, Dejun Dong Ant Group, Xianjin Fu Ant Group, Jinzhou Su Ant Group, Fuxiong Pan Ant Group, Pengshuai Luo Ant Group, Youzheng Feng Ant Group, Ruoxiang Hu Ant Group, Hanyang Guo School of Software Engineering, Sun Yat-sen University, Jing Fan Ant Group, Xiao Xiao Sourcebrella Inc., Peng Di Ant Group & UNSW Sydney | ||