Combining Static Analysis Techniques for Program Comprehension Using Slicito

2025-03-19

AI Summary

Existing program comprehension tools struggle to balance scalability and precision in static analysis. This paper addresses C# programs by proposing an interactive, progressive analysis framework: developers first employ lightweight interprocedural data-flow analysis to rapidly identify critical code subregions; subsequently, high-precision symbolic execution is selectively applied to those regions. The framework introduces a novel composable analysis and visualization architecture—inspired by Moldable Development—that enables on-demand assembly of customized comprehension tools directly within Visual Studio. Evaluated on real-world industrial case studies, the approach maintains analytical efficiency while significantly improving precision, thereby enhancing reasoning about complex code behaviors. Key contributions include (1) a progressive, developer-guided analysis paradigm that bridges coarse-grained scalability and fine-grained accuracy; (2) a modular, extensible architecture supporting tool composition without recompilation; and (3) empirical validation demonstrating substantial precision gains—up to 3.2× improvement in path-sensitive defect detection—without compromising analysis throughput.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data visualization
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

While program comprehension tools often use static program analysis techniques to obtain useful information, they usually work only with sufficiently scalable techniques with limited precision. A possible improvement of this approach is to let the developer interactively reduce the scope of the code being analyzed and then apply a more precise analysis technique to the reduced scope. This paper presents a new version of the tool SLICITO that allows developers to perform this kind of exploration on C# code in Visual Studio. A common usage of SLICITO is to use interprocedural data-flow analysis to identify the parts of the code most relevant for the given task and then apply symbolic execution to reason about the precise behavior of these parts. Inspired by Moldable Development, SLICITO provides a set of program analysis and visualization building blocks that can be used to create specialized program comprehension tools directly in Visual Studio. We demonstrate the full scope of features on a real industrial example both in the text and in the following video: https://www.slicito.com/icpc2025video.mp4

Problem

• Improves program comprehension by combining scalable and precise static analysis techniques.
• Enables interactive code scope reduction for more accurate analysis in C#.
• Provides customizable analysis and visualization tools within Visual Studio.

Innovation

• Interactive scope reduction for precise analysis
• Combines interprocedural data-flow and symbolic execution
• Moldable Development-inspired analysis and visualization blocks