More about HKUST
Leveraging Existing Test Cases to Analyze JavaScript Libraries
PhD Thesis Proposal Defence
Title: "Leveraging Existing Test Cases to Analyze JavaScript Libraries"
by
Mr. Wai Ting CHEUNG
Abstract:
JavaScript is one of the most popular programming languages in 2017. Many
JavaScript libraries such as Bootstrap, AngularJS, and jQuery are widely
used in practice. However, existing research on JavaScript is not yet
ready for analysis of libraries. Pattern-based static analyzers such as
Lint identify known buggy patterns from the source code, but pattern
matching can hardly identify bugs that require semantics-based analysis
and has a large number of warnings and false positives. While
semantics-based static analyzers can detect bugs that pattern-based static
analyzers cannot, analyzing JavaScript libraries using existing
semantics-based static analyzers for JavaScript faces several challenges:
JavaScript functions not reachable during analysis, numerous combinations
of inputs due to dynamic typing, and insufficient aids for debugging.
Therefore, research on analyzing JavaScript libraries mostly focuses on
the design of their techniques on popular libraries such as jQuery.
To address these limitations, this work proposes a combination of static
and dynamic analysis techniques that leverages existing test cases to call
the library functions so that they are reachable during static analysis.
Also, our technique extends existing test cases and utilizes them to
filter out false warnings and to reveal real errors that serve as
debugging aids for developers. The proposed approach acts as a milestone
that drives the JavaScript community to expand their scope of research by
analyzing a large variety of JavaScript libraries in their work.
Date: Friday, 28 April 2017
Time: 3:30pm - 5:30pm
Venue: Room 4475
(lifts 25/26)
Committee Members: Dr. Sunghun Kim (Supervisor)
Dr. Xiaojuan Ma (Chairperson)
Prof. Shing-Chi Cheung
Prof. Frederick Lochovsky
**** ALL are Welcome ****