JavaScript contains many dynamic features that allegedly ease the task of programming modern web applications. The flexibility of the language has a price. It becomes challenging to reason about the behavior of JavaScript programs without actually running them. For many kinds of programming errors that cause compilation errors or runtime errors in other languages, JavaScript programs keep on running, often with surprising consequences. As a consequence, JavaScript programmers must rely on tedious testing to a much greater extent than necessary with statically typed languages. The TAJS project aims to develop efficient static analysis techniques that can reason soundly about JavaScript web applications, for example to catch likely errors during programming instead of at runtime. This talk presents the status of the project and gives an overview of the dataflow analysis techniques we have developed, such as lazy propagation to handle large abstract states, and integrated refactoring of code that uses 'eval'.

Anders Møller is associate professor at Aarhus University where he is manager of the Center for Advanced Software Analysis (http://cs.au.dk/CASA/)

이광근교수(x1857, 02-880-1857, 010-3895-9374)