Center for Advanced Software Analysis is an umbrella for a series of research projects related to program analysis and web technologies - LAWP, TRIA, and SADL - located at Department of Computer Science, Aarhus University.

The center is supported by grants (totalling 2.2 M euros) from the Danish Research Council for Technology and Production Sciences and the Danish Council for Independent Research. The research activities are additionally financially supported by IBM Research and Google.

Rich internet applications are computer programs that use web browsers as platform. This is a popular approach to software development that has many advantages compared to alternative architectures, but it also poses unique challenges to the software developers due to the dynamic nature of the underlying technologies, including JavaScript, DOM, AJAX, and JSON.

The TRIA research project aims to develop formal languages for specifying essential requirements and assumptions together with program analysis tools for verifying consistency between the components, in collaboration with the software company Trifork.

Software developers increasingly choose dynamic programming languages despite their well-known drawbacks compared to languages that are statically typed. The flexibility of dynamic languages often makes them more suitable for rapid development of small and medium-sized software applications, whereas statically typed languages permit static type checking that catches many programming errors early in development where they are cheap to correct.

The SADL project aims to lead the way in unifying the two kinds of programming languages by means of static analysis and thereby give the programmers the best from both worlds.

Errors in programs are a well-known problem, both to programmers and users. Modern Web-based software systems are being built from components that in complicated ways communicate with each other and the users. With the increasing prevalence and complexity of the Web, the consequences of programming errors are becoming problematic.

The LAWP research project aims to develop high-level languages and static analysis techniques for XML transformations and server-based web applications, in order to prevent programming errors.

• TAJS - type analysis for JavaScript
• Artemis - automated testing for JavaScript
• JSRefactor - refactoring for JavaScript
• ServletValidator - program analysis for Servlets
• XACT - type-safe XML transformations
• JWIG - high-level Web programming
• JSA - string analysis for Java

• Automated Testing with Targeted Event Sequence Generation, Jensen, Prasad, and Møller, ISSTA 2013
• Efficient Construction of Approximate Call Graphs for JavaScript IDE Services, Feldthaus, Schäfer, Sridharan, Dolby, and Tip, ICSE 2013
• Remedying the Eval that Men Do, Jensen, Jonsson, and Møller, ISSTA 2012 (ACM SIGSOFT Distinguished Paper)
• Improving Tools for JavaScript Programmers (Position Paper), Andreasen, Feldthaus, Jensen, Jensen, Jonsson, Madsen, and Møller, STOP 2012
• Automated Detection of Client-State Manipulation Vulnerabilities, Møller and Schwarz, ICSE 2012 (ACM SIGSOFT Distinguished Paper)
• Tool-supported Refactoring for JavaScript, Feldthaus, Millstein, Møller, Schäfer, and Tip, OOPSLA 2011
• Modeling the HTML DOM and Browser API in Static Analysis of JavaScript Web Applications, Jensen, Madsen, and Møller, ESEC/FSE 2011
• A Framework for Automated Testing of JavaScript Web Applications, Artzi, Dolby, Jensen, Møller, and Tip, ICSE 2011
• HTML Validation of Context-Free Languages, Møller and Schwarz, FoSSaCS 2011
• XML Graphs in Program Analysis, Møller and Schwartzbach, Science of Computer Programming, 76(6)
• Interprocedural Analysis with Lazy Propagation, Jensen, Møller, and Thiemann, SAS 2010
• Type Analysis for JavaScript, Jensen, Møller, and Thiemann, SAS 2009

If you are interested in joining CASA as a postdoc, assistant professor, or PhD student, please contact Associate Professor Anders Møller (amoeller@cs.au.dk) and supply a CV and a description of your experience with static analysis, web technologies, and scripting languages.

• Anders Møller
• Mathias Schwarz
• Jan Midtgaard
• Magnus Madsen
• Asger Feldthaus
• Esben Andreasen
• Casper S. Jensen
• Fabio Strocco
• Christian Budde Christensen

• Simon Holm Jensen (PhD)
• Peter A. Jonsson (postdoc)
• Jacob H.C. Kragh (programmer)
• Kristoffer Just Andersen (programmer)
• Matthias Diehn Ingesman (programmer)