Compilers & related

The most of actual core developers of Phalanger project are in our company. All of them have strong academic background in automata theory, compilers, optimization and dynamic languages. Our specialization is domain of dynamic languages that can be interpreted, compiled or combination of both. Compilation of dynamic language is hard job, because behavior is known just at time of the execution, not during the compilation in contrast to the static languages. Hence, it needs robust and efficient runtime.

Implementing dynamic language compiler involves a lot from computer science and it needs continuous research. Basically we are involved in these areas:

• Microsoft Intermediate Language (MSIL) (.NET instructions)
• Code optimization
• Debuggers
• Intelligent tests
• Syntactic and semantic analysis
• Integrated Development Environments (IDE)
• Interoperability methos between static and dynamic languages

Publications

• Adam Abonyi, Daniel Balaš, Miloslav Beňo, Jakub Míšek, Filip Zavoral:
  Phalanger Improvements, Technical report 2009/5, Department of Software Engineering, Faculty of Mathematics and Physics, Charles University in Prague, 2009
• Jan Benda, Tomas Matousek, Ladislav Prosek:
  Phalanger: Compiling and Running PHP Applications on the Microsoft .NET Platform, .NET Technologies 2006

Syntactic and Semantic Prediction for PHP Language Integration

Phalanger IntelliSense is an implementation of the language integration for Microsoft Visual Studio development environment. There are used own techniques for analyzing dynamic source code and offering possible completions of an expression to the user. The integration processes the PHP language code and integrates it with the .NET static assemblies.

The text prediction and source code analysis works well even for non static expressions without a need of execution.Based on practical experiences, resulting suggestions can cover most of developer’s common tasks.

Publications

• Jakub Míšek, Filip Zavoral:
  Syntactic and Semantic Prediction in Dynamic Languages appears in the Proceedings of 7th ACIS International Conference on Software Engineering Research, Management and Applications, SERA 2009, Studies in Computational Intelligence, Springer Verlag, 2009
• Daniel Balaš, Jakub Míšek, Filip Zavoral:
  Phalanger IntelliSense: Syntactic and Semantic Prediction appears in Informačné technológie – Aplikácia a Teória, ITAT 2009
• Jakub Míšek:
  IntelliSense Implementation of a Dynamic Language, Charles University in Prague, 2009

Dynamic Language Runtime

The Microsoft .NET Framework was from the beginning designed to support broad range of languages on a Common Language Runtime (CLR). CLR provides shared services such as garbage collection, JIT and tools integration. The other benefit is that these languages can work together and use libraries written in any of them as well as .NET Base class library (BCL).
The CLR didn’t have the support for dynamic languages. Their dynamic nature makes the compilation uneasy and places high demands on the language runtime. Unlike static languages as C# which don’t require runtime support other than CLR itself.
The new Dynamic Language Runtime (DLR) makes a difference. It adds a lot of support for dynamic languages on .NET, that makes implementing the dynamic languages much easier and it also enables the interoperability between the dynamic languages built on DLR and standard static languages on .NET.

Publications

• Miloslav Beňo
  Implementing the Dynamic Languages using DLR Technology, Charles University in Prague, 2010