Finally, after many years, a new version of Lush(*) is maturing. The most significant change will not be directly apparent to Lush users: Lush2 has a completely revamped memory managed based on a new garbage collector module. This allowed us to simplify much the bridge between the interpreter and compiled code, and to remove a few annoying limitations imposed by the compiler in the old system. The remaining tasks in the refactoring category, scheduled for post-2.0 releases, are a compiler rewrite/refactoring and the definition of a runtime-API. Some important new features planned for future releases are improved compiler support for array operations, parallel implementation of array operations, and interfaces to GUI toolkits. The Lush project is looking for new developers with interest in data-parallel programming, compiler design, scientific computing, or data visualization.
(*) Lush is a Lisp dialect with extensions for object-oriented and array-oriented programming. It is intended as a programming environment for prototyping numerically intensive applications. Unlike alternatives like Python or SciLab, Lush is designed for easy integration of existing C/C++/Fortran codes.
Log in to post a comment.