Feldspar: Application and Implementation
Paper in proceeding, 2012

The Feldspar project aims to develop a domain specific language for Digital Signal Processing algorithm design. From functional descriptions, imperative code (currently C) is generated. The project partners are Ericsson, Chalmers and ELTE, Budapest. The background and motivation for the project have been documented elsewhere [3]. We aim to raise the level of abstraction at which algorithm developers and implementors work, and to generate, from Feldspar descriptions, the kind of code that is currently written by hand. These lecture notes first give a brief introduction to Feldspar and the style of programming that it encourages. Next, we document the implementation of Feldspar as a domain specific language (DSL), embedded in Haskell. The implementation is built using a library called Syntactic that was built for this purpose, but also designed to be of use to other implementors of embedded domain specific languages. We show the implementation of Feldspar in sufficient detail to give the reader an understanding of how the use of the Syntactic library enables the modular construction of an embedded DSL. For those readers who would like to apply these techniques to their own DSL embedded in Haskell, further instructions are given in section 5. The programming examples are available in the CEFP directory of the Feldspar package, version 0.5.0.1: http://hackage.haskell.org/package/ feldspar-language-0.5.0.1 The code can be fetched by running: > cabal unpack feldspar-language-0.5.0.1 All code is written in Haskell, and has been tested using the Glasgow Haskell Compiler (GHC), version 7.0.2, and the packages syntactic-0.8 feldspar-language-0.5.0.1 feldspar-compiler-0.5.0.1

Author

Emil Axelsson

Chalmers, Computer Science and Engineering (Chalmers), Software Technology (Chalmers)

Mary Sheeran

Chalmers, Computer Science and Engineering (Chalmers), Software Technology (Chalmers)

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

03029743 (ISSN) 16113349 (eISSN)

Vol. 7241 402-439

4th Summer School on Central European Functional Programming School, CEFP 2011
Budapest, Hungary,

Areas of Advance

Information and Communication Technology

Subject Categories

Software Engineering

DOI

10.1007/978-3-642-32096-5_8

More information

Latest update

7/20/2021