# Re: Separating algorithms from implementations (long)

## "Joachim Durchholz" <joachim_d@gmx.de>

8 Sep 2000 02:23:45 -0400

*From comp.compilers*

| List of all articles for this month |

**From: ** | "Joachim Durchholz" <joachim_d@gmx.de> |

**Newsgroups: ** | comp.graphics.algorithms,comp.compilers,comp.dsp |

**Date: ** | 8 Sep 2000 02:23:45 -0400 |

**Organization: ** | Compilers Central |

**References: ** | 00-08-124 |

**Keywords: ** | design |

Toby Sharp <TSharp@Serif.com> wrote:

*>*

*> The crux of the idea is to have a new language which describes*

*> algorithms separately from implementations. The language also provides*

*> a way to give extra information about supplied input and required*

*> output, so that highly optimised code can be created. A compiler goes*

*> into great detail to find the best implementation for the supplied*

*> algorithm information.*

An interesting approach is Design by Contract as exercised in Eiffel.

Essentially, you specify the algorithm in the preconditions and

postconditions of a routine, then you write an imperative

implementation in the routine's body. (Preconditions and

postconditions are functional.)

Don't expect a full solution, but it might give you ideas.

Regards,

Joachim

Post a followup to this message

Return to the
comp.compilers page.

Search the
comp.compilers archives again.