Logic Programming 06-25433 Autumn 2012

Lecture 11: Prolog's execution strategy explained more precisely The execution strategy has been presented as simply placing all matching clauses onto the stack. In fact, Prolog is slightly more selective and a knowledge of this allows more efficient programs to be written. This lecture includes: revision of the elementary Prolog execution strategy; optimizations to the elementary model based on indexing; optimizations to the elementary model based on determinism; the if...then...else construct and the proper use of the cut. It answers questions from previous lectures about: how to select a good order of arguments for a procedure; how to order clauses in a procedure.

How to prepare   No specific preparation for this lecture.

Video   Recording of the lecture (Vimeo)

Slides   If you really want the slides from the lecture ... 1 slide per page 6 slides per page but don't forget printing the slides is not the same as understanding them.

Summary   Summary and programs with commentary from the lecture

Programs from the lecture   Example 1 - appending lists Example 2 - computing a factorial (without an accumulator) Example 3 - computing a factorial (with an accumulator) Example 4 - deciding if an input is "a" Example 5 - computing a factorial (with disjunction) Example 6 - finding the maximum of two numbers (correct) Example 7 - finding the maximum of two numbers (incorrect)

Individual study suggestions and further reading   There are useful sections in the SICStus Prolog manual on: Indexing Last Clause Determinacy Detection Last Call Optimization if...then...else compilation Efficiency is covered in: Covington. Chapter 4, sections 4.12-4.14, pp 107-113. O'Keefe. Chapter 3, pp 74-111- especially on cuts: sections 3.8-3.13, pp 88-101 and if...then...else: section 3.16, pp 108-109. Sterling & Shapiro. Chapter 11, pp 189-213.

Assessed work associated with this lecture   Tutorial sheet 4