The Square-Sum Chain problem

[Joachim Schimpf]

A couple of days ago, the Square-Sum Chain problem appeared as a
programming challenge on Reddit https://redd.it/7t6fnc, with
further explanations at https://www.youtube.com/watch?v=G1m7goLCJDY
and related Prolog-specific discussion at
https://stackoverflow.com/questions/48532121/infinite-loop-in-prolog-or-just-very-slow

The challenge is to order numbers 1 to N such that every two consecutive
numbers in the sequence add up to a square number. For N=15 a solution is
8 1 15 10 6 3 13 12 4 5 11 14 2 7 9.

I went a bit overboard exploring a couple of approaches in ECLiPSe,
which some of you may find interesting. Read about it at

http://eclipseclp.org/wiki/Examples/SquareSumChain


Cheers,
Joachim

[Julio Di Egidio]

Thanks, makes one think. I would propose (in my ever limited knowledge of
this language) that Prolog+CLP is a multi-paradigm language...

Julio

[burs...@gmail.com]

Where would you recommend looking for a SAT Solver in
ECLiPSe Prolog? Are there efforts under way to translate

for example a CLP(FD) problem into a SAT problem?