constraint based killer sudoku solver performance improvements

[Blockheads Oi Oi]

I have a working program based on [1] that sets up all different
constraints for each row, column and box and then sets exact sum
constraints for each cage. It'll run in around 0.2 secs for a simple
problem, but a tough one takes 2 hours 45 minutes. I did some research
into improving the performance and found [2] but can't work out how to
implement the constraints given. Can someone please help, assuming that
it's even possible.

[1] http://pypi.python.org/pypi/python-constraint/1.1
[2] http://4c.ucc.ie/~hsimonis/sudoku.pdf
--
Cheers.

Mark Lawrence.

[Frank Millman]

I don't have an answer, but are you aware of this -

http://www.ics.uci.edu/~eppstein/PADS/Sudoku.py

It is a sudoko solver written in pure python.

I don't know what you call a tough problem, but this one solves the hardest
one I have thrown at it in the blink of an eye. It also outputs a full trace
of the reasoning it used to arrive at a solution.

Frank Millman

[Blockheads Oi Oi]

I'd looked at this years back and forgotten all about it so thanks for
the reminder :) Some of the code names directly match ideas given in my
[2] above so I'll take another look.
--
Cheers.

Mark Lawrence.

[Blockheads Oi Oi]

Just for the record I ended up using gecode-python and 2 hours 45
minutes became 2 milli seconds with the same model. As most politicains
have been known to say, lessons will be learned :)

--
Cheers.

Mark Lawrence.