On 23 March 2012 10:01, Nathan Hüsken <nathan.huesken@posteo.de> wrote:

If I understand correctly, in this solution it is assumed that that a
word must be a complete line (row column or diagonal), correct?

No it doesn't. See the "search" function and the use of isInfixOf there.

I do not understand why a set is used. Couldn't just a list be used
here, or is there some performance advantage I do not see?

You certainly can use a list in this case. However, if you do not care about the number of times a word appears in a collection (which seemed to be the case here), then using a Set might be a better idea.

You can implement the same thing using lists instead of sets and see it for yourself :)

I find it very difficult to estimate the performance of an haskell
program. The other solution of Lorenzo Bolla utilizes Data.Vector.
Does that give a performance advantage in this case?

I don't use a lot of Vectors, so no comment here.

Cheers,

Ozgur