Re: [Haskell-cafe] Re: Functional specification of DFS

2007-10-18 Thread Vimal
> First, the suspicion that lazy evaluation may lead to a non-termination > of such algorithm is almost surely wrong, and in any case it is against > my religious beliefs [some twisted smiley here]. > Sorry to have clubbed both lazy evaluation and non-termination. I meant that Lazy eval could have

[Haskell-cafe] Re: Functional specification of DFS

2007-10-18 Thread jerzy . karczmarczuk
Vimal writes: ... Now, I face some problem which i think is due to Lazy evaluation. I tried adding strictness in as many meaningful places as possible, but it doesnt work :( I have sat with the code for a long time, and yet I am not able to come up with a convincing reason as to why it doesn

[Haskell-cafe] Re: Functional specification of DFS

2007-10-18 Thread Vimal
> So add another argument containing all nodes seen in any path, and > maintain that properly. You're going to need to make your DFS > > routine > tail-recursive. > The problem is efficiency. I don't know if it can be done in > linear time > without state (and something like arrays with O(1) acces