Hello GHC, Tuesday, August 29, 2006, 10:54:07 PM, you wrote:
It makes sense to me that the above behaviour is seen: length is now a good consumer, but it generates 1+(1+(1+(1+... as it is consuming, and this causes a stack overflow. I don't think we can fix this while staying with fold/build fusion, so it looks to me like the patch should be reverted and the whole problem looked at post-6.6.
in general, for any function we can either implement 1) good consumer based on using foldr 2) tail-recursion are you agree? in particular, filterM, replicateM and other *M list operations (their problems was mentioned earlier) can be implemented in first or second way, but not both may be this can be fixed with some smart rules, which selects first or second implementation depending on context? -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com