Re: How lazy is DData.Seq?

Am Montag, 10. Mai 2004 00:14 schrieben Sie:

--- Wolfgang Jeltsch wrote:

...

Hi,

meanwhile, I'm not sure if I need lazyness in my concrete example. But the general point is the following:

I have functions for converting values of certain data types to strings. I don't use [Char] but Seq Char as the result type of these functions to allow for efficient concatenation. For one concrete type I want to test if the values of this type have a certain property.

I use the to-string conversion function to check for this property because a value has the property in question iff the first character of the string representation is a letter. So I want to do something like isAlpha $ head $ toString value. toString value is constructed by multiple applications of append, so it may in the end be something like append (append a (append b c)) (append (append d e) f). In this example I'd want the terms append b c and append (append d e) f) to not be evaluated at all.

Ok, I think was mistaken in this respect.

What do you mean with this? Do you mean: "I think *I* was mistaken [...]"?

The current version behaves ok for sequences mainly constructed with fromList, cons or snoc; a sequence made up of 'append' is strict.

Would it be possible to change this behavior?

Cheers, JP.

Wolfgang