Andrei Alexandrescu:
> This is a great example that I don't have the time to look into now.

It's a nice example, and I know you don't have a lot of time to implement it 
now.


> Currently Phobos doesn't have the means to compute the cross-product of 
> ranges.

The idea is to have standard means in Phobos to implement that kind delayed 
access to the lazily generated sequence.
I have already shown this topic here:
http://www.digitalmars.com/webnews/newsgroups.php?art_group=digitalmars.D&article_id=103682

Partially related: in the meantime someone has implemented my idea of 
vectorized lazyness, it's called "Chunked Sequences" in Clojure 1.1:
http://www.infoq.com/news/2009/12/clojure-11-rc1-transients
More (don't save this, click on it):
http://clojure.googlegroups.com/web/chunks.pdf

Despite your good intelligence Phobos2 will need some tuning and refinement, 
they are far from being battle-tested :-) "Small" examples like those two ones 
may help.

Bye,
bearophile

Reply via email to