[Haskell-cafe] Re: Fwd: Semantics of iteratees, enumerators,
apfelmus at quantentunnel.de
Fri Aug 27 05:18:15 EDT 2010
Daniel Fischer wrote:
> Heinrich Apfelmus wrote:
>> Daniel Fischer wrote:
>>> For many hashing or de/encryption algorithms, chunking is more natural
>>> than single-character access.
>> Even when the chunk lengths are unpredictable? After all, unlike with
>> fread in C, you can't request the next chunk to have a certain length
>> with Iteratees.
> Well, I just gave an example where one would want chunking for reasons
> other than performance. That iteratees don't provide the desired
> functionality is a different matter.
> For performance reasons, one would still be likely to want the I/O to
> happen in larger chunks than the processing, so it's kind of moot.
Yes, I/O should happen in chunks, but I thought that the Enumerator
implementations could buffer I/O while still presenting single
characters to the Iteratees.
More information about the Haskell-Cafe