[Haskell-cafe] Re: Haskell, Queries and Monad Comprehension

[Günther Schmidt]

> 2010/5/23 Günther Schmidt<gue.schmidt at web.de>:
>> is there anybody currently using Haskell to construct or implement a query
>> language?
>
> I've a half-baked, type-indexed (in HList style) implementation of
> relational algebra lying around somewhere, if that counts as a "query
> language". I was experimenting with using it as a sort of abstract
> collection interface, which actually worked rather nicely I think, but
> I didn't have time to flesh it out completely. In particular, only
> very simple queries and limited kinds of relation composition were
> supported. Definitely just toy code, though, and dreadfully
> inefficient; if you're looking for an "actual implementation" meaning
> "usable interface to an external persistence layer" then disregard
> this.
>

Well you're on the radar now :)

BTW: I'm working on alternative, non-type-indexed version of extensible 
records (Which I have gathered to be an essential ingredient when you 
want to roll a Relational Algebra lib). There will be no first-class 
labels, but there also will be no need for singleton-types either (which 
is what you need with type-indexed solutions). And a bit less overhead 
all-around so it's kind making up for not having 1st class labels.