January 2004 - Libraries

Re: Database interfaces?
by Dominic Steinitz 17 Jan '04

17 Jan '04

In the end, the HSQL toolkit provided enough so I didn't do anything further. It's good news that the HaskellDB project is being revived. Dominic. > > Date: Wed, 14 Jan 2004 14:53:12 -0800 (PST) > From: Christopher Milton <cmiltonperl(a)yahoo.com> > To: libraries(a)haskell.org > Subject: Re: Database interfaces? > Message-ID: <20040114225312.69079.qmail(a)web20808.mail.yahoo.com> > In-Reply-To: <opr1reytgunrnk38(a)mail.xs4all.nl> > Content-Type: text/plain; charset=us-ascii > MIME-Version: 1.0 > Precedence: list > Message: 2 > > --- Daan Leijen <daanleijen(a)xs4all.nl> wrote: > > On Tue, 13 Jan 2004 19:01:41 -0800 (PST), Christopher Milton > > <cmiltonperl(a)yahoo.com> wrote: > > > What happened to the Haskell database interfaces project? > > > > Hi Christopher, > > > > I don't know exactly what the "Haskell database interfaces project" is, but > > here are some developments that you might want to know about: > > Dominic Steinitz was leading an effort back in Jan/Feb 2003, but it seemed > to fizzle out, or at least I never heard about it again. > > > 1) Krasimir Angelov's HSQL toolkit provides (untyped) access to ODBC, MySQL > > and > > PostgresSQL databases. It doesn't return meta information yet. > > <http://htoolkit.sourceforge.net/> > > > > 2) wxHaskell has extensive ODBC support (but only tested on windows as yet). > > also untyped but with meta-information. > > <http://wxhaskell.sourceforge.net/doc/Graphics.UI.WXCore.Db.html> > > > > 3) Haskell/DB is being revived to provide typed access to databases. This > > is a student project at Chalmers. You can reach them at > > <dp03-7(a)mdstud.chalmers.se>, and I think it was Anders Höckersten who > > is doing it? > > Now that Haskell has functional dependencies, you can mimick just enough > > of TREX with nested cartesian products to use HaskellDB without TREX. > > I think they use HSQL for the low-level access :-( (that is meant > > funny:-) > > Typed access, that's exactly it. I'll have to contact the folks at Chalmers. > > Thank you. > > Chris > > ===== > Christopher Milton > cmiltonperl(a)yahoo.com

1 0

help on project
by CHOPRA A. (230365) 16 Jan '04

16 Jan '04

Hi there.. I am undertaking a project in haskell to represent Proofs graphically using Haskell School of expression . i was wondering how to go ahead with the class definition and module structure. if any one has any idea about how i should approach this project, please email back thanks aloke

1 0

Adding FGL to the hierarchical libraries
by Martin Erwig 16 Jan '04

16 Jan '04

Hi there, Abe Egnor and Isaac Jones at Aetion Technologies have refactored the modules of the FGL (http://eecs.oregonstate.edu/~erwig/fgl/haskell/) into the new hierarchical name space. It is therefore possible now to add the FGL directly to the hierarchical library structure. I don't really know how that works, but I would like to contribute the FGL to the hierarchical libraries. (I haven't released the new version yet, but plan to do so soon.) I have attached the modules to this email. If you want to add them, please go ahead. If you have any questions, please let me know. Thanks, Martin

12 18

RE: collections in hierarchical libraries
by Simon Marlow 15 Jan '04

15 Jan '04

> No one seems to support Edison. However, there is an important > difference between Edison and DData: DData provides implementations > of specific structures, while Edison tried to provide general > collection > interfaces *and* implementations. I believe that designing > good generic > collection interfaces is very hard and may even be impossible > to do "right" > in current Haskell's. So, I guess we will be stuck with concrete > data structures for a while :-) I second Daan's opinion here. The Haskell community has been searching for good generic collection interfaces for a while and hasn't come up with anything that is obviously the right thing. In the meantime we're stuck with poor data structure support because everyone is waiting for the wonderful generic collection library to come along. Having a well-designed and consistent concrete data structure library is certainly a lot better than nothing. Indeed, it would be more useful in a sense because the barrier to adoption would be low, and it wouldn't have an impact on the existing libraries. If such a thing were to materialise, I'm sure we'd import it into the hierarchical libraries that are shipped with GHC, Hugs & nhc98. Perhaps DData is pretty close to what we're after. Would the interested parties care to thrash out the details and present it to the wider community for comments? Cheers, Simon

3 2

Fwd: Ejacu.late Like A P0RN Star
by Glasgow-haskell-bugs 15 Jan '04

15 Jan '04

1 0

Database interfaces?
by Christopher Milton 14 Jan '04

14 Jan '04

What happened to the Haskell database interfaces project? ===== Christopher Milton cmiltonperl(a)yahoo.com

2 2

Modern way of losing Regina we;ght natural fantasies
by Whalen 14 Jan '04

14 Jan '04

Hello outbound, At last you have an opportunity to purchase good directly from manufactures. You save your money purchasing quality products from our plant's store. Today we present you FatBlast product. What is FatBlast actually? Fatblast is an advanced fat-binding supplement that removes fat from the foods you eat! Formulated with the powerful fat-binding fiber Chitosan, the proprietary blend of all-natural compounds... Our corporation was the first one who started selling this product on the web in the year 2004. Try our FDA approved product tday tints Read about our dscounts and special bonses: http://www.tabletok.com/fly/index.php?pid=pharmaboss cadenced abutment perceives containers, roughness pulley exhume saint disliking obscurer stumped Nazism begrudges diffuses weave skunk overtakers.

1 0

Re: graphs and trees again
by Ross Paterson 14 Jan '04

14 Jan '04

[moving to libraries] On Tue, Jan 13, 2004 at 05:46:29PM +0100, Wolfgang Jeltsch wrote: > for my studies I recently needed graph and tree handling code. Because > nothing I found seemed to satisfy my needs, I finally started writing my own > graph and tree module. I was especially disappointed with Data.Graph and > Data.Tree. The reasons are: > * The Implementation of the types is not hidden so that code using > Data.Graph and Data.Tree can get very implementation-dependent. > * Vertices are ints which is too low level in my opintion. I think it > would be better to allow different types for vertices because this seems > to closer reflect what applications demand. > * I'd prefer to be able to use arbitrary sets of vertices instead of only > continuous ranges. > * There are only very few functions for trees and forests. The Data.Tree module in base is certainly quite small. You've made two changes: * made the types abstract, but provided the equivalent constructor and destructor functions. I'm not sure that gains anything. * added upwards and downwards accumulators (except that your "upwards" accumulator is actually a variant downwards accumulator, with quadratic performance). It might be worthwhile to add fold and accumulators. I was also thinking of adding unfolds (pure, and monadic depth-first and breadth-first).

2 3

Re: Herba.l Products of the year, 2003
by Fred 14 Jan '04

14 Jan '04

1 0

Hi how are you?
by Tcb 12 Jan '04

12 Jan '04

1 0