Version française
Home     About     Download     Resources     Contact us    
Browse thread
Re: [Caml-list] OCaml standard library improvement
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: -- (:)
From: Diego Olivier Fernandez Pons <Diego-Olivier.FERNANDEZ-PONS@c...>
Subject: Re: [Caml-list] OCaml standard library improvement
    Bonjour,

On Mon, 24 Feb 2003, Benjamin C. Pierce wrote:

> I'm all in favor of improving the standard library, but...
>
>   1) I agree that such a project will be most useful if there is some
>      kind of strong unifying vision
>
>   2) I suggest looking at the SML Basis Library for ideas -- the people
>      that designed it thought a *lot* about consistent naming and such.
>      (In fact, would it be a bad idea just to *copy* the SML basis APIs
>      verbatim?)

I am afraid I am not familiar with the whole SML/NJ library (rather
than SML Basis Library which is quite sparse), the only thing I can
truly say is that :

- SML/NJ library already copies in some points Caml standard library :

e.g. The pretty printing (smlnj-lib/PP) is a port of Pierre Weis one

- SML/NJ library may not be the best library to copy since :

smlnj-lib/Regexp is not superior to Marché's Regexp (classical RE lib,
compiled to dfa or directly interpreted), Vouillon's LibRE (lazy
automata building), or to Galax (Glushkov construction)
(I have not yet seen Leroy's Str new implementation)

smlnj-lib/Util is not really superior to the data structures already
avaible in Caml (standard library, JCF's Patricia trees, ...)

smlnj-lib/HashCons is mostly equivalent to JCF ones (I remember having
read somewhere that JCF actually used some tricks found in the SML/NJ
library)

smlnj-lib/Doc is not superior to OCamlDoc

Concerning SML-Basis, if a common naming for all SML libraries was a
good idea, I am afraid SML-Basis is really not enough, just providing
some quite trivial minimal interfaces. And I do not really feel that
SML-Basis has done, even for the few functions it provides, the 'best'
choices avaible.

I agree that looking in other libraries for ideas is a good thing,
but this seems to have already be done concerning SML libraries.


        Diego Olivier


-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners