From mboxrd@z Thu Jan  1 00:00:00 1970
Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id AAA02493; Sat, 9 Aug 2003 00:13:22 +0200 (MET DST)
Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id AAA20927 for <caml-list@pauillac.inria.fr>; Sat, 9 Aug 2003 00:13:21 +0200 (MET DST)
Received: from epexch01.qlogic.org ([63.170.40.3])
	by concorde.inria.fr (8.11.1/8.11.1) with ESMTP id h78MDKf22055
	for <caml-list@pauillac.inria.fr>; Sat, 9 Aug 2003 00:13:20 +0200 (MET DST)
Received: from epmailtmp.qlogic.org ([10.20.33.254]) by epexch01.qlogic.org with Microsoft SMTPSVC(5.0.2195.5329);
	 Fri, 8 Aug 2003 17:13:28 -0500
Received: from eagle.ancor.com ([10.20.33.146]) by epmailtmp.qlogic.org with Microsoft SMTPSVC(5.0.2195.4905);
	 Fri, 8 Aug 2003 17:13:19 -0500
Received: from localhost (bhurt@localhost)
	by eagle.ancor.com (8.11.6/8.11.6) with ESMTP id h78MDJa01743;
	Fri, 8 Aug 2003 17:13:19 -0500
X-Authentication-Warning: eagle.ancor.com: bhurt owned process doing -bs
Date: Fri, 8 Aug 2003 17:13:18 -0500 (CDT)
From: Brian Hurt <bhurt@spnz.org>
X-X-Sender:  <bhurt@eagle.ancor.com>
To: Matt Gushee <matt@gushee.net>
cc: <caml-list@pauillac.inria.fr>
Subject: Re: [Caml-list] Multi-keyed lookup table?
In-Reply-To: <20030808213037.GA21525@swordfish>
Message-ID: <Pine.LNX.4.33.0308081708240.2616-100000@eagle.ancor.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-OriginalArrivalTime: 08 Aug 2003 22:13:19.0795 (UTC) FILETIME=[463A7830:01C35DFA]
X-Loop: caml-list@inria.fr
X-Spam: no; 0.00; caml-list:01 gushee:01 font:99 fallback:01 preferable:01 match:02 tree:02 coding:03 wrote:03 data:03 let:04 implement:05 efficient:05 seems:05 aug:05 
Sender: owner-caml-list@pauillac.inria.fr
Precedence: bulk

On Fri, 8 Aug 2003, Matt Gushee wrote:

> On Thu, Aug 07, 2003 at 03:16:49PM -0500, Brian Hurt wrote:
> > 
> > In the general case, this is hard.  In this specific case, you might 
> > consider just hard coding your levels.  So you'd end up with a data 
> > structure like:
> > 
> >              All font
> >              /   |   \
> >             /    |    \
> >          medium bold light   <-- pick the weight
> >                 / | \
> >               [ ..... ]
> 
> Interesting idea, and not one that I had considered at all. It seems
> quite efficient, too. On the other hand, it looks like the fixed search
> path would make it rather hard to implement fallback rules in case an
> exact match isn't found. It seems to me that that's fairly important for
> fonts: there are many situations where it is preferable to produce some
> output with some fonts not quite right rather than producing nothing.

You certainly can handle * rules by just concatenating results from the 
different children.  You could also do something like:

    match weight with
        "medium" ->
            let t = get_style tree.medium ... in
            match t with
                _ :: _ -> (* we have at least one font *) t
                | [] ->
                    (* try getting something in bold *)
                    let t = get_style tree.bold in
                    ...


Brian


-------------------
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