LATEX-L Archives

Mailing list for the LaTeX3 project

LATEX-L@LISTSERV.UNI-HEIDELBERG.DE

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Date:
Wed, 9 Jul 2003 12:10:08 +0200
Content-Type:
text/plain
Parts/Attachments:
text/plain (60 lines)
Hello,

On Tue, Jul 08, 2003 at 07:33:42PM +0200, Joachim Schrod wrote:
> That's why the ant work of Achim is so interesting, including his
> distinction between markup and programming syntax.[*] It's an open
> question if O'Caml programs are really good for this problem domain
> (I'm not really convinced),

Neither am I. I think at least some syntactic sugar is needed to
simplify the code. For example, at the moment reading the arguments of a
command is quite cumbersome:

The code for \setcounter{counter}{number} is

  value set_counter ps = do
  {
    let ctr = read_expanded_argument ps in
    let arg = Parser.str_to_int (read_expanded_argument ps) in

    ParseState.set_counter ps ctr arg
  };

It would be much neater to write something like:

  command \setcounter (ctr : expanded) (arg : int) =
  {
    ParseState.set_counter ps ctr arg
  };

On the other hand, the advantage of using OCaml is that every command
has access to all internals of ant.

I'm thinking of implementing a special purpose programming language for
style files, but that's a lot of work and I won't manage to make it as
fast as compiled OCaml code. So probably I'll just write a preprocessor
to add some syntactic sugar to OCaml.

> but the framework is powerful enough to plug in other style sheets and
> to experiment with it.

Yes, it would be no problem at all to attach, say, an XML parser to ant,
or a graphical interface like Lyx.

> [*] Disclaimer: I'm biased here and think that ant has not got the
> acknowledgement in the TeX community it deserves. All talk about NTS,
> and here we have a system with concepts that is better already. The
> concepts are nearer towards the original goals for NTS, too.

To be honest, NTS has the big advantage of being complete. ant still
lacks support of insertions and a lot of smaller things.

Achim
--
________________________________________________________________________
                                                              | \_____/ |
   Achim Blumensath                                          \O/ \___/\ |
   Mathematische Grundlagen der Informatik                   =o=  \ /\ \|
   www-mgi.informatik.rwth-aachen.de/~blume                  /"\   o----|
____________________________________________________________________\___|

ATOM RSS1 RSS2