Re: [Texmacs-dev] literate programming for TeXmacs

[Felix Breuer]

On Sun, 22 Jan 2006 21:31:04 +0100
Joris van der Hoeven <address@hidden> wrote:

> Hi Felix and David,
> 
> On Mon, Jan 16, 2006 at 12:19:33PM +0000, Felix Breuer wrote:
> > almost a year ago I published a working version of a small literate 
> > programming
> > system for use with TeXmacs.
> [...]
> > I started working on a proper plugin for TeXmacs, but then realized that for
> [...]
> 
> I followed a bit of your discussion.
> Let me quickly tell you a few thoughts:
> 
> 1) I think that source files with TeXmacs comments (a la David) and
>    TeXmacs files with special serializing/export (a la Felix)
>    are both of independent interest.

To avoid misunderstandings: I think there were actually three concepts
in discussion, namely

1) David's original idea: Use TeXmacs to work on source files that have
   tm markup as comments. 
2) My original idea: Generate (multiple) source files (without comments)
   from a single TeXmacs document.
3) What I am currently suggesting: Serialize a TeXmacs document as a
   source file where TeXmacs markup is commented. (This is a variant
   of David's suggestion.)

I am in favour of 3) because it seems to be the easiest way to get
a literate programming system that works both ways.


> 2) In both cases, notice that you may wish to override the standard
>    load and save routines. Using the contextual overloading mechanism
>    this can be done extremely elegantly, just by redefining
>    load-buffer/save-buffer in case that the user included a certain
>    style package. The overloaded routine would typically do some
>    (un)commenting or additional serialization (i.e. generate
>    the necessary "source" files when you save your document).

I will have a look. This might take some time, though. It's not that
I hate Scheme, or something. I can read and write Scheme (even though
I have not worked on a project with it) and I think it is a very elegant
language. It's just that I am having a hard time reading and
understanding TeXmacs' API.


> 3) The handling of TeXmacs comments in source files should be
>    very easy to implement. Something like
> 
>       /* TeXmacs 1.0.5.7 markup
>       ....
>       end TeXmacs markup */
> 
>    would probably be a good marker in the case of C/C++ files.

Can the string "*/" appear in TeXmacs markup?

 
> 4) For the roadmap after version 1.1, I plan to include an internal
>    linking mechanism based on tree obervers. This should be very robust
>    and completely outdate the current linking/referencing mechanism,
>    mutators and the Proclus plug-in. Also this will allow you to think
>    of documents as graphs rather than trees. However, I have no idea when
>    I will have time.

That is very good news. Given this plan for tree observers, I will put
my plans for clones on hold for the time being. I understand that tree
observers may be a long way off, but I think that focusing on an easy
literate serialization mechanism may gain the most. If I, at some point,
think I need clones ASAP I can still start implementing them as an 
intermediate solution.

 
> 5) Independently from me, I think that it would be good to develop
>    a "TeXmacs file system" as suggested a month or so ago on this list.
>    This should be done in a server based way and the development can be
>    done quite independently from TeXmacs itself (Felix: I do strongly
>    recommend the use of Scheme). I think that such a file system would
>    be of a great help for litterate programming too: it basically
>    would consider the file system to be a big document.

That is a great idea. I have been thinking about something like this
myself after reading an article on 9P (the Plan 9 protocol). Can you
point me to the relevant posting? I had a look but couldn't find that
email.


Thanks,
Felix