🚀 go-pugleaf

Nice idea.  I'd be perfectly happy if HTML+ was an exact clone of the
Version 2 TeXInfo (yes, even including @b and @i), but including an
extra primitive for embedding.  We are using a system (called TIM)
similar to what you suggest for documentation.  It uses a superset of
TeXInfo to which macros have been added.  These macros allow
domain-specific markup to be defined, along with a mapping of that
markup to straight-TeXInfo features.  For example, straight TeXInfo
has the construct @file to mark filenames.  We might add @executable
to mark the names of *executable* files, with a macro of the form

	@macro executable file

Or to differentiate between examples of code and examples of user
interaction, we might use

	@macro codeexample example
	@macro transcript example

Finally, we can use markup that is "thrown away" in the printed or
hypertext form, with

	@macro foo noop

So in our sources we retain full domain-specific markup, but by
running a tim-to-TeXInfo converter, we have the ability to generate
both Postscript (via TeX and dvi), and GNU Info hypertext.  I am
confident that soon there will be a good TeXInfo-to-HTML+ converter,
too.

TIM also provides a few other things that make TeXInfo a bit easier to
use in large documents, but it's still basically TeXInfo.  The biggest
problem with it is the lack of a good way to include drawings and
pictures.  In systems documentation, pictures can sometimes help.

TIM will eventually be posted to {somewhere}, including the GNU
Project.

Bill
--
 Bill Janssen      janssen@parc.xerox.com      (415) 812-4763
 Xerox Palo Alto Research Center      FAX: (415) 812-4777
 3333 Coyote Hill Road, Palo Alto, California   94304

mosedale@genome.Stanford.EDU (Dan Mosedale) writes:

[...]
>So the best of all worlds, it would seem, would be to have a
>translator which could translate from TeXinfo (or LaTeXinfo) to HTML.
>Has anyone thought about writing one of these?
[...]
>Assuming that there is no such translator, which scheme would you use
>for system documentation? 

A TeXinfo to HTML translator already exists. It is one of the demos
in the PHYTHON package:
	ftp.cwi.nl:/pub/python/python0.9.8.tar.Z

Heiko
--
Heiko Schlichting               heiko@Chemie.FU-Berlin.DE
Freie Universitaet Berlin, Institut fuer Kristallographie

Dan Mosedale (mosedale@genome.Stanford.EDU) wrote:
: translator which could translate from TeXinfo (or LaTeXinfo) to HTML.
: Has anyone thought about writing one of these?  And, if not, how
: difficult would it be -- could HTML be construed as (or written in
: such a style that it would be) a subset of TeXinfo?

There's a link to a bunch of TeXinfo documents on the Ohio State
WWW server - Mosaic has a pointer to it, and the doc id is
	http://www.cis.ohio-state.edu:85/info/dir
if you want to look at it.

That gateway looks to be a start of a work in progress - i say that
mostly because some obvious tags like @i() aren't done.

--Ed

emv@garnet.msen.com (Edward Vielmetti) writes:
>There's a link to a bunch of TeXinfo documents on the Ohio State
>WWW server - Mosaic has a pointer to it, and the doc id is
>	http://www.cis.ohio-state.edu:85/info/dir
>if you want to look at it.

Unfortunately some of the GNU documents pointed to are very old and
obsolete, for example, the g++ one, which is for g++ version 1.
Since xmosaic thinks this is the "official" interface to TexInfo
it could be cleaned up.

>That gateway looks to be a start of a work in progress - i say that
>mostly because some obvious tags like @i() aren't done.

The program that does the translation goes from the .info file to
html, not from the .texi file, so there isn't sufficient information to do
the @i() command.  Still, it's kind of cool: it works on the fly and runs
off of the standard info tree (the same one Emacs, xinfo, tkinfo etc look
at).  A better job could be done if the .texi file were the starting
point.





--
Joe Buck	jbuck@ohm.EECS.Berkeley.EDU

In article <1vluk9$eh3@agate.berkeley.edu> jbuck@forney.eecs.berkeley.edu (Joe Buck) writes:
>The program that does the translation goes from the .info file to
>html, not from the .texi file, so there isn't sufficient information to do
>the @i() command.  Still, it's kind of cool: it works on the fly and runs
>off of the standard info tree (the same one Emacs, xinfo, tkinfo etc look
>at).  A better job could be done if the .texi file were the starting
>point.

Earlier this year, I spent about two weeks hacking support for a +html
option in an old version of makeinfo.  Unfortunately, it did not
survive the transition to the then-current version of makeinfo.  If
somebody were to pay me for it, I would do the same thing for the
current makeinfo.

If you want to see what the output looked like, you can point your web
client at http://trantor.emba.uvm.edu:9101/texinfo?Top.  (Warning:
this server is liable to go down at any time...)

-GAWollman

-- 
Garrett A. Wollman   | Shashish is simple, it's discreet, it's brief. ... 
wollman@emba.uvm.edu | Shashish is the bonding of hearts in spite of distance.
uvm-gen!wollman      | It is a bond more powerful than absence.  We like people
UVM disagrees.       | who like Shashish.  - Claude McKenzie + Florent Vollant

In article <1vltnm$pke@nigel.msen.com>, emv@garnet.msen.com (Edward
Vielmetti) writes:
|> There's a link to a bunch of TeXinfo documents on the Ohio State
|> WWW server - Mosaic has a pointer to it, and the doc id is
|> 	http://www.cis.ohio-state.edu:85/info/dir
|> if you want to look at it.
|> 
|> That gateway looks to be a start of a work in progress - i say that
|> mostly because some obvious tags like @i() aren't done.

Does someone know how I could establish a TeXinfo -> HTML interface
locally? Most of our machines are allowed local TCP connections only and
can't go out to the Internet. However, I would like to provide such a
service here. The same applies to a manual pages interface.

-- 
  Dipl.-Inform. Rainer Klute          I R B :  immer richtig beraten
  Universit�t Dortmund, IRB
  Postfach 500500             |)|/    Tel.: +49 231 755-4663
D-W4600 Dortmund 50           |\|\    Fax : +49 231 755-2386

  new address after June 30th: Universit�t Dortmund, D-44221 Dortmund

In article <1vluk9$eh3@agate.berkeley.edu>
jbuck@forney.eecs.berkeley.edu (Joe Buck) writes:

   >There's a link to a bunch of TeXinfo documents on the Ohio State
   >WWW server - Mosaic has a pointer to it, and the doc id is
   >	http://www.cis.ohio-state.edu:85/info/dir
   >if you want to look at it.

   Unfortunately some of the GNU documents pointed to are very old and
   obsolete, for example, the g++ one, which is for g++ version 1.
   Since xmosaic thinks this is the "official" interface to TexInfo
   it could be cleaned up.

There is nothing "official" about it.  As explained in the docs,
Mosaic just points to certain available and interesting network
resources to demonstrate what the WWW/Mosaic technology is capable of.
If you have strong opinions about what should be at the other end of
that menubar entry, put a server up yourself.  In the meantime, thanks
to Ohio State for allowing their server to be used publicly.

Marc

--
Marc Andreessen
Software Development Group
National Center for Supercomputing Applications
marca@ncsa.uiuc.edu

In article <1vltnm$pke@nigel.msen.com> emv@garnet.msen.com (Edward Vielmetti) writes:
>There's a link to a bunch of TeXinfo documents on the Ohio State
>WWW server - Mosaic has a pointer to it, and the doc id is
>	http://www.cis.ohio-state.edu:85/info/dir
>if you want to look at it.
>
>That gateway looks to be a start of a work in progress - i say that
>mostly because some obvious tags like @i() aren't done.

That gateway uses the info files as the source for the documents, not
the texinfo source.  It converts info files to html on the fly.

I wrote my own info file to html converter, because the one Ohio State is
using doesn't take advantage of the tag information for large files, and
just does a simple linear search.  You can try it out on my
<A HREF="http://herman.cmf.nrl.navy.mil/info">experimental server</A>.
I like the idea of converting info to html on the fly, because it doesn't
require an extra step to make stuff available in the Web - if you install
a new package and update the info directory, then it's immediately available.

  Bill
-- 
Bill Fenner                  fenner@cmf.nrl.navy.mil