[l2h] Problem with \HTMLcode and APPLET attributes

[Les Kitchen]

Hi LaTeX2HTMLarians,

I'm doing some lecture notes about Java applets, and obviously
want to conditionalize between the online HTML version, which
has a live applet and a link to the source code, and the printed
LaTeX version, which has just a screen-grab of the applet with
the source code included verbatim.  (Both versions show the
literal mark-up needed.)

I could do that fine with the right combination of latexonly,
htmlonly and rawhtml environments, but that requires needlessly
repeating the applet's basename and the width and height.  So I
decided to abstract that into a macro, parametrized by the
applet's basename, width and height.  That required using
\HTMLcode to parameterize the generated raw HTML.  Here I get
problems.  This simple LaTeX file:

================================================================
\documentclass{article}
\usepackage{html}
\begin{document}
\HTMLcode[CODE=foo,WIDTH=256,HEIGHT=64]{APPLET}{}
\end{document}
================================================================

sends LaTeX2HTML into an infinite loop (recursion?) endlessly
printing:

*** attributes CODE,HEIGHT,WIDTH are required for <APPLET> ***.

This happens both with 2002-1 and 2K.1beta, on both MacOSX (with
fink) and Red Hat 7.2 (x86).

Wrapping a <DIV> tag around the applet -- as above, but with

\HTMLcode{DIV}{\HTMLcode[CODE=foo,WIDTH=256,HEIGHT=64]{APPLET}{}}

doesn't loop.  It produces the above "*** attributes" message
just once, but at the end gives the warnings:

*********** WARNINGS ***********  
HEIGHT=64 is an invalid value in the <APPLET> tag
CODE=foo is an invalid value in the <APPLET> tag
WIDTH=256 is an invalid value in the <APPLET> tag

I've tried different variants in the attributes (spacing,
quotes), and tried having the applet body non-empty.  But it's
much the same result.  The -debug and -verbosity flags haven't
shed any light on what's going on.

So I'm stumped.  Any ideas?

My workaround (as often when LaTeX(2HTML) can't quite do what I
want) is to generate the bulky original LaTeX (latexonly,
htmlonly, rawhtml) with appropriate substitutions via a Perl
script, and \input it.  But it's not the way I'd like to do it,
especially when it looks like LaTeX2HTML *should* be able to do
just what I want.


                                        Les.


Les Kitchen, Senior Lecturer                <URL:http://www.cs.mu.oz.au/~ljk>
Department of Computer Science and Software Engineering       ljk@cs.mu.oz.au
The University of Melbourne                      phone: +61-3-8344-9101,-9104
Computer Vision & Machine Intelligence Lab               fax: +61-3-9348-1184