Hello All!
I just started to test l2h98_1p3 and ... after converting two of my documents,
I switched back to l2h98_1p1.
Test environment: Linux 2.0, generating for HTML 3.2
1. In \begin{center} ... \end{center} the end of the center-environment
is lost in HTML. The same happens with \begin{quotation} ... \end{quotation}.
2. As generating user I'm now always getting "bin" in the address-section.
3. There is still the big footnote-problem which was present in all l2h98_1
versions:
The Latex-problem is about footnotes in the tabular-environment. My solution
was, to put \footnotemarks in the tabular and use \footnotetext outside while
messing around with the footnote-counter. This works fine with Latex.
latex2html has two problems with this. First it is unable to identify the
right footnotemark with the right footnotetext. This is okay, Latex needs
a little bit of help with this and from what I understand of the structure
of l2h, messing around with counters is really deadly to l2h.
The idea to put the footnotetext into footnode.html anyway is very good, but,
and this is the second problem, l2h has problems compiling it. It reports
unknown commands like: leq begin overline vdash. The resulting text looks
like this:
</PRE>
<DT><A NAME="foot67">...</A>
<DD>Dies entspricht tex2html_wrap_inline$a$ëtex2html_wrap_inline$d$,
tex2html_wrap_inline$b$etex2html_wrap_inline$c$. Da die Syllogistik im
klassischen Sinne weder tex2html_wrap_inline$0$ noch
tex2html_wrap_inline$1$kennt, sollte man vielleicht lieber
tex2html_wrap_inline$a d$ und tex2html_wrap_inline$b
c$ o.ä. schreiben. Wir bevorzugen i.a. die termnegationsfreie
Schreibweise oder eine solche, die den Zusammenhang von syllogistischer und
minimaler Lösung zu verdeutlichen geeignet erscheint.
<PRE>.
This compiling-problem vanishes if you use \footnote inside the tabular
instead of the \footnotemark -- \footnotetext -- pair, but then you don't
see the footnote in Latex.
------------------------------------ SCHNIPP -----------------------------
\documentclass[10pt]{article}
\usepackage[german]{babel}
\begin{document}
\begin{tabular}{lllll}
1 & aa & $a \leq b$, $X$ $\vdash$ $c \leq d$ & $X_{\rm{s}}^{\rm{1}}$: & $c \leq a$, $b
\leq d$ \\
& & & $X_{\rm{s}}^{\rm{2}}$: & $1 \leq a +
d$, $b \cdot c \leq 0$\footnotemark\\
& & & $X_{\rm{m}}$: & $c \leq a + d$, $b
\cdot c \leq d$\footnotemark\\ \\
2 & ao & $a \leq b$, $X$ $\vdash$ $c \not\leq d$ & $X_{\rm{s}}^{\rm{1}}$: & $a
\not\leq d$, $b \leq c$ \\
& & & $X_{\rm{s}}^{\rm{2}}$: & $a \cdot
c \not\leq 0$, $b \cdot d \leq 0$ \\
& & & $X_{\rm{s}}^{\rm{3}}$: & $1 \leq
a + c$, $1 \not\leq b + d$ \\
& & & $X_{\rm{s}}^{\rm{4}}$: & $d \leq
a$, $c \not\leq b$ \\
& & & $X_{\rm{m}}$: & $( a \cdot
\overline{b} ) + ( c \cdot \overline{d} ) \not\leq 0$ \\ \\
3 & oa & $a \not\leq b$, $X$ $\vdash$ $c \leq d$ & & Hier scheint es
keine L"osung zu geben\footnotemark. \\ \\
4 & oo & $a \not\leq b$, $X$ $\vdash$ $c \not\leq d$ & $X_{\rm{s}}^{\rm{1}}$: & $a
\leq c$, $d \leq b$ \\
& & & $X_{\rm{s}}^{\rm{2}}$: & $a
\cdot d \leq 0$, $1 \leq b + c$ \\
& & & $X^{*}$: & $a \leq b + c$, $a
\cdot d \leq b$
\end{tabular}
\addtocounter{footnote}{-2}\footnotetext{Dies entspricht $a$\,\"{e}\,$d$,
$b$\,e\,$c$. Da die Syllogistik im klassischen Sinne weder $0$ noch $1$
kennt, sollte man vielleicht lieber $\overline{a} \leq d$ und $b \leq
\overline{c}$ o."a.\ schreiben. Wir bevorzugen i.a.\ die termnegationsfreie
Schreibweise oder eine solche, die den Zusammenhang von syllogistischer und
minimaler L"osung zu verdeutlichen geeignet erscheint.}
\addtocounter{footnote}{1}\footnotetext{Diese L"osung gilt auch schon f"ur
distributive Verb"ande bzw.\ Begriffslogiken. Distributivit"at ist aber wohl
erforderlich, da man zum bequemen Nachweis der L"osungseigenschaft die sog.\
Schnittregel zu ben"otigen scheint.
Die angegebene L"osung erledigt auch gleich den allgemeinen Fall mit n
Pr"amissen und m Konklusionen:
$a_{\rm{1}} \leq b_{\rm{1}}$, \ldots , $a_{\rm{n}} \leq b_{\rm{n}}$,
$X$ $\vdash$ $c_{\rm{1}} \leq d_{\rm{1}}$, \ldots , $c_{\rm{m}} \leq
d_{\rm{m}}$, da man Subsumtionen stets in "aquivalente $0$--Subsumtionen
umformen und diese dann in eine einzige zusammenfassen kann (Beweis mit
vollst"andiger Induktion).}
\addtocounter{footnote}{1}\footnotetext{Abgesehen von der trivialen L"osung
$X_{\rm{}} = B$, hier also $X_{\rm{m}} := c \leq d$. $X_{\rm{m}} = B$ ist im
"Ubrigen eine minimale L"osung f"ur alle vier F"alle, hier allerdings
die einzig m"ogliche.}
\end{document}
---------------------------------- SCHNIPP ---------------------------------
4. There is a problem with selfdefined lists, which was also present in
l2h98_1p1, but is now getting worse. In this instance it is
a "distributed" list, I am trying to preserve a counter through some
sublists, starting with a counter that was used in some other list
before. May be there is a better way to do this in LaTeX, but it works:
----------------------------------- SCHNIPP ---------------------------------
\begin{list}{xx}{\setlength{\parsep}{0mm}}
\item[BL$_{2}^{\vdash}$]
\newcounter{sets2}
\begin{list}{\arabic{sets2})}{\usecounter{sets2}\setlength{\parsep}{0mm}}
\setcounter{sets2}{\value{sets}}
\item $ab \leq 0$, $a \not\leq 0$~~$\vdash$~~$b \leq 0$
\item $ab \not\leq 0$~~$\vdash$~~$ a = b $
\item $a \not\leq b$~~$\vdash$~~$b \leq a$, $a = \overline{b}$
\item $\vdash$~~$1 \leq ( a \leq b ) + ( b \leq a )$
\item $a \not\leq b$~~$\dashv \vdash$~~$1 \leq a$, $b \leq 0$
\end{list}
\vspace{.5\baselineskip}
\item[BL$_{\rm{u}}^{\vdash}$]
\newcounter{sets3}
\begin{list}{\arabic{sets3})}{\usecounter{sets3}\setlength{\parsep}{0mm}}
\setcounter{sets3}{\value{sets2}}
\item $\vdash$~~$( A \leq B ) = ( \overline{A} + B )$
\item $\vdash$~~$A+\overline{A}$
\item $AB \leq C$~~$\dashv \vdash$~~$A \leq ( B \leq C )$
\item $\vdash$~~$( A \leq B ) + ( B \leq A )$
\item $A \not\leq 0$~~$\dashv \vdash$~~$1 \leq A$
\end{list}
\end{list}
--------------------------- SCHNIPP -----------------------------------
Okay, in l2h98_1p1 l2h is not able to preserve the counter, but there is
also some residue from the conversion, this "tex2html_nowrap" before each
item of the sublists:
<P>
<DL COMPACT>
<P>
<DT>BL
<!-- MATH: $_{2}^{\vdash}$ -->
<IMG
WIDTH="13" HEIGHT="34" ALIGN="MIDDLE" BORDER="0"
SRC="img109.gif"
ALT="$_{2}^{\vdash}$">
<DD><P>
<DL COMPACT>
<P>
<DT>tex2html_nowrap
1)
<DD><IMG
WIDTH="49" HEIGHT="29" ALIGN="MIDDLE" BORDER="0"
SRC="img51.gif"
ALT="$ab \leq 0$">,
...
Now with l2h98_1p3 I am getting this
*** no brace for \end , before:
*** using "" as the argument instead; is this correct? ***
during conversion when setting the counters. At the end of the
conversion I am getting:
? brace missing for \end
? brace missing for \end
? brace missing for \end
? brace missing for \end
? brace missing for \end
? brace missing for \end
Unknown commands: begin item
Done.
The tex2html_nowrap-problem is still there (looks even better with
surounding brackets) and the inline-images are at the wrong places
in the text.
That's it for now ...
Regards
Andreas Otte
([EMAIL PROTECTED])
