Re: [Patch] LaTeX: export TOC and endnotes

From: Xun Sun <xun.sun.cn_at_gmail.com>
Date: Tue Apr 01 2008 - 16:24:39 CEST

Hi Dom,

On Tue, Apr 1, 2008 at 10:04 PM, Dominic Lachowicz
<domlachowicz@gmail.com> wrote:
> Hi Xun,
>
>
> > The problem that concerned me is that endnotes may not be included in
> > every LaTeX system, although my experience is only with fully-featured
> > LaTeX systems.
>
> I'd say that it is entirely reasonable to require that package,
> otherwise the user risks data loss for documents that contain
> endnotes, complex tables, unicode text, etc...
>
>
> > On the other hand, the dynamic feature I'm proposing does not deserve
> > the effort of the two-pass analysis you proposed. It this happens to
> > be the only option, I guess I would give up...
>
> Of course, conditionally/dynamically requiring these packages is fine too.
>
> If you have a better proposal than mine, that's fine. If you want to
> unconditionally require the "endnotes" and the "multirow" packages,
> that's fine too. But writing the code to do an analysis of the
> document is on the order of 20 lines of code, so I wouldn't let that
> discourage you. I'll even write it for you if you'd like. We use this
> technique extensively in the OpenDocument exporter.

I was not worrying about the programming effort actually. Instead the
run-time effort.

The approach I was originally thinking about is to extend the IE_Exp
class (IE_Exp_LaTeX in the comment of the patch was a typo) to allow
inseart text at the beginning of the stream -- the current
implementation only allows write() at the end. Thus when we finish
analyzing the document, we can then require the proper packages to be
included. Not sure whether this works or not. If it does, it should
also be around 20 lines of code.

>
> Thanks!
> Dom
>

-- 
Thanks & regards
Xun Sun
Received on Tue Apr 1 16:33:50 2008

This archive was generated by hypermail 2.1.8 : Tue Apr 01 2008 - 16:33:50 CEST